Дисертації з теми "Hybird Artificial intelligence"
Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Hybird Artificial intelligence.
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Hybird Artificial intelligence".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Liu, Ziming. "Méthodes hybrides d'intelligence artificielle pour les applications de navigation autonome." Electronic Thesis or Diss., Université Côte d'Azur, 2024. http://www.theses.fr/2024COAZ4004.
Анотація:
La navigation autonome est une tâche difficile qui a un large éventail d'applications dans le monde réel. Le système de navigation autonome peut être utilisé sur différentes plateformes, telles que les voitures, les drones et les robots. Ces systèmes autonomes réduiront considérablement le travail humain et amélioreront l'efficacité du système de transport actuel. Certains systèmes autonomes ont été utilisés dans des scénarios réels, comme les robots de livraison et les robots de service. Dans le monde réel, les systèmes autonomes doivent construire des représentations de l'environnement et se localiser pour interagir avec l'environnement. Différents capteurs peuvent être utilisés pour atteindre ces objectifs. Parmi eux, le capteur caméra est le meilleur choix entre le coût et la fiabilité. Actuellement, la navigation autonome visuelle a connu des améliorations significatives grâce à l'apprentissage profond. Les méthodes d'apprentissage profond présentent des avantages pour la perception de l'environnement. Cependant, elles ne sont pas robustes pour la localisation visuelle où les méthodes basées sur des modèles ont des résultats plus fiables. Afin d'utiliser les avantages des méthodes basées sur les données et sur les modèles, une méthode hybride d'odométrie visuelle est étudiée dans cette thèse. Tout d'abord, des méthodes d'optimisation efficaces sont essentielles pour les méthodes basées sur les modèles et les méthodes basées sur les données qui partagent la même théorie d'optimisation. Actuellement, la plupart des réseaux d'apprentissage profond sont encore formés avec des optimiseurs de premier ordre inefficaces. Par conséquent, cette thèse propose d'étendre les méthodes d'optimisation efficaces basées sur les modèles pour former les réseaux d'apprentissage profond. La méthode Gaussienne-Newton et les méthodes efficaces de second ordre sont appliquées pour l'optimisation de l'apprentissage profond. Deuxièmement, la méthode d'odométrie visuelle basée sur un modèle repose sur des informations préalables sur la profondeur, l'estimation robuste et précise de la profondeur est essentielle pour la performance du module d'odométrie visuelle. Sur la base de la théorie traditionnelle de la vision par ordinateur, la vision stéréo peut calculer la profondeur avec l'échelle correcte, ce qui est plus fiable que les solutions monoculaires. Toutefois, les réseaux stéréoscopiques 2D-3D actuels à deux niveaux présentent des problèmes d'annotations de profondeur et d'écart entre les domaines de disparité. En conséquence, un réseau stéréo supervisé par la pose et un réseau stéréo adaptatif sont étudiés. Toutefois, les performances des réseaux en deux étapes sont limitées par la qualité des caractéristiques 2D qui construisent le volume de coût de l'appariement stéréo. Au lieu de cela, un nouveau réseau stéréo 3D en une étape est proposé pour apprendre les caractéristiques et l'appariement stéréo implicitement en une seule étape. Troisièmement, pour assurer la robustesse du système, le réseau stéréo et le module d'odométrie visuelle directe dense sont combinés pour créer une odométrie visuelle directe dense hybride stéréo (HDVO). L'odométrie visuelle directe dense est plus fiable que la méthode basée sur les caractéristiques, car elle est optimisée à partir des informations globales de l'image. L'HDVO est optimisée avec la perte de minimisation photométrique. Cependant, cette perte souffre de bruits provenant de la zone d'occlusion, de la zone de texture homogène et des objets dynamiques. Cette thèse étudie la suppression des valeurs de perte bruitées à l'aide de masques binaires. De plus, pour réduire les effets des objets dynamiques, les résultats de la segmentation sémantique sont utilisés pour améliorer ces masques. Enfin, pour être généralisée à un nouveau domaine de données, une méthode d'entraînement test-temps pour l'odométrie visuelle est exploréeAutonomous driving is a challenging task that has a wide range of applications in the real world. The autonomous driving system can be used in different platforms, such as cars, drones, and robots. These autonomous systems will reduce a lot of human labor and improve the efficiency of the current transportation system. Some autonomous systems have been used in real scenarios, such as delivery robots, and service robots. In the real world, autonomous systems need to build environment representations and localize themselves to interact with the environment. There are different sensors can be used for these objectives. Among them, the camera sensor is the best choice between cost and reliability. Currently, visual autonomous driving has achieved significant improvement with deep learning. Deep learning methods have advantages for environment perception. However, they are not robust for visual localization where model-based methods have more reliable results. To utilize the advantages of both data-based and model-based methods, a hybrid visual odometry method is explored in this thesis. Firstly, efficient optimization methods are critical for both model-based and data-based methods which share the same optimization theory. Currently, most deep learning networks are still trained with inefficient first-order optimizers. Therefore, this thesis proposes to extend efficient model-based optimization methods to train deep learning networks. The Gaussian-Newton and the efficient second-order methods are applied for deep learning optimization. Secondly, the model-based visual odometry method is based on the prior depth information, the robust and accurate depth estimation is critical for the performance of visual odometry module. Based on traditional computer vision theory, stereo vision can compute the depth with the correct scale, which is more reliable than monocular solutions. However, the current two-stage 2D-3D stereo networks have the problems of depth annotations and disparity domain gap. Correspondingly, a pose-supervised stereo network and an adaptive stereo network are investigated. However, the performance of two-stage networks is limited by the quality of 2D features that build stereo-matching cost volume. Instead, a new one-stage 3D stereo network is proposed to learn features and stereo-matching implicitly in a single stage. Thirdly, to keep robust, the stereo network and the dense direct visual odometry module are combined to build a stereo hybrid dense direct visual odometry (HDVO). Dense direct visual odometry is more reliable than the feature-based method because it is optimized with global image information. The HDVO is optimized with the photometric minimization loss. However, this loss suffers noises from the occlusion area, homogeneous texture area, and dynamic objects. This thesis explores removing noisy loss values with binary masks. Moreover, to reduce the effects of dynamic objects, semantic segmentation results are used to improve these masks. Finally, to be generalized for a new data domain, a test-time training method for visual odometry is explored. These proposed methods have been evaluated on public autonomous driving benchmarks, and show state-of-the-art performances
2
Wen, Chien-Hsien. "Applying artificial intelligence hybrid techniques in wastewater treatment." Ohio : Ohio University, 1997. http://www.ohiolink.edu/etd/view.cgi?ohiou1184357721.
3
Castorina, Giovanni. "Artificial intelligence based hybrid systems for financial forecasting." Thesis, University of the West of England, Bristol, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.365146.
Анотація:
Current research carried out on financial forecasting has highlighted some limitations of classical econometric methods based on the assumption that the investigated time series can be described as stationary stochastic processes with Gaussian probability density functions. Chaotic behaviour, fractal characteristics and non-linear dynamics have been emerging in different aspects of the financial forecasting problem. The objective of this thesis is to take a system level perspective of the financial forecasting problem and to explore a number of approaches to enhance more 'traditional' decision making flows for stock market forecasting, with particular emphasis on stock selection and timing. To achieve this purpose, a number of stock selection and timing computational 'modules' are investigated. From a computational point of view, the investigation performed in this work encompass techniques such as artificial neural networks, genetic algorithms, chaos theory and fractal geometry, as well as more traditional methods such as clustering, screening, ranking, and statistics based models. From a financial data point of view, this research takes advantage of both fundamental and technical information to enhance the stock selection and timing processes and to cover several investment horizons. Three computational modules are proposed. First, a multivariate stock ranking module which uses fundamental information and is optimised through genetic algorithms. Second, a multivariate forecasting module which uses technical information and is based on artificial neural networks. Third, a univariate price time series forecasting module based on artificial neural networks. In addition, an integrated flow that takes advantage of some synergies and complementary properties of the devised modules is proposed. The effectiveness of the developed modules and the viability of the proposed integrated flow are evaluated over a number of investment horizons using (out-of-sample) historical data.
4
Rodic, Daniel. "A Hybrid search heuristic-exhaustive search approach for rule extraction." Pretoria : [s.n.], 2000. http://upetd.up.ac.za/thesis/available/etd-05292006-110006/.
5
Natsheh, Emad Maher. "Hybrid power systems energy management based on artificial intelligence." Thesis, Manchester Metropolitan University, 2013. http://e-space.mmu.ac.uk/314015/.
Анотація:
This thesis presents a novel adaptive scheme for energy management in stand-alone hybrid power systems. The proposed management system is designed to manage the power flow between the hybrid power system and energy storage elements in order to satisfy the load requirements based on artificial neural network (ANN) and fuzzy logic controllers. The neural network controller is employed to achieve the maximum power point (MPP) for different types of photovoltaic (PV) panels, based on Levenberg Marquardt learning algorithm. The statistical analysis of the results indicates that the R2 value for the testing set was 0.99. The advance fuzzy logic controller is developed to distribute the power among the hybrid system and to manage the charge and discharge current flow for performance optimization. The developed management system performance was assessed using a hybrid system comprises PV panels, wind turbine, battery storage, and proton exchange membrane fuel cell (PEMFC). To improve the generating performance of the PEMFC and prolong its life, stack temperature is controlled by a fuzzy logic controller. Moreover, perturb and observe (P&O) algorithm with two different controller techniques - the linear PI and the non-linear passivity-based controller (PBC) - are provided for a comparison with the proposed MPPT controller system. The comparison revealed the robustness of the proposed PV control system for solar irradiance and load resistance changes. Real-time measured parameters and practical load profiles are used as inputs for the developed management system. The proposed model and its control strategy offer a proper tool for optimizing the hybrid power system performance, such as the one used in smart-house applications. The research work also led to a new approach in monitoring PV power stations. The monitoring system enables system degradation early detection by calculating the residual difference between the model predicted and the actual measured power parameters. Measurements were taken over 21 month’s period; using hourly average irradiance and cell temperature. Good agreement was achieved between the theoretical simulation and the real time measurement taken the online grid connected solar power plant.
