Showing papers on "Hybrid neural network published in 2011"
••
TL;DR: In this article, a wind power forecasting strategy composed of a feature selection component and a forecasting engine is proposed, which applies an irrelevancy filter and a redundancy filter to the set of candidate inputs.
Abstract: Following the growing share of wind energy in electric power systems, several wind power forecasting techniques have been reported in the literature in recent years. In this paper, a wind power forecasting strategy composed of a feature selection component and a forecasting engine is proposed. The feature selection component applies an irrelevancy filter and a redundancy filter to the set of candidate inputs. The forecasting engine includes a new enhanced particle swarm optimization component and a hybrid neural network. The proposed wind power forecasting strategy is applied to real-life data from wind power producers in Alberta, Canada and Oklahoma, U.S. The presented numerical results demonstrate the efficiency of the proposed strategy, compared to some other existing wind power forecasting methods.
202 citations
••
TL;DR: Two hybrid neural networks derived from fuzzy neural networks (FNN) using the fuzzified wavelet features as the inputs to FNN and fuzzy neural network (FNCI) employing the Choquet integral as the outputs of FNN are presented.
Abstract: This paper presents two hybrid neural networks derived from fuzzy neural networks (FNN): wavelet fuzzy neural network (WFNN) using the fuzzified wavelet features as the inputs to FNN and fuzzy neural network (FNCI) employing the Choquet integral as the outputs of FNN. The learning through FNCI is simplified by the use of q-measure and the speed of convergence of the parameters is increased by reinforced learning. The underlying fuzzy models of these hybrid networks are a modified form of fuzzy rules of Takagi-Sugeno model. The number of fuzzy rules is found from a fuzzy curve corresponding to each input-output by counting the total number of peaks and troughs in the curve. The models can forecast hourly load with a lead time of 1 h as they deal with short-term load forecasting. The results of the two hybrid networks using Indian utility data are compared with ANFIS and other conventional methods. The performance of the proposed WFNN is found superior to all the other compared methods.
118 citations
••
TL;DR: The proposed multi-agent signal control was found to produce a significant improvement in the traffic conditions of the road network reducing the total travel time experienced by vehicles simulated under dual and multiple peak traffic scenarios.
72 citations
••
24 Jul 2011TL;DR: A neural network (NN) model is implemented and a hierarchical model is suggested for enhanced estimation of the classification efficiency, if that data was classified using support vector machines (SVM) and an encoding technique is proposed that can identify illegal consumers with better efficiency and faster classification of data.
Abstract: Total losses in transmission and distribution (T&D) of electrical energy including nontechnical losses (NTL) are huge and are affecting the good interest of utility company and its customers. In this context, importance of customer load profile evaluation for detection of illegal consumers is explained in this paper. Classification of the customers based on load profile evaluation using SVMLIB requires us to choose training function and related parameters. Selecting these parameters would consume a lot of time and is not suggestible evaluation of real time electricity consumption patterns, as, the suspicious profiles are to be predicted instantly. In light of this issue, this paper implements a neural network (NN) model and suggests a hierarchical model for enhanced estimation of the classification efficiency, if that data was classified using support vector machines (SVM). In addition, this paper proposes an encoding technique that can identify illegal consumers with better efficiency and faster classification of data.
71 citations
•
21 Jan 2011-World Academy of Science, Engineering and Technology, International Journal of Mechanical, Aerospace, Industrial, Mechatronic and Manufacturing Engineering
TL;DR: Results show that presented method can introduce a closer form to the analytic solution than other numerical methods and can be easily extended to solve a wide range of problems.
Abstract: This study presents a hybrid neural network and Gravitational Search Algorithm (HNGSA) method to solve well known Wessinger's equation. To aim this purpose, gravitational search algorithm (GSA) technique is applied to train a multi-layer perceptron neural network, which is used as approximation solution of the Wessinger's equation. A trial solution of the differential equation is written as sum of two parts. The first part satisfies the initial/ boundary conditions and does not contain any adjustable parameters and the second part which is constructed so as not to affect the initial/boundary conditions. The second part involves adjustable parameters (the weights and biases) for a multi-layer perceptron neural network. In order to demonstrate the presented method, the obtained results of the proposed method are compared with some known numerical methods. The given results show that presented method can introduce a closer form to the analytic solution than other numerical methods. Present method can be easily extended to solve a wide range of problems. Keywords—Neural Networks; Gravitational Search Algorithm (GSR); Wessinger's Equation.
