電氣自動(dòng)化自動(dòng)控制畢業(yè)設(shè)計(jì)外文翻譯(編輯修改稿)

　

【文章內(nèi)容簡(jiǎn)介】 algorithm is used to optimize the gains of a PIbased coupled control scheme. Responses to step changes in generated power showed good steadystate behavior, but there were large oscillations due to the integral action. In [5] a robust control approach showed good results, nevertheless, in addition to being putationally intensive, the design required frequency response information and a linearized model of the plant. Perhaps, the main disadvantage of most advanced methods is the requirement of precise mathematical models for design and during operation. High dimension and plexity of power units have precluded the implementation of full scope mathematical model based centralized control the other hand, this schemes. On situation has motivated the prolonged existence of multiloop configurations based on conventional PID control algorithms with of a great variety of plementary pensation schemes, for which more general and systematic design approaches are needed. This paper presents a simple, effective and systematic approach to implement a pensated multiloop CC strategy through the use of fuzzy systems. The main concernis to design a high performance and robust control strategy for wide range operation without the precisemathematicalmodel requirement. Instead, an approximate offline learned linguistic model is used. The approach is to substitute the conventional PID controllers by simple PIDlike fuzzy with wider operating ranges and better disturbance rejection properties, and to add a fullrange nonlinear MIMO fuzzy pensator to reduce the effects of interaction among control loops along the whole FFPU operating range. After this introduction, part II describes the configuration of the proposed coordinated control strategy. Then, parts III and IV give the details about the fuzzyPID controllers and the MIMO fuzzy pensator respectively. Part V provides some simulations results, and part VI concludes this work. Practical CC implementations present many feedforward pensations to the fundamental control loops [6]. Their objective is to provide a means to minimize the effect of interaction between the control loops due to coupled dynamics in the process. There is no general or widely accepted method to implement them. Most realizations have evolved through various decades of research and practical experience, and some are confidential industrial property of their developers. The keypoint is that the design of a mathematicalmodelbased interaction pensator might be intractable or is so plex that precludes its realization. In this research, our approach has been to keep the fundamental control loops free of the above mentioned kind of pensations, and to include a pensator which may be obtained systematically, without the hassle of the precisemathematicalmodel requirement. The CC scheme proposed in this work updates the PIDbased multiloop coordinated turbinefollower control scheme in two steps (Fig. 2). First, the conventional PID controllers are substituted by simple PIDlike fuzzy controllers. The objective of this change is to take advantage of their better performance and superior robustness to variation in plant parameters. Thus, the scheme is expected to have wider ranges of operation with better control loop performance. The characteristics and properties of different versions of PIDlike fuzzy controllers are provided in section III, where their performance is pared to that of a conventional PID controller in various categories. Second, a MIMO fuzzy pensator is inserted between the fuzzyPID controllers and the power unit. The proposed pensator is a fuzzy system that implements a linguistic model that can be learned offline from inputoutput plant data. The objective of this addition is to provide a way to deal with the interaction effects along the whole operating range of the FFPU. It is expected that this pensator will decrease the interaction effects among control loops at levels manageable by the PIDlike fuzzy controllers with lessened control of the fuzzy pensator is described in section IV. It should be noted that the proposed configuration incorporates two different kinds of knowledge into the control system. First, the PIDlike fuzzy controllers account for explicit knowledge of operators regarding how to control a process. Second, the fuzzy pensator accounts for implicit