6
Abbas, Syed Murtuza. "Advanced Hybrid Simulation Model based on Phenomenology and Artificial Intelligence." University of Cincinnati / OhioLINK, 2015. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1427963095.
7
Chhabra, Rupanshi. "Control Power Optimization using Artificial Intelligence for Hybrid Wing Body Aircraft." Thesis, Virginia Tech, 2015. http://hdl.handle.net/10919/56580.
Анотація:
Traditional methods of control allocation optimization have shown difficulties in exploiting the full potential of controlling a large array of control surfaces. This research investigates the potential of employing artificial intelligence methods like neurocomputing to the control allocation optimization problem of Hybrid Wing Body (HWB) aircraft concepts for minimizing control power, hinge moments, and actuator forces, while keeping the system weights within acceptable limits. The main objective is to develop a proof-of-concept process suitable to demonstrate the potential of using neurocomputing for optimizing actuation power for aircraft featuring multiple independently actuated control surfaces and wing flexibility. An aeroelastic Open Rotor Engine Integration and Optimization (OREIO) model was used to generate a database of hinge moment and actuation power characteristics for an array of control surface deflections. Artificial neural network incorporating a genetic algorithm then performs control allocation optimization for an example aircraft. The results showed that for the half-span model, the optimization results (for the sum of the required hinge moment) are improved by more than 11%, whereas for the full-span model, the same approach improved the result by nearly 14% over the best MSC Nastran solution by using the neural network optimization process. The results were improved by 23% and 27% over the case where only the elevator is used for both half-span and full-span models, respectively. The methods developed and applied here can be used for a wide variety of aircraft configurations.Master of Science
8
Schlobach, Klaus Stefan. "Knowledge discovery in hybrid knowledge representation systems." Thesis, King's College London (University of London), 2002. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.272023.
9
Viademonte, da Rosa Sérgio I. (Sérgio Ivan) 1964. "A hybrid model for intelligent decision support : combining data mining and artificial neural networks." Monash University, School of Information Management and Systems, 2004. http://arrow.monash.edu.au/hdl/1959.1/5159.
10
Wakelam, Mark. "Intelligent hybrid approach for integrated design." Thesis, University of Nottingham, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.263942.
11
Ortiz, Paul. "Conception d’un système hybride de stockage de l’énergie pour la réduction des émissions carbone dans l’habitat individuel." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0208.
Анотація:
Répondre aux objectifs de l'UE sur la réduction des émissions carbone, nécessite des propositions innovantes dans le domaine de l'habitat qui rejette 300 millions de tonnes de CO2 par an. RED WOLF a pour but d'augmenter l'usage de l'énergie renouvelable pour réduire l'émission CO2 de maisons équipées de panneaux hotoVoltaïques non raccordés au Gaz (PVnG). Un Système de Stockage Hybrides (SSH) sera développé et déployé dans 100 maisons PVnG. Elles seront équipées de batteries avec des Radiateurs à Accumulation à faible coût capables de stocker de la chaleur pour réchauffer à la demande les habitations sur un horizon de 24h. Les SSHs stockeront à la fois l'énergie produite in situ par les PVHs et les énergies vertes (vent, solaire) transportées par le réseau électrique lorsque la demande sera faible. Les pics de consommation (120% de la moyenne quotidienne) ont lieu entre 6h et 9h et entre 16h et 20h quand la production d'électricité est fortement carbonée. Actuellement, durant les journées venteuses ou ensoleillées, la production des parcs éoliens et solaires est réduite et l'énergie propre est donc gaspillée : Chaque année, dans les régions de l'ENO (Est-Nord-Ouest), jusqu'à 6% de l'énergie de l'éolien est perdue. De plus, les maisons PVnG reversent leur production électrique locale dans le réseau électrique indépendamment des besoins de celui-ci, ce qui augmente le déséquilibre entre la demande et la production sachant que la capacité des PV augmente de 3GW/an dans la région ENO. Dans le cadre de cette thèse, la recherche sera divisée en trois parties: (i) comprendre le comportement des habitants afin d'obtenir le modèle de consommation énergétique de chaque maison. Ce modèle sera utilisé afin de dimensionner les batteries et gérer intelligemment le basculement entre utilisation entre "énergie produite localement" et celle du réseau national ; (ii) concevoir un système de stockage hybrides afin d'analyser des flux de données provenant de sources d'informations multiples (p. ex., smart grid, météo, température ambiante, comportement des habitants, niveau des batteries), et ce à des fins de limiter l'émission carbone des habitations individuelles ; (iii) réaliser une analyse globale pour évaluer les stratégies de gestion des données (Cloud and Fog) et des réseaux (SDN, Internet des Objets), et ce, afin de trouver le meilleur compromis entre "Qualité de Service" et "impact environnemental". Les phases expérimentales se baseront sur des sites pilotes (disposant de maisons PVnG) situés au Royaume Uni, en France, en Irlande, au Luxembourg et Pays-Bas. En définitif, le système de stockage hybride proposé vise à être commercialisé dans les pays partenaires, et diffusé dans toute la région ENOPrivate homes are increasingly fitted with PhotoVoltaics (PV) for increasing the renewable energy use. Typical issues of this type of installation are : (1) the production of green energy is not necessary in line with the inhabitants' energy consumption (2) a peak of energy demand on smart grid increases the carbon emission. Mitigating carbon emission in using efficiently PV is the main objective of this research, which is part of the INTERREG NW Europe RED WOLF project (H2020 programme). This project is led led by Leeds Beckett University and involves 21 partners across Europe, including University of Lorraine. The RED WOLF project aims at installing PV solar panels and home batteries across UK, France, Netherlands, and Ireland, with the overall goal to design a Smart Storage Driver (SDS) system capables at storing solar energy produced at home, and using it when a peak of energy demand is detected on smart grid. To this end, the research is divided into three steps: (i) first, it is necessary to understand how inhabitants behave in order to obtain an energy consumption pattern for each home. This pattern will be used to size the batteries and to smartly manage the switching between the use of local and smart grid energy. Specific instrumentation will be deployed to achieve this step ; (ii) second, it is necessary to design a SDS system aiming at analysing data coming from multiple sources (e.g., smart grid, weather, indoor temperature, inhabitant behaviour, battery charge level) with one major target : the limitation of carbon emission ; (iii) finally, a global analysis ofdata management (Cloud and Fog) and network management (SDN, IoT) must be carried out in order to find the best trade off between Quality of Service offered for inhabitants and environmental impact
12
Khan, Laiq. "Hybrid AI paradigms applied to power system damping controls." Thesis, University of Strathclyde, 2003. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.273412.
13
Benbernou, Reda. "Application of artificial intelligence to seismic horizon mapping : An integrated hybrid approach." Thesis, University of Reading, 2010. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.529953.
14
Akinade, Olugbenga Olawale. "BIM-based software for construction waste analytics using artificial intelligence hybrid models." Thesis, University of the West of England, Bristol, 2017. http://eprints.uwe.ac.uk/31762/.
Анотація:
The Construction industry generates about 30% of the total waste in the UK. Current high landfill cost and severe environmental impact of waste reveals the need to reduce waste generated from construction activities. Although literature reveals that the best approach to Construction Waste (CW) management is minimization at the design stage, current tools are not robust enough to support architects and design engineers. Review of extant literature reveals that the key limitations of existing CW management tools are that they are not integrated with the design process and that they lack Building Information Modelling (BIM) compliance. This is because the tools are external to design BIM tools used by architects and design engineers. This study therefore investigates BIM-based strategies for CW management and develops Artificial Intelligent (AI) hybrid models to predict CW at the design stage. The model was then integrated into Autodesk Revit as an add-in (BIMWaste) to provide CW analytics. Based on a critical realism paradigm, the study adopts exploratory sequential mixed methods, which combines both qualitative and quantitative methods into a single study. The study starts with the review of extant literature and (FGIs) with industry practitioners. The transcripts of the FGIs were subjected to thematic analysis to identify prevalent themes from the quotations. The factors from literature review and FGIs were then combined and put together in a questionnaire survey and distributed to industry practitioners. The questionnaire responses were subjected to rigorous statistical process to identify key strategies for BIM-based approach to waste efficient design coordination. Results of factor analysis revealed five groups of BIM strategies for CW management, which are: (i) improved collaboration for waste management, (ii) waste-driven design process and solutions, (iii) lifecycle waste analytics, (iv) Innovative technologies for waste intelligence and analytics, and (v) improved documentation for waste management. The results improve the understanding of BIM functionalities and how they could improve the effectiveness of existing CW management tools. Thereafter, the key strategies were developed into a holistic BIM framework for CW management. This was done to incorporate industrial and technological requirements for BIM enabled waste management into an integrated system. The framework guided the development of AI hybrid models and BIM based tool for CW management. Adaptive Neuro-Fuzzy Inference System (ANFIS) model was developed for CW prediction and mathematical models were developed for CW minimisation. Based on historical Construction Waste Record (CWR) from 117 building projects, the model development reveals that two key predictors of CW are “GFA” and “Construction Type”. The final models were then incorporated into Autodesk Revit to enable the prediction of CW from building designs. The performance of the final tool was tested using a test plan and two test cases. The results show that the tool performs well and that it predicts CW according to waste types, element types, and building levels. The study generated several implications that would be of interest to several stakeholders in the construction industry. Particularly, the study provides a clear direction on how CW management strategies could be integrated into BIM platform to streamline the CW analytics.
15
Chaiyaratana, Nachol. "Neuro-genetic based hybrid frameworks with applications in biomedicine and robotics." Thesis, University of Sheffield, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.298963.
