Literatura científica selecionada sobre o tema "Catastrophe de gradient"
Crie uma referência precisa em APA, MLA, Chicago, Harvard, e outros estilos
Consulte a lista de atuais artigos, livros, teses, anais de congressos e outras fontes científicas relevantes para o tema "Catastrophe de gradient".
Ao lado de cada fonte na lista de referências, há um botão "Adicionar à bibliografia". Clique e geraremos automaticamente a citação bibliográfica do trabalho escolhido no estilo de citação de que você precisa: APA, MLA, Harvard, Chicago, Vancouver, etc.
Você também pode baixar o texto completo da publicação científica em formato .pdf e ler o resumo do trabalho online se estiver presente nos metadados.
Artigos de revistas sobre o assunto "Catastrophe de gradient"
1
LeBrasseur, Nicole. "Catastrophe gradient drives anaphase B". Journal of Cell Biology 177, n.º 6 (18 de junho de 2007): 944a. http://dx.doi.org/10.1083/jcb.1776iti5.
Texto completo da fonte
2
Zakharov, S. V., e A. M. Il'in. "From weak discontinuity to gradient catastrophe". Sbornik: Mathematics 192, n.º 10 (31 de outubro de 2001): 1417–33. http://dx.doi.org/10.1070/sm2001v192n10abeh000599.
Texto completo da fonte
3
Konopelchenko, B. G., e G. Ortenzi. "Gradient catastrophe and flutter in vortex filament dynamics". Journal of Physics A: Mathematical and Theoretical 44, n.º 43 (7 de outubro de 2011): 432001. http://dx.doi.org/10.1088/1751-8113/44/43/432001.
Texto completo da fonte
4
Masoero, Davide, e Andrea Raimondo. "Semiclassical Limit for Generalized KdV Equations Before the Gradient Catastrophe". Letters in Mathematical Physics 103, n.º 5 (24 de janeiro de 2013): 559–83. http://dx.doi.org/10.1007/s11005-013-0605-x.
Texto completo da fonte
5
Radko, Timour. "Applicability and failure of the flux-gradient laws in double-diffusive convection". Journal of Fluid Mechanics 750 (30 de maio de 2014): 33–72. http://dx.doi.org/10.1017/jfm.2014.244.
Texto completo da fonteResumo:
AbstractDouble-diffusive flux-gradient laws are commonly used to describe the development of large-scale structures driven by salt fingers – thermohaline staircases, collective instability waves and intrusions. The flux-gradient model assumes that the vertical transport is uniquely determined by the local background temperature and salinity gradients. While flux-gradient laws adequately capture mixing characteristics on scales that greatly exceed those of primary double-diffusive instabilities, their accuracy rapidly deteriorates when the scale separation between primary and secondary instabilities is reduced. This study examines conditions for the breakdown of the flux-gradient laws using a combination of analytical arguments and direct numerical simulations. The applicability (failure) of the flux-gradient laws at large (small) scales is illustrated through the example of layering instability, which results in the spontaneous formation of thermohaline staircases from uniform temperature and salinity gradients. Our inquiry is focused on the properties of the ‘point-of-failure’ scale ($\def \xmlpi #1{}\def \mathsfbi #1{\boldsymbol {\mathsf {#1}}}\let \le =\leqslant \let \leq =\leqslant \let \ge =\geqslant \let \geq =\geqslant \def \Pr {\mathit {Pr}}\def \Fr {\mathit {Fr}}\def \Rey {\mathit {Re}}H_{pof}$) at which the vertical transport becomes significantly affected by the non-uniformity of the background stratification. It is hypothesized that$H_{pof} $can control some key characteristics of secondary double-diffusive phenomena, such as the thickness of high-gradient interfaces in thermohaline staircases. A more general parametrization of the vertical transport – the flux-gradient-aberrancy law – is proposed, which includes the selective damping of relatively short wavelengths that are inadequately represented by the flux-gradient models. The new formulation is free from the unphysical behaviour of the flux-gradient laws at small scales (e.g. the ultraviolet catastrophe) and can be readily implemented in theoretical and large-scale numerical models of double-diffusive convection.
6
Konopelchenko, B. G., e G. Ortenzi. "Quasi-Classical Approximation in Vortex Filament Dynamics. Integrable Systems, Gradient Catastrophe, and Flutter". Studies in Applied Mathematics 130, n.º 2 (5 de setembro de 2012): 167–99. http://dx.doi.org/10.1111/j.1467-9590.2012.00563.x.
Texto completo da fonte
7
Zakharov, S. V. "Asymptotic solution of a Cauchy problem in a neighbourhood of a gradient catastrophe". Sbornik: Mathematics 197, n.º 6 (30 de junho de 2006): 835–51. http://dx.doi.org/10.1070/sm2006v197n06abeh003780.
