Computational Intelligence in Time Series Forecasting: Theory and Engineering Applications

ISBN-13: 9781852339487 / Angielski / Twarda / 2005 / 372 str.

Ajoy K. Palit; Dobrivoje Popovic

Opis

Foresight can be crucial in process and production control, production-and-resources planning and in management decision making generally. Although forecasting the future from accumulated historical data has become a standard and reliable method in production and financial engineering, as well as in business and management, the use of time series analysis in the on-line milieu of most industrial plants has been more problematic because of the time and computational effort required. The advent of intelligent computational technologies such as the neural network and the genetic algorithm promotes the efficient solution of on-line forecasting problems. Their most outstanding successes include:

• prediction of nonlinear time series and the nonlinear combination of forecasts using neural networks;
• prediction of chaotic time series and of output data for second-order nonlinear plant using fuzzy logic.

The power of intelligent technologies applied individually and in combination, has created advanced forecasting methodologies, exemplified in Computational Intellingence in Time Series Forecasting by particular systems and processes. The authors give a comprehensive exposition of the improvements on offer in quality, model building and predictive control, and the selection of appropriate tools from the plethora available using such examples as:

• forecasting of electrical load and of output data for nonlinear plant with neuro-fuzzy networks;
• temperature prediction and correction in pyrometer reading, tool-wear monitoring and materials property prediction using hybrid intelligent technologies;
• evolutionary training of neuro-fuzzy networks by the use of genetic algorithms and prediction of chaotic time series;
• isolated use of neural networks and fuzzy logic in the nonlinear combination of traditional forecasts of temperature series obtained from a pilot-scale chemical reactor with temporarily disconnected controller.

Application-oriented engineers in process control, manufacturing, the production industries and research centres will find much to interest them in Computational Intelligence in Time Series Forecasting and the book is suitable for industrial training purposes. It will also serve as valuable reference material for experimental researchers.

Recenzje

From the reviews:

This is a monograph whose aim is of special and singular interest: to present systematic and comprehensive methods and techniques of computational intelligence and soft computing for solving forecasting and prediction problems ... of time series. The book is designed to be largely self-contained and is devoted to offer researchers, practicing engineers, and applications-oriented professionals a reference volume and a valuable guide for the design, building and execution of forecasting and prediction experiments ... . The entire monograph is sensibly structured ... .

Zentralblatt MATH 1095 (2006) (Reviewer: Neculai Curteanu)

Spis treści

Part I: Introduction Computational Intelligence: An Introduction Traditional Problem Definition Part II: Basic Intelligent Computational Technologies Neural Networks Approach Fuzzy Logic Approach Evolutionary Computation Part III: Hybrid Computational Technologies Neuro-fuzzy Approach Transparent Fuzzy/Neuro-fuzzy Modeling Evolving Neural and Fuzzy Systems Adaptive Genetic Algorithms Part IV: Recent Developments The State of the Art and Development Trends

Blurb

Foresight in an engineering enterprise can make the difference between success and failure and can be vital to the effective control of industrial systems. Forecasting the future from accumulated historical data is a tried and tested method in areas such as engineering finance. Applying time series analysis in the on-line milieu of most industrial plants has been more problematic because of the time and computational effort required. The advent of soft computing tools such as the neural network and the genetic algorithm offers a solution.

Chapter by chapter, Computational Intelligence in Time Series Forecasting harnesses the power of intelligent technologies individually and in combination. Examples of the particular systems and processes susceptible to each technique are investigated, cultivating a comprehensive exposition of the improvements on offer in quality, model building and predictive control, and the selection of appropriate tools from the plethora available; these include:

• forecasting electrical load, chemical reactor behaviour and high-speed-network congestion using fuzzy logic;

• prediction of airline passenger patterns and of output data for nonlinear plant with combination neuro-fuzzy networks;

• evolutionary modelling and anticipation of stock performance by the use of genetic algorithms.

Application-oriented engineers in process control, manufacturing, the production industries and research centres will find much to interest them in Computational Intelligence in Time Series Forecasting and the book is suitable for industrial training purposes. It will also serve as valuable reference material for experimental researchers.

 

Advances in Industrial Control aims to report and encourage the transfer of technology in control engineering. The rapid development of control technology has an impact on all areas of the control discipline. The series offers an opportunity for researchers to present an extended exposition of new work in all aspects of industrial control.