53 citations
••
01 Jan 2011TL;DR: It has been analyzed that the feed forward neural network by two Evolutionary algorithms makes better generalization accuracy in character recognition problems and can solve challenging problem most reliably and efficiently.
Abstract: This paper describes the performance evaluation for the feed forward neural network with three different soft computing techniques to recognition of hand written English alphabets. Evolutionary algorithms for the hybrid neural network are showing the numerous potential in the field of pattern recognition. We have taken five trials and two networks of each of the algorithm: back propagation, Evolutionary algorithm, and Hybrid Evolutionary algorithm respectively. These algorithms have been taken the definite lead on the conventional approaches of neural network for pattern recognition. It has been analyzed that the feed forward neural network by two Evolutionary algorithms makes better generalization accuracy in character recognition problems. The problem of not convergence the weight in conventional backpropagation has also eliminated by using the soft computing techniques. It has been observed that, there are more than one converge weight matrix in character recognition for every training set. The results of the experiments show that the hybrid evolutionary algorithm can solve challenging problem most reliably and efficiently. These algorithms have also been compared on the basis of time and space complexity for the training set.
47 citations
••
01 Jan 2011TL;DR: This study incorporates desirability functions into a hybrid neural network/genetic algorithm approach to optimize the parameter design of dynamic multiresponse with continuous values of parameters and reveals that the approach has higher performance than the traditional experimental design.
Abstract: Engineers have widely applied the Taguchi method, a traditional approach for robust experimental design, to a variety of quality engineering problems for enhancing system robustness. However, the Taguchi method is unable to deal with dynamic multiresponse owing to increasing complexity of the product or design process. Although several alternative approaches have been presented to resolve this problem, they cannot effectively treat situations in which the control factors have continuous values. This study incorporates desirability functions into a hybrid neural network/genetic algorithm approach to optimize the parameter design of dynamic multiresponse with continuous values of parameters. The objective is to find the optimal combination of control factors to simultaneously maximize robustness of each response. The proposed approach is based on three stages which (1) use neural networks for constructing a response function model of a dynamic multiresponse system, (2) use exponential desirability functions for evaluating overall performance of a specific factor combination, and (3) use a genetic algorithm to optimize parameter design. Effectiveness of the proposed approach is illustrated with a simulated example. Analysis results reveal that the approach has higher performance than the traditional experimental design.
42 citations
••
TL;DR: An evolutionary fuzzy hybrid neural network (EFHNN) is developed to enhance the effectiveness of assessing subcontractor performance in the construction industry and shows that the proposed EFHNN may be deployed effectively to achieve optimal mapping of input factors and subcontractors performance output.
30 citations
••
TL;DR: Results of real-time rainfall-inundation forecasting help the emergency manager set operational responses, which are beneficial for flood warning preparations, and indicate that the RiHNNs with fewer weights can have about the same performance as a feed-forward neural network.
Abstract: . This study attempts to achieve real-time rainfall-inundation forecasting in lowland regions, based on a synthetic potential inundation database. With the principal component analysis and a feed-forward neural network, a rainfall-inundation hybrid neural network (RiHNN) is proposed to forecast 1-h-ahead inundation depth as hydrographs at specific representative locations using spatial rainfall intensities and accumulations. A systematic procedure is presented to construct the RiHNN, which combines the merits of detailed hydraulic modeling in flood-prone lowlands via a two-dimensional overland-flow model and time-saving calculation in a real-time rainfall-inundation forecasting via ANN model. Analytical results from the RiHNNs with various principal components indicate that the RiHNNs with fewer weights can have about the same performance as a feed-forward neural network. The RiHNNs evaluated through four types of real/synthetic rainfall events also show to fit inundation-depth hydrographs well with high rainfall. Moreover, the results of real-time rainfall-inundation forecasting help the emergency manager set operational responses, which are beneficial for flood warning preparations.
26 citations
••
01 Jun 2011TL;DR: A new hybrid adaptive neural network (ANN) with modified adaptive smoothing errors (MASE) based on genetic algorithm (GA) employing modified adaptive relaxation (MAR) are presented in this paper to construct learning system for complex problem solving in fluid dynamics.
