Dissertations / Theses on the topic 'Neural-genetic algorithm'
To see the other types of publications on this topic, follow the link: Neural-genetic algorithm.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Neural-genetic algorithm.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Tong, Dong Ling. "Genetic algorithm-neural network : feature extraction for bioinformatics data." Thesis, Bournemouth University, 2010. http://eprints.bournemouth.ac.uk/15788/.
Full textAbstract:
With the advance of gene expression data in the bioinformatics field, the questions which frequently arise, for both computer and medical scientists, are which genes are significantly involved in discriminating cancer classes and which genes are significant with respect to a specific cancer pathology. Numerous computational analysis models have been developed to identify informative genes from the microarray data, however, the integrity of the reported genes is still uncertain. This is mainly due to the misconception of the objectives of microarray study. Furthermore, the application of various preprocessing techniques in the microarray data has jeopardised the quality of the microarray data. As a result, the integrity of the findings has been compromised by the improper use of techniques and the ill-conceived objectives of the study. This research proposes an innovative hybridised model based on genetic algorithms (GAs) and artificial neural networks (ANNs), to extract the highly differentially expressed genes for a specific cancer pathology. The proposed method can efficiently extract the informative genes from the original data set and this has reduced the gene variability errors incurred by the preprocessing techniques. The novelty of the research comes from two perspectives. Firstly, the research emphasises on extracting informative features from a high dimensional and highly complex data set, rather than to improve classification results. Secondly, the use of ANN to compute the fitness function of GA which is rare in the context of feature extraction. Two benchmark microarray data have been taken to research the prominent genes expressed in the tumour development and the results show that the genes respond to different stages of tumourigenesis (i.e. different fitness precision levels) which may be useful for early malignancy detection. The extraction ability of the proposed model is validated based on the expected results in the synthetic data sets. In addition, two bioassay data have been used to examine the efficiency of the proposed model to extract significant features from the large, imbalanced and multiple data representation bioassay data.
2
Blomström, Karl. "Benchmarking an artificial neural network tuned by a genetic algorithm." Thesis, Umeå universitet, Institutionen för datavetenskap, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-58253.
Full textAbstract:
This thesis starts with a brief introduction to neural networks and the tuning of neural networks using genetic algorithms. An improved genetic algorithm is benchmarked using the technical paper Proben1 as a starting point. The benefits of using a genetic algorithm as well as results of the benchmark tests in comparison to a resilient backpropagation algorithm are discussed. The improved genetic algorithm is not a universal solution to all classification problems. Even though it outperforms the resilient backpropagation algorithm slightly in these benchmark tests more benchmarking on more vast solution domains must be performed.
3
Murnion, Shane D. "Neural and genetic algorithm applications in GIS and remote sensing." Thesis, Queen's University Belfast, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.337024.
Full text
4
McMurtrey, Shannon Dale. "Training and Optimizing Distributed Neural Networks Using a Genetic Algorithm." NSUWorks, 2010. http://nsuworks.nova.edu/gscis_etd/243.
Full textAbstract:
Parallelizing neural networks is an active area of research. Current approaches surround the parallelization of the widely used back-propagation (BP) algorithm, which has a large amount of communication overhead, making it less than ideal for parallelization. An algorithm that does not depend on the calculation of derivatives, and the backward propagation of errors, better lends itself to a parallel implementation.
One well known training algorithm for neural networks explicitly incorporates network structure in the objective function to be minimized which yields simpler neural networks. Prior work has implemented this using a modified genetic algorithm in a serial fashion that is not scalable, thus limiting its usefulness.
This dissertation created a parallel version of the algorithm. The performance of the proposed algorithm is compared against the existing algorithm using a variety of syn-thetic and real world problems. Computational experiments with benchmark datasets in-dicate that the parallel algorithm proposed in this research outperforms the serial version from prior research in finding better minima in the same time as well as identifying a simpler architecture.
5
Reiling, Anthony J. "Convolutional Neural Network Optimization Using Genetic Algorithms." University of Dayton / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1512662981172387.
Full text
6
Kopel, Ariel. "NEURAL NETWORKS PERFORMANCE AND STRUCTURE OPTIMIZATION USING GENETIC ALGORITHMS." DigitalCommons@CalPoly, 2012. https://digitalcommons.calpoly.edu/theses/840.
Full textAbstract:
Artificial Neural networks have found many applications in various fields such as function approximation, time-series prediction, and adaptive control. The performance of a neural network depends on many factors, including the network structure, the selection of activation functions, the learning rate of the training algorithm, and initial synaptic weight values, etc.
Genetic algorithms are inspired by Charles Darwin’s theory of natural selection (“survival of the fittest”). They are heuristic search techniques that are based on aspects of natural evolution, such as inheritance, mutation, selection, and crossover.
This research utilizes a genetic algorithm to optimize multi-layer feedforward neural network performance and structure. The goal is to minimize both the function of output errors and the number of connections of network. The algorithm is modeled in C++ and tested on several different data sets. Computer simulation results show that the proposed algorithm can successfully determine the appropriate network size for optimal performance. This research also includes studies of the effects of population size, crossover type, probability of bit mutation, and the error scaling factor.
7
MacLeod, Christopher. "The synthesis of artificial neural networks using single string evolutionary techniques." Thesis, Robert Gordon University, 1999. http://hdl.handle.net/10059/367.
Full textAbstract:
The research presented in this thesis is concerned with optimising the structure of Artificial Neural Networks. These techniques are based on computer modelling of biological evolution or foetal development. They are known as Evolutionary, Genetic or Embryological methods. Specifically, Embryological techniques are used to grow Artificial Neural Network topologies. The Embryological Algorithm is an alternative to the popular Genetic Algorithm, which is widely used to achieve similar results. The algorithm grows in the sense that the network structure is added to incrementally and thus changes from a simple form to a more complex form. This is unlike the Genetic Algorithm, which causes the structure of the network to evolve in an unstructured or random way. The thesis outlines the following original work: The operation of the Embryological Algorithm is described and compared with the Genetic Algorithm. The results of an exhaustive literature search in the subject area are reported. The growth strategies which may be used to evolve Artificial Neural Network structure are listed. These growth strategies are integrated into an algorithm for network growth. Experimental results obtained from using such a system are described and there is a discussion of the applications of the approach. Consideration is given of the advantages and disadvantages of this technique and suggestions are made for future work in the area. A new learning algorithm based on Taguchi methods is also described. The report concludes that the method of incremental growth is a useful and powerful technique for defining neural network structures and is more efficient than its alternatives. Recommendations are also made with regard to the types of network to which this approach is best suited. Finally, the report contains a discussion of two important aspects of Genetic or Evolutionary techniques related to the above. These are Modular networks (and their synthesis) and the functionality of the network itself.
8
Deane, Jason. "Scheduling online advertisements using information retrieval and neural network/genetic algorithm based metaheuristics." [Gainesville, Fla.] : University of Florida, 2006. http://purl.fcla.edu/fcla/etd/UFE0015400.
Full text
9
Fischer, Manfred M., and Yee Leung. "A Genetic Algorithm Based Evolutionary Computational Neural Network for Modelling Spatial Interaction Data." WU Vienna University of Economics and Business, 1998. http://epub.wu.ac.at/4151/1/WSG_DP_6198.pdf.
Full textAbstract:
Building a feedforward computational neural network model (CNN) involves two distinct
tasks: determination of the network topology and weight estimation. The specification of a
problem adequate network topology is a key issue and the primary focus of this contribution.
