Font Size: a A A

Study On Optimal Fuzzy Decentralized Control Method (FDC) For Complex Uncertain Mechanical Servo System

Posted on:2013-08-11Degree:DoctorType:Dissertation
Country:ChinaCandidate:D XiongFull Text:PDF
GTID:1222330377961082Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
In this dissertation, results in robust and adaptive control of complex mechanical system withuncertainties(linear and nonlinear) are presented via the framework of the new optimal robust decen-tralized control method for complex system based on fuzzy theory(FDC). The proposed controllers areintuitively appealing and treat uncertainty by fuzzy arithmetic instead of the typical skills in managinguncertainty. The basic concepts of fuzzy set theory of complex uncertain system are organized. Aclass of mechanical servo uncertain interconnected system is chose to implemented with FDC due toits characters in structure, operation features and mechanism. This system contains uncertainty, con-sists of multiple subsystems, and faces information structure constraints. The recent research statuson control for uncertainty system is presented. FDC is considered for such a complex system with un-certainty. The universality and applicability of the framework based on FDC are further explored andextended by management of high dimensional uncertainty. This attempt develops FDC to a more gen-eral situation, say FDCH, that it is versatile from linear uncertainty to high dimensional uncertainty.The uncertainty effects are taken into consideration by both linear and nonlinear. This series of controlmethods presented in this work(FDC, FDCH) establish a new theoretic schemes to tackle the controlproblems in uncertain systems in a creative perspective. The uncertainty may be due to the unknownparameters and the uncertainty is bounded with uncertain bound. The bound is high dimensionalitythat may be due to the unknown parameters and input disturbance. The available information of theuncertainty is prescribed by fuzzy characteristics. That is, the uncertainty is assumed to be withincertain fuzzy sets.Since FDC or FDCH is applied and therefore the problem of system control is switched into theminimization of a performance index, which reflects the system’s average fuzzy characteristics. Sys-tem optimization is solved by obtain an optimal parameter. In addition, it is robust and optimal: whileit guarantees certain performance regardless of the uncertainty, it also minimizes a performance index.We cast the problem into an optimal gain design setting. We show how to obtain the optimal gainsolutions of a complex mechanical servo system. With regard to the minimization phase of the design,the solution to this optimization system performance index is proved to be always exists and is unique.The closed-form solution and cost are explicitly listed.Considering the application of FDC and FDCH to mechanical servo system, an uncertain flexibleinverted pendulum balancing system is employed to test the control framework. Also, another twotypical control schemes are implemented to the physical system as well. The results show that FDC is capable of guarantee the deterministic performance and minimize the average fuzzy performance.Comparing with the other control schemes, advantages of FDC are substantiated by the results. Thehigh precision control requirements in mechanical servo system can be obtained. For further illumina-tion and exploration of FDC and FDCH to general engineering area, control problem in well-knownautomated highway system(AHS) is reconsidered. The control framework related to high dimension-ality uncertainty(FDCH) is then employed to AHS. The implementation is brought to improve thefeasibility and reliability of FDCH in normal engineering system. The results thus show that the algo-rithms are implementable to systems as complex as AHS. Simulation results are used to corroboratethe theoretical findings of this work.
Keywords/Search Tags:mechanical servo system, complex and uncertain system, decentralized control, fuzzy set theory, optimization
PDF Full Text Request
Related items