16
Rudnick, Alexander James. "Cross-Lingual Word Sense Disambiguation for Low-Resource Hybrid Machine Translation." Thesis, Indiana University, 2019. http://pqdtopen.proquest.com/#viewpdf?dispub=13422906.
Анотація:
This thesis argues that cross-lingual word sense disambiguation (CL-WSD) can be used to improve lexical selection for machine translation when translating from a resource-rich language into an under-resourced one, especially when relatively little bitext is available. In CL-WSD, we perform word sense disambiguation, considering the senses of a word to be its possible translations into some target language, rather than using a sense inventory developed manually by lexicographers.
Using explicitly trained classifiers that make use of source-language context and of resources for the source language can help machine translation systems make better decisions when selecting target-language words. This is especially the case when the alternative is hand-written lexical selection rules developed by researchers with linguistic knowledge of the source and target languages, but also true when lexical selection would be performed by a statistical machine translation system, when there is a relatively small amount of available target-language text for training language models.
In this work, I present the Chipa system for CL-WSD and apply it to the task of translating from Spanish to Guarani and Quechua, two indigenous languages of South America. I demonstrate several extensions to the basic Chipa system, including techniques that allow us to benefit from the wealth of available unannotated Spanish text and existing text analysis tools for Spanish, as well as approaches for learning from bitext resources that pair Spanish with languages unrelated to our intended target languages. Finally, I provide proof-of-concept integrations of Chipa with existing machine translation systems, of two completely different architectures.
17
Liga, Davide <1990>. "Hybrid Artificial Intelligence to Extract Patterns and Rules from Argumentative and Legal Texts." Doctoral thesis, Alma Mater Studiorum - Università di Bologna, 2022. http://amsdottorato.unibo.it/9996/5/Davide%20Liga%20-%20unibo.pdf.
Анотація:
This Thesis is composed of a collection of works written in the period 2019-2022, whose aim is to find methodologies of Artificial Intelligence (AI) and Machine Learning to detect and classify patterns and rules in argumentative and legal texts.
We define our approach “hybrid”, since we aimed at designing hybrid combinations of symbolic and sub-symbolic AI, involving both “top-down” structured knowledge and “bottom-up” data-driven knowledge.
A first group of works is dedicated to the classification of argumentative patterns. Following the Waltonian model of argument and the related theory of Argumentation Schemes, these works focused on the detection of argumentative support and opposition, showing that argumentative evidences can be classified at fine-grained levels without resorting to highly engineered features.
To show this, our methods involved not only traditional approaches such as TFIDF, but also some novel methods based on Tree Kernel algorithms.
After the encouraging results of this first phase, we explored the use of a some emerging methodologies promoted by actors like Google, which have deeply changed NLP since 2018-19 — i.e., Transfer Learning and language models. These new methodologies markedly improved our previous results, providing us with best-performing NLP tools.
Using Transfer Learning, we also performed a Sequence Labelling task to recognize the exact span of argumentative components (i.e., claims and premises), thus connecting portions of natural language to portions of arguments (i.e., to the logical-inferential dimension).
The last part of our work was finally dedicated to the employment of Transfer Learning methods for the detection of rules and deontic modalities.
In this case, we explored a hybrid approach which combines structured knowledge coming from two LegalXML formats (i.e., Akoma Ntoso and LegalRuleML) with sub-symbolic knowledge coming from pre-trained (and then fine-tuned) neural architectures.
18
Shadabi, Fariba. "Medical outcome prediction : a hybrid artificial neural networks approach /." Canberra, 2007. http://erl.canberra.edu.au/public/adt-AUC20070816.130444/index.html.
Анотація:
Thesis (PhD) -- University of Canberra, 2007.Thesis submitted in fulfilment of the requirements of the Degree of Doctor of Philosophy in Information Sciences and Engineering, University of Canberra, January 2007. Bibliography: leaves 110-127.
19
Hare, Matthew Peter. "Weaver - a hybrid artificial intelligence laboratory for modelling complex, knowledge- and data-poor domains." Thesis, University of Aberdeen, 1999. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.287609.
Анотація:
Weaver is a hybrid knowledge discovery environment which fills a current gap in Artificial Intelligence (AI) applications, namely tools designed for the development and exploration of existing knowledge in complex, knowledge and data-poor domains. Such domains are typified by incomplete and conflicting knowledge, and data which are very hard to collect. Without the support of robust domain theory, many experimental and modelling assumptions have to be made whose impact on field work and model design are uncertain or simply unknown. Compositional modelling, experimental simulation, inductive learning, and experimental reformulation tools are integrated within a methodology analogous to Popper's scientific method of critical discussion. The purpose of Weaver is to provide a 'laboratory' environment in which a scientist can develop domain theory through an iterative process of in silico experimentation, theory proposal, criticism, and theory refinement. After refinement within Weaver, this domain theory may be used to guide field work and model design. Weaver is a pragmatic response to tool development in complex, knowledge- and data- poor domains. In the compositional modelling tool, a domain-independent algorithm for dynamic multiple scale bridging has been developed. The multiple perspective simulation tool provides an object class library for the construction of multiple simulations that can be flexibly and easily altered. The experimental reformulator uses a simple domain-independent heuristic search to help guide the scientist in selecting the experimental simulations that need to be carried out in order to critically test and refine the domain theory. An example of Weaver's use in an ecological domain is provided in the exploration of the possible causes of population cycles in red grouse (Lagopus, lagopus scoticus). The problem of AI tool validation in complex, knowledge- and data-poor domains is also discussed.
20
MAINO, CLAUDIO. "Hybrid and Electric Vehicles Optimal Design and Real-time Control based on Artificial Intelligence." Doctoral thesis, Politecnico di Torino, 2022. http://hdl.handle.net/11583/2971994.
21
Topalli, Ayca Kumluca. "Hybrid Learning Algorithm For Intelligent Short-term Load Forecasting." Phd thesis, METU, 2003. http://etd.lib.metu.edu.tr/upload/627505/index.pdf.
Анотація:
Short-term load forecasting (STLF) is an important part of the power generation
process. For years, it has been achieved by traditional approaches stochastic like
time seriesbut, new methods based on artificial intelligence emerged recently in literature and started to replace the old ones in the industry. In order to follow the latest developments and to have a modern system, it is aimed to make a research on STLF in Turkey, by neural networks. For this purpose, a method is proposed to forecast Turkey&rsquo
s total electric load one day in advance. A hybrid learning scheme that combines off-line learning with real-time forecasting is developed to make use of the available past data for adapting the weights and to further adjust these connections according to the changing conditions. It is also suggested to tune the step size iteratively for better accuracy. Since a single neural network model cannot cover all load types, data are clustered due to the differences in their characteristics. Apart from this, special days are extracted from the normal training sets and handled separately. In this way, a solution is proposed for all load types, including working days, weekends and special holidays. For the selection of input parameters, a technique based on principal component analysis is suggested. A traditional ARMA model is constructed for the same data as a benchmark and results are compared. Proposed method gives lower percent errors all the time, especially for holiday loads. The average error for year 2002 data is obtained as 1.60%.
22
Ali, Sadaqat. "Energy management of multi-source DC microgrid systems for residential applications." Electronic Thesis or Diss., Université de Lorraine, 2023. http://www.theses.fr/2023LORR0159.
