基于S7 - 300的PLC模糊PID控制器调节因子 英语论文及其翻译

基于S7 - 300的PLC模糊PID控制器调节因子 英语论文及其翻译 A Fuzzy-PID Controller with adjustable factor based on S7-300 PLC Abstract: A Fuzzy-PID controller with adjustable factor is designed in this paper. Scale factor?s self-adjust will come true. Fuzzy control algorithm is finished in STEP7 software, and then downloaded in S7-300 PLC. WinCC software will be used to control the change-trend in real time. Data communication between S7-300 PLC and WinCC is achieved by MPI. The research shows that this Fuzzy-PID controller has better robust capability and stability. It?s an effective method in controlling complex long time-varying delay systems. Keywords: fuzzy-PID, adjustable factor, temperature control, MPI. 1 INTRODUCTION Temperature control is very important in industrial production. The most common temperature control objects in modern industry are boiler, electric furnace, the control system of steam plant and distillation column (F. G. Hinskey, 2004). Temperature control system generally has the characteristic of large inertia and delay, so it?s difficult to establish mathematical model exactly. In industrial production process, some control methods have been employed, such as PID control (Bolat, E.D., Erkan, K., Postalcioglu, S., 2005), Smith predictive control (He, S.- Z.,Xu, F.-L., Tan, S., 1992), Model predictive control, Fuzzy control (Chia-Feng Juang, Jung-Shing Chen, Hao- Jung Huang , 2004), Robust control (Ingram, J.E., Hodel, A.S., Kirkici, H., 1997) Neural network (Khalid, M., Omatu, S. etc 1992). PID controller is still widely used in process control field for its many advantages. But for the time- varying process with large time-delay, traditional PID algorithm has many shortcomings: the control accuracy is low, the structure is difficult to stabilize and the algorithm is more sensitive in the match degree of the models. Therefore, industrial process control which has large time-delay is still a recognized difficult problem at present. And for large lag, time-varying process whose object parameters changed as working condition and environment changed, it is more difficult to control it. And for large lag, time-varying process whose object parameters change as working condition and environment change, it is more difficult to control it. Fuzzy control has the characteristic that doesn?t charged with the object model and with strong robust, but conventional fuzzy control can not overcome negative effects caused by large-lag very well. In this page we?ll give a design of a hybrid fuzzy controller. The project is aided by the key Scientific and Technological Project (Industry Part) of Jiangsu Province (BE2006090), the Science and Technology Innovation Fund of Jiangsu University (1293000240) and the Natural Fund for Colleges and Universities in Jiangsu Province (05KDJ470048). 2 THE SELECT AND IMPLEMENT OF CONTROL METHOD Commonly used two-dimensional fuzzy control system always takes systematic error e and the error rate ec as input variables. This kind of control system can be divided into two categories: Fuzzy PD control and Fuzzy PI control. Fuzzy PD control takes u as output while Fuzzy PI control takes Δu as output (Han- Xiong Li, Gatland. H.B., 1996). In this page, we choose fuzzy PI controller as shown in Figure 1 (Chu Jing etc., 1999). eEt t Ke ΔUt Fuzzy ut KDeduc- u ect +tion -1 ECt 1-Z -1 Kc Zu(t-1) Figure 1 Fuzzy PI Control’s block diagram In this fuzzy controller, u is control variable, is controlled variable, SV is reference input, the input of t fuzzy controller is error E and error difference EC, the output isΔut. K and K are the quantify factors of ttec error and error rate respective. K is the proportion factor of fuzzy PI controller. The fuzzy control algorithm u has been brought into effect in Step7 (Liao Changchu , 2005) and downloaded in S7-300PLC, the monitor picture and tendency chart have been established by monitor software WinCC (Kun Zhe, 2004) and used to monitor