Up to now, this issue has been either completely neglected in spatial application domains, or
tackled by search heuristics (see Fischer and Gopal 1994). With the view of modelling
interactions over geographic space, this paper considers this problem as a global
optimization problem and proposes a novel approach that embeds backpropagation learning
into the evolutionary paradigm of genetic algorithms. This is accomplished by interweaving a
genetic search for finding an optimal CNN topology with gradient-based backpropagation
learning for determining the network parameters. Thus, the model builder will be relieved of
the burden of identifying appropriate CNN-topologies that will allow a problem to be solved
with simple, but powerful learning mechanisms, such as backpropagation of gradient descent
errors. The approach has been applied to the family of three inputs, single hidden layer,
single output feedforward CNN models using interregional telecommunication traffic data for
Austria, to illustrate its performance and to evaluate its robustness. (authors' abstract)Series: Discussion Papers of the Institute for Economic Geography and GIScience
10
Stivason, Charles T. "Industry Based Fundamental Analysis: Using Neural Networks and a Dual-Layered Genetic Algorithm Approach." Diss., Virginia Tech, 1998. http://hdl.handle.net/10919/40422.
Full textAbstract:
This research tests the ability of artificial learning methodologies to map market returns better than logistic regression. The learning methodologies used are neural networks and dual-layered genetic algorithms. These methodologies are used to develop a trading strategy to generate excess returns. The excess returns are compared to test the trading strategy's effectiveness. Market-adjusted and size-adjusted excess returns are calculated.
Using a trading strategy based approach the logistic regression models generated greater returns than the neural network and dual-layered genetic algorithm models. It appears that the noise in the financial markets prevents the artificial learning methodologies from properly mapping the market returns. The results confirm the findings that fundamental analysis can be used to generate excess returns.Ph. D.
11
Buys, Stefan. "Genetic algorithm for Artificial Neural Network training for the purpose of Automated Part Recognition." Thesis, Nelson Mandela Metropolitan University, 2012. http://hdl.handle.net/10948/d1008356.
Full textAbstract:
Object or part recognition is of major interest in industrial environments. Current methods implement expensive camera based solutions. There is a need for a cost effective alternative to be developed. One of the proposed methods is to overcome the hardware, camera, problem by implementing a software solution. Artificial Neural Networks (ANN) are to be used as the underlying intelligent software as they have high tolerance for noise and have the ability to generalize. A colleague has implemented a basic ANN based system comprising of an ANN and three cost effective laser distance sensors. However, the system is only able to identify 3 different parts and needed hard coding changes made by trial and error. This is not practical for industrial use in a production environment where there are a large quantity of different parts to be identified that change relatively regularly. The ability to easily train more parts is required. Difficulties associated with traditional mathematically guided training methods are discussed, which leads to the development of a Genetic Algorithm (GA) based evolutionary training method that overcomes these difficulties and makes accurate part recognition possible. An ANN hybridised with GA training is introduced and a general solution encoding scheme which is used to encode the required ANN connection weights. Experimental tests were performed in order to determine the ideal GA performance and control parameters as studies have indicated that different GA control parameters can lead to large differences in training accuracy. After performing these tests, the training accuracy was analyzed by investigation into GA performance as well as hardware based part recognition performance. This analysis identified the ideal GA control parameters when training an ANN for the purpose of part recognition and showed that the ANN generally trained well and could generalize well on data not presented to it during training.
12
Cheng, Martin Chun-Sheng, and pjcheng@ozemail com au. "Dynamical Near Optimal Training for Interval Type-2 Fuzzy Neural Network (T2FNN) with Genetic Algorithm." Griffith University. School of Microelectronic Engineering, 2003. http://www4.gu.edu.au:8080/adt-root/public/adt-QGU20030722.172812.
Full textAbstract:
Type-2 fuzzy logic system (FLS) cascaded with neural network, called type-2 fuzzy neural network (T2FNN), is presented in this paper to handle uncertainty with dynamical optimal learning. A T2FNN consists of type-2 fuzzy linguistic process as the antecedent part and the two-layer interval neural network as the consequent part. A general T2FNN is computational intensive due to the complexity of type 2 to type 1 reduction. Therefore the interval T2FNN is adopted in this paper to simplify the computational process. The dynamical optimal training algorithm for the two-layer consequent part of interval T2FNN is first developed. The stable and optimal left and right learning rates for the interval neural network, in the sense of maximum error reduction, can be derived for each iteration in the training process (back propagation). It can also be shown both learning rates can not be both negative. Further, due to variation of the initial MF parameters, i.e. the spread level of uncertain means or deviations of interval Gaussian MFs, the performance of back propagation training process may be affected. To achieve better total performance, a genetic algorithm (GA) is designed to search better-fit spread rate for uncertain means and near optimal learnings for the antecedent part. Several examples are fully illustrated. Excellent results are obtained for the truck backing-up control and the identification of nonlinear system, which yield more improved performance than those using type-1 FNN.
13
Kothari, Bhavin Chandrakant. "Structural optimisation of artificial neural networks by the genetic algorithm using a new encoding scheme." Thesis, Brunel University, 1997. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.389263.
Full text
14
Zhang, Xiaohui. "Development and Testing of a Combined Neural-Genetic Algorithm to Identify CO2 Sequestration Candidacy Wells." Thesis, University of Louisiana at Lafayette, 2015. http://pqdtopen.proquest.com/#viewpdf?dispub=1594272.
Full textAbstract:
This study was motivated by how to use statistical tool to identify the candidacy wells for CO2 Capture and Sequestration based on the idea of using Artificial Neural Networks to predict the leakage index of a well. A Combined Neural-Genetic Algorithm was introduced to avoid BP neural network getting a local minimum because Genetic Algorithm simulates the survival of the fittest among individuals over consecutive generation. Based on the algorithm, 1356 lines of C code were composed using Microsoft Visual Studio 2010. The Combined Neural-Genetic Algorithm developed in this thesis is able to handle large size of data sample with at least 10 factors. Several parameters were considered as factors that may have an effect on the performance of Combined Neural-Genetic Algorithm, including the population size, max epoch, error goal, probability of crossover, probability of mutation, number of neurons in hidden layer, number of factors and size of data sample. The accuracy of the BP neural network and the CPU time are the two major parameters to evaluate the performance of the Combined Neural-Genetic Algorithm. A sensitivity analysis was performed to identify the effect these factor have on the performance. Based on the result of the sensitivity analysis, some recommendations are provided about initializing these factors.
15
Spittle, Mark Charles. "Complexity reduction in artificial neural networks with an emphasis on genetic algorithm based optimisation techniques." Thesis, Cardiff University, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.389853.
Full text
16
Dehaven, Ryan Swords. "Smarter NEAT Nets." DigitalCommons@CalPoly, 2013. https://digitalcommons.calpoly.edu/theses/1024.
Full textAbstract:
This paper discusses a modification to improve usability and functionality of a ge- netic neural net algorithm called NEAT (NeuroEvolution of Augmenting Topolo- gies). The modification aims to accomplish its goal by automatically changing parameters used by the algorithm with little input from a user. The advan- tage of the modification is to reduce the guesswork needed to setup a successful experiment with NEAT that produces a usable Artificial Intelligence (AI). The modified algorithm is tested against the unmodified NEAT with several different setups and the results are discussed. The algorithm shows strengths in some areas but can increase the runtime of NEAT due to the addition of parameters into the solution search space.
17
Leong, Sio Hong. "Kinematics control of redundant manipulators using CMAC neural networks combined with Descent Gradient Optimizers & Genetic Algorithm Optimizers." Thesis, University of Macau, 2003. http://umaclib3.umac.mo/record=b1446170.
Full text
18
Sahebi, Mahmod Reza. "Understanding microwave backscattering of bare soils by using the inversion of surface parameters, neural networks and genetic algorithm." Thèse, Université de Sherbrooke, 2003. http://savoirs.usherbrooke.ca/handle/11143/2736.