Анотація:
Comparé au réseau électrique alternatif (AC), le réseau électrique en courant continu (DC) a démontré de nombreux avantages tels que son interface naturelle avec les RES, les systèmes de stockage d'énergie et les charges en courant continu, une efficacité supérieure avec moins d'étapes de conversion, et un contrôle plus simple sans effet de peau et sans considérations sur le flux de puissance réactive. Le micro-réseaux DC reste une technologie relativement nouvelle, et ses architectures de réseau, stratégies de contrôle, techniques de stabilisation méritent d'énormes efforts de recherche. Dans ce contexte, cette thèse porte sur les problèmes de gestion de l'énergie d'un réseau électrique en courant continu (DC) multi-source dédié aux applications résidentielles. Le réseau électrique en courant continu (DC) est composé de générateurs distribués (panneaux solaires), d'un système de stockage d'énergie hybride (HESS) avec des batteries et un supercondensateur (SC), et de charges en courant continu, interconnectées via des convertisseurs de puissance DC/DC. L'objectif principal de cette recherche est de développer une stratégie avancée de gestion de l'énergie (EMS) d'améliorer l'efficacité opérationnelle du système tout en renforçant sa fiabilité et sa durabilité. Une plateforme de simulation hiérarchique de réseau électrique DC a été développée sous MATLAB/Simulink. Elle est composée de deux couches avec des échelles de temps différentes : une couche de contrôle de niveau local (échelle de temps de quelques secondes à quelques minutes en raison des comportements de commutation des convertisseurs) pour les contrôles des composants locaux, et une couche de contrôle de niveau système (avec une échelle de temps de quelques jours à quelques mois avec un test accéléré) pour la validation à long terme de l'EMS et son évaluation de performance. Dans la couche de contrôle de niveau local, les panneaux solaires, les batteries et le supercondensateur ont été modélisés et contrôlés séparément. Différents modes de contrôle tels que le contrôle de courant, le contrôle de tension et le contrôle du point de puissance maximale (MPPT) ont été mis en œuvre. Un filtre passe-bas (LPF) a été appliqué pour diviser la puissance totale du HESS : basse et haute fréquence pour les batteries et le supercondensateur. Différentes fréquences de coupure du LPF pour le partage de puissance a également été étudiée. Un EMS hybride bi-niveau combiné et un dimensionnement automatique ont été proposés et validés. Il couvre principalement cinq scénarios d'exploitation, notamment la réduction de la production des panneaux solaires, la réduction de la charge et trois scénarios via le contrôle du HESS associé à la rétention du contrôle de l'état de charge (SOC) du supercondensateur. Une fonction objective prenant en compte à la fois le coût en capital (CAPEX) et les coûts d'exploitation (OPEX) a été conçue pour l'évaluation des performances de l'EMS. L'interaction entre l'HESS et l'EMS a été étudiée conjointement sur la base d'un ensemble de données ouvertes de profils de consommation électrique résidentielle couvrant à la fois l'été et l'hiver. Finalement, une plateforme expérimentale de réseau électrique à courant continu (DC) multi-source a été développée pour valider en temps réel l'EMS. Elle est composée de quatre batteries lithium-ion, d'un supercondensateur, d'une alimentation électrique à courant continu programmable, d'une charge à courant continu programmable, de convertisseurs DC/DC correspondants et d'un contrôleur en temps réel (dSPACE/Microlabbox). Des tests accélérés ont été réalisés pour vérifier l'EMS proposé dans différents scénarios d'exploitation en intégrant des panneaux solaires réels et les profils de consommation de charge. Les plateformes de simulation hiérarchique de réseau électrique en courant continu (DC) et expérimentale, peuvent être utilisées de manière générale pour vérifier et évaluer divers EMSCompared to the alternating current (AC) electrical grid, the direct current (DC) electrical grid has demonstrated numerous advantages, such as its natural interface with renewable energy sources (RES), energy storage systems, and DC loads. It offers superior efficiency with fewer conversion steps, simpler control without skin effect or reactive power considerations. DC microgrids remain a relatively new technology, and their network architectures, control strategies, and stabilization techniques require significant research efforts. In this context, this thesis focuses on energy management issues in a multi-source DC electrical grid dedicated to residential applications. The DC electrical grid consists of distributed generators (solar panels), a hybrid energy storage system (HESS) with batteries and a supercapacitor (SC), and DC loads interconnected via DC/DC power converters. The primary objective of this research is to develop an advanced energy management strategy (EMS) to enhance the operational efficiency of the system while improving its reliability and sustainability. A hierarchical simulation platform of the DC electrical grid has been developed using MATLAB/Simulink. It comprises two layers with different time scales: a local control layer (time scale of a few seconds to minutes due to converter switching behavior) for controlling local components, and a system-level control layer (time scale of a few days to months with accelerated testing) for long-term validation and performance evaluation of the EMS. In the local control layer, solar panels, batteries, and the supercapacitor have been modeled and controlled separately. Various control modes, such as current control, voltage control, and maximum power point tracking (MPPT), have been implemented. A low-pass filter (LPF) has been applied to divide the total HESS power into low and high frequencies for the batteries and supercapacitor. Different LPF cutoff frequencies for power sharing have also been studied. A combined hybrid bi-level EMS and automatic sizing have been proposed and validated. It mainly covers five operational scenarios, including solar panel production reduction, load reduction, and three scenarios involving HESS control combined with supercapacitor state of charge (SOC) control retention. An objective function that considers both capital expenditure (CAPEX) and operating costs (OPEX) has been designed for EMS performance evaluation. The interaction between the HESS and EMS has been jointly studied based on an open dataset of residential electrical consumption profiles covering both summer and winter seasons. Finally, an experimental platform of a multi-source DC electrical grid has been developed to validate the EMS in real-time. It comprises four lithium-ion batteries, a supercapacitor, a programmable DC power supply, a programmable DC load, corresponding DC/DC converters, and a real-time controller (dSPACE/Microlabbox). Accelerated tests have been conducted to verify the proposed EMS in different operational scenarios by integrating real solar panels and load consumption profiles. The hierarchical simulation and experimental DC electrical grid platforms can be generally used to verify and evaluate various EMS
23
Gibson, Paul Martyn. "The application of hybrid neural network models to archaeofaunal ageing and interpretation." Thesis, University of York, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.296383.
24
Adegbindin, Moustaine Kolawole Agnide. "Control Power Optimization using Artificial Intelligence for Forward Swept Wing and Hybrid Wing Body Aircraft." Thesis, Virginia Tech, 2017. http://hdl.handle.net/10919/74950.
Анотація:
Many futuristic aircraft such as the Hybrid Wing Body have numerous control surfaces that can result in large hinge moments, high actuation power demands, and large actuator forces/moments. Also, there is no unique relationship between control inputs and the aircraft response. Distinct sets of control surface deflections may result in the same aircraft response, but with large differences in actuation power. An Artificial Neural Network and a Genetic Algorithm were used here for the control allocation optimization problem of a Hybrid Wing Body to minimize the Sum of Absolute Values of Hinge Moments for a 2.5-G pull-up maneuver. To test the versatility of the same optimization process for different aircraft configurations, the present work also investigates its application on the Forward Swept Wing aircraft. A method to improve the robustness of the process is also presented. Constraints on the load factor and longitudinal pitch rate were added to the optimization to preserve the trim constraints on the control deflections. Another method was developed using stability derivatives. This new method provided better results, and the computational time was reduced by two orders of magnitude. A hybrid scheme combining both methods was also developed to provide a real-time estimate of the optimum control deflection schedules to trim the airplane and minimize the actuation power for changing flight conditions (Mach number, altitude and load factor) in a pull-up maneuver. Finally, the stability derivatives method and the hybrid scheme were applied for an antisymmetric, steady roll maneuver.Master of Science
25
Scott, Lawrence Gill. "Explanations in hybrid expert systems." Thesis, University of British Columbia, 1990. http://hdl.handle.net/2429/28741.
Анотація:
This thesis addresses the problem of providing explanations for expert systems implemented in a shell that supports a hybrid knowledge representation architecture. Hybrid representations combine rules and frames and are the predominant architecture in intermediate and high-end commercial expert system shells. The main point of the thesis is that frames can be endowed with explanation capabilities on a par with rules. The point is illustrated by a partial specification for an expert system shell and sample explanations which could be generated by an expert system coded to that specification.
As background information, the thesis introduces expert systems and the standard knowledge representation schemes that support them: rule-only schemes, and hybrid schemes that combine rules with frames. Explanations for expert systems are introduced in the context of rules, since rules are the only representation for which explanations are supported, either in commercial tools or in the preponderance of research.
The problem addressed by the thesis, how to produce explanations for hybrid architectures, is analyzed in two dimensions. Research was surveyed in three areas for guiding principles toward solving the problem: frame logic, metalevel architectures, and reflective architectures. With the few principles that were discovered in hand, the problem is then
analyzed into a small number of subproblems, mainly concerning high-level architectural decisions.
The solution proposed to the problem is described in two ways. First a partial specification for expert system shell functionality is offered, which describes, first, object structures and, then, behaviors at three points in time—object compilation time, execution time, and explanation generation time. The second component of the description is a set of extended examples which illustrate explanation generation in a hypothetical expert system. The solution adopts principles of reflective architectures, storing metainformation for explanations
in metaobjects which are distinct from the object-level objects they explain. The most novel contribution of the solution is a scheme for relating all the ways that objects' slot values may be computed to the goal tree construct introduced by the seminal Mycin expert system.
The final chapter explores potential problems with the solution and the possibility of producing better explanations for hybrid expert system shell architectures.Science, Faculty of
Computer Science, Department of
Graduate
26
Pacheco, Roberto Carlos dos Santos. "A Hybrid intelligent system for diagnosing and solving financial problems." reponame:Repositório Institucional da UFSC, 1996. http://repositorio.ufsc.br/xmlui/handle/123456789/76393.
Анотація:
Tese (doutorado) - Universidade Federal de Santa Catarina, Centro Tecnologico. Programa de Pós-Graduação em Engenharia de ProduçãoMade available in DSpace on 2012-10-16T09:55:39Z (GMT). No. of bitstreams: 1 107010.pdf: 764081 bytes, checksum: e2d86b02036f32b8786711b88ac93359 (MD5)
27
Dinu, Razvan. "Web Agents : towards online hybrid multi-agent systems." Thesis, Montpellier 2, 2012. http://www.theses.fr/2012MON20126/document.
Анотація:
Multi-agent systems have been used in a wide range of applications from computer-based simulations and mobile robots to agent-oriented programming and intelligent systems in real environments. However, the largest environment in which software agents can interact is, without any doubt, the World Wide Web and ever since its birth agents have been used in various applications such as search engines, e-commerce, and most recently the semantic web. However, agents have yet to be used on the Web in a way that leverages the full power of artificial intelligence and multi-agent systems, which have the potential of making life much easier for humans. This thesis investigates how this can be changed, and how agents can be brought to the core of the online experience in the sense that we want people to talk and interact with agents instead of "just using yet another application or website". We analyze what makes it hard to develop intelligent agents on the web and we propose a web agent model (WAM) inspired by recent results in multi-agent systems. Nowadays, a simple conceptual model is the key for widespread adoption of new technologies and this is why we have chosen the MASQ meta-model as the basis for our approach, which provides the best compromise in terms of simplicity of concepts, generality and applicability to the web. Since until now the model was introduced only in an informal way, we also provide a clear formalization of the MASQ meta-model.Next, we identify the three main challenges that need to be addressed when building web agents: integration of bodies, web semantics and user friendliness. We focus our attention on the first two and we propose a set of principles to guide the development of what we call strong web agents. Finally, we validate our proposal through the implementation of an award winning platform called Kleenk. Our work is just a step towards fulfilling the vision of having intelligent web agents mediate the interaction with the increasingly complex World Wide WebMulti-agent systems have been used in a wide range of applications from computer-based simulations and mobile robots to agent-oriented programming and intelligent systems in real environments. However, the largest environment in which software agents can interact is, without any doubt, the World Wide Web and ever since its birth agents have been used in various applications such as search engines, e-commerce, and most recently the semantic web. However, agents have yet to be used on the Web in a way that leverages the full power of artificial intelligence and multi-agent systems, which have the potential of making life much easier for humans. This thesis investigates how this can be changed, and how agents can be brought to the core of the online experience in the sense that we want people to talk and interact with agents instead of "just using yet another application or website". We analyze what makes it hard to develop intelligent agents on the web and we propose a web agent model (WAM) inspired by recent results in multi-agent systems. Nowadays, a simple conceptual model is the key for widespread adoption of new technologies and this is why we have chosen the MASQ meta-model as the basis for our approach, which provides the best compromise in terms of simplicity of concepts, generality and applicability to the web. Since until now the model was introduced only in an informal way, we also provide a clear formalization of the MASQ meta-model.Next, we identify the three main challenges that need to be addressed when building web agents: integration of bodies, web semantics and user friendliness. We focus our attention on the first two and we propose a set of principles to guide the development of what we call strong web agents. Finally, we validate our proposal through the implementation of an award winning platform called Kleenk. Our work is just a step towards fulfilling the vision of having intelligent web agents mediate the interaction with the increasingly complex World Wide Web
28
Burdelis, Mauricio Alexandre Parente. "Ajuste de taxas de mutação e de cruzamento de algoritmos genéticos utilizando-se inferências nebulosas." Universidade de São Paulo, 2009. http://www.teses.usp.br/teses/disponiveis/3/3141/tde-14082009-180444/.