Abstract: A new hybrid adaptive neural network (ANN) with modified adaptive smoothing errors (MASE) based on genetic algorithm (GA) employing modified adaptive relaxation (MAR) are presented in this paper to construct learning system for complex problem solving in fluid dynamics. This system can predict an incompressible viscous fluid flow represents by stream function (@j) through symmetrical backward-facing steps channels. The proposed learning system is constructed as an intelligent computing technique by enforcing three stages run simultaneously; the first stage concerns to construct finite-element method (FEM) employing a new approach named modified adaptive incremental loading (MAIL) to build-up in run-time a dataset driven that contains an effective patterns represented by @j for specific Reynolds number (Re), these patterns are associated to three kinds of clusters. The second stage is pertained a new hybrid neural network with new modification of adaptive smoothing errors and the third stage illustrated to modifying the numerical values of neural network connection weights through certain training algorithm with new optimization approach. The present simulation results of the proposed learning system are in good agreement with the available previous works and it is fast enough and stable.
23 citations
••
01 Jun 2011TL;DR: A cybernetic system of combining the vector autoregression (VAR) and genetic algorithm (GA) with neural network (NN) is proposed to take advantage of the lead–lag dynamics, to make the NN forecasting process more transparent and to improve the Nn’s prediction capability.
Abstract: The internal structure of a complex system can manifest itself with correlations among its components. In global business, the interactions between different markets cause collective lead–lag behavior having special statistical properties which reflect the underlying dynamics. In this work, a cybernetic system of combining the vector autoregression (VAR) and genetic algorithm (GA) with neural network (NN) is proposed to take advantage of the lead–lag dynamics, to make the NN forecasting process more transparent and to improve the NN’s prediction capability. Two business case studies are carried out to demonstrate the advantages of our proposed system. The first one is the tourism demand forecasting for the Hong Kong market. Another business case study is the modeling and forecasting of Asian Pacific stock markets. The multivariable time series data is investigated with the VAR analysis, and then the NN is fed with the relevant variables determined by the VAR analysis for forecasting. Lastly, GA is used to cope with the time-dependent nature of the co-relationships among the variables. Experimental results show that our system is more robust and makes more accurate prediction than the benchmark NN. The contribution of this paper lies in the novel application of the forecasting modules and the high degree of transparency of the forecasting process.
••
TL;DR: The use of a hybrid neural network for automatic formulation of fuzzy control rules is considered, in the testing of internal combustion engines, and the topology of this network is determined.
Abstract: The use of a hybrid neural network for automatic formulation of fuzzy control rules is considered, in the testing of internal combustion engines. The topology of this network is determined. On the basis of the resulting fuzzy rules, the quality of combustion-engine control is assessed.
•
26 Jan 2011-World Academy of Science, Engineering and Technology, International Journal of Mechanical, Aerospace, Industrial, Mechatronic and Manufacturing Engineering
TL;DR: Fluid flow and heat transfer of vertical full cone embedded in porous media is studied and the obtained solution represents a remarkable accuracy.
Abstract: Fluid flow and heat transfer of vertical full cone embedded in porous media is studied in this paper. Nonlinear differential equation arising from similarity solution of inverted cone (subjected to wall temperature boundary conditions) embedded in porous medium is solved using a hybrid neural networkparticle swarm optimization method. To aim this purpose, a trial solution of the differential equation is defined as sum of two parts. The first part satisfies the initial/ boundary conditions and does contain an adjustable parameter and the second part which is constructed so as not to affect the initial/boundary conditions and involves adjustable parameters (the weights and biases) for a multi-layer perceptron neural network. Particle swarm optimization (PSO) is applied to find adjustable parameters of trial solution (in first and second part). The obtained solution in comparison with the numerical ones represents a remarkable accuracy. Keywords—Porous Media, Ordinary Differential Equations (ODE), Particle Swarm Optimization (PSO), Neural Network (NN).
••
15 Apr 2011TL;DR: The comparison results revealed that the suggested model could increase the forecasted accuracy and prolong the length time of prediction.
Abstract: Hydrologic time series forecasting is very an important area in water resource. Based on the multi-time scale and the nonlinear characteristics of the rainfall-runoff time series, a new hybrid neural network (NN) has been suggested by Genetic Algorithm (GA) selection the lag period of time series for NN input variables, optimization neural network architecture and connection weights. The evolved neural network architecture and connection weights are then input into a new neural network. The new neural network is trained using back -- propagation (BP) algorithm for hydrologic time series forecasting. The ensemble strategy is implemented using the quadratic programming. The present model absorbs some merits of GA and artificial neural network. Case studies, the short and long term prediction of hydrological time series, have been researched. The comparison results revealed that the suggested model could increase the forecasted accuracy and prolong the length time of prediction.