Full textAbstract:
Estimates of the physical parameters of the soil surface, namely moisture content and surface roughness, are important for hydrological and agricultural studies, as they appear to be the two major parameters for runoff forecasting in an agricultural watershed. Radar has high potentiality for the remote measurement of soil surface parameters. In particular, the investigation of the radar backscattering response of bare soil surfaces is an important issue in remote sensing because of its capacity for retrieving the desired physical parameters of the surface. The objective of this study is to formulate and to constrain a methodology for solving the inverse problem for the operational retrieval of soil surface roughness and moisture. To separate the effects of the different parameters on the measured signal over complex areas, multi-technique concepts (multi-polarization, multi-angular, multi-sensor, multi-frequency, and multi-temporal) are the main solution. In this work, based on a simulation study, three different configurations, multi-polarization, multi-frequency and multi-angular, are compared to obtain the best configuration for estimating surface parameters and the multi-angular configuration gives the best results. Based on these results, this study was continued according to five different phases: (1) A new index, the NBRI (Normalized radar Backscatter soil Roughness Index), using the multi-angular approach was presented. This index can estimate and classify surface roughness in agricultural fields using two radar images with different incidence angles. (2) A new linear empirical model to estimate soil surface moisture using RADARSAT-1 data was proposed. This model can provide soil moisture with reduced errors of estimation compared to other linear models. (3) Inversion of the surface parameters using nonlinear classical methods. In this case, the Newton-Raphson method, an iterative numerical method, was used in the retrieval algorithm to solve the inverse problem. (4) In this phase, the neural network technique, with a dynamic learning method, was applied to invert the soil surface parameters from the radar data. The results were obtained through performance testing on two different input schemes (one and two data series) and two different databases (theoretical and empirical). The advantage of the multi-angular set with measured data is apparent. These results are the best in this study. (5) Finally, a novel genetic algorithm (GA) was developed to retrieve soil surface parameters. In this study, it is shown that the genetic algorithms, as an optimization technique, can estimate simultaneously soil moisture and surface roughness from only one radar image.
19
Liu, Jingrong. "Design and Analysis of Intelligent Fuzzy Tension Controllers for Rolling Mills." Thesis, University of Waterloo, 2002. http://hdl.handle.net/10012/848.
Full textAbstract:
This thesis presents a fuzzy logic controller aimed at maintaining constant tension between two adjacent stands in tandem rolling mills. The fuzzy tension controller monitors tension variation by resorting to electric current comparison of different operation modes and sets the reference for speed controller of the upstream stand. Based on modeling the rolling stand as a single input single output linear discrete system, which works in the normal mode and is subject to internal and external noise, the element settings and parameter selections in the design of the fuzzy controller are discussed. To improve the performance of the fuzzy controller, a dynamic fuzzy controller is proposed. By switching the fuzzy controller elements in relation to the step response, both transient and stationary performances are enhanced. To endow the fuzzy controller with intelligence of generalization, flexibility and adaptivity, self-learning techniques are introduced to obtain fuzzy controller parameters. With the inclusion of supervision and concern for conventional control criteria, the parameters of the fuzzy inference system are tuned by a backward propagation algorithm or their optimal values are located by means of a genetic algorithm. In simulations, the neuro-fuzzy tension controller exhibits the real-time applicability, while the genetic fuzzy tension controller reveals an outstanding global optimization ability.
20
Sabih, Ann Faik. "Cognitive smart agents for optimising OpenFlow rules in software defined networks." Thesis, Brunel University, 2017. http://bura.brunel.ac.uk/handle/2438/15743.
Full textAbstract:
This research provides a robust solution based on artificial intelligence (AI) techniques to overcome the challenges in Software Defined Networks (SDNs) that can jeopardise the overall performance of the network. The proposed approach, presented in the form of an intelligent agent appended to the SDN network, comprises of a new hybrid intelligent mechanism that optimises the performance of SDN based on heuristic optimisation methods under an Artificial Neural Network (ANN) paradigm. Evolutionary optimisation techniques, including Particle Swarm Optimisation (PSO) and Genetic Algorithms (GAs) are deployed to find the best set of inputs that give the maximum performance of an SDN-based network. The ANN model is trained and applied as a predictor of SDN behaviour according to effective traffic parameters. The parameters that were used in this study include round-trip time and throughput, which were obtained from the flow table rules of each switch. A POX controller and OpenFlow switches, which characterise the behaviour of an SDN, have been modelled with three different topologies. Generalisation of the prediction model has been tested with new raw data that were unseen in the training stage. The simulation results show a reasonably good performance of the network in terms of obtaining a Mean Square Error (MSE) that is less than 10−6 [superscript]. Following the attainment of the predicted ANN model, utilisation with PSO and GA optimisers was conducted to achieve the best performance of the SDN-based network. The PSO approach combined with the predicted SDN model was identified as being comparatively better than the GA approach in terms of their performance indices and computational efficiency. Overall, this research demonstrates that building an intelligent agent will enhance the overall performance of the SDN network. Three different SDN topologies have been implemented to study the impact of the proposed approach with the findings demonstrating a reduction in the packets dropped ratio (PDR) by 28-31%. Moreover, the packets sent to the SDN controller were also reduced by 35-36%, depending on the generated traffic. The developed approach minimised the round-trip time (RTT) by 23% and enhanced the throughput by 10%. Finally, in the event where SDN controller fails, the optimised intelligent agent can immediately take over and control of the entire network.
21
Li, Keh-Tsong, and 李克聰. "Neural Network combined with Genetic Algorithm-Evolutionary Neural Network." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/37236508646662658444.
Full textAbstract:
碩士國立交通大學
電機與控制工程系
87
This thesis presents a Real-Coded Rank-Based Genetic Algorithm (RCRBGA), which is represented by a chromosome containing parameters in floating-point. The use of rank-based fitness increases the population diversity. The offspring are generated by the rank-based reproduction, real parametric crossover and mutation in the evolving process. Besides, an Evolutionary Neural Network (ENN) which combines RCRBGA and Back-Propagation (BP) is introduced. ENN applies the learning concept to the evolution process, like the behavior of human beings. It not only improves the disadvantage of easily slumping in to local minima of BP but also overcomes the defect of genetic algorithm, which can't efficiently converge to minima. Finally, the search ability of RCRBGA is demonstrated by an example, linear state-feedback controller via pole-assignment method. In addition, ENN applies to a classifying problem of the modified XOR to show its advantage.
22
Chen, lily, and 陳麗莉. "A Neural Genetic Algorithm for Product Design." Thesis, 1997. http://ndltd.ncl.edu.tw/handle/76419743128612100659.
Full text
23
Fong-Hang, Liao, and 廖鴻翰. "Construct Neural Network Model Using Genetic Algorithm." Thesis, 1998. http://ndltd.ncl.edu.tw/handle/60094674695662600817.