Анотація:
Neste trabalho foi realizada uma proposta de utilização de Sistemas de Inferência Nebulosos para controlar, em tempo de execução, parâmetros de Algoritmos Genéticos. Esta utilização busca melhorar o desempenho de Algoritmos Genéticos diminuindo, ao mesmo tempo: a média de iterações necessárias para que um Algoritmo Genético encontre o valor ótimo global procurado; bem como diminuindo o número de execuções do mesmo que não são capazes de encontrar o valor ótimo global procurado, nem mesmo para quantidades elevadas de iterações. Para isso, foram analisados os resultados de diversos experimentos com Algoritmos Genéticos, resolvendo instâncias dos problemas de Minimização de Funções e do Caixeiro Viajante, sob diferentes configurações de parâmetros. Com base nos resultados obtidos a partir destes experimentos, foi proposto um modelo com a troca de valores de parâmetros de Algoritmos Genéticos, em tempo de execução, pela utilização de Sistemas de Inferência Nebulosos, de forma a melhorar o desempenho do sistema, minimizando ambas as medidas citadas anteriormente.This work addressed a proposal of the application of Fuzzy Systems to adjust parameters of Genetic Algorithms, during execution time. This application attempts to improve the performance of Genetic Algorithms by diminishing, at the same time: the average number of necessary generations for a Genetic Algorithm to find the desired global optimum value, as well as diminishing the number of executions of a Genetic Algorithm that are not capable of finding the desired global optimum value even for high numbers of generations. For that purpose, the results of many experiments with Genetic Algorithms were analyzed; addressing instances of the Function Minimization and the Travelling Salesman problems, under different parameter configurations. With the results obtained from these experiments, a model was proposed, for the exchange of parameter values of Genetic Algorithms, in execution time, by using Fuzzy Systems, in order to improve the performance of the system, minimizing both of the measures previously cited.
29
Fialho, Álvaro Roberto Silvestre. "Exploração de relações entre as técnicas nebulosas e evolutivas da inteligência computacional." Universidade de São Paulo, 2007. http://www.teses.usp.br/teses/disponiveis/3/3141/tde-26072007-173902/.
Анотація:
Neste trabalho foi realizada uma busca por relações, regras e transformações entre duas metodologias constituintes da Inteligência Computacional - a Computação Nebulosa e a Computação Evolutiva. Com a organização e sistematização da existência de tais transformações, obtém-se uma mudança na modelagem de soluções que as utilizam de forma conjunta, possibilitando que teorias e modelos bem estabelecidos em uma das metodologias possam ser aproveitados pela outra de uma forma mais robusta, correta por construção, intrínseca e transparente. Um modelo foi proposto para direcionar o trabalho de pesquisa. Através da análise desse modelo e de uma revisão bibliográfica realizada, transformações pontuais entre as metodologias foram elencadas, e posteriormente consolidadas por meio de experimentos práticos: uma Base de Conhecimento (BC) de um Controlador Lógico Nebuloso foi criada e modificada, conforme a necessidade, através de um Algoritmo Genético (AG). Com a abordagem desenvolvida, além da criação de BCs a partir de pouquíssimo conhecimento sobre o domínio do problema, tornou-se possível a inserção de novos \"comportamentos desejados\" em BCs já existentes, automaticamente, através de AGs. Os resultados desses experimentos, realizados sobre uma plataforma computacional especificada e implementada para este fim, foram apresentados e analisados.This work addressed a search of relations, rules and transformations between two Computational Intelligence constituent methodologies - Fuzzy Computing and Evolutionary Computing. The existence of these relations changes the actual way of solutions modeling that uses these methodologies, allowing the utilization of well established theories and models of one technique by the other in a more robust, intrinsic and transparent way. Besides the research and systematization of points that indicate the existence of relations between the two methodologies, a model to guide these exploration was proposed. By this model analysis and by the bibliographic revision made, punctual transformations were pointed out, and further consolidated through practical experiments: a Knowledge Base (KB) of a Fuzzy Logic Controller was created and modified automatically by a Genetic Algorithm. With the developed approach, besides the creation of KBs, it became possible to automatically insert new \"desired behaviors\" to existent KBs. The results of such experiments, realized through a computational platform specified and implemented to this task, were presented and analyzed.
30
Bontorin, alves Guilherme. "Intelligent multielectrode arrays : improving spatiotemporal performances in hybrid (living-artificial), real-time, closed-loop systems." Thesis, Bordeaux 1, 2010. http://www.theses.fr/2010BOR14056/document.
Анотація:
Cette thèse présente un système bioélectronique prometteur, l’Hynet. Ce Réseau Hybride (vivant-artificiel) est conçu pour l’étude du comportement à long terme des cellules électrogénératrices, comme les neurones et les cellules betas, en deux aspects : l’individuel et en réseau. Il est basé sur une boucle fermée et sur la communication en temps réel entre la culture cellulaire et une unité artificielle (Matériel, Logiciel). Le premier Hynet utilise des Matrices d’électrodes (MEA) commerciales qui limitent les performances spatiotemporelles du Hynet. Une nouvelle Matrice d’électrodes intelligente (iMEA) est développée. Ce nouveau circuit intégré, analogique et mixte, fournit une interface à forte densité, à forte échelle et adaptative avec la culture. Le nouveau système améliore le traitement des données en temps réel et une acquisition faible bruit du signal extracellulaireThis thesis presents a promising new bioelectronics system, the Hynet. The Hynet is a Hybrid (living-artificial) Network, developed to study the long-term behavior of electrogenic cells (such as Neurons or Beta-cells), both individually and in a network. It is based on real-time closed-loop communication between a cell culture (bioware) and an artificial processing unit (hardware and software). In the first version of our Hynet, we use commercial Multielectrode Arrays (MEA) that limits its spatiotemporal performances. A new Intelligent Multielectrode Array (iMEA) is therefore developed. This new analog/mixed integrated circuit provides a large-scale, high-density, and adaptive interface with the Bioware, which improves the real-time data processing and the low-noise acquisition of the extracellular signal
Esta dissertação de doutorado apresenta um sistema bioeletrônico auspicioso, o Hynet. Esta Rede Híbrida (viva e artificial), é concebida para o estudo do comportamento à longo prazo de células eletrogeneradoras (como neurônios ou células beta), em dois aspectos : individual e em redes. Ele é baseado na comunicação bidirecional, em laço fechado e em tempo real entre uma cultura celular (Bioware) e uma unidade artificial (Hardware ou Software). Um primeiro Hynet é apresentado, mas o uso de Matrizes de Eletrodos (MEA) comerciais limita a performance do sistema. Finalmente, uma nova Matriz de Eletrodos Inteligente (iMEA) é desenvolvida. Este novo circuito integrado fornece uma interface adaptativa, em alta densidade e grande escala, com o Bioware. O novo sistema melhora o processamento de dados em tempo real e a aquisição baixo ruído do sinal extracelular
31
Osman, Norhaslinda Yasmin, and n/a. "The development of a predictive damage condition model of light structures on expansive soils using hybrid artificial intelligence techniques." Swinburne University of Technology, 2007. http://adt.lib.swin.edu.au./public/adt-VSWT20071002.131831.
Анотація:
Expansive soils have damage light structures due to movement of soil which was a common problem all around the world. Soils exhibiting expansive properties were common throughout Australia. The damage to light structures founded on expansive soils in Victoria occurred mainly in properties built on quaternary basaltic clays and Tertiary to Ordovician clays. A review of existing literature in the area of expansive soils showed a lack of a thorough scientific diagnostic of the damage to light structures founded on expansive soils. Very few studies had been performed on damage to light structures on expansive soils in Victoria. There were no models so far to predict damage condition to light structures. More over, most of the reports on damage to light structures on expansive soils in Victoria were poorly documented. The aim of this research project was to develop a model to predict the damage condition of light structure on expansive soils in Victoria. A hybrid Neural Network trained with Genetic Algorithm was adopted for the de-velopment of the Predictive Damage Condition model. The Neural Network and Ge-netic Algorithm toolboxes from MATLAB� version 7.1 were used. The development of a Predictive Damage Condition model was driven by the shortage of defined quanti-tative studies and methods of selecting the factors that influenced the damage to light structure on expansive soils. The data used was based on information extracted from the Building Housing Commission which was recorded by different engineering companies based only on the tenants complain and site investigation of the properties. A series of factors that were believed to be dominant in influencing damage to light structures were chosen including: structural type, foundation, the presence of vegetation, soil type, age, and climate change. The model showed that it was able to resolve the problems facing light structures on expansive soils. First and foremost, the Predictive Damage Condition model was able to predict the damage condition or damage class using different combinations of fac-tors. It was also possible to identify the factors contributing to the damage of the struc-ture and to assess their relative importance in causing damage to light structures on expansive soil. It was found that the construction footing and vegetation were the most important among all the other input parameters. Change in Thornthwaite Moisture In-dex or climate was ranked second. Construction wall and age, were ranked third and fourth respectively while both region and geology were ranked fifth. In addition, Change in Thornthwaite Moisture Index was noted to have the strongest correlation with other input parameters.