••
23 Mar 2011TL;DR: In this paper, a hybrid neural network and fuzzy control is proposed for automatic generation control in thermal power systems, where a recurrent neural network is employed to forecast controller and system's future output, based on the current Area Control Error (ACE) and the predicted change of ACE.
Abstract: The AGC of reheat interconnected two area power systems are characterized by non-linearity and uncertainty. A hybrid neural network and fuzzy control is proposed for automatic generation control in power systems. Recurrent neural network is employed to forecast controller and system's future output, based on the current Area Control Error (ACE) and the predicted change-of-ACE. The Control Performance Standard (CPS) criterion is adapted to the fuzzy controller design, thus improves the dynamic quality of system. The system was simulated and the frequency deviations in area 1 and area 2 and tie-line power deviations for 1% step-load disturbance in area 1 were obtained. The comparison of frequency deviations and tie-line power deviations for the two area interconnected thermal power system integral controller with Redox Flow Batteries (RFB) reveals that the system with hybrid fuzzy neural controller enhances a better stability than that of system without hybrid fuzzy neural controller.
••
TL;DR: Developing a hybrid neural network that will be able to predict tourists' overall satisfaction of their travel experience and prioritizing the travel attributes based on their proportional impact on tourists' Overall satisfaction in Iran.
Abstract: Purpose – The purpose of this paper is to contribute to the tourism management literature by: first, developing a hybrid neural network that will be able to predict tourists' overall satisfaction of their travel experience; and second, prioritizing the travel attributes based on their proportional impact on tourists' overall satisfaction of their travel experience in Iran.Design/methodology/approach – A total of 1,870 questionnaires were distributed amongst foreign tourists in the departure lounge of “Imam Khomeini International Airport” over a period of three months. The data were used to develop a hybrid neural network in which the “rough set” is used to reduce travel attributes and the neural network to predict tourists' overall satisfaction of travel experience. After the model proved its predictive accuracy, using the sensitivity analysis of the neural network travel attributes were prioritized based on their impact on tourists' overall satisfaction.Findings – The results were quite promising in that...
••
TL;DR: A new hybrid neural network approach, Fuzzy ART K-Means Clustering Technique (FAKMCT), to solve the part machine grouping problem in CMS considering operation time and the results support the better performance of the proposed algorithm.
Abstract: Cellular manufacturing system (CMS) is regarded as an efficient production strategy for batch type of production. Literature suggests, since the last two decades neural network has been intensively used in cell formation while production factor such as operation time is merely considered. This paper presents a new hybrid neural network approach, Fuzzy ART K-Means Clustering Technique (FAKMCT), to solve the part machine grouping problem in CMS considering operation time. The performance of the proposed technique is tested with problems from open literature and the results are compared to the existing clustering models such as simple K-means algorithm and modified ART1 algorithm as found in the recent literature. The results support the better performance of the proposed algorithm. The novelty of this study lies in the simple and efficient methodology to produce quick solutions with least computational efforts.
••
TL;DR: This work employs two different types of KBHNN architecture with an effort to understand the suitability and applicability of the hybrid network in case of predictions for an ultrafiltration (UF) process.
••
TL;DR: In this paper, a novel Structure Approaching Hybrid Neural Network (SAHNN) approach to model batch reactors is presented, which involves the use of approximate mechanistic equations to characterize unmeasured state variables.
Abstract: A novel Structure Approaching Hybrid Neural Network (SAHNN) approach to model batch reactors is presented. The Virtual Supervisor−Artificial Immune Algorithm method is utilized for the training of SAHNN, especially for the batch processes with partial unmeasurable state variables. SAHNN involves the use of approximate mechanistic equations to characterize unmeasured state variables. Since the main interest in batch process operation is on the end-of-batch product quality, an extended integral square error control index based on the SAHNN model is applied to track the desired temperature profile of a batch process. This approach introduces model mismatches and unmeasured disturbances into the optimal control strategy and provides a feedback channel for control. The performance of robustness and antidisturbances of the control system are then enhanced. The simulation result indicates that the SAHNN model and model-based optimal control strategy of the batch process are effective.
••
06 May 2011TL;DR: The testing results are proved that the detection rate of the multiple classifiers intrusion detection system based on HNNA learning algorithm is higher than the IDS that use LM and improved GA learning algorithm, and the false negative rate is less.
