【正文】 if the quantisation is too coarse the controller may oscillate around the reference or even bee unstable. Nonlinear scaling is an option . In the FL Smidth controller the operator is asked to enter three typical numbers for a small, medium and large measurement respectively (Holmblad amp。 Pedrycz, 1993) or fuzzy modelbased control (see later). 4. Based on learning. The selfanising controller is an example of a controller that finds the rules itself. Neural works is another possibility. There is no design procedure in fuzzy control such as rootlocus design and the frequency response design, pole placement design, or stability margins, because the rules are often nonlinear. Therefore we will settle for describing the basic ponents and functions of fuzzy controllers, in order to recognise and understand the various options in mercial software packages for fuzzy controller design. There is much literature on fuzzy control and many mercial software tools (MIT, 1995), but there is no agreement on the terminology, which is confusing. There 西北工業(yè)大學明德學院本科畢業(yè)設(shè)計 5 are efforts, however, to standardise the terminology, and the following makes use of a draft of a standard from the International Electrotechnical Committee (IEC, 1996). Throughout, letters denoting matrices are in bold upper case, for example A。 Reinfrank (1993)or Passino amp。 Ostergaard, 1982). The most mon approach to establishing such a collection of rules of thumb, is to question experts or operators using a carefully anized questionnaire. 2. Based on the operator’s control action rules can be deduced from observations of an operator’s control actions or a log rules express inputoutput relationships. Figure 2： Fuzzyparameter adaptive control 3. Based on a fuzzy model of the process. A linguistic rule base may be viewed as an inverse model of the controlled process. Thus the fuzzy control rules might be obtained by inverting a fuzzy model of the process. This method is restricted to relatively low order systems, but it provides an explicit solution assuming that fuzzy models of the open and closed loop systems are available (Braaeamp。 and operations are in bold, for example min. Structure of a fuzzy controller There are specific ponents characteristic of a fuzzy controller to support a design the block diagram in Fig. 3, the controller is between a preprocessing block and a postprocessing block. The following explains the diagram block by block. Figure 3: Blocks of a fuzzy controller. Preprocessing The inputs are most often hard or crisp measurements from some measuring equipment, rather than linguistic. A preprocessor, the first block in Fig. 3, conditions the measurements before they enter the controller. Examples of preprocessing are: 1. Quantisation in connection with sampling or rounding to integers。雖然有許多入門的課本講述了模糊控制 ,但關(guān)于如何對一個簡單的模糊控制器設(shè)立參數(shù)還沒有多少指導(dǎo) . 本文所講到的方法就是關(guān)于 PID 控制的設(shè)計步驟 ,它分三步 : 1. 先以一個 PID控制器開始。拿下面的一個模糊控制器為例。 在一臺包含規(guī)則庫的控制器里，控制策略多少會以自然語言的形式儲存。 再前反饋控制系統(tǒng)中，一個可測量的擾動需被彌補。傳感器測量被用來作為時間變量，支配著控制器參數(shù)的改變。 4. 均衡考慮長期或短期的趨勢 。 如果誤差是零 ,誤差變化率是正 ,那末輸出是正中 。這種形式可能更適合想快點了解規(guī)則庫的有經(jīng)驗的使用者。 零、正、負都是模糊集的語言量，同樣還有負大、負中、正大、正中。 如果誤差是負 ,誤差變化率是零，那末輸出是負中 。在線性系統(tǒng)里，判斷系統(tǒng)穩(wěn)定性是相對直接，比如判斷 系統(tǒng)所用的特征根是否都在負平面的左半軸。 在參數(shù)適應(yīng)控制計劃里，模糊規(guī)則也被用來糾正調(diào)諧參數(shù)。 模糊控制器能被用在各種控制計劃里（國際電工委員會， 1996）最明顯的例子是直接控制，在那里，模糊控制器位于反饋控制系統(tǒng)的前相通道里。這些規(guī)則是以如果 那末形式出現(xiàn)的，通常如果那邊叫做條件，那末一邊叫做結(jié)論。接下來簡短的概括在一個簡單的模糊控制器的主要設(shè)計方案，用到的術(shù)語是以現(xiàn)行的國際標準用語。 5. A bination of several measurements to obtain key indicators。 Sugeno – all in Lee, 1990。 the controller C may be a linear PID controller, while the fuzzy controller F is a supplementary nonlinear controller. Fuzzy rules are also used to correct tuning parameters in parameter adaptive control schemes (Fig. 2). If a nonlinear plant changes operating point, it may be possible to change the parameters of the controller according to each operating point. This is called gain