32
Osman, Norhaslinda Yasmin. "The development of a predictive damage condition model of light structures on expansive soils using hybrid artificial intelligence techniques." Australasian Digital Thesis Program, 2007. http://adt.lib.swin.edu.au/public/adt-VSWT20070731.124824/index.html.
Анотація:
Thesis (PhD) - Swinburne University of Technology, Faculty of Engineering and Industrial Sciences, 2007.Submitted in fulfilment of requirements for the degree of Doctor of Philosophy, Faculty of Engineering and Industrial Sciences, Swinburne University of Technology, 2007. Typescript. Includes bibliographical references (p. 174-202).
33
Barla-Szabo, Daniel. "A study of gradient based particle swarm optimisers." Diss., University of Pretoria, 2010. http://hdl.handle.net/2263/29927.
Анотація:
Gradient-based optimisers are a natural way to solve optimisation problems, and have long been used for their efficacy in exploiting the search space. Particle swarm optimisers (PSOs), when using reasonable algorithm parameters, are considered to have good exploration characteristics. This thesis proposes a specific way of constructing hybrid gradient PSOs. Heterogeneous, hybrid gradient PSOs are constructed by allowing the gradient algorithm to optimise local best particles, while the PSO algorithm governs the behaviour of the rest of the swarm. This approach allows the distinct algorithms to concentrate on performing the separate tasks of exploration and exploitation. Two new PSOs, the Gradient Descent PSO, which combines the Gradient Descent and PSO algorithms, and the LeapFrog PSO, which combines the LeapFrog and PSO algorithms, are introduced. The GDPSO represents arguably the simplest hybrid gradient PSO possible, while the LeapFrog PSO incorporates the more sophisticated LFOP1(b) algorithm, exhibiting a heuristic algorithm design and dynamic time step adjustment mechanism. The strong tendency of these hybrids to prematurely converge is examined, and it is shown that by modifying algorithm parameters and delaying the introduction of gradient information, it is possible to retain strong exploration capabilities of the original PSO algorithm while also benefiting from the exploitation of the gradient algorithms.Dissertation (MSc)--University of Pretoria, 2010.
Computer Science
unrestricted
34
Ostheimer, Julia. "Human-in-the-loop Computing : Design Principles for Machine Learning Algorithms of Hybrid Intelligence." Thesis, Linnéuniversitetet, Institutionen för informatik (IK), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-94051.
Анотація:
Artificial intelligence (AI) is revolutionizing contemporary industries and being applied in application domains ranging from recommendation systems to self-driving cars. In scenarios in which humans are interacting with an AI, inaccurate algorithms could lead to human mistreatment or even harmful events. Human-in-the-loop computing is a machine learning approach desiring hybrid intelligence, the combination of human and machine intelligence, to achieve accurate and interpretable results. This thesis applies human-in-the-loop computing in a Design Science Research project with a Swedish manufacturing company to make operational processes more efficient. The thesis aims to investigate emerging design principles useful for designing machine learning algorithms of hybrid intelligence. Hereby, the thesis has two key contributions: First, a theoretical framework is built that comprises general design knowledge originating from Information Systems (IS) research. Second, the analysis of empirical findings leads to the review of general IS design principles and to the formulation of useful design principles for human-in-the-loop computing. Whereas the principle of AI-readiness improves the likelihood of strategical AI success, the principle of hybrid intelligence shows how useful it can be to trigger a demand for human-in-the-loop computing in involved stakeholders. The principle of use case-marketing might help designers to promote the customer benefits of applying human-in-the-loop computing in a research setting. By utilizing the principle of power relationship and the principle of human-AI trust, designers can demonstrate the humans’ power over AI and build a trusting human-machine relationship. Future research is encouraged to extend and specify the formulated design principles and employ human-in-the-loop computing in different research settings. With regard to technological advancements in brain-machine interfaces, human-in-the-loop computing might even become much more critical in the future.
35
Wong, Yin-cheung Eugene, and 黃彥璋. "A hybrid evolutionary algorithm for optimization of maritime logisticsoperations." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2010. http://hub.hku.hk/bib/B44526763.
36
Pimentel, Noeli Antônia. "SISTEMA TUTOR INTELIGENTE HÍBRIDO COM PERSONALIZAÇÃO ESTRUTURADA PELO MÉTODO DAS DIFERENÇAS FINITAS." Pontifícia Universidade Católica de Goiás, 2013. http://localhost:8080/tede/handle/tede/2446.
Анотація:
Made available in DSpace on 2016-08-10T10:40:19Z (GMT). No. of bitstreams: 1
Noeli Antonia Pimentel Vaz.pdf: 27939265 bytes, checksum: a285e763b07345fb2801fce0e9ed4cfe (MD5)
Previous issue date: 2013-05-15The generation of a pattern of navigation in Virtual Learning Environments (VLE) closer to the characteristics of the student is a question relevant of research. This paper takes as its starting point a Hybrid Intelligent Tutoring System (HITS) already proposed in the literature that uses the Artificial Neural Networks MLP (Multi Layer Perceptron) and rules of faculty experts, and adds a methodology using Partial Differential Equations (PDE) calculated by the method of finite differences, in order to improve the navigation decision to lead the student in a personalized way, adding to their biggest gains learning process. The original HITS developed for web platform has been adapted using technologies that enable the evolution system and the integration of the system settings for navigation mid-level, system without adaptive capacity, and the smart navigation were conducted to verify the effectiveness of the proposed system. Data obtained were evaluated by statistical tests with 5% level of significance, which proved the efficiency of the proposed model.
A geração de um padrão de navegação em Ambientes Virtuais de Aprendizagem (AVA) mais próximo das características do estudante é uma questão relevante de pesquisa. Esta dissertação toma como ponto de partida um Sistema Tutor Inteligente (STI) Híbrido já proposto na literatura que utiliza Redes Neurais Artificiais do tipo MLP (Multi Layer Perceptron) e regras de docentes especialistas; e agrega uma metodologia utilizando Equações Diferenciais Parciais (EDP) calculadas pelo método das diferenças finitas, com o objetivo de melhorar a decisão de navegação, para conduzir o estudante de forma personalizada, agregando melhores ganhos ao seu processo de aprendizado. O STI original, desenvolvido para plataforma web, foi adaptado utilizando tecnologias que possibilitem a evolução do sistema e a integração a Ambientes Virtuais de Aprendizagem, como o Moodle, por exemplo. Experimentos com o sistema em configurações para a navegação nível médio (sistema sem capacidade adaptativa) e a navegação inteligente foram realizados para verificação da efetividade do sistema proposto. Os dados obtidos foram avaliados por meio de testes estatísticos com nível de significância de 5%, que comprovam a eficiência do modelo proposto.
37
Sieberg, Philipp Maximilian [Verfasser], and Dieter [Akademischer Betreuer] Schramm. "Hybrid Methods in Vehicle Dynamics State Estimation and Control – Exploiting Potentials and Ensuring Reliability of Artificial Intelligence / Philipp Maximilian Sieberg ; Betreuer: Dieter Schramm." Duisburg, 2021. http://d-nb.info/1236501829/34.
38
Thorstensen, Evgenij. "Hybrid tractability of constraint satisfaction problems with global constraints." Thesis, University of Oxford, 2013. http://ora.ox.ac.uk/objects/uuid:05707b54-69e3-40eb-97e7-63b1a178c701.
Анотація:
A wide range of problems can be modelled as constraint satisfaction problems (CSPs), that is, a set of constraints that must be satisfied simultaneously. Constraints can either be represented extensionally, by explicitly listing allowed combinations of values, or intensionally, whether by an equation, propositional logic formula, or other means. Intensionally represented constraints, known as global constraints, are a powerful modelling technique, and many modern CSP solvers provide them. We give examples to show how problems that deal with product configuration can be modelled with such constraints, and how this approach relates to other modelling formalisms. The complexity of CSPs with extensionally represented constraints is well understood, and there are several known techniques that can be used to identify tractable classes of such problems. For CSPs with global constraints, however, many of these techniques fail, and far fewer tractable classes are known. In order to remedy this state of affairs, we undertake a systematic review of research into the tractability of CSPs. In particular, we look at CSPs with extensionally represented constraints in order to understand why many of the techniques that give tractable classes for this case fail for CSPs with global constraints. The above investigation leads to two discoveries. First, many restrictions on how the constraints of a CSP interact implicitly rely on a property of extensionally represented constraints to guarantee tractability. We identify this property as being a bound on the number of solutions in key parts of the instance, and find classes of global constraints that also possess this property. For such classes, we show that many known tractability results apply. Furthermore, global constraints allow us to treat entire CSP instances as constraints. We combine this observation with the above result, and obtain new tractable classes of CSPs by dividing a CSP into smaller CSPs drawn from known tractable classes. Second, for CSPs that simply do not possess the above property, we look at how the constraints of an instance overlap, and how assignments to the overlapping parts extend to the rest of the problem. We show that assignments that extend in the same way can be identified. Combined with a new structural restriction, this observation leads to a second set of tractable classes. We conclude with a summary, as well as some observations about potential for future work in this area.
39
Ma, Tan. "Hybrid Power System Intelligent Operation and Protection Involving Plug-in Electric Vehicles." FIU Digital Commons, 2015. http://digitalcommons.fiu.edu/etd/1760.