Full textAbstract:
碩士大葉大學
電機工程研究所
86
In this thesis a new Genetic Algorithm to optimize weights and topology of Neural Networks is presented and compared with other learning methods, such as gradient-descent learning algorithm, and other evolutionary system. Since the characteristics of topology space and weight space (one of them is in integer space and the other is in real space) are absolutely different, it is very difficult to optimize both of them at the same time. Cascade-Correlation algorithm (CCA) is a popular supervised learning architecture that dynamically grows layers of hidden neurons, so that the network topology (size, depth) can be determined with network weights at the same time. CCA is a gradient-descent based learning algorithm. It is known that the gradient methods always foiled by local minimum problem. On the other hand, it is powerful on local search but insufficient on global search. Genetic algorithm (GA) is a computationally intensive optimization method. The rewards from the huge computational power are some very desirable properties. One of them is that a global search is performed during the optimization. But unfortunately, GA is insufficient on local search for the reason of poor fine-tuning. In order to optimize network topology and weights more efficiently. It is possible to combine the advantages of them (CCA and GA), and avoid the disadvantages. For the reason, Genetic Algorithm Based Correlation-Construction (GABCC) was developed. The basic concept of GABCC is ''adaptability''. By adaptive GA operators(selection, crossover, mutation), we can evaluate the population more efficiently. Adaptive GA operators have good abilities both on global search and local search. For the reason, it gets better performance than traditional gradient methods shown in the benchmarks we have tested. More over, GABCC combine both of the adaptive GA and correlation-constructed method. So that weights and topology of network can be optimized at the same time
24
Hsu, Tung-Jung, and 許東榮. "Integrating Genetic Algorithm with Neural Network for." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/37047274475860922509.
Full textAbstract:
碩士國立勤益科技大學
工業工程與管理系
101
Taiwan, which possesses cutting-edge industries, lacking for natural resources, and dominating the field in semiconductors, optoelectronics, information, communications, electronics precision manufacturing technology. In recent years, with the rapid growth of international trade and the competitive environment, inbound and outbound passengers and volume of imported goods are increasing. In addition to execution levied on tariffs and preventing smuggling, Customs officers must perform border control measures, such as national security, quarantine, environmental protection, protection of intellectual property rights, etc. Using scientific management methods, construction of preventing smuggling expert system, knowledge management, and purchasing high-tech equipment can relieve customs officers on tasks burden. This study integrated genetic algorithms with neural network to construct a smuggling case classification and import tariffs prediction model which provide Customs expert system a decision-making tool and enhance the performance.
25
Chen, Tsung-Hung, and 陳宗宏. "Neural networks assess liquefaction of sand -Genetic algorithm." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/44542373651175737769.
Full text
26
Lee, I.-Ting, and 李宜庭. "Evolution of Neural Circuit Models by Genetic Algorithm." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/bx28bc.
Full text
27
Li, Hai-Han, and 李海涵. "An Improved Algorithm Applied in Training Neural Network-Combined with Genetic Algorithm." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/et2442.
Full textAbstract:
碩士國立臺北科技大學
商業自動化與管理研究所
93
Gradient steepest descent (GSD) is often used to train the back-propagation neural network (BPN) because of its excellent performance of reducing training errors; however, it also has some drawbacks such as slow convergence and local optimum problem. Many improved methods are proposed to amend the aforementioned demerits; for example, momentum can be employed to accelerate convergence, and global search methods, e.g. probabilistic climbing search and taboo search (TS), etc. are introduced to fix the local optimum problem. Nevertheless, some weaknesses exist in those methods. For instance, added momentum may sometimes not work well in speeding up convergence; probabilistic climbing methods assume that error function follows a certain distribution, which may not always be true. While TS might approximate the global solutions, its quality of solution remains unstable on account of too many random variables and it often requires heavy computation. This paper proposes an improved method to hasten convergence and decrease the training errors effectively without much more training time. Even so, whatever algorithms which are mentioned above encounter bottleneck of achieving more accuracy of training. That is, diminishing of training errors becomes stagnant at some convergence level. If evolutionary algorithms e.g. genetic algorithm (GA) is combined, training accuracy may be in theory refined indefinitely to its maximum precision with proper evolving strategies. This paper lays emphasis on evolving strategies instead of evolving operators. It’s preliminarily proved in this paper that during the long evolution process, influence of evolving strategies is greater than that of evolving operators. The training time of combining GA would not grow as a result of parallel processing.
28
Yang, Guo-Feng, and 楊國鋒. "Face Detection Using Genetic Algorithm and Artificial Neural Network." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/18758722183870591957.
Full textAbstract:
碩士元智大學
資訊工程學系
96
Human face represents one of the most common patterns in our vision. Therefore, automatic recognition of human faces is an essential task in many applications such as criminal identification and security checks. The first important step of automatic human face recognition is to detect face in a given unknown picture. However, the task of automatic face detection in a complex background is difficult to cope with. In this thesis, discriminating features are selected by genetic algorithm with neural network so as to design an accurate face detector. Moreover, verification on face skin has been involved to increase the accuracy of face detection. Experimental results prove the effectiveness of the proposed text detection method.
29
Feng, Yen-Ru, and 馮彥儒. "Text Detection Using Genetic Algorithm and Artificial Neural Network." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/62282555625352186744.
Full textAbstract:
碩士元智大學
資訊工程學系
96
The text embedded in images and video streams imply tremendous information. Thus, text extraction from image or video streams has been widely applied in a variety of application fields, such as document analysis, content-based retrieval and intelligent transportation system, etc. However, texts are often embedded in an image and may vary in language, font, size, and deformation, which, in turn enhance the difficulty of text detection problem. In this thesis, discriminating features are selected by genetic algorithm with neural network. Moreover, fusion of pyramid images has been involved to detect texts in variation of sizes. Experimental results prove the effectiveness of the proposed text detection method.
30
Ying-Yi, Wang. "A Hybrid Neural-genetic Algorithm for Reservoir Water Quality Management." 2006. http://www.cetd.com.tw/ec/thesisdetail.aspx?etdun=U0001-2007200610412000.
Full text
31
Shia, Yu-Lung, and 夏裕龍. "Apply Genetic Algorithm And Neural Network To Forecast Taiwan Weather." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/78336942646795301517.
Full text
32
Chang, Chia-Tsang, and 張家瑲. "Application of Neural Network and Genetic Algorithm to System Identification." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/73351083036139978352.
Full textAbstract:
碩士朝陽科技大學
營建工程系碩士班
91
Taiwan is a high seismic zone since it is located at the active arc-continent collision region between the Luzon arc of the Philippine Sea plate and the Eurasian plate. The Chi-Chi Earthquake is the largest inland earthquake occurred in Taiwan during this century. Due to the great damage caused by this earthquake, more and more emphases have been put on the earthquake resistant design of buildings. Dynamic behavior of buildings under earthquakes should be considered in the process of design. In order to realize the dynamic behavior of structural systems subjected to earthquakes, we can determine dynamic models and parameters through various system identification techniques. In this study, it is intended to develop new identification techniques by combining the advantages of both neural network (NN) and genetic algorithm (GA). Firstly, the time history of the ground acceleration and the system parameters of a variety of SDOF systems are used as the input data of neural network, and the time history of the relative acceleration of the respective SDOF systems as the neural network outputs. After the training of the neural network, the network topology used to evaluate the time history of the relative acceleration of the SDOF systems will be captured. This network topology is then employed to replace the procedure for solving the governing (differential) equation when GA is used to identify the system parameters. Furthermore, this topology is used in the identification of the MDOF system subjected to the single input by mode superposition technique. On the other hand, the starting weights of NN are randomly selected and the optimization algorithm used in the training of NN may get stuck in the local minimal. GA is a search method based on natural selection and genetics and is different from conventional optimization methods in several ways. The GA is a parallel and global search technique that searches multiple points, so it is more likely to obtain a global solution. In this regard, a new algorithm of combining GA and NN is proposed here. The GA is employed to search for the starting weights and the NN is used to obtain the network topology. Through the iterative process of selection, reproduction, cross over and mutation, the optimal weight can then be obtained. This proposed algorithm is applied to the Duffing oscillator and Wen’s degrading nonlinear systems. Finally, the accuracy of this method is illustrated by comparing the results of the predicted response with the measured one.
33
Wang, Ying-Yi, and 王英義. "A Hybrid Neural-genetic Algorithm for Reservoir Water Quality Management." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/87142282187697168993.
Full textAbstract:
博士國立臺灣大學
土木工程學研究所
94
There has been concern over the water quality in Feitsui Reservoir, particularly since the beginning of Taipei-Ilan highway construction in 1991. In the present study, a combined artificial neural network (ANN) and genetic algorithms (GAs) approach was proposed for water quality management of Feitsui Reservoir in Taiwan. First, two simplified water quality models based on ANN were developed and used as universal approximators to imitate the cause-and-effect relationships between phosphorus loads from the watershed and water quality concentrations (total phosphorus and chlorophyll a, respectively) in Feitsui Reservoir. A six-year (1992-1997) record of water quality data was used for network training, and additional data collected in 1998-2000 was used for model verification. The performance and validity of the proposed ANN models were evaluated using two conventional water quality models, including a total phosphorus model and an eutrophication model (WASP/EUTRO). Further, a GA with water quality prediction produced by the ANN model was used to optimize the control of watershed nutrient loads. The GA was applied to the problem of reservoir water quality management to provide an alternative when searching for an optimal control strategy. The study results reveal that the ANN model can effectively simulate the dynamics of reservoir water quality, indicating that an ANN model can replace the conventional water quality model in this water quality management analysis, and the GA is able to identify control schemes that improve the current trophic levels to achieve water quality standards. Finally, the time-variable control schemes derived from the ANN-GA method were applied to the WASP/EUTRO model to assess the impact on eutrophication in Feitsui Reservoir following phosphorus load reductions in its watershed. The modeling results suggest that adequate control of phosphorus loads into the reservoir is needed for preserving the water quality of Feitsui Reservoir from eutrophication. In practice, the time-varying reductions in phosphorus loads for controlling reservoir eutrophication can be achieved by way of the joint reduction of point and nonpoint source pollution loads.
34
Huang, Shin-Mao, and 黃鑫茂. "A Novel Neural Network Training Technique by Using Genetic Algorithm." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/75933188485584344214.
Full textAbstract:
碩士國立交通大學
電機與控制工程系
88
This thesis investigates a novel neural network training technique, which employs the genetic algorithm to finding the initial values of the neural network. It is represented by a chromosome containing parameters in floating-point, so that the convergence rate to the minima becomes faster. This hybrid algorithm can overcome not only the drawback of easily slumping into local minima of back-propagation but also the genetic algorithm’s defect that can’t efficiently converge to the minima of the neighborhood. Further, the thesis shows that a gene changing one by one is better than that changing totally at once. Finally, the results of computer simulations reveal that this algorithm has a better convergence property, the time of global searching is obviously decreased.
35
Chang, Chen-Chi, and 張錦基. "Measuring body fat using regression analysis﹐artificial neural network and genetic algorithm neural network." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/60672346971399257994.
Full textAbstract:
碩士淡江大學
資訊工程學系碩士在職專班
98
Body fat mass is one of the health indicators. Measuring it is helpful to understand the relationship between body fat and diseases. Although, cadaver dissection provides the most accurate method to assess the value. But, it is not appropriate for the people who are living. Additionally, some accurate methods, such as underwater weighting, isotope dilution, bioelectrical impedance analysis , are complicated and costly incredibly. Therefore, Young Men''s Christian Association (YMCA) and the United States army tried to develop instruments for gauging body fat. Furthermore, World Health Organization (WHO) suggested that using body mass index (BMI) instead of body fat. However, evaluating BMI is not considered distribution of human body fat tissue component and specific region. It is doubtful about the accuracy. The purpose of this study is constructing a more precise predict model by multiple regression analysis, artificial neural network, genetic algorithm neural network, the parameters are age, weight, height, neck circumference, chest circumference, abdomen circumference, hip circumference, and so on. 252 males’ body measurement indicators were database which were collected by Dr. A. Garth Fisher who was in Human Performance Research Center , Brigham Young University , Provo. The result is genetic algorithm neural network RMSE: Root Mean Square Error (RMSE 4.0854) > artificial neural network 5 variables model (RMSE 4.3330) > artificial neural network 12 variables model (RMSE 4.3783) > multiple regression analysis 12 variables model (RMSE 4.3981) > multiple regression analysis 5 variables model (RMSE 4.4620) > YMCA body fat model (RMSE 4.7757) > US Army body fat Model (RMSE 7.7336)。
36
Hippolyte, Djonon Tsague. "MACHINE CONDITION MONITORING USING NEURAL NETWORKS: FEATURE SELECTION USING GENETIC ALGORITHM." Thesis, 2007. http://hdl.handle.net/10539/2127.
Full textAbstract:
Student Number : 9800233A -
MSc dissertation -
School of Electrical and Information Engineering -
Faculty of Engineering and the Built EnvironmentCondition monitoring of machinery has increased in importance as more engineering processes are automated and the manpower required to operate and supervise plants is reduced. The monitoring of the condition of machinery can significantly reduce the cost of maintenance. Firstly, it can allow an early detection of potential catastrophic fault, which could be extremely expensive to repair. Secondly, it allows the implementation of conditions based maintenance rather than periodic or failure based maintenance [1]. In these cases, significant savings can be made by delaying schedule maintenance until convenient or necessary. Although there are numerous efficient methods for modeling of mechanical systems, they all suffer the disadvantage that they are only valid for a particular machine. Changes within the design or the operational mode of the machine normally require a manual adaptation. Using Neural Networks to model technical systems eliminates this major disadvantage. The basis for a successful model is an adequate knowledge base on which the network is "trained". Without prior knowledge of the machines systematic behavior or its history, training of a neural Network is not possible. Therefore, it is a pre-requisite that the knowledge base contains a complete behavior of the machine covering the respective operational modes whereby, not all rather the most important modes are required. Neural networks have a proven ability in the area of nonlinear pattern classification. After being trained, they contain expert knowledge and can correctly identify the different causes of bearing vibration. The capacity of artificial neural networks to mimic and automate human expertise is what makes them ideally suited for handling nonlinear systems. Neural networks are able to learn expert knowledge by being trained using a representative set of data [2]-[6]. At the beginning of a neural network’s training session, the neural network fault detector’s diagnosis of the motor’s condition will not be accurate. An error quantity is measured and used to adjust the neural network’s internal parameters in order to produce a more accurate output. This process is repeated until a suitable error is achieved. Once the network is sufficiently trained and the parameters have been saved, the neural network contains all the necessary knowledge to perform the fault detection. One of the most important aspects of achieving good neural network performance has proven to be the proper selection of training features. The curse of dimensionality states that, as a rule of thumb, the required cardinality of the training set for accurate training increases exponentially with the input dimension [7]. Thus feature selection which is a process of identifying those features that contribute most to the discrimination ability of the neural network is required. Proposed methods for selecting an appropriate subset of features are numerous [8]-[11]. Methods based on generating a single solution, such as the popular forward step wise approach, can fail to select features which do poorly alone but offer valuable information together. Approaches that maintain a population of solutions, such as genetic algorithms (GA) are more likely to speedily perform efficient searches in high dimensional spaces, with strong interdependencies among the features. The emphasis in using the genetic algorithm for feature selection is to reduce the computational load on the training system while still allowing near optimal results to be found relatively quickly. To obtain accurate measure of the condition of machinery, a wide range of approaches can be employed to select features indicative of condition. By comparing these features with features for known normal and probable fault conditions, the machine’s condition can be estimated. The most common approach is that of analysis in the frequency domain by applying a Fast Fourier Transform (FFT) to the time domain history data. The idea is simply to measure the energy (mean square value) of the vibrations. As the machine condition deteriorates, this measure is expected to increase. The method is able to reveal the harmonics around the fundamental frequency of the machine and other predominant frequency component (such as the cage frequency) [12]. Frequency analysis is well established and may be used to detect, diagnose and discriminate a variety of induction motor faults such as broken rotor bars, cage faults, phase imbalance, inner and outer race faults. However, as common in the monitoring of any industrial machine, background noise in recorded data can make spectra difficult to interpret. In addition, the accuracy of a spectrum is limited due to energy leakage [12- 14]. Like many of the new techniques now finding application in machinery condition monitoring, Higher Order Statistics was originally confined to the realms of non-linear structural dynamics. It has of recent however found successful application to the identification of abnormal operation of diesel engines and helicopter gearboxes [5, 7]. Higher Order Statistics provide convenient basis for comparison of data between different measurement instances and are sufficiently robust for on-line use. They are fast in computation compared with frequency or time-domain analysis. Furthermore, they give a more robust assessment than lower orders and can be used to calculate higher order spectra. This dissertation reports work which attempts to extend this capability to induction motors. The aim of this project is therefore to examine the use of Genetic Algorithms to select the most significant input features from a large set of possible features in machine condition monitoring contexts. The results show the effectiveness of the selected features from the acquired raw and preprocessed signals in diagnosis of machine condition. This project consists of the following tasks: #1; Using Fast Fourier transform and higher order signals techniques to preprocess data samples. #1; Create an intelligent engine using computational intelligence methods. The aim of this engine will be to recognize faulty bearings and assess the fault severity from sensor data. #1; Train the neural network using a back propagation algorithm. #1; Implement a feature selection algorithm using genetic algorithms to minimize the number of selected features and to maximize the performance of the neural network. #1; Retrain the neural network with the reduced set of features from genetic algorithm and compare the two approaches. #1; Investigate the effect of increasing the number of hidden nodes in the performance of the computational intelligence engine. #1; Evaluate the performance of the system using confusion matrices. The output of the design is the estimate of fault type and its severity, quantified on a scale between 0-3. Where, 0 corresponds to the absence of the specific fault and 3 the presence of a severe machine bearing fault. This research should make contribution to many sectors of industry such as electricity supply companies, and the railroad industry due to their need of techniques that are capable of accurately recognizing the development of a fault condition within a machine system component. Quality control of electric motors is an essential part of the manufacturing process as competition increases, the need for reliable and economical quality control becomes even more pressing. To this effect, this research project will contribute in the area of faults detection in the production line of electric motor.
37
Chen, Chi-Wei, and 陳啟瑋. "Hybrid Genetic Algorithm and Neural Network to Design A PID Controller." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/91856841436605951017.
Full textAbstract:
碩士國立臺灣海洋大學
機械與機電工程學系
98
The PID controller operation is simple and easy to design, and it has been used widely in industrial applications. The performance of controller depends on the control parameters. As the PID controller parameter set dependence of experience or experiment to determine, it is difficult to get the best parameters. The back-propagation neural network uses the steepest gradient decent method to adjust weights. The initial weights of neural network generate by random or experience. This will cause time-consuming and poor reliability. This paper proposes a genetic algorithm to optimize the parameters by the selection, crossover and mutation. Through the set of fitness function, and find the best initial weight in the system. According to the simulation results, hybrid genetic algorithm and neural network PID controller can self-adjust the parameters, and both have good robustness and adaptability on linear and nonlinear systems.
38
Shen, Tzung-Tza, and 沈宗澤. "Training Artificial Neural Network Using Genetic Algorithm and Conjugate Gradient Method." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/18262883491045855458.
Full textAbstract:
碩士國立成功大學
航空太空工程學系
89
The purpose of this study is to combine the conjugate gradient method(CG) and the genetic algorithm(GA) for the training of artificial neural networks(ANN). The back-propagation artificial neural network is a broadly used artificial neural network in many areas. It usually adopts the steepest descent method(SD) to search for a set of connection weights that minimizes the training error. But the convergence of the steepest descent method is very slow and easy to trap into a local optimal. In order to speed up the convergence, the conjugate gradient method searches the optimal weights along a set of conjugate directions in stead of steepest descent ones. But it still has the drawback of trapping into local optimals. The genetic algorithm is a global optimization method based on the Darwin’s principle of ‘’Survival of the fittest’’. The genetic algorithm always searches for the global optimal. In this study, we develop a hybrid method which combines the conjugate gradient method and the genetic algorithm to improve the convergence and successful rate for the training of artificial neural networks.
39
Lee, Jung-Che, and 李榮哲. "Implementation of FPGA-Based Artificial Neural Network Combined with Genetic Algorithm." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/92274078038329124679.
Full textAbstract:
碩士國立交通大學
電控工程研究所
100
This thesis is aimed to implement the hardware structure of the genetic algorithm (GA), which is applied to search the optimal weights for the FPGA-based artificial neural network (ANN). In contrast with the traditional gradient algorithm, GA uses multi-point population to search the optimum, which is suitable to implement on FPGA in binary code without complex computation. There are two modules proposed for GA hardware to speed up searching, CMU and SU. The CMU generates one crossover mask and two mutation masks at the same time, not in order, to reduce a lot of execution clock cycles. The SU finds the best individual in each generation and saves it as the next generation parent to always keep the elite in the population. The hardware includes three crossover operations, one-point crossover, two-point crossover and uniform crossover. The users can choose one of them and define the crossover rate and mutation rate to deal with different problems. As for the forward calculation of ANN, the multilayer architecture is realized by the layer multiplexing method to reduce the resource since it only requires a single layer to be used repeatedly. The success of the GA hardware architecture is demonstrated by three experiments on Altera DE2-70 FPGA board with 50 MHz operation frequency, including two-dimensional optimal searching, M-G curve prediction fitting and edge detection.
40
LAI, I.-CHIEN, and 賴以建. "GENETIC ALGORITHM, NEURAL NETWORK AND DECISION TREE IN PRE-WARNING MODELS." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/34939994824622402263.
Full textAbstract:
碩士國立臺北大學
企業管理學系
91
In the past, the pre-warning models for Financial Crisis are usually established based on traditional statistical methods such as Discriminant Analysis. However, it is often questionable whether the financial data satisfies the assumptions of such models. Therefore, this study investigates the construction of pre-warning model through nonlinear methods such as Genetic Algorithm and Neural Network. In additional, since the reference value for the key indicator that influences business failure most cannot be extracted from the pre-warning model, this study starts with using Decision Tree technique to extract this reference value. Based upon this, the objectives of this thesis include the following: 1.Identify the chromosome that influences business failure most through Genetic Algorithm’s strong searching capability. 2.Construct financial pre-warning models from Neural Network and traditional Discriminant Analysis techniques, and evaluate their pre-warning performance by comparing the ability to predict business failure three years before its occurrence. 3.Extract the reference value and the key descriptive indicator that influences business failure most through Decision Tree technique, thus enabling the investing public and associated authority to constantly monitor the key financial factors. The main characteristic of the Genetic Algorithm used in this study is its massive parallel optimizing ability. The analyses on the actual data show that: 1.Identify the Genetic component (chromosome) that influences business failure most through Genetic Algorithm: after 500 generations, the optimal chromosome combinations are Operating Income Ratio, Sales per Share, Earnings before Interest/Equity, Net Present Value per Stock (A), Net Present Value per Stock (B), Retained Profit Ratio, Cash Flow Adequacy Ratio, Times Interest Earned, Fixed Asset Turnover Ratio, and Operating Expense Ratio. 2.By employing the key indicators obtained from Genetic Algorithm, both Neural Network model and Discriminant Analysis model can accurately predict business failure (on average, for three-year ago prediction, hit ratio: 0.9500 compared with 0.9055). The hit ratios for both models are the same (0.9667) for one-year ago prediction. However, the hit ratios for two- and three-year ago predictions are higher for Neural Network model (0.9500 and 0.9333 compared with 0.9166 and 0.8333). This indicates that Neural Network pre-warning model has higher probability to successfully predict business failure earlier. 3.The Decision Tree cannot effectively distinguish the samples of successful and failure business. The following results are observed. When the Retained Profit Ratio of a business is larger than 0.9931, the business failure rate is about 88%. When the Retained Profit Ratio is larger than 0.9931 and Operating Income Ratio is lower than 0.0098, the business failure rate is as high as 97.33%.