Texto completo da fonte
8
Kolesnykov, A. V., S. V. Semenova e О. О. Makovetska. "SIMULATION OF COMPOSITES STRUCTURE FORMATION PROCESSES USING THE METHODS OF THE CATASTROPHE THEORY". Modern construction and architecture, n.º 6 (25 de dezembro de 2023): 90–98. http://dx.doi.org/10.31650/2786-6696-2023-6-90-98.
Texto completo da fonteResumo:
The paper examines the possibility to model the structure formation processes in binder materials using some methods of catastrophe theory. Events at the level of coexistence and interaction of macroscopic and microscopic phenomena are analyzed. The dynamics of transformations at this level of material organization is considered using the method of structural potential, similar to the thermodynamic one, with the possibility of its empirical identification based on the processing of microscopic images. The transition to the potential functions of the theory of catastrophes is carried out by means of geometric parameterization ‒ selection of areas of the volume occupied by material phases and areas associated with interface boundaries. The model of structure formation is presented as a phenomenon in stochastic gradient systems caused by the emergence of features of the structural potential and their transformation upon changing the controlling physicochemical parameters. The possibility of adapting the apparatus of phase diagrams of three-component systems to the considered tasks and its affinity with catastrophe theory models is shown. The structural-phase diagram is constructed by analogy to the Gibbs-Roseboom method on a triangle, while the coordinates are parts of components with different geometric structures ‒ cavities, continuous material at a given scale and interfaces with the surrounding transformed material. From the potential functions of the theory of catastrophes, those that correspond to the analyzed system are reasonably selected - umbilical functions, and in the simplified version ‒ potentials of one variable. A scheme for the study of materials using methods of structural potentials has been developed, which is based on the determination of mesoscopic scales for the material being studied, on the transition to the frequency characteristics of the image, then to the structural potential, which determines the range of structures realized in the material and, in the future, to the transition from structural potentials to physical properties.
9
Aksenov, A. V., e K. P. Druzhkov. "Construction of Exact Solutions of the System of One-Dimensional Gas Dynamics Equations without Gradient Catastrophe". Известия Российской академии наук. Механика жидкости и газа, n.º 1 (1 de janeiro de 2023): 135–43. http://dx.doi.org/10.31857/s0568528122600734.
Texto completo da fonteResumo:
The system of equations that describes one-dimensional polytropic gas flows is considered. The invariants up to the second order of characteristics of the considered system of equations are classified. The method of reducing the Cauchy problems to systems of ordinary differential equations is proposed. Examples of the solutions without gradient catastrophe are constructed using invariants of characteristics supplementary to the Riemann invariants.
10
Beisenbi, М. А., e Zh O. Basheyeva. "Solving output control problems using Lyapunov gradient-velocity vector function". International Journal of Electrical and Computer Engineering (IJECE) 9, n.º 4 (1 de agosto de 2019): 2874. http://dx.doi.org/10.11591/ijece.v9i4.pp2874-2879.
Texto completo da fonteResumo:
This paper describes a controller and observer parameter definition approach in one input-one output (closed-loop) control systems using Lyapunov gradient-velocity vector function. Construction of the vector function is based on the gradient nature of the control systems and the parity of the vector functions with the potential function from the theory of catastrophe. Investigation of the closed-loop control system’s stability and solution of the problem of controller (determining the coefficient of magnitude matrix) and observer (calculation of the matrix elements of the observing equipment) synthesis is based on the direct methods of Lyapunov. The approach allows to select parameters based on the requested characteristics of the system.
Teses / dissertações sobre o assunto "Catastrophe de gradient"
1
Demiquel, Antoine. "Control of nonlinear modulated waves in flexible mechanical metamaterials". Electronic Thesis or Diss., Le Mans, 2024. https://cyberdoc-int.univ-lemans.fr/Theses/2024/2024LEMA1015.pdf.