Анотація:
Two key solutions to reduce the greenhouse gas emissions and increase the overall energy efficiency are to maximize the utilization of renewable energy resources (RERs) to generate energy for load consumption and to shift to low or zero emission plug-in electric vehicles (PEVs) for transportation. The present U.S. aging and overburdened power grid infrastructure is under a tremendous pressure to handle the issues involved in penetration of RERS and PEVs. The future power grid should be designed with for the effective utilization of distributed RERs and distributed generations to intelligently respond to varying customer demand including PEVs with high level of security, stability and reliability. This dissertation develops and verifies such a hybrid AC-DC power system. The system will operate in a distributed manner incorporating multiple components in both AC and DC styles and work in both grid-connected and islanding modes.
The verification was performed on a laboratory-based hybrid AC-DC power system testbed as hardware/software platform. In this system, RERs emulators together with their maximum power point tracking technology and power electronics converters were designed to test different energy harvesting algorithms. The Energy storage devices including lithium-ion batteries and ultra-capacitors were used to optimize the performance of the hybrid power system. A lithium-ion battery smart energy management system with thermal and state of charge self-balancing was proposed to protect the energy storage system. A grid connected DC PEVs parking garage emulator, with five lithium-ion batteries was also designed with the smart charging functions that can emulate the future vehicle-to-grid (V2G), vehicle-to-vehicle (V2V) and vehicle-to-house (V2H) services. This includes grid voltage and frequency regulations, spinning reserves, micro grid islanding detection and energy resource support.
The results show successful integration of the developed techniques for control and energy management of future hybrid AC-DC power systems with high penetration of RERs and PEVs.
40
Hula, Tomáš. "Experimenty s rojovou inteligencí (swarm intelligence)." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2008. http://www.nusl.cz/ntk/nusl-235936.
Анотація:
This work deals with the issue of swarm intelligence as a subdiscipline of artificial intelligence. It describes biological background of the dilemma briefly and presents the principles of searching paths in ant colonies as well. There is also adduced combinatorial optimization and two selected tasks are defined in detail: Travelling Salesman Problem and Quadratic Assignment Problem. The main part of this work consists of description of swarm intelligence methods for solving mentioned problems and evaluation of experiments that were made on these methods. There were tested Ant System, Ant Colony System, Hybrid Ant System and Max-Min Ant System algorithm. Within the work there were also designed and tested my own method Genetic Ant System which enriches the basic Ant System i.a. with development of unit parameters based on genetical principles. The results of described methods were compared together with the ones of classical artificial intelligence within the frame of both solved problems.
41
MERCORIO, FABIO. "Model Checking for the Analysis and Control of Complex and Non-deterministic Systems." Doctoral thesis, Università Degli Studi di L'Aquila, 2012. http://hdl.handle.net/10281/36432.
Анотація:
In the last two decades, the use of intelligent planning algorithms, complex controllers,and automated verification processes is growing apace, having a great impact in many industrial fields, as robotics, manufacturing processes, and embedded systems, which now are present in an increasing number of everyday products and appliances. Moreover, many processes take place in an environment having variable and unpredictable influences on the system dynamics, making the problem of dealing with these non-deterministic behaviours a very significant concern. As a result, a growing synergy between Control and
AI Planning communities has been established, with the aim to develop algorithms and tools able to cope with such systems. In particular, it is interesting to evaluate robustness, verify the correctness and compute plans to execute activities. To this aim, formal methods (and in particular model checking) are well-suited to deal with these issues.
For several years, both control and planning problems have been addressed only through symbolic model checking, which has been successfully applied to a wide class of systems.
Nevertheless, there are still some open issues in dealing with Discrete Time Hybrid Systems (DTHS), whose state description involves both continuous and discrete variables, as well as systems with a complex nonlinear dynamics, for which symbolic approaches are hard to apply. To this regard, we focus on the use of explicit model checking, which is
based on the explicit enumeration of the system states, to deal with control and planning problems in both deterministic and non-deterministic domains.
Nevertheless, the explicit approach is strongly affected by the so called state explosion problem. In order to mitigate this problem, a first contribution is the developing of a disk-based algorithm for the UPMurphi tool: a universal planner for continuous domains
built on top of the Murphi model checker. We exploit the use of disk-based approach to analyse and control systems having a huge state space, showing a number of benchmarks and real world planning and control case studies. Moreover, we extend the use model checking to database data quality problems, using formal methods for the verification of data consistency defined over a set of data items, and evaluating the results on a real application of a Public Administration database provided by the C.R.I.S.P. research centre.
Finally, we tackle with non-deterministic systems in which an action may have different outcomes, unpredictable at planning time, addressing the problem to synthesise a plan able to reach a goal in spite of the non-determinism, i.e., strong plan. Many approaches have been applied in literature, mainly based on symbolic model checking. As a novel contribution, we present an algorithm able to synthesise strong plans (if any) with minimum cost with respect to a given cost function (that is minimising the non-deterministic worst-case execution), analysing its complexity, correctness and completeness. Finally, we describe the implementation of the algorithm into UPMurphi and we test it on two continuous non-deterministic case studies.
42
Cheng, Iunniang. "Hybrid Methods for Feature Selection." TopSCHOLAR®, 2013. http://digitalcommons.wku.edu/theses/1244.
Анотація:
Feature selection is one of the important data preprocessing steps in data mining. The feature selection problem involves finding a feature subset such that a classification model built only with this subset would have better predictive accuracy than model built with a complete set of features. In this study, we propose two hybrid methods for feature selection. The best features are selected through either the hybrid methods or existing feature selection methods. Next, the reduced dataset is used to build classification models using five classifiers. The classification accuracy was evaluated in terms of the area under the Receiver Operating Characteristic (ROC) curve (AUC) performance metric. The proposed methods have been shown empirically to improve the performance of existing feature selection methods.
43
Baumgartner, Dustin. "Global-Local Hybrid Classification Ensembles: Robust Performance with a Reduced Complexity." Connect to full text in OhioLINK ETD Center, 2009. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=toledo1241034194.
Анотація:
Thesis (M.S.)--University of Toledo, 2009.Typescript. "Submitted as partial fulfillment of the requirements for The Master of Science in Engineering." "A thesis entitled"--at head of title. Bibliography: leaves 158-164.
44
Fetter, Karl Christian. "Natural Selection For Disease Resistance In Hybrid Poplars Targets Stomatal Patterning Traits And Regulatory Genes." ScholarWorks @ UVM, 2019. https://scholarworks.uvm.edu/graddis/1162.
Анотація:
The evolution of disease resistance in plants occurs within a framework of interacting
phenotypes, balancing natural selection for life-history traits along a continuum of
fast-growing and poorly defended, or slow-growing and well-defended lifestyles. Plant
populations connected by gene flow are physiologically limited to evolving along a
single axis of the spectrum of the growth-defense trade-off, and strong local selection
can purge phenotypic variance from a population or species, making it difficult to
detect variation linked to the trade-off. Hybridization between two species that have
evolved different growth-defense trade-off optima can reveal trade-offs hidden in either
species by introducing phenotypic and genetic variance. Here, I investigated the
phenotypic and genetic basis for variation of disease resistance in a set of naturally
formed hybrid poplars.
The focal species of this dissertation were the balsam poplar (Populus balsamifera),
black balsam poplar (P. trichocarpa), narrowleaf cottonwood (P. angustifolia), and
eastern cottonwood (P. deltoides). Vegetative cuttings of samples were collected from
natural populations and clonally replicated in a common garden. Ecophysiology and
stomata traits, and the severity of poplar leaf rust disease (Melampsora medusae)
were collected. To overcome the methodological bottleneck of manually phenotyping
stomata density for thousands of cuticle micrographs, I developed a publicly available
tool to automatically identify and count stomata. To identify stomata, a deep con-
volutional neural network was trained on over 4,000 cuticle images of over 700 plant
species. The neural network had an accuracy of 94.2% when applied to new cuticle
images and phenotyped hundreds of micrographs in a matter of minutes.
To understand how disease severity, stomata, and ecophysiology traits changed
as a result of hybridization, statistical models were fit that included the expected
proportion of the genome from either parental species in a hybrid. These models in-
dicated that the ratio of stomata on the upper surface of the leaf to the total number
of stomata was strongly linked to disease, was highly heritable, and wass sensitive
to hybridization. I further investigated the genomic basis of stomata-linked disease
variation by performing an association genetic analysis that explicitly incorporated
admixture. Positive selection in genes involved in guard cell regulation, immune sys-
tem negative regulation, detoxification, lipid biosynthesis, and cell wall homeostasis
were identified.
Together, my dissertation incorporated advances in image-based phenotyping with
evolutionary theory, directed at understanding how disease frequency changes when
hybridization alters the genomes of a population.
45
Salamon, András Z. "Transformations of representation in constraint satisfaction." Thesis, University of Oxford, 2013. http://ora.ox.ac.uk/objects/uuid:5d641fff-4d95-43b2-9ff8-73395d782ad8.
Анотація:
In this thesis I study constraint satisfaction problems or CSPs. These require determining whether values can be assigned to variables so that all constraints are satisfied. An important challenge is to identify tractable CSPs which can be solved efficiently. CSP instances have usually been grouped together by restricting either the allowed combinations of values, or the way the variables are allowed to interact. Such restrictions sometimes yield tractable CSPs. A weakness of this method is that it cannot explain why all-different constraints form a tractable CSP. In this common type of constraint, all variables must be assigned values that are different from each other. New techniques are therefore needed to explain why such CSPs can be solved efficiently. My main contribution is an investigation of such hybrid CSPs which cannot be defined with either one of these kinds of restrictions. The main technique I use is a transformation of a CSP instance to the microstructure representation. This represents an instance as a collection of sets, and a solution of the instance corresponds to an independent set in the clause structure. For the common case where all constraints involve only two variables, I show how the microstructure can be used to define CSPs that are tractable because their clause structures fall within classes of graphs for which an independent set of specified size can be found efficiently. Such tractable hereditary classes are defined by using the technique of excluded induced subgraphs, such as classes of graphs that contain neither odd cycles with five or more vertices, nor their complements. I also develop finer grained techniques, by allowing vertices of the microstructure representation to be assigned colours, and the variables to be ordered. I show that these techniques define a new tractable CSP that forbids an ordered vertex-coloured subgraph in the microstructure representation.