Abstract: Based on the advantages and disadvantages of the improved GA and LM algorithm, in this paper, the Hybrid Neural Network Algorithm (HNNA) is presented. Firstly, the algorithms use the advantage of the improved GA with strong whole searching capacity to search global optimal point in the whole question domain. Then, it adopts the strongpoint of the LM algorithm with fast local searching to fine search near the global optimal point. The paper used respectively the three algorithms, namely the Improved GA, LM algorithm and HNNA, to adjust the input and output parameters of the ANN model, and adopt the theories of the fusion of the multi-classifiers to structure the Intrusion Detection System. By repeatedly experiment, it is found that the HNNA is better in stability and convergence precision than LM algorithm and improved GA from the training result. The testing results are also proved that the detection rate of the multiple classifiers intrusion detection system based on HNNA learning algorithm, including all attack categories that has a few or many training samples, is higher than the IDS that use LM and improved GA learning algorithm, and the false negative rate is less. So, the HNNA is proved to be feasible in theory and practice.
••
11 Mar 2011TL;DR: Simulation analysis demonstrated that this network model can attain higher categories of precision and coordination makes the optimized BP network not to trap into the local minima, and good generalization characteristic.
Abstract: After studying the disadvantage of BP neural network which has low convergent speed and trap into local minima easily, an idea of designing a new hybrid neural network model. By using Artificial Bee Colony Algorithm (ABC) to expand the updated space of weight and using the fitness functions to decide the better weight. On the basis, make the acquired better value as the weight of BP neural network. Both are coordination makes the optimized BP network not to trap into the local minima, and good generalization characteristic. Simulation analysis demonstrated that this network model can attain higher categories of precision.
••
TL;DR: The enhanced hybrid network is proposed that organizes the middle layer effectively by using the enhanced ART1 network adjusting the vigilance parameter dynamically according to the similarity between patterns to improve the recognition success rate.
Abstract: The hybrid neural networks have characteristics such as fast learning times, generality, and simplicity, and are mainly used to classify learning data and to model non-linear systems. The middle layer of a hybrid neural network clusters the learning vectors by grouping homogenous vectors in the same cluster. In the clustering procedure, the homogeneity between learning vectors is represented as the distance between the vectors. Therefore, if the distances between a learning vector and all vectors in a cluster are smaller than a given constant radius, the learning vector is added to the cluster. However, the usage of a constant radius in clustering is the primary source of errors and therefore decreases the recognition success rate. To improve the recognition success rate, we proposed the enhanced hybrid network that organizes the middle layer effectively by using the enhanced ART1 network adjusting the vigilance parameter dynamically according to the similarity between patterns. The results of experiments on a large number of calling card images showed that the proposed algorithm greatly improves the character extraction and recognition compared with conventional recognition algorithms.
••
17 Jul 2011TL;DR: A hybrid neural network based on the combination of GA and BP algorithms is proposed, which made fully use of GA's global searching to improve the learning ability of BP neural network.
Abstract: BP algorithm can be applied in comprehensive evaluation. A hybrid neural network based on the combination of GA and BP algorithms is proposed. The algorithm made fully use of GA's global searching to improve the learning ability of BP neural network. Then, the method is used in comprehensive evaluation, which the genetic algorithm can improve the weights of the neural network and enhance the training precision of the neural network. The experimental results show that the method is valid and feasible.
••
TL;DR: A new hybrid neural network approach, Fuzzy ART-Centroid Linkage Clustering Technique (FACLCT), to solve the part-machine grouping problems in cellular manufacturing systems considering operation time is presented.
Abstract: The design of Cellular Manufacturing Systems (CMS) has attained the significant interest of academicians, researchers and practitioners over the last three decades. CMS is regarded as an efficient production strategy for batch type of production. Literature suggests that since the last two decades neural network based methods have been intensively used in cell formation problems while production factor such as operation time is merely considered. This paper presents a new hybrid neural network approach, Fuzzy ART-Centroid Linkage Clustering Technique (FACLCT), to solve the part-machine grouping problems in cellular manufacturing systems considering operation time. The performance of the proposed technique is tested with problems from open literature and the results are compared with the existing clustering models such as simple C-Linkage, K-Means, modified ART1 and genetic algorithm and achieved better performance. The novelty of this study lies in the simple and efficient methodology to produce quick solutions with least computational efforts.
••
TL;DR: The results show that the hybrid NN approach can significantly improve modeling precision in comparison with the conventional modeling method and is superior to NN black-box modeling because the former possesses smaller network scale, higher convergence speed, higher model precision and better generalization performance.