Texto completo da fonteResumo:
Ce travail est consacré à l'étude des ondes modulées se propageant le long de métamatériaux mécaniques flexibles nonlinéaires (FlexMM). Ces structures sont des matériaux architecturés constitués d'éléments souples très déformables connectés à des éléments plus rigides. Leur capacité à subir de grandes déformations locales favorise l'apparition de phénomènes d'ondes non linéaires. En utilisant une approche par éléments discrets, nous formulons des équations discrètes non linéaires qui décrivent les déplacements longitudinaux et rotationnels de chaque cellule unitaire et leur couplage mutuel. Une analyse multi-échelles est employée afin d'obtenir une équation de Schrödinger non linéaire (NLS) effective décrivant les ondes modulées pour le degré de liberté rotationnel du FlexMM. En nous appuyant sur l'équation NLS, nous identifions divers types de phénomènes d'ondes non linéaires dans le FlexMM. En particulier, nous avons observé que des ondes planes faiblement non linéaires peuvent être modulationellement stables ou instables en fonction des paramètres du système et de l'excitation utilisée. De plus, nous avons trouvé que les FlexMMs supportent des solitons-enveloppe vectoriels où le degré de liberté rotationnel des unités peut prendre la forme de solitons dits "bright" ou "dark" et, en raison du couplage, le degré de liberté de déplacement longitudinal présente un comportement de type "kink". Enfin, nous abordons le phénomène de "catastrophe de gradient", qui prédit l'émergence de structures similaires aux solitons de Peregrine dans la limite semi-classique de l'équation NLS, dans la structure FlexMM. Grâce à nos prédictions analytiques et à l'utilisation de simulations numériques, nous pouvons déterminer les conditions requises et les valeurs des paramètres physiques pour observer ces phénomènes dans les FlexMMsThis work is dedicated to the investigation of modulated waves propagating along nonlinear flexible mechanical metamaterials (FlexMM). These structures are architected materials consisting of highly deformable soft elements connected to stiffer ones. Their capacity to undergo large local deformations promotes the occurrence of nonlinear wave phenomena. Using a lump element approach, we formulate nonlinear discrete equations that describe the longitudinal land rotational displacements of each unit cell and their mutual coupling. A multiple scales analysis is employed in order to derive an effective nonlinear Schrödinger (NLS) equation describing envelope waves for the rotational degree of freedom of FlexMM. Leveraging on the NLS equation we identify various type of nonlinear waves phenomena in FlexMM. In particular we observed that weakly nonlinear plane waves can be modulationally stable or unstable depending of the system and excitation parameters. Moreover we have found that the FlexMMs support envelope vector solitons where the units rotational degree of freedom might take the form of bright or dark soliton and due to coupling, the longitudinal displacement degree of freedom has a kink-like behavior. Finally, we address the phenomenon of "gradient catastrophe", which predicts the emergence of Peregrine soliton-like structures in the semiclassical limit of the NLS equation, in FlexMM. Through our analytical predictions and by using numerical simulations, we can determine the required conditions and the values of the physical parameters in order to observe these phenomena in FlexMMs
Capítulos de livros sobre o assunto "Catastrophe de gradient"
1
Knobel, Roger. "Gradient catastrophes and breaking times". In The Student Mathematical Library, 137–44. Providence, Rhode Island: American Mathematical Society, 1999. http://dx.doi.org/10.1090/stml/003/18.
Texto completo da fonte
2
Giannini, Federico, Giacomo Ziffer e Emanuele Della Valle. "cPNN: Continuous Progressive Neural Networks for Evolving Streaming Time Series". In Advances in Knowledge Discovery and Data Mining, 328–40. Cham: Springer Nature Switzerland, 2023. http://dx.doi.org/10.1007/978-3-031-33383-5_26.
Texto completo da fonteResumo:
AbstractDealing with an unbounded data stream involves overcoming the assumption that data is identically distributed and independent. A data stream can, in fact, exhibit temporal dependencies (i.e., be a time series), and data can change distribution over time (concept drift). The two problems are deeply discussed, and existing solutions address them separately: a joint solution is absent. In addition, learning multiple concepts implies remembering the past (a.k.a. avoiding catastrophic forgetting in Neural Networks’ terminology). This work proposes Continuous Progressive Neural Networks (cPNN), a solution that tames concept drifts, handles temporal dependencies, and bypasses catastrophic forgetting. cPNN is a continuous version of Progressive Neural Networks, a methodology for remembering old concepts and transferring past knowledge to fit the new concepts quickly. We base our method on Recurrent Neural Networks and exploit the Stochastic Gradient Descent applied to data streams with temporal dependencies. Results of an ablation study show a quick adaptation of cPNN to new concepts and robustness to drifts.
3
Sun, Jian, Stefan Löhnert e Tengfei Lyu. "Prediction of Fatigue Lifetime Using a Wavelet Transformation Induced Multi-time Scaling Method and Xfem". In Regeneration of Complex Capital Goods, 307–26. Cham: Springer International Publishing, 2024. http://dx.doi.org/10.1007/978-3-031-51395-4_16.
Texto completo da fonteResumo:
AbstractFatigue lifetime prediction due to dynamic crack growth is a significant issue for design and manufacture of engineering components. Damage accumulation in a material on the micro-scale is a main physical mechanism governing crack initiation. For high-cycle fatigue, the number of loading cycles leading to catastrophic fatigue failure can be of the order of millions. The simulation of such processes would be extremely expensive and time-consuming using conventional single time scale methods. In order to overcome this challenging requirement, a wavelet-transformation based multi-time scaling method is successfully adopted in this research to accelerate the prediction of accumulated damage for a large number loading cycles. In this work, the WATMUS technique is coupled with gradient- enhanced damage and the extended finite element method to simulate dynamic crack propagation for a turbine blade.