the change trends of controlled plant, the data communication between S7-300 PLC and WinCC is built by MPI net. In this page we choose AE2000A process control equipment?s boiler temperature as controlled plant. The fuzzy control algorithm was realized by inquiring a two-dimensional table on-line. The process can be divided into the following three steps: Step 1: Calculate system's error and error rate according to the sampling signal and the given value in the control circuit. Then fuzzed the error and error rate according to these two equations: K = n/e and emax K=m/ccmax Step 2: Inquire the two-dimensional table according to the fuzzified error and error rate. In Step7, there?s no special instruction for inquiring two-dimensional table. As we known that the data structure in microprocessor is linear, so we written a two-dimensional polling routine based on this characteristic. In the two-dimension array which has n×m factors, the physical address of cell data α[i][j]is: (first address + i×n + j). According to the absolute physical address and Step7 STL instructions? characteristic, we can get the value of cell data α[i][j]. Step 3: In order to control the controlled plant we should defuzzy the fuzzy control variable Δu which we got from step 2. The defuzzification equation is: K =Δu/h. umax 3 THE DESIGN OF SELF-ADJUSTING FUZZY CONTROLLER The fuzzy controller is composed of the following four elements: 1. A rule-base (a set of If-Then rules), which contains a fuzzy logic quantification of the expert's linguistic description of how to achieve good control. 2. An inference mechanism (also called an "inference engine" or "fuzzy inference" module), which emulates the expert's decision making in interpreting and applying knowledge about how best to control the plant. 3. A fuzzification interface, which converts controller inputs into information that the inference mechanism can easily use to activate and apply rules. 4. A defuzzification interface, which converts the conclusions of the inference mechanism into actual inputs for the process. The fuzzy control rule is: IF E=A which can be described by the THEN IF EC=BTHEN Δu=C j ,i ijfuzzy relationshipR, that is R = ? A B C , when the error and error rate are taken from the fuzzy subset A 1Z i i and B separately,we can get the output variable Δu = (A×B?R) through fuzzy deduction rules. The 1M“center of mass” defuzzification (Sun Zengqi etc., 2004) is: Z,,(z)z/,(z) ,0cttctt,1 We can get a query table from the fuzzy controller which we designed in MATLAB?s fuzzy toolbox (Zhang Guoliang, Zeng Jing,Ke Xizheng , 2002), as shown in Table 1. 3.1 The modification of temperature fuzzy controller’s query table Because of the particularity of experiment installation, we need to adjust the temperature fuzzy controller?s query table.The boiler?s electric heating silk is three-phase resistance wire and the three-phase electric heating tube's current is controlled by SCR?s conductive angle. Through experiment we knew that when the max value of PLC?s analog output module was 27648, the value of electric heating tube?s ammeter was 4.2A. There?s no current display until the value of PLC?s analog output module was about 12500 and then the resistance wire started heating. At the beginning of the test, the temperature value was rising and the fuzzy controller?s query value was floating between [6,-6] and [6,6], that?s just the data in the last line of Table 1. If the inquired value is too small, the quantified output value will be very little and the transferred analog output value will be too little to reach the SCR?s conductive value, so the SCR can?t be conducted and the resistance wire can?t work. According to analysis based on control theory, we know that larger control effect is needed in the rising stage so as to make actual value reach set value rapidly. So, we just modified the last line of Table 1, the new modified query table as shown in Table 2. Store this query table in the