H_∞Filter Design For Networked Control Systems Based On The Fuzzy Models

As the development of science and the progress of technology, traditional control theoryis very hard to solve the problem of modeling, analyzing and designing for complex nonlinearsystems. On the other hand, with the rapid development of computer and sensor technology,it laid the foundation of hardware for us to achieve a variety of complicated nonlinear controlalgorithm. Therefore, since the1980s, the control problem of complex nonlinear system hasreceived more and more attention at home and abroad, and the fuzzy control technology isappeared and developed in this context. The fuzzy control technology has the characteristicssuch as the controller design is simple, applicable for many nonlinear systems, use humannatural language to qualitative description and the strong robustness.In this thesis, we choose T-S fuzzy model to describe the controlled plant, because it canapplied the process of thought, experience and judgment of people to the control of complexsystems. On the other hand, for the condition of network limited channel bandwidth and theinevitable existence of data packet loss, introduced the logarithmic quantizer, applied Stochasticvariable satisfying Bernoulli random binary distribution and Markov jump model to describe thedata packet loss respectively. The main task of our work are listed as follows:1. For the fuzzy systems, when there exists measurement missing, we design a H∞filterthat the control system is asymptotically mean-square stability and satisfies the prescribed H∞performance constraint. A numerical example is given to solve the parametric matrices of theH∞controller, and influence of associated factors to the H∞performance in the example provethe feasibility of the proposed approach.2. The problem of Networked H∞filtering for T-S fuzzy time delay systems with Quan-tization and Data Dropouts is investigated. Stochastic variable satisfying Bernoulli binary dis-tribution is introduced to model the lossy measurements. Then, a full order filter is designed tocope with data packet loss and lighten quantization effects such that the filtering error system isstable and satisfies the prescribed H∞filtering performance. Further, the filter parameters arederived by solving a linear matrix inequality (LMI). Finally, a numerical example and waveformare recorded to prove the effectiveness and reliability of the approach method.3. We concern with the issue of H∞Filtering for T-S Fuzzy Systems with unideal com-munication channel using quantized measurements at last. The missing data are described bytwo state Markov chain. The purpose is to design a nonlinear H∞filter such that the filteringerror system is stochastically stable, and satisfies the prescribed H∞performance. By usingLyapunov stability theory, a new stability conditions dependent on delay is obtained. Then, the filter parameters are derived by solving a linear matrix inequality (LMI). Finally, using numer-ical simulation to verify the effectiveness of the method.