4
Messaoudi, S., e A. Al Shehri. "Gradient Catastrophe in Heat Propagation with second Sound". In Lecture Notes in Pure and Applied Mathematics, 273–81. CRC Press, 2005. http://dx.doi.org/10.1201/9781420026511.pt3.
Texto completo da fonte
5
Camassa, Roberto. "Global gradient catastrophe in a shallow water model: evolution unfolding by stretched coordinates". In Nonlinear Systems and Their Remarkable Mathematical Structures, 445–58. Chapman and Hall/CRC, 2019. http://dx.doi.org/10.1201/9780429263743-15.
Texto completo da fonte
6
Bader, Richard F. W. "Mathematical Models Of Structural Change". In Atoms in Molecules, 110–29. Oxford University PressOxford, 1990. http://dx.doi.org/10.1093/oso/9780198551683.003.0004.
Texto completo da fonteResumo:
Abstract The definition of structure obtained through the application of the equivalence relation to the gradient vector fields of the charge density leads to the partitioning of nuclear configuration space into a finite number of structural regions. The boundaries of these regions denote the configurations of the unstable structures. This information enables one to construct a system’s structure diagram, a diagram which details all of its stable structures and all of the unstable structures that are transitional in form between those of neighbouring structural regimes. The collection of the structurally unstable points in nuclear configuration space R, is called the catastrophe set C. As a corollary of the theorem of structural stability stated by Palis and Smale (1970) one finds that only two types of catastrophe points are possible a configuration X∈ R is structurally unstable if and only if its associated charge density ρ (r; X) exhibits (a) at least one degenerate critical point, also known as a singularity in the charge density or (b) a non-transversal intersection of the stable and unstable manifolds of a pair of its critical points. A configuration satisfying either (a) or (b) is called a bifurcation or conflict point, respectively.
7
Hallam, Tony. "Oxygen deficiency in the oceans". In Catastrophes and Lesser Calamities. Oxford University Press, 2004. http://dx.doi.org/10.1093/oso/9780198524977.003.0009.
Texto completo da fonteResumo:
We are all very much aware that oxygen deprivation leads quickly to death, and this is true not just of our own species but of virtually the whole organic world. There are indeed very few exceptions, such as the anaerobic bacteria that derive their energy from reducing sulphates to sulphides, which flourish in the absence of free oxygen. (As these organisms do not leave a fossil record they provide no clues for the geological detective.) Today the atmosphere never lacks oxygen, except in artificially enclosed conditions, but oxygen deficiency can be lethal in certain marine environments and thus must be explored as a possible factor in causing mass extinctions. Mixing with atmospheric winds ensures that the surface waters of the ocean, down to the greatest depth attained by storm waves, always contain plenty of oxygen. Most of the oceans and marginal seas today contain oxygen throughout their depth, but in certain circumstances an oxygen deficiency can occur in the lower parts of the ocean. In parts of some tropical oceans, for instance, the oxygen content decreases with depth until near the ocean bottom, where under the influence of currents driven by cold water from around Antarctica, the oxygen content increases again. This gives rise to a zone in the ocean known as the oxygen minimum zone. The rapid deep ocean circulation is today driven ultimately by the presence of polar ice on Antarctica, which is the main cause of the strong sea-water temperature gradient from the tropics to the poles. For long periods in the Earth’s history substantial polar ice caps were lacking, and many geologists believe that during those periods latitudinal ocean currents were more sluggish. The deep ocean must then have been largely deficient in oxygen, if not completely lacking in oxygen (anoxic). (Sea water with a content of one or more millilitres of oxygen per litre of water is called oxic; 0.1 ml or less is anoxic; and for any intermediate value the water is dysoxic.) Certain parts of the sea bed where the overlying water is deficient in oxygen are enriched in organic matter derived principally from the plankton.
8
"Community Ecology of Stream Fishes: Concepts, Approaches, and Techniques". In Community Ecology of Stream Fishes: Concepts, Approaches, and Techniques, editado por Gary D. Grossman e John L. Sabo. American Fisheries Society, 2010. http://dx.doi.org/10.47886/9781934874141.ch20.