memory of S7-CPU315-2DP. In real-time control process, the program searches this query table directly and gets the control value Δu according to the value of fuzzfied error ij and error rate , then multiply it by the proportional factor K,this result can be used to control the u controlled plant as output value. Error rate ec Δu -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 -6 -5.8 -5.5 -5.45 -5.5 -5.48 -5.09 -4.22 -3.73 -2.8 -2.33 -19. -1.00 0.35 -5 -5.5 -5.59 -5.46 -5.59 -5.5 -5.09 -4.2 -3.6 -2.67 -2.29 -1.61 0.12 0.61 E r -4 -5.45 -5.46 -5.22 -5.19 -5.17 -4.26 -4.25 -3.02 -2.33 -1.1 0.15 1.06 1.19 r o -3 -5.5 -5.59 -5.19 -5.09 -4.88 -3.68 -3.11 -2.26 -1.39 0.30 1.1 2.15 2.33 r -2 -5.48 -5.5 -5.17 -4.88 -4.72 -3.47 -2.7 -1.96 0.18 0.15 1.19 2.32 2.8 e -1 -5.09 -5.09 -4.43 -3.68 -3.47 -2.28 -1.09 0.30 0.97 1.29 2.79 3.52 3.73 0 -4.33 -4.14 -2.83 -2.24 -2.00 -1.13 0 1.13 1.89 2.24 3.43 4.14 4.33 1 -3.73 -3.52 -1.57 -2.09 -0.93 0.3 1.16 3.03 4.00 4.11 4.15 5.09 5.09 2 -2.8 -2.33 -1.18 -0.30 0.56 1.53 2.23 3.05 4.20 4.22 5.17 5.5 5.48 3 -2.33 -2.1 -0.8 0.3 1.39 2.34 3.11 3.11 4.22 4.84 5.19 5.59 5.5 4 -1.19 -0.80 0.3 1.56 1.74 3.02 3.25 4.26 4.47 5.15 5.22 5.46 5.45 5 -0.78 0.12 1.1 2.19 2.70 3.48 4.14 4.92 4.92 5.5 5.46 5.59 5.5 6 0.45 0.78 1.9 2.37 2.8 3.73 4.22 5.09 5.48 5.5 5.45 5.5 5.8 Table 1 The query table of fuzzy PID controller’s control variable Δu Error Error Rate ec e 3.0 3.0 3.5 3.5 4.0 4.5 4.5 5.09 5.48 5.5 5.45 5.5 5.8 Table 2 The modified query table of Δu 3.2 The design of adjustable factor The proportional factor on-line self-adjustment method was employed in this fuzzy controller. As conventional control, fuzzy control is still has contradiction between its static and dynamic characteristics. So, if we adjust the three parameters simultaneously, the control algorithm will be too complex. From controller?s structure we can find that the causality of adjusting Ku should be clearer and we still can reach our purpose of adjusting K and K finally. In order to get the best control performance, we chose ec setting K and K off-line while setting K on-line (Chu Jing etc., 1999). The principle block diagram has ecu shown in Figure Query Table2 vd e+ Control- Ke Quanti-Query v’ led K KuouficationTable1 - Plant d ec Kc dt Rules Fuzzy Set Set Figure 2 Block diagram of self-adjusting fuzzy controller From tests we got error?s changing trend and then drew out its changing curve as shown in Figure 3. e a e b f 0 t c d Figure 3 Error changing curve In point a, e(t)，0, larger and de/dt，0, in order to get rid of error rapidly we need stronger control effect, so we make K larger. uIn point b, e(t) is nearly reaching steady value and de/dt，0, to avoid e(t) dashing over the set value and lead to new fluctuation, we hope K can be smaller. uIn point c, e(t) ， 0 and de/dt ， 0 , for accelerating the convergence speed of e , we need K larger. uuIn point d, e(t) ，0 , de/dt，0, the control effect should be weaker to void big overturning, so K should be usmaller. Similarly, we can analyse K values in other points. In this way, we can get a group of fuzzy rules about u’K?s values, according to these rules, a query table about Ks value can be built. The general form of these uurules is: ’if E=A and EC=B, then K=C (i=0, 1, 2,......n) iiui???Where K should satisfy the equation: K = KK . In this equation K means setting off-line and uuuuouo?Kmeans searching in fuzzy value table. u ?We obtained Ks query table in the same way as we got the fuzzy controller?s query table, both generated uoff-line in MATLAB as shown in Table 3. Error Error Rate ec e -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 -6 6 5.6 5 5 4 4 4 3 2 2 2 2 1 -5 5.6 5.5 5 4 4 3 3 2 2 2 1.5 1 1 -4 5 5 4 4 3 3 2.5 2 2 1.5 1 1 1 -3 5 5 4 4 3 2.5 2 1.5 1 1 1 2 2 -2 5 5 4 3.5 3 2.5 2 1.5 1 1 1.5 2 3 -1 4 4 3.5 3 2.5 2 2 1 1 1.5 2 2 3 0 4 3 3 2.5 2 1 1 1 2 2.5 3 3 4 1 3 2 2 2 1 1 2 2 2.5 2 3.5 4 4 2 3 2 1.5 1 1 1.5 2 2.5 3 3.5 4 5 5 3 2 2 1 1 1 1.5 2 2.5 