46
Osório, Fernando Santos. "Inss : un système hybride neuro-symbolique pour l'apprentissage automatique constructif." Grenoble INPG, 1998. https://tel.archives-ouvertes.fr/tel-00004899.
Анотація:
Plusieurs méthodes ont été développées par l'Intelligence Artificielle pour reproduire certains aspects de l'intelligence humaine. Ces méthodes permettent de simuler les processus de raisonnement en s'appuyant sur les connaissances de base disponibles. Chaque méthode comporte des points forts, mais aussi des limitations. La réalisation de systèmes hybrides est une démarche courante Qui permet de combiner les points forts de chaque approche, et d'obtenir ainsi des performances plus élevées ou un champ d'application plus large. Un autre aspect très important du développement des systèmes hybrides intelligents est leur capacité d'acquérir de nouvelles connaissances à partir de plusieurs sources différentes et de les faire évoluer. Dans cette thèse, nous avons développé des recherches sur les systèmes hybrides neuro-symboliques, et en particulier sur l'acquisition incrémentale de connaissances à partir de connaissances théoriques (règles) et empiriques (exemples). Un nouveau système hybride, nommé système INSS - Incremental Neuro-Symbolic System, a été étudié et réalisé. Ce système permet le transfert de connaissances déclaratives (règles symboliques) d'un module symbolique vers un module connexionniste (réseau de neurones artificiel - RNA) à travers un convertisseur de règles en réseau. Les connaissances du réseau ainsi obtenu sont affinées par un processus d'apprentissage à partir d'exemples. Ce raffinement se fait soit par ajout de nouvelles connaissances, soit par correction des incohérences, grâce à l'utilisation d'un réseau constructif de type Cascade-Correlation. Une méthode d'extraction incrémentale de règles a été intégrée au système INSS, ainsi que des algorithmes de validation des connaissances qui ont permis de mieux coupler les modules connexionniste et symbolique. Le système d'apprentissage automatique INSS a été conçu pour l'acquisition constructive (incrémentale) de connaissances. Le système a été testé sur plusieurs applications, en utilisant des problèmes académiques et des problèmes réels (diagnostic médical, modélisation cognitive et contrôle d'un robot autonome). Les résultats montrent que le système INSS a des performances supérieures et de nombreux avantages par rapport aux autres systèmes hybrides du même typeVarious Artificial Intelligence methods have been developed to reproduce intelligent human behaviour. These methods allow to reproduce some human reasoning process using the available knowledge. Each method has its advantages, but also some drawbacks. Hybrid systems combine different approaches in order to take advantage of their respective strengths. These hybrid intelligent systems also present the ability to acquire new knowledge from different sources and so to improve their application performance. This thesis presents our research in the field of hybrid neuro-symbolic systems, and in particular the study of machine learning tools used for constructive knowledge acquisition. We are interested in the automatic acquisition of theoretical knowledge (rules) and empirical knowledge (examples). We present a new hybrid system we implemented: INSS - Incremental Neuro-Symbolic System. This system allows knowledge transfer from the symbolic module to the connectionist module (Artificial Neural Network - ANN), through symbolic rule compilation into an ANN. We can refine the initial ANN knowledge through neural learning using a set of examples. The incremental ANN learning method used, the Cascade-Correlation algorithm, allows us to change or to add new knowledge to the network. Then, the system can also extract modified (or new) symbolic rules from the ANN and validate them. INSS is a hybrid machine learning system that implements a constructive knowledge acquisition method. We conclude by showing the results we obtained with this system in different application domains: ANN artificial problems(The Monk's Problems), computer aided medical diagnosis (Toxic Comas), a cognitive modelling task (The Balance Scale Problem) and autonomous robot control. The results we obtained show the improved performance of INSS and its advantages over others hybrid neuro-symbolic systems
47
Alsalama, Ahmed. "A Hybrid Recommendation System Based on Association Rules." TopSCHOLAR®, 2013. http://digitalcommons.wku.edu/theses/1250.
Анотація:
Recommendation systems are widely used in e-commerce applications. Theengine of a current recommendation system recommends items to a particular user based on user preferences and previous high ratings. Various recommendation schemes such as collaborative filtering and content-based approaches are used to build a recommendation system. Most of current recommendation systems were developed to fit a certain domain such as books, articles, and movies. We propose a hybrid framework recommendation system to be applied on two dimensional spaces (User × Item) with a large number of users and a small number of items. Moreover, our proposed framework makes use of both favorite and non-favorite items of a particular user. The proposed framework is built upon the integration of association rules mining and the content-based approach. The results of experiments show that our proposed framework can provide accurate recommendations to users.
48
ISAKSSON, LARS JOHANNES. "HYBRID DEEP LEARNING AND RADIOMICS MODELS FOR ASSESSMENT OF CLINICALLY RELEVANT PROSTATE CANCER." Doctoral thesis, Università degli Studi di Milano, 2022. https://hdl.handle.net/2434/946529.
Анотація:
Precision medicine holds the potential to revolutionize healthcare by providing every patient with personalized treatments and decisions tailored to his or her individual needs. This might be enabled by the large influx of potentially diagnostic information from new sources such as genetics and modern imaging techniques, provided the relevant information can be extracted. One such framework that has started to demonstrate promise in radiology, especially in the assessment of cancer, is radiomics; the practice of characterizing images by extracting a substantial amount of quantitative mathematical descriptors. This success has largely been enabled by artificial intelligence (AI) and machine learning developments that are capable of handling the big data arrays. Using radiomics, researchers have been able to build prediction models capable of assisting and informing doctors in important decisions such as risk assessment or the choice of treatment. But even though radiomics has shown promise in preliminary studies, there is still a long way to go before radiomics and related AI applications can become routine tools in clinics. The road from patient admission to release is long, and all its intricate steps need to be studied in detail to establish the AI models' benefits and safety.
Deep learning is an incredibly powerful AI technique that has revolutionized many areas of science and industry such as recommender systems and protein folding. The technique has demonstrated particular capabilities in image analysis, such as the ability to drive cars autonomously and generate realistic-looking images from scratch. However, the recent advances in deep learning have largely been segregated from the radiomics domain, even though they can synergize with radiomics by performing complementary tasks such as image segmentation and denoising. There is considerable potential for DL and radiomics to cooperatively reinforce each other that so far has been majorly unexplored.
This thesis investigates the application of radiomics and deep learning in the context of prostate cancer. It focuses on the clinical perspective of where machine learning implementations are most likely to have a beneficial real-world impact. A key contribution is the deployment aspect: the models are not simply proofs of concept but are conceived and applied in a practical scenario, from patient admission to treatment decision. The specific areas studied include automatic organ segmentation in medical images, automatic quality assurance of segmentations, image processing, and radiomic feature analysis. Finally, a comprehensive study is performed on predicting essential pathological variables with AI, which so far has not been studied previously. Taken together, the methods outlined in this thesis constitute a concrete pathway of how AI can be used to bolster the steps along the patient's clinical trajectory. Successful applications of these methods hold the potential to reduce the workload of clinicians and improve patient outcomes.
49
Leite, Patrícia Teixeira. "Aplicação de técnicas de inteligência artificial no planejamento da operação de sistemas hidrotérmicos de potência." Universidade de São Paulo, 2003. http://www.teses.usp.br/teses/disponiveis/18/18133/tde-23042015-084714/.
Анотація:
Neste trabalho foi investigado um novo modelo baseado em inteligência artificial como ferramenta para a resolução do problema de planejamento da operação de sistemas hidrotérmicos de potência. Esta abordagem, que utiliza os princípios da evolução genética, tem se destacado com alta eficiência na solução de problemas de otimização. Para atender a todas as características do problema foram feitas algumas adaptações dos operadores genéticos tradicionais de recombinação e mutação, sendo o problema codificado usando uma cadeia de números reais, e não binários como normalmente é apresentado na literatura. Para isto, foram realizados vários testes visando moldar a técnica ao problema em questão, levando em conta suas características específicas. O algoritmo proposto também foi aplicado em vários testes com usinas pertencentes ao sistema hidroelétrico brasileiro e mostrou o bom desempenho desta abordagem em determinar uma operação ótima, garantindo, da melhor forma possível, o atendimento da demanda por um custo mínimo e com confiabilidade. As aplicações incluíram sistemas complexos, de grande porte, com até 35 usinas hidroelétricas, onde foram obtidos resultados satisfatórios.The present thesis investigates a new model based on artificial inteligence as a tool to solve the problem of the operational planning of hydrothermal systems. This approach, which uses the principle of genetic evolution, has been very successful and efficient in the solution of optimization problems. To represent all the characteristics of the problem some adaptations of the traditional genetic operators of recombination and mutation were made. The problem used a string of real numbers instead of binary as usually presented in the literature. Thus, several tests were performed in order to adapt the technique to the problem, taking into account its specific characteristics. The proposed algorithm has been applied in several tests in real hydrothermal systems, with plants belonging to the brazilian southeast system. The results achieved so far have indicated that the proposed approach can be an effective alternative or a complementary technique for the planning of hidrothermal system, as it determines an operation strategy for each power plant and minimizes the expected value of the operative cost along the planning horizon. The applications include large systems, with up to 35 hydroelectric plants, where good results were obtained.
50
Green, Robert C. II. "Novel Computational Methods for the Reliability Evaluation of Composite Power Systems using Computational Intelligence and High Performance Computing Techniques." University of Toledo / OhioLINK, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=toledo1338894641.