Texto completo da fonteResumo:
<em>Abstract.—</em>Stochastic dynamics are central to theory, data analysis, and understanding in the fields of hydrology and population ecology. More importantly, hydrologic variability has been identified as a key process affecting biodiversity and coexistence in stream fish assemblages. Until recently, however, we have lacked tools by which hydrologic variability can be directly linked to measures of community stability. Herein, we show how a modification of Fourier analysis of daily average discharge data can be used to quantify aspects of hydrologic variability for three reference streams and then linked to measures of fish assemblage stability in Coweeta Creek, North Carolina; Sagehen Creek, California; and Otter Creek, Indiana) via multivariate autoregressive (MAR) models. Specifically, we define the magnitude of catastrophic variability as the standard deviation of <em>residual </em>flows referenced to a long-term annual trend, and individual catastrophic events as flows greater than (floods) or less than (droughts) two times this magnitude (i.e., 2 𝛔). We then directly link the magnitude of annual <em>residual </em>flows with MAR models that quantify the relationship between flows and the stability of fish assemblages from the same or nearby streams. Our results confirm that these streams represent a gradient in the stability properties of fish assemblages; Sagehen Creek is the most stable, whereas Otter Creek is the least stable. The timing of catastrophic high and low flows is most predictable in Sagehen Creek and least predictable in Big Raccon Creek (reference stream for Otter Creek), whereas the magnitude and frequency of <em>catastrophic </em>events varied in a manner less consistent with the gradient in fish community stability. Nevertheless, the stability of fish communities covaried significantly with both residual flow magnitudes (high- and low-flow events). Although this technique is not without limitations (e.g., it is most relevant to resident species), it appears to be a promising new tool for linking hydrologic variability directly to fish assemblage stability and, more broadly, for quantifying links between flow regulation and the viability of native aquatic faunas.
9
Newnham, Robert E. "Tensors and physical properties". In Properties of Materials. Oxford University Press, 2004. http://dx.doi.org/10.1093/oso/9780198520757.003.0007.
Texto completo da fonteResumo:
In this chapter we introduce the tensor description of physical properties along with Neumann’s Principle relating symmetry to physical properties. As pointed out in the introduction, many different types of anisotropic properties are described in this book, but all have one thing in common: a physical property is a relationship between two measured quantities. Four examples are illustrated in Fig. 5.1. Elasticity is one of the standard equilibrium properties treated in crystal physics courses. The elastic compliance coefficients relate mechanical strain, the dependent variable, to mechanical stress, the independent variable. For small stresses and strains, the relationship is linear, but higher order elastic constants are needed to describe the departures from Hooke’s Law. Thermal conductivity is typical of the many transport properties in which a gradient leads to flow. Here the dependent variable is heat flow and the independent variable is a temperature gradient. Again the relationship is linear for small temperature gradients. Hysteretic materials such as ferromagnetic iron exhibit more complex physical properties involving domain wall motion. In this case magnetization is the dependent variable responsive to an applied magnetic field. The resulting magnetic susceptibility depends on the past history of the material. If the sample is initially unmagnetized, the magnetization will often involve only reversible domain wall motion for small magnetic fields. In this case the susceptibility is anhysteretic, but for large fields the wall motion is only partly reversible leading to hysteresis. The fourth class of properties leads to permanent changes involving irreversible processes. Under very high electric fields, dielectric materials undergo an electric breakdown process with catastrophic current flow. Under small fields Ohm’s Law governs the relationship between current density and electric field with a well-defined resistivity, but high fields lead to chemical, thermal, and mechanical changes that permanently alter the sample. Irreversible processes are sometimes anisotropic but they will not be discussed in this book. Measured quantities such as stress and strain can be represented by tensors, and so can physical properties like elastic compliance that relate these measurements. This is why tensors are so useful in describing anisotropy.
10
Nagy, Laszlo, e Georg Grabherr. "The alpine environment: energy and climate". In The Biology of Alpine Habitats, 52–75. Oxford University PressOxford, 2009. http://dx.doi.org/10.1093/oso/9780198567035.003.0004.
Texto completo da fonteResumo:
Abstract Radiation and thermal energy in alpine environments is as important for life conditions as in any other life zone, and in addition, mountain environments are characterized by a high kinetic potential, or relief energy. The complex alpine landscape results in a distribution pattern in direct incident solar irradiance that creates a broad range of thermal conditions within the alpine zone. Relief energy manifests in high-energy al-pine mass movement events. These either develop over a long time period (non-catastrophic events) and form permanent landscape features, such as scree, rock glacier, or block fields, or are caused by a sudden descent of material that radically alters landscapes and ecological conditions in the short-term (catastrophic events, such as rock slide, debris slide, mudflow). The historic or contemporaneous movements of high-energy material, including water and ice have traversed the altitude gradients of mountains and contributed to the creation of azonal habitats (see Chapter 6). Topography related phenomena and climate determine largely the alpine life zone and its habitats. The discussion of habitats and their underlying ecological conditions (energy and climate, hydrology, soils, and vegetation) follows a model framework that builds up from mechanistic energy budgets to vegetation feedbacks and considers habitat-forming factors alone, and in combination for habitat characteristics. The logical framework follows that of layers and overlays of layers applied in geographical information systems.
Trabalhos de conferências sobre o assunto "Catastrophe de gradient"
1
Elyas, Mohamed, Sherif Aly, Uche Achinanya, Sergey Prosvirkin, Shayma AlSaffar, Muthafar Mohammad, Mohammad Siddiqui e Awrad Fahad. "Well Integrity Catastrophe Avoided Through Advanced Well Integrity and Reservoir Monitoring Analysis, a Case Study". In Abu Dhabi International Petroleum Exhibition & Conference. SPE, 2021. http://dx.doi.org/10.2118/208055-ms.
Texto completo da fonteResumo:
Abstract Well integrity is one of the main challenges that are facing operators, finding the source of the well problem and isolating it before a catastrophic event occurs. This study demonstrates the power of integrating different reservoir monitoring and well integrity logs to evaluate well integrity, identify the underlying cause of the potential failure, and providing a potential corrective solution. Recently, some Injector/producer wells reported migration of injection fluids/gas into shallower sections, charging these formations and increasing the risk of compromised well integrity. Characterization of the well issues required integration of multi-detector pulsed-neutron, well integrity (multi finger caliper, multi-barrier corrosion, cement evaluation, and casing thickness measurements), high precision temperature logs and spectral noise logs. After data integration, detailed analysis was performed to specifically find the unique issues in each well and assess possible corrective actions. The integrated well integrity logs clearly showed different 9.625-inch and 13.375-inch casings leak points. The reservoir monitoring logs showed lateral and vertical gas and water movements across Wara, Tayarat, Rus, and Radhuma formations. Cement evaluation loges showed no primary cement behind the first barrier casing which was the root cause of the problem. Therefore, the proposed solution, was a cement squeeze. Post squeeze, re-logging occurred, validating zonal isolation and a return of a standard geothermal gradient across the Tayarat formation. Most importantly, the cement evaluation identified good bond from the squeeze point clear to surface, isolating all formations. All these wells were returned to service (injector/producer), daily annular pressure monitoring confirmed that no further pressure build up was seen. Kuwait Oil Company managed to avoid a catastrophic well integrity event on these wells and utilized the approach presented to take the proper corrective actions, and validate that the action taken resolved the initial well integrity issues. Consequently, the wells were returned to service, and the company avoided a costly high probability blowout.
2
Luo, Kangyang, Xiang Li, Yunshi Lan e Ming Gao. "GradMA: A Gradient-Memory-based Accelerated Federated Learning with Alleviated Catastrophic Forgetting". In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2023. http://dx.doi.org/10.1109/cvpr52729.2023.00361.
Texto completo da fonte
3
Yin, Hao, He Xu, Yuhan Zhao e Feng Sun. "Fault Diagnosis of Control Valve Based on Fusion of Deep Learning and Elastic Weight Consolidation". In BATH/ASME 2022 Symposium on Fluid Power and Motion Control. American Society of Mechanical Engineers, 2022. http://dx.doi.org/10.1115/fpmc2022-89359.
Texto completo da fonteResumo:
Abstract Deep neural network learning is a commonly used method for fault diagnosis of the control valve. However, the catastrophic forgetting problem of deep learning in multi-task affects the fault diagnosis accuracy. Moreover, the traditional training model can be improved by using parameter constraint control or adding a few parameters, but it has many limitations. Therefore, this paper proposed a fusion of elastic weight consolidation algorithm and residual shrinkage network method, sharing common feature layers. According to the weight of the same or similar parameters of the previous task, the correct solution of the current task could be obtained, and the forgetting degree of the previous task could be reduced. It improved the generalization ability of the training model. The control valve data were collected and compared with the stochastic gradient descent algorithm in different valve openings. The results indicate that this method has a high accuracy for the condition identification of the control valve. This method can effectively alleviate the problem of the catastrophic forgetting of deep learning in multi-task identification of control valve.
4
Rios, Amanda, e Laurent Itti. "Closed-Loop Memory GAN for Continual Learning". In Twenty-Eighth International Joint Conference on Artificial Intelligence {IJCAI-19}. California: International Joint Conferences on Artificial Intelligence Organization, 2019. http://dx.doi.org/10.24963/ijcai.2019/462.
Texto completo da fonteResumo:
Sequential learning of tasks using gradient descent leads to an unremitting decline in the accuracy of tasks for which training data is no longer available, termed catastrophic forgetting. Generative models have been explored as a means to approximate the distribution of old tasks and bypass storage of real data. Here we propose a cumulative closed-loop memory replay GAN (CloGAN) provided with external regularization by a small memory unit selected for maximum sample diversity. We evaluate incremental class learning using a notoriously hard paradigm, single-headed learning, in which each task is a disjoint subset of classes in the overall dataset, and performance is evaluated on all previous classes. First, we show that when constructing a dynamic memory unit to preserve sample heterogeneity, model performance asymptotically approaches training on the full dataset. We then show that using a stochastic generator to continuously output fresh new images during training increases performance significantly further meanwhile generating quality images. We compare our approach to several baselines including fine-tuning by gradient descent (FGD), Elastic Weight Consolidation (EWC), Deep Generative Replay (DGR) and Memory Replay GAN (MeRGAN). Our method has very low long-term memory cost, the memory unit, as well as negligible intermediate memory storage.