3 4 4 5 5 4 1 1 1 1.5 2 2 2.5 3 3 4 4 5 5 5 1 1 1.5 2 2 2 3 3 4 4 5 5.5 5.6 6 1 2 2 2 2 3 4 4 4 5 5 5.6 6 ’ Table 3 The query table of Ku 4 RESULTS AND ANALYSES OF THE EXPERIMENT Because temperature is influenced by outside environment, the initial temperatures of different times are different. In order to increase experiment?s comparability, we made temperature changed in a same range in every experiment. The sampling time of temperature controller is t=2s. Figure 4 Temperature response curve of PID controller in 4? Figure 5 Temperature response curve of fuzzy controller in 4? Figure 6 Temperature response curve of fuzzy PID controller in 4 ? The analyses of these controllers are as following: (a) PID controller. The temperature variation range is 14?,18?, dashed line represents set value. Through Figure 4 we can see the rise-time is t=180s and the overshoot is γ σ%=15.25%, stable range is between ?0.45. (b) Fuzzy controller. The temperature variation range is 18.1?,22.1?.Through Figure 5 we can see the rise-time is t =126s and the overshoot is σ%=10%, stable range is between ?0.16. This system is stable and the rise-time is shorter. (c) Fuzzy PID controller. The temperature variation range is 22?,26?.Through Figure 6 we can know this system is stable and the rise-time is short too, what?s more, it has a better steady precision than conventional fuzzy controller. The risetime is t= 106s and the overshoot is σ%=8.6%, stable range is γ between ?0.06 、 5 CONCLUSION In this page, we took AE2000A central control system?s boiler temperature as controlled plant, gave the analysis and comparison on control results based on PID controller, Fuzzy controller and Fuzzy PID controller. We found that Fuzzy PID controller with adjustable factors has obvious advantages over the other two. It has a better dynamic-static response characteristic and stronger robustness, so it can get rid of system?s residual error. We can get the conclusion that Fuzzy PID control is an effective method in dealing with time-varying process control problems with large time- delay. REFERENCES Bolat, E.D., Erkan, K., Postalcioglu, S.(2005) „Experimental Autotuning PID Control of Temperature Using Microcontroller .? Computer as a Tool, 2005. EUROCON 2005.The International Conference on Volume 1, 21-24 Nov. 2005 Page(s):266 - 269 Chia-Feng Juang, Jung-Shing Chen, Hao-Jung Huang (2004) „ Temperature control by hardware implemented recurrent fuzzy controller? Fuzzy Systems, 2004. Proceedings. 2004 IEEE International Conference on Volume 2, 25-29 July 2004 Page(s):795 - 799 vol.2 Chu Jing etc (1999) „Principle and Application of Fuzzy Control? Beijing: China Machine Press, May. 1999. F. G. Hinskey (2004), „PROCESS CONTROL SYSTEMS- APPLICATION, DESIGN, AND TUNING? (Third Edition), Chinese Version. Translated by Xiao Deyun, Lv Boming. Beijing: Tsinghua University Press. Han-Xiong Li, Gatland. H.B (1996) „Conventional fuzzy control and its enhancement? Systems, Man and Cybernetics, Part B, IEEE Transactions on Volume 26,Issue 5, Oct. 1996 Page(s):791 – 797. He, S.-Z., Xu, F.-L., Tan, S (1992). „A new adaptive Smith predictor controller? TENCON '92. Technology Enabling Tomorrow : Computers, Communications and Automation towards the 21st Century. 1992 IEEE Region 10 InternationalConference. 11-13 Nov. 1992 Page(s):1038 - 1042 vol.2 Ingram, J.E., Hodel, A.S., Kirkici, H. (1997) „Robust temperature control in the measurement of high temperature vapor pressures? Industrial Electronics, Control and Instrumentation,1997. IECON 97. 23rd International Conference on Volume 1, 9-14 Nov. 1997 Page(s):149 - 154 vol.1 Khalid, M., Omatu, S. (1992) „A neural network controller for a temperature control system? Control Systems Magazine, IEEEVolume 12, Issue 3, June 1992 Page(s):58 - 64 Digital Object Identifier 10.1109/37.165518 Kun Zhe (2004) „WinCC-Explaining the Profound of SIMENS WinCC in Simple Language? Beijing: Beijing University of Aeronautics and Astronautics Press,Oