Neurofuzzy modelling combines the attractive attributes of fuzzy systems and neuralnetworks, and is ideally suited to data modelling. The resulting models possess the ability to learn empirical data, and their behavi...
详细信息
Neurofuzzy modelling combines the attractive attributes of fuzzy systems and neuralnetworks, and is ideally suited to data modelling. The resulting models possess the ability to learn empirical data, and their behaviour can be described by a series of humanly understandable fuzzy rules. However, conventional neurofuzzy modelling is essentially restricted to low dimensional problems for which good quality expert knowledge and empirical data are available. This observation has motivated the development of constructive neurofuzzy modelling techniques, which iteratively identify parsimonious neurofuzzy models based on a combination of available a priori knowledge and empirical data. Bayesian inferencing techniques are adapted to perform local regularisation producing a method for successfully controlling superfluous model parameters, further improving model generalisation and data interpretation by the generation of valid models. This paper examines the merits of this approach by applying the techniques to a real world data set. The technique successfully produces an accurate transparent model and highlights inadequacies in the data.
Batch processes may constantly be changing, sometimes in unpredictable ways. For this reason, the operators need to keep a close watch on things. Batch-plant control stations have traditionally been located very close...
详细信息
Batch processes may constantly be changing, sometimes in unpredictable ways. For this reason, the operators need to keep a close watch on things. Batch-plant control stations have traditionally been located very close to the process, because operators can do a better job of sensing abnormalities if they can see, hear and smell what is going on. The reaction process can be considered as unstable, the instability being understood as the characteristic making the self abandoned system evolve towards unpermitted temperature values with the associated danger of accidents. The temperature also plays a decisive role in the final product quality, each time the reactor changes from the desired temperature the quality diminishes, the appearance of undesired species increases, and the overall performance of the reaction decreases which directly affects the economic performance.< >
The principal constituents of soft computing are the fuzzy logic (FL), artificial neuralnetworks (NN) and probabilistic reasoning (PR). It is generally regarded that FL primarily deals with imprecision, NN with learn...
详细信息
The principal constituents of soft computing are the fuzzy logic (FL), artificial neuralnetworks (NN) and probabilistic reasoning (PR). It is generally regarded that FL primarily deals with imprecision, NN with learning, and PR with uncertainty. They have, however, overlapping boundaries and are known to be complementary rather than competitive to each other in many applications. Here, two control algorithms, one implemented by fuzzy logic and the other by a neural network, are used as the basis to highlight salient features of soft computing. A DC motor servo system with the proposed soft computing based algorithms is discussed. The fuzzy logic control employs the principles of fuzzy logic to calculate an optimal output action based on input conditions, and a knowledge base expressed in linguistic forms, thereby performing a parallel operation to control the output with a high degree of robustness against parameter change. In the neural network control, focus is on how neuralnetworks can overcome deadzone-plus-saturation nonlinearity commonly found in the power driver of a DC servo motor. Simulation results have been performed to establish the validity of these control algorithms.< >
The importance of models of power systems has long been recognized A set of accurate models can be obtained through field tests by means of modern identification methods. In this paper a new way to establish power sys...
详细信息
ISBN:
(纸本)0852969120
The importance of models of power systems has long been recognized A set of accurate models can be obtained through field tests by means of modern identification methods. In this paper a new way to establish power system models with the artificial neuralnetworks (ANN) is presented. Both power generator using fast backpropagation neuralnetworks(FBP) and excitation system model using. radial basis function network(RBFN) are developed The simulation results of field and laboratory tests demonstrate that the application of developed ANN approach to power generator and excitation system modeling with fast training procedure and high precision is promising.
One of the important problems to be solved for neural network applications is to find a suitable network structure solving the given task. To reduce the engineering efforts for the architecture design a data driven al...