41
Yang, Chia-Rong, and 楊佳榮. "Simulated Annealing, Genetic Algorithm, and Neural Network for Seismic Velocity Picking." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/09117340783882246780.
Full textAbstract:
碩士國立交通大學
生醫工程研究所
101
Velocity picking is an important step for seismic data processing. It is to pick several time-velocity pairs forming a polyline in a semblance image to represent the time and velocity relation in layers. Conventionally the geophysicists did it, but it took much time. We transfer it to a combinatorial problem which is finding the best combination from the set of candidate points. We define an objective function of energy that includes total semblance value of picked points, and constraints on the number of picked points, interval velocity, and velocity slope. We adopt three optimization methods: simulated annealing (SA), genetic algorithm (GA), and Hopfield neural network (HNN), to find the optimal solution of objective function and obtain the best polyline consisting of picked peak points respectively. In SA, the random system state represents a solution, and the higher energy random system state has a certain probability to be accepted and skip the local minimum. After annealing, the lowest energy system state is the best polyline. It is a global optimal solution. In GA, an individual represents a solution. Several individuals evolve many generations. The highest fitness individual in the last generation is the best polyline. It is also a global optimal solution. In HNN, the neurons of network represent a solution. We derive the equation of motion from the objective function and use asynchronous updating to renew the neurons of network. Finally, the network converges. The stable network state represents the best polyline. In GA, we find the maximum of the objective function. In SA and HNN, we find the minimum of the objective function. In the implementations of SA and HNN, we change the objective function to a negative function and find the minimum. In the parameter settings of SA and GA, we find the best parameter settings by sequential method. We have experiments on simulation data and Nankai real seismic data. We have 22 common midpoint gathers (CMP gathers) of simulated seismic data and 15 CMP gathers of Nankai real seismic data for experiments. We evaluate the performance by comparing the mean difference between the picking result of each adopted method and that of human. The experiments show that GA has the best result on the simulated and real seismic data experiments. The best picking results by three methods are further used to do normal move-out (NMO) correction and stacking. The results show that both of the signals of the simulated and real seismic data are enhanced. The results of velocity picking by three optimization methods will be helpful for further seismic data processing and interpretation.
42
Zhu, Yuqing. "Nonlinear system identification using a genetic algorithm and recurrent artificial neural networks." Thesis, 2006. http://spectrum.library.concordia.ca/9060/1/MR20771.pdf.
Full textAbstract:
In this study, the application of Recurrent Artificial Neural Network (RANN) in nonlinear system identification has been extensively explored. Three RANN-based identification models have been presented to describe the behavior of the nonlinear systems. The approximation accuracy of RANN-based models relies on two key factors: architecture and weights. Due to its inherent property of parallelism and evolutionary mechanism, a Genetic Algorithm (GA) becomes a promising technique to obtain good neural network architecture. A GA is developed to approach the optimal architecture of a RANN with multiple hidden layers in this study. In order to approach the optimal architecture of Neural Networks in the sense of minimizing the identification error, an effective encoding scheme is in demand. A new Direct Matrix Mapping Encoding (DMME) method is proposed to represent the architecture of a neural network. A modified Back-propagation (BP) algorithm, in the sense of not only tuning NN weights but tuning other adjustable parameters as well, is utilized to tune the weights of RANNs and other parameters. The RANN with optimized or approximately optimized architecture and trained weights have been applied to the identification of nonlinear dynamic systems with unknown nonlinearities, which is a challenge in the control community. The effectiveness of these models and identification algorithms are extensively verified in the identification of several complex nonlinear systems such as a "smart" actuator preceded by hysteresis and friction-plague harmonic drive.
43
CHIH, LIN CHIEN, and 林建智. "A study on integrated application of genetic algorithm and artificial neural network." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/26610585807673509540.
Full text
44
Wei, Chih-hsiu, and 韋至修. "Fuzzy Clustering by Distributed Genetic Algorithm and Multi-Synapse Neural Network Approaches." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/15901631622604488212.
Full textAbstract:
博士國立臺灣科技大學
電機工程系
90
The area of research in this dissertation is fuzzy c-partition clustering, which is understood to be the grouping of similar objects with the concept of fuzzy set theory to incorporate the uncertainty of the final classification results. There are three parts in this dissertation. The first part is an overview of fuzzy c-partition clustering. In the second part, two distributed approaches of genetic search strategies for fuzzy clustering are proposed to surmount the problem of huge search space in the traditional combination of evolutionary algorithms and fuzzy c-partition clustering. The distributed optimization approaches proposed can divide the huge search space into many small ones, which in effect will lower the size of the total search space. The benefit of our approaches is especially shown in clusters with shell shapes, of which the basins of attraction of local minima are very small. In the third part, a new neural architecture, the multi-synapse neural network, is developed for constrained optimization problems, whose objective functions may include high order, logarithmic, sinusoidal forms, unlike the traditional Hopfield networks which can only handle quadratic form optimization. Meanwhile, based on the application of this new architecture, a fuzzy bidirectional associative clustering network (FBACN) is proposed for fuzzy c-partition clustering according to the objective-functional method. It is well known that fuzzy c-means is a milestone algorithm in the area of fuzzy c-partition clustering. All of the following objective-functional-based fuzzy c-partition algorithms incorporate the formulas of fuzzy c-means as the prime mover in their algorithms. However, when an application of fuzzy c-partition has sophisticated constraints, the necessity of analytical solutions in a single iteration step becomes a fatal issue of the existing algorithms. The largest advantage of FBACN is that it does not need analytical solutions. For the problems on which some prior information is known, we bring a concept of the combination of part crisp and part fuzzy clustering. Basically, the FBACN is composed of two layers of recurrent networks. Layer 1 can be a Hopfield network or a multi-synapse neural network based on whether its objective function is a quadratic form or a high order form. Yet layer 2 is definitely a multi-synapse neural network. Three examples are given in part III. The first two are the famous butterfly and Anderson’s Iris data sets, which are usually utilized as benchmarks. The last one is a data set with two concentric circles used to demonstrate the constrained fuzzy c-partition.
45
Lee, Ming-chang, and 李明璋. "Artificial Neural Network with Genetic Algorithm for Nonlinear Model of Machining Processes." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/88047810592371927245.
Full textAbstract:
碩士國立高雄第一科技大學
機械與自動化工程研究所
100
This study an artificial neural network (ANN) model with hybrid Taguchi-genetic algorithm (HTGA) is applied in a nonlinear multiple-input multiple-output (MIMO) model of machining processes. The HTGA in the MIMO ANN model optimizes parameters (i.e., weights of links and biases governing ) input-output relationships in the ANN by directly minimizing root-mean-squared error (RMSE), which is a key performance criterion. Experimental results show that, for nonlinear modeling of machining processes, the proposed MIMO HTGA-based ANN model has better prediction accuracy compared to conventional MIMO-based ANN models with backpropagation that are included in the Matlab toolbox.
46
Ho, Cheng-Yi, and 何承懌. "Optimal Chiller Loading by Genetic Algorithm based on Artificial Neural Network model." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/d96phv.