Abstract: A hybrid neural network (NN) approach is proposed and applied to modeling of transducers in the paper. The modeling procedures are also presented in detail. First, the simulated studies on the modeling of single input?single output and multi input?multi output transducers are conducted respectively by use of the developed hybrid NN scheme. Secondly, the hybrid NN modeling approach is utilized to characterize a six-axis force sensor prototype based on the measured data. The results show that the hybrid NN approach can significantly improve modeling precision in comparison with the conventional modeling method. In addition, the method is superior to NN black-box modeling because the former possesses smaller network scale, higher convergence speed, higher model precision and better generalization performance.
••
TL;DR: The proposed method leads the mixed integer quadratic bilevel programming problem to a global optimal solution and a numerical example is used to illustrate the application of the method in a power system environment, which shows that the algorithm is feasible and advantageous.
Abstract: In this paper, a hybrid neural network approach to solve mixed integer quadratic bilevel programming problems is proposed. Bilevel programming problems arise when one optimization problem, the uppe...
••
19 Sep 2011
TL;DR: Computer simulation results in a chemical reactor indicate that the proposed evolutionary neural network fault diagnosis system works effectively and is superior to the conventional back propagation (BP)neural network.
Abstract: Rapid and accurate fault diagnosis remains a problem in the case of multiple fault for the large and complex chemical system. A novel evolutionary neural network for fault diagnosis is suggested. Which adopts three-layer feed — forward neural network with dual genetic algorithm (GA)loops embedded in its training. The dual GA loops are designed for optimizing both topology and connection weights of the neural network and establishing global optimal neural network for fault diagnosis. Computer simulation results in a chemical reactor indicate that the proposed evolutionary neural network fault diagnosis system works effectively and is superior to the conventional back propagation(BP)neural network.
••
01 Dec 2011TL;DR: An integrated multi-task control system using artificial intelligence technologies is proposed to improve the efficiency and reliability of a hybrid fuel-cell with gas turbine power plant.
Abstract: Development of Smart Grid requires power plants to be more intelligent, efficient, and reliable, which raises new challenges of the control system design for modern power plants. Regarding these requirements, an integrated multi-task control system using artificial intelligence technologies is proposed to improve the efficiency and reliability of a hybrid fuel-cell with gas turbine power plant. The integrated control system consists of a hybrid Neural Network plant model with online learning ability, an Optimal Reference Governor generating optimal setpoints as local control references, and a Fault Diagnosis and Accommodation system to detect internal plant faults and to regulate the plant during plant failures. The three subsystems are integrated to provide compressive management for the power plant. The hybrid fuel-cell power plant is introduced; the structure and strategies of the control system are discussed, and simulation results are presented.
••
21 Jun 2011TL;DR: The simulation result indicates that the hybrid network model and model-based multiobjective optimal algorithm are effective in polymerizing of PET with maximum yield and the best quality.
Abstract: A multiobjective intelligence optimal approach in polymerizing of PET with maximum yield and the best quality is proposed. The hybrid neural network based on B-spline and diagonal recursive neural network is used to model the PET process qualities, i.e. the Intrinsic Viscosity and Molecular Weight distribution. Then a hybrid NSGAII-PSO optimal algorithm with penalty functions is applied to solve the multiobjective optimal problem in order to get the best operation conditions. The simulation result indicates that the hybrid network model and model-based multiobjective optimal algorithm are effective.
••
TL;DR: In this paper, a methodology for modeling the dependence of the PA66 S-N curves on the material parameters, the material state, and the operating conditions is presented, where a multilayer perceptron neural network is combined with an analytical model of the SN curve.
Abstract: The fatigue damage to polymers generally depends on the material properties as well as on the mechanical, thermal, chemical, and other environmental influences. In this article, a methodology for modeling the dependence of the PA66 S-N curves on the material parameters, the material state, and the operating conditions is presented. The core of the presented methodology is a multilayer perceptron neural network combined with an analytical model of the PA66 S-N curve. Such a hybrid approach simultaneously utilizes the good approximation capabilities of the multilayer perceptron and knowledge of the phenomenon under consideration, because the analytical model for the S-N curves was estimated on the basis of the existing experimental data from the literature. The article presents the theoretical background of the applied methodology. The applicability and uncertainty of the presented methodology were assessed for the available data from the literature. The results show that it was possible to approximate the PA66 S-N curves for different input parameters if the space of the input parameters was adequately covered by the corresponding S-N curves.