详细信息
One of the important problems to be solved for neural network applications is to find a suitable network structure solving the given task. To reduce the engineering efforts for the architecture design a data driven algorithm is desirable which constructs a network structure during the learning process. There are different approaches for structure adaptation with evolutionary algorithms, growth algorithms and others. To solve large problems successfully it is necessary to divide the problem into subproblems and to solve them separately by experts. This is a fundamental principle of nature. To implement this principle in artificial neuralnetworks there are different approaches, but these algorithms yield fixed network structures. The authors propose a learning architecture for growing complex artificial neuralnetworks which tries to include both sides of the coin, structure adaptation and task decomposition. The growing process is controlled by self-observation or reflexion. The algorithm generates a feedforward network bottom up by cyclically inserting cascaded hidden layers. Inputs of a hidden layer unit are locally restricted with respect to the input space by using a new kind of activation function, combining the local characteristics of radial basis function units with sigmoid units. Contrary to the cascade-correlation learning architecture the authors introduce different correlation measures to train the network units featuring different goals. The task decomposition between subnetworks is done by maximizing the anticorrelation between the hidden layer units output and a connection routing algorithm which only connects cooperative units of different layers. These features resemble the TACOMA (TAsk decomposition, COrrelation Measures and local Attention neurons) learning architecture. Self-observation is done by transforming the errors and the network structure to the input space. So it is possible to infer from errors to structure and reverse.< >
Today, artificial neuralnetworks (ANN) are used in various areas of application, in order to reproduce interdependencies between input and output information, similar to the information processing of the human brain....
详细信息
Today, artificial neuralnetworks (ANN) are used in various areas of application, in order to reproduce interdependencies between input and output information, similar to the information processing of the human brain. Especially where analytic modelling and calculation is impossible, or in case of unacceptable calculation times, they present a promising way of problem solving. In this project, the evaluation of the thermal capacity of energy cable, with load curves as input data, is discussed by using an ANN-approach. The presented work deals with the calculation of the temperature during normal and faulty operation. The configuration of the ANN is chosen according to this problems and in addition, the input data, especially for training purposes, is analysed. The results underling sufficient exactness for load limit calculations concerning the thermal behaviour of energy cables.
The proceedings contains 73 papers from the Fourth International Conference on advances in Power System control, Operation and Management. Topics discussed include: operation development;power research;intelligent con...
详细信息
The proceedings contains 73 papers from the Fourth International Conference on advances in Power System control, Operation and Management. Topics discussed include: operation development;power research;intelligent controlsystems;change management;flexible alternating current transmission systems;power system planning;neuralnetworks;integrated fuzzy logic generator controller;adaptive variable window algorithm;digital distance protection;high impedance fault protection;power frequency model;voltage support devices;power system voltage stability;electric load forecasting;and distributed feeder expansion planning method.
Describes work started in order to investigate the use of neuralnetworks for application in adaptive or learning controlsystems. neuralnetworks have learning capabilities and they can be used to realize non-linear ...
详细信息
Describes work started in order to investigate the use of neuralnetworks for application in adaptive or learning controlsystems. neuralnetworks have learning capabilities and they can be used to realize non-linear mappings. These are attractive features which could make them useful building blocks for non-linear adaptive or learning controllers.< >
The application of neuralnetworks (NNs) as a direct inverse controller for general nonlinear systems is considered Since little knowledge of the nonlinear plant is normally available, it is difficult to obtain an ana...
详细信息
The application of neuralnetworks (NNs) as a direct inverse controller for general nonlinear systems is considered Since little knowledge of the nonlinear plant is normally available, it is difficult to obtain an analytical expression for the plant Jacobian. Thus, an emulator is required as a channel to compute the derivative of the system output with respect to its input for NN training. neural network training using genetic algorithms (GAs) offer several advantages. No understanding of the plant model is required. Also, since no derivative computations are involved, it is less likely for these algorithms to get trapped in local minima. The scheme imitates nature's cleansing phenomena of natural selection and survival of the fittest to generate individual controllers with the best fitness values. A hybrid coding method and several appropriate modifications of the classical genetic algorithms for NN control purposes are discussed. To overcome the difficulties of saturation and fluctuation in the controller output, the output of the NN controller is obtained as the sum of several small sigmoidal functions. This effectively increases the linear range of operation of controller output without affecting the nonlinear feature of a sigmoidal function It is noted in this case that, better control is achieved. Fuzzy logic with dynamic features is used to provide an optimal direction for genetic search. It, thus, speeds up the process of convergence by bringing the chromosomes near to the problem space and bringing more exploration amongst the most desirable ones. The method is demonstrated with the control of a single-link flexible manipulator.
暂无评论