5
Patil, Sangram, Aum Patil, Vishwadeep Handikherkar, Sumit Desai, Vikas M. Phalle e Faruk S. Kazi. "Remaining Useful Life (RUL) Prediction of Rolling Element Bearing Using Random Forest and Gradient Boosting Technique". In ASME 2018 International Mechanical Engineering Congress and Exposition. American Society of Mechanical Engineers, 2018. http://dx.doi.org/10.1115/imece2018-87623.
Texto completo da fonteResumo:
Rolling element bearings are very important and highly utilized in many industries. Their catastrophic failure due to fluctuating working conditions leads to unscheduled breakdown and increases accidental economical losses. Thus these issues have triggered a need for reliable and automatic prognostics methodology which will prevent a potentially expensive maintenance program. Accordingly, Remaining Useful Life (RUL) prediction based on artificial intelligence is an attractive methodology for several researchers. In this study, data-driven condition monitoring approach is implemented for predicting RUL of bearing under a certain load and speed. The approach demonstrates the use of ensemble regression techniques like Random Forest and Gradient Boosting for prediction of RUL with time-domain features which are extracted from given vibration signals. The extracted features are ranked using Decision Tree (DT) based ranking technique and training and testing feature vectors are produced and fed as an input to ensemble technique. Hyper-parameters are tuned for these models by using exhaustive parameter search and performance of these models is further verified by plotting respective learning curves. For the present work FEMTO bearing data-set provided by IEEE PHM Data Challenge 2012 is used. Weibull Hazard Rate Function for each bearing from learning data set is used to find target values i.e. projected RUL of the bearings. Results of proposed models are compared with well-established data-driven approaches from literature and are found to be better than all the models applied on this data-set, thereby demonstrating the reliability of the proposed model.
6
Bui, Huy, Nelly De Nicolias, Rebecca Nye e John Estrada. "Machine Learning Applications to Improve Pore Pressure Prediction in Hazardous Drilling Environments". In Offshore Technology Conference. OTC, 2022. http://dx.doi.org/10.4043/31781-ms.
Texto completo da fonteResumo:
Abstract Pore pressure analysis can be imperative while drilling, especially in certain offshore environments where abnormal pore pressures can cause serious problems such as fluid influx, kicks, and blowouts. In order to avoid such events, the prediction in real-time, or preferably ahead of time, is required. These events can have a catastrophic impact on operations and the safety of the entire platform. Real-time wellbore logs can be used to assist in the prediction of pore pressure, however, due to the high cost of downhole data acquisition and risks associated with a tool getting stuck or lost in hole, comprehensive well logs are not always available. In the absence of a measured sonic log, a predicted acoustic log can be used for input into the pore pressure prediction avoiding the risk and cost of downhole wellbore logs, however, accuracy is extremely important. A Gradient Boosting model is trained and validated on 3 different sets of features to predict acoustic wellbore logs, followed by a physics-based model for pore pressure prediction. The physics-based model is built using Eaton, density extrapolation, and other optimization methods to ensure speed and accuracy. Lastly, the model takes the predicted acoustic data to derive the pore pressure gradient, calibrated by other drilling parameters. The Gradient Boosting model reduces any impact from the lack of data availability, significantly reduces the root mean square error (RMSE) and increases the overall accuracy. The results are then calibrated to drilling events to ensure the predictions are within the range of actual recorded event data. The result from a recent case study shows that pore pressure prediction using predicted acoustic logs correlates closely with recorded drilling events. The client successfully estimated pore pressures using predicted acoustic logs, reducing the for the need to acquire costly logging data downhole. The Gradient Boosting model provides a solution to predict acoustic logs and pore pressure that is highly accurate in real-time. The drilling event calibration method then helps to avoid physical factors that cannot be captured by the model, increasing the overall reliability of the workflow. The method allows for pore pressure analysis to be carried out accurately, regardless of the downhole logs acquired.
7
Bui, Huy, Nelly De Nicolias, Rebecca Nye e John Estrada. "Machine Learning Applications to Improve Pore Pressure Prediction in Hazardous Drilling Environments". In Offshore Technology Conference. OTC, 2022. http://dx.doi.org/10.4043/31781-ms.
