Skip to main content
Elsevier
View Cart

Diffuse Algorithms for Neural and Neuro-Fuzzy Networks

With Applications in Control Engineering and Signal Processing

  • 1st Edition - February 10, 2017
  • Latest edition
  • Author: Boris.A Skorohod
  • Language: English

Diffuse Algorithms for Neural and Neuro-Fuzzy Networks: With Applications in Control Engineering and Signal Processing presents new approaches to training neural and neuro-fuz… Read more

World Book Day celebration

Where learning shapes lives

Up to 25% off trusted resources that support research, study, and discovery.

Explore resources

Description

Diffuse Algorithms for Neural and Neuro-Fuzzy Networks: With Applications in Control Engineering and Signal Processing presents new approaches to training neural and neuro-fuzzy networks. This book is divided into six chapters. Chapter 1 consists of plants models reviews, problems statements, and known results that are relevant to the subject matter of this book. Chapter 2 considers the RLS behavior on a finite interval. The theoretical results are illustrated by examples of solving problems of identification, control, and signal processing.

Properties of the bias, the matrix of second-order moments and the normalized average squared error of the RLS algorithm on a finite time interval are studied in Chapter 3. Chapter 4 deals with the problem of multilayer neural and neuro-fuzzy networks training with simultaneous estimation of the hidden and output layers parameters. The theoretical results are illustrated with the examples of pattern recognition, identification of nonlinear static, and dynamic plants.

Chapter 5 considers the estimation problem of the state and the parameters of the discrete dynamic plants in the absence of a priori statistical information about initial conditions or its incompletion. The Kalman filter and the extended Kalman filter diffuse analogues are obtained. Finally, Chapter 6 provides examples of the use of diffuse algorithms for solving problems in various engineering applications. This book is ideal for researchers and graduate students in control, signal processing, and machine learning.

Key features

  • Presents a new approach to training which can be applied to solve the control, identification, signal processing, and classification problems arising in practice
  • Offers an improvement from the existing learning techniques in control, robotics, and machine learning
  • Provides examples of the use of diffuse algorithms for solving problems in various engineering applications

Readership

Researchers and graduate students in control, signal processing, machine learning

Table of contents

1 Introduction1.1 Separable models of plants and training problems associated with them1.1.1 Separable least squares method1.1.2 Perceptron with one hidden layer1.1.3 Radial basis neural network1.1.4 Neuro-fuzzy network1.1.5 Plants models with time delays1.1.6 Systems with partly unknown dynamics1.1.7 Recurrent neural network1.1.8 Neurocontrol1.2 The recursive least squares algorithm with diffuse and soft initializations1.3 Diffuse initialization of the Kalman filter2 Diffuse algorithms for estimating parameters of linear regression2.1 Problem statement2.2 Soft and diffuse initializations2.3 Examples of application2.3.1 Identification of nonlinear dynamic plants2.3.2 Supervisory control2.3.3 Estimation with a sliding window3 Statistical analysis of fluctuations of least squares algorithm on final time interval3.1 Problem statement3.2 Properties of normalized root mean square estimation error3.3 Fluctuations in soft initialization3.4 Fluctuations in diffuse initialization4 Diffuse neural and neuro-fuzzy networks training algorithms4.1 Problem statement4.2 Training with use of soft and diffuse initializations4.3 Training in absence of a priori information about parameters of output layer4.4 Convergence of diffuse training algorithms4.4.1 Finite training set4.4.2 Infinite training set4.5. Iterative versions of diffuse training algorithms4.6 Diffuse training algorithm of recurrent neural network4.7 Analysis of training algorithms with small noise measurements4.8 Examples of application4.8.1 Identification of nonlinear static plants4.8.2 Identification of nonlinear dynamic plants4.8.3 Example of classification task5 Diffuse Kalman filter5.1 Problem statement5.2 Estimation in the absence or incomplete a priori information about initial conditions5.3 Estimation with diffuse initialization 5.4 Systems state recovery in a finite number of steps 5.5 Filtering with the sliding window5.6 Diffuse analog of the extended Kalman filter 5.7 Recurrent neural network training5.8 Systems with partly unknown dynamics6 Applications of diffuse algorithms6.1 Identification of the mobile robot dynamics6.2 Modeling of hysteretic deformation by neural networks6.3 Harmonics tracking of electric power networkReferences

Product details

  • Edition: 1
  • Latest edition
  • Published: February 10, 2017
  • Language: English

About the author

BS

Boris.A Skorohod

Professor Skorohod was born in Dnepropetrovsk, Soviet Union in 1951. He received M.S. and Ph.D in electrical engineering from Sevastopol State University, Sevastopol, in 1973 and 1980, respectively.

From 1980 to 1985, he was a Senior Researcher with the Department of Technical Cybernetics, Sevastopol State University, from 1985 to 1991, he was the Head of the Laboratory, Sevastopol State University and since 1992, he has been professor in the Department of Informatics and control in Technical systems, Sevastopol State University, Sevastopol, Russia. His interests include control systems, intelligent algorithms, neural networks and fuzzy logic.

Affiliations and expertise
Informatics and Control in Technical Systems Department, Sevastopol State University, Russia

View book on ScienceDirect

Read Diffuse Algorithms for Neural and Neuro-Fuzzy Networks on ScienceDirect