Hybrid Soft Computing Approach to Identification and Control of Nonlinear Systems

Created by W.Langdon from gp-bibliography.bib Revision:1.8129

@PhdThesis{YuehuiChen:thesis,

• author = "Yuehui Chen",
• title = "Hybrid Soft Computing Approach to Identification and Control of Nonlinear Systems",
• school = "Department of Computer Science, Kumamoto University",
• year = "2001",
• address = "Japan",
• month = mar,
• email = "CHEN Yuehui ",
• keywords = "genetic algorithms, genetic programming, PIPE Algorithm",
• URL = "http://www31.freeweb.ne.jp/computer/chen_yh/thesis.pdf.001",
• URL = "http://www31.freeweb.ne.jp/computer/chen_yh/thesis.pdf.002",
• URL = "http://www31.freeweb.ne.jp/computer/chen_yh/thesis.pdf.003",
• URL = "http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/papers/yuehui.chen/YuehuiChenThesis.pdf",
• size = "182 pages",
• abstract = "

  Recently, complex industrial plants such as mobile robots, flexible manufacturing system etc., are often required to perform complex tasks with high precision under ill-defined conditions, and conventional control techniques may not be quite effective in these systems. Soft computing approaches are some computational models inspired by the simulated human and/or natural intelligence, and includes fuzzy logic, artificial neural networks, genetic and evolutionary algorithms. There have been many successful researches for the identification and control of nonlinear systems by using various soft computing techniques with different computational architectures. The experiences gained over the past decade indicate that it can be more effective to use the various soft computing approaches in a combined manner. But there is no common recognition about how to combine them in an effective way, and a unified framework of hybrid soft computing models in which various soft computing models can be developed, evolved and evaluated has not been established.",

• abstract = "In this research, a unified framework of hybrid soft computing models is proposed and it is applied to the identification and control of industrial plants. First, a scheme for identification and control of nonlinear systems using probabilistic incremental program evolution algorithm (PIPE) is proposed. Based on the modified PIPE (MPIPE) and some parameter tuning strategies, a unified framework of hybrid soft computing models is constructed for the identification of nonlinear systems, and then the hybrid soft computing based controller design principles and methods are developed. As an application, the proposed methods are applied to the identification and control of the thrust force (cutting torque) in the drilling system.

  This dissertation consists of six chapters as follows:

  In Chapter 1, the background and the current state of soft computing researches, and the purpose of the thesis are described briefly.

  In chapter 2, the basic elements of soft computing technique are discussed, including the evolutionary algorithms and random search algorithm, neural networks and fuzzy logic systems. The problems and disadvantages of the soft computing approaches are pointed out and their modification and improvements are given.",

• abstract = "In chapter 3, in order to cope with the problems of architecture selection and parameter optimization of soft computing models simultaneously, a unified framework is constructed in which various hybrid soft computing models can be developed, evolved and evaluated. In the proposed method, the architecture of the hybrid soft computing models is evolved by MPIPE and the parameters used in soft computing models are optimized by hybrid or non-hybrid parameter optimization strategy, respectively. The effectiveness of the proposed method has been confirmed by simulation studies.

  In chapter 4, some common soft computing based controller design principles are discussed briefly. Then a new control scheme for nonlinear systems based on PIPE algorithm is proposed. Finally, based on the basis function networks a unified framework for control of affine and non-affine nonlinear systems is presented with guaranteed stability analysis. The simulation and experimental results show the effectiveness of the proposed controller.",

• abstract = "In chapter 5, the soft computing based identification and control schemes developed in chapter 3 and 4 are applied to the drilling system. In order to control thrust force (cutting torque) in the process of drill, a number of thrust force (cutting torque) identification methods are developed, and then thrust force (cutting torque) soft model based neural control scheme are presented. Real time implementations show that the soft computing approaches based control schemes are efficient and effective.

  Finally in chapter 6, the results obtained in previous chapter are summarized, and some topics for future research in this direction are given.

  In this research, the applicability of PIPE algorithm to identification and control of nonlinear systems is confirmed. Based on the MPIPE and some parameter tuning strategies, a unified framework of hybrid soft computing models is constructed. Simulation and experiments results for the identification and control of nonlinear systems show the effectiveness of the proposed methods. The key point of the research is that various soft computing based identification and control schemes can be re-evaluated in a unified framework and then it is valuable for the proposed approach in order to construct the unified soft computing theories and applications.",

• notes = "my PDF reader barfed 20 July 2001. url_2 ok",

}

Genetic Programming entries for Yuehui Chen

Citations