Texto completo da fonteResumo:
Abstract Pore pressure analysis can be imperative while drilling, especially in certain offshore environments where abnormal pore pressures can cause serious problems such as fluid influx, kicks, and blowouts. In order to avoid such events, the prediction in real-time, or preferably ahead of time, is required. These events can have a catastrophic impact on operations and the safety of the entire platform. Real-time wellbore logs can be used to assist in the prediction of pore pressure, however, due to the high cost of downhole data acquisition and risks associated with a tool getting stuck or lost in hole, comprehensive well logs are not always available. In the absence of a measured sonic log, a predicted acoustic log can be used for input into the pore pressure prediction avoiding the risk and cost of downhole wellbore logs, however, accuracy is extremely important. A Gradient Boosting model is trained and validated on 3 different sets of features to predict acoustic wellbore logs, followed by a physics-based model for pore pressure prediction. The physics-based model is built using Eaton, density extrapolation, and other optimization methods to ensure speed and accuracy. Lastly, the model takes the predicted acoustic data to derive the pore pressure gradient, calibrated by other drilling parameters. The Gradient Boosting model reduces any impact from the lack of data availability, significantly reduces the root mean square error (RMSE) and increases the overall accuracy. The results are then calibrated to drilling events to ensure the predictions are within the range of actual recorded event data. The result from a recent case study shows that pore pressure prediction using predicted acoustic logs correlates closely with recorded drilling events. The client successfully estimated pore pressures using predicted acoustic logs, reducing the for the need to acquire costly logging data downhole. The Gradient Boosting model provides a solution to predict acoustic logs and pore pressure that is highly accurate in real-time. The drilling event calibration method then helps to avoid physical factors that cannot be captured by the model, increasing the overall reliability of the workflow. The method allows for pore pressure analysis to be carried out accurately, regardless of the downhole logs acquired.
8
Wang, Shuai, Yibing Zhan, Yong Luo, Han Hu, Wei Yu, Yonggang Wen e Dacheng Tao. "Joint Input and Output Coordination for Class-Incremental Learning". In Thirty-Third International Joint Conference on Artificial Intelligence {IJCAI-24}. California: International Joint Conferences on Artificial Intelligence Organization, 2024. http://dx.doi.org/10.24963/ijcai.2024/565.
Texto completo da fonteResumo:
Incremental learning is nontrivial due to severe catastrophic forgetting. Although storing a small amount of data on old tasks during incremental learning is a feasible solution, current strategies still do not 1) adequately address the class bias problem, and 2) alleviate the mutual interference between new and old tasks, and 3) consider the problem of class bias within tasks. In light of the above issues, we analyze the cause of class bias in incremental learning, as well as the drawbacks of existing approaches, and propose a joint input and output coordination (JIOC) mechanism to address these issues. This mechanism assigns different weights to different categories of data according to the gradient of the output score, and uses knowledge distillation (KD) to reduce the mutual interference between the outputs of old and new tasks. The proposed mechanism is general and flexible, and can be incorporated into different incremental learning approaches that use memory storage. Extensive experiments show that our mechanism can significantly improve their performance.
9
Li, Siyi, Tianbo Liu, Chi Zhang, Dit-Yan Yeung e Shaojie Shen. "Learning Unmanned Aerial Vehicle Control for Autonomous Target Following". In Twenty-Seventh International Joint Conference on Artificial Intelligence {IJCAI-18}. California: International Joint Conferences on Artificial Intelligence Organization, 2018. http://dx.doi.org/10.24963/ijcai.2018/685.
Texto completo da fonteResumo:
While deep reinforcement learning (RL) methods have achieved unprecedented successes in a range of challenging problems, their applicability has been mainly limited to simulation or game domains due to the high sample complexity of the trial-and-error learning process. However, real-world robotic applications often need a data-efficient learning process with safety-critical constraints. In this paper, we consider the challenging problem of learning unmanned aerial vehicle (UAV) control for tracking a moving target. To acquire a strategy that combines perception and control, we represent the policy by a convolutional neural network. We develop a hierarchical approach that combines a model-free policy gradient method with a conventional feedback proportional-integral-derivative (PID) controller to enable stable learning without catastrophic failure. The neural network is trained by a combination of supervised learning from raw images and reinforcement learning from games of self-play. We show that the proposed approach can learn a target following policy in a simulator efficiently and the learned behavior can be successfully transferred to the DJI quadrotor platform for real-world UAV control.
10
Szolwinski, M. P., G. Harish e T. N. Farris. "The Development and Validation of Design-Oriented Metrics for Fretting Fatigue in Titanium Engine Components". In ASME 1998 International Mechanical Engineering Congress and Exposition. American Society of Mechanical Engineers, 1998. http://dx.doi.org/10.1115/imece1998-1136.
Texto completo da fonteResumo:
Abstract The area of contact between the blade root and disk in high-performance turbomachinery has been identified as a critical area for the nucleation of fatigue damage leading to premature and often catastrophic componential failures. The combination of high cycle aerodynamic and low cycle inertial loads lead to distributions of both normal and shear tractions at the blade/disk interface, leading to an aggressive tribological damage mechanism known as fretting. The interaction of small-scale, oscillatory relative displacements or slip at the contact surfaces and the sharp gradients of near-surface contact stresses induced by fretting contact is responsible for the nucleation of near-surface fatigue damage in these components. The paper will summarize efforts targeted at the development and validation of design-oriented tools for assessing quantitatively the threat of fretting fatigue to the integrity and subsequent performance of jet engine components.