Full textAbstract:
碩士國立臺北科技大學
能源與冷凍空調工程系碩士班
98
In large HVAC systems, the chiller is usually the most power-consuming component. Although different chillers have similar capacities and performance at the initial stage of operation, due to factors such as varying amount of water distribution, different installation locations, pump supply efficiency, chiller initiation sequence, operating time and so forth after specific amount of operation time, different chillers gradually exhibit varying levels of operational performance. In order to determine the characteristics of chiller operation, one must monitor the status of operation and temperature settings for chillers under different configurations while using relevant parameters to build the power consumption models for chillers. With sufficient understanding of the characteristics of various chiller operation in HVAC systems, it is possible to minimize power consumption by the system by keeping various chillers operating at optimal working conditions through chiller control whilst satisfying the required cooling load. In this research, the author has adopted a neural network and regression analysis to construct models of power consumption for chillers in various case studies in order to compare their R2 and average error. With the models completed, appropriate genetic algorithms were applied to compute the optimal load distribution; through the reproduction, crossover, mutation of the genetic algorithms and the coding/decoding of relevant parameters during the computation process, the author was able to derive the combination of the lowest power consumption for the chiller control (under the premise of satisfying the cooling load requirements). Results of the research revealed that chiller power consumption model constructed from neural network turned out to offer better accuracy compared to model constructed from regression analysis. Not only that, the chiller power consumption model constructed from the neural network also offered better number of converging generations and results in the computation of optimized load distribution using a genetic algorithm.
47
Lin, Yong-Cing, and 林永青. "A Study of Flight Time Using Artificial Neural Networks and Genetic Algorithm." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/v2d4ej.
Full textAbstract:
碩士國立臺北科技大學
工業工程與管理系所
93
Since the adoption of open-air policy by the government in 1987, people make more frequent use of air travel to do various business or tourism activities. The volume of air traffic has greatly increased, along with the occurrences of traffic jam in the air. Delays of landings or take-offs and the congestions in the approach air space have become commonplace, exacerbating the already heavy workload of air-traffic controllers and the inadequacies of ATC system. Therefore, a study of flight time in ATC operation to help alleviate airspace congestions has become more and more urgent and important. Taking international airway A1 as an example, this study makes use of the known entry time, flight altitude, speed, penetrating and descending as the input of artificial neural networks; the time between departure and transfer point as the output of Artificial Neural Networks, to establish artificial neural network. Applying artificial neural networks and genetic algorithm to the study to simulate the result of actual flight, one can precisely estimate the flight time, thereby making it an efficient air-traffic-control instrument. It can help controllers handle different time segments of air traffic, thus upgrading the quality of air traffic control service.
48
Chen, Yuan-Wen, and 陳淵文. "Using Neural Network and Genetic Algorithm to Implement Artificial Intelligence of Starcraft." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/31382797757704133886.
Full textAbstract:
碩士國立中正大學
資訊工程研究所
102
StarCraft is a Real-Time War Strategy video game developed by Blizzard Entertainment in 1998. Real Time Strategy Games are one of the most popular game schemes in PC markets and offer a dynamic environment that involves several interacting agents. The core strategies that need to be developed in these games are unit micro management, building order, resource management, and the game main tactic. The player must reason about high-level strategy and planning while having effective tactics. Unfortunately, current games only use scripted and fixed behaviors for their artificial intelligence, and the player can easily learn the counter measures to defeat the AI. Enabling an artificial agent to deal with such a task entails breaking down the complexity of this environment. In this paper, we describe a system based on neural networks that controls what units should do in the game StarCraft. The system combined with genetic algorithm which can learn better way to play this game.
49
Lu, Chin-Lung, and 呂金龍. "Integrating genetic algorithm and neural networks to predict the manufacturing process quality." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/23235695152575215410.
Full textAbstract:
碩士國立勤益科技大學
工業工程與管理系
100
The society demands change rapidly that consumers are attention to the product quality and require better product quality. The product quality is important to obtain the consumer’s favorite. It is result continue to improve by comparing quality and appearance in industry. The factories that is able to satisfy consumers will get more orders and earn more profit. The product is unable to satisfy consumers .The consumer market will disappear. Therefore, it is important to keep and improve the product quality. The steel ball is the basic materials of bearing industry. The Production patterns of steel ball are mass production and high product quality requirements. The manufacturing process of steel balls is: cold-pressing, light-grinding, heat treatment and fine grinding. Each assignment just produces a change, and then influence the balls quality. The steel balls are one of important parts of bearing which has different size, hardness and loads are installed in different types of bearings. In order to improve the efficiency of quality control. This study used the process capability analysis that the criteria of the steel balls classification in the manufacturing process. The number of steel balls process data and quality inspection data is used to classify and construction predict model. Applying the neural network constantly modify error and the genetic algorithm escape local solution to approximation of the optimal solution. In order to construct a prediction model of the steel balls process quality. The model is rapidly response to the process quality of each batch of steel balls to satisfy customers demand and obtain orders to enable promote the enterprise competition.
50
Huang, Hao-Fan, and 黃皓汎. "Clear Air Turbulence Avoidance Strategy via Genetic Algorithm & Neural Network Methods." Thesis, 2002. http://ndltd.ncl.edu.tw/handle/54893682778818764720.
Full textAbstract:
碩士淡江大學
航空太空工程學系
90
In modern airline’s operation, clear air turbulence (CAT) remains one of the most influential factors in flight safety and flight quality consideration. In this research we use Matlab to create 3-D turbulence based on the real turbulence profiles, and prediction parameters (indices) T1, T2 and T3. The T1 factor is to define the turbulence intensity, the T2 and T3 factors are the response of aircraft in linear acceleration and three angular accelerations. Finally we use the genetic algorithm and combining the genetic algorithm (GA) and annealed neural network (ANN) methods to search the optimum escape trajectory. Results show moderate success that the computational time was shortened by 25% and with the same quality of solutions. It is hoped that the concepts and techniques implemented in this work could be used in future airborne Doppler radar research and flight simulation practice. In this work we first simulate turbulence/gust like three-dimensional wind profiles. The method is to use the Matlab tool and directly combine more then fifty trigonometric function waves. Comparing with real wind velocity profiles, the simulated wind show similar fluctuating behavior and can be used in our flight simulation. Secondly, to quantify the severity of CAT phenomenon, a set of prediction parameters（T1, T2, T3）have been proposed, T1 is three-dimensional turbulence acceleration, T2 is aircraft response in linear translation, and T3 is aircraft response in angular motion. These simulated T values show excellent agreement with real turbulence/gust T values. Thirdly, the classical rigid body, mass/mass distribution fixed flight dynamics equations are solved by standard 4th order Runge-Kutta method. To achieve an optimum flight trajectory in order to avoid the severity of CAT, two methods are employed as the steering tools, namely, the genetic algorithm and the genetic algorithm plus annealed neural network modification method. In our work the real-value GA approach is chosen due to its computation efficiency and similarity it the natural world. Our GA process is implemented as follow: both of T1+ T2+ T3, and T1+ T2+ T3+root mean squares of three Euler angles are assigned as the objective functions. And in the last, to further improve the computation efficiency of our work, the neural network method is added to our GA scheme. The model we selected is annealed neural network. It is relatively new and gives accurate data in a less timely fashion. Results show that this combination of GA and annealed neural network do improve the computation efficiency by 25%. When the CAT avoidance strategy is implemented and optimum flight trajectory achieved, it is obvious that direction attitude angles are also kept minimum. Thus represent a high degree of ride comfort and flight quality. It is hoped that the concepts proposed in this work will improve future passenger flight safety, and we no longer need to worry about clear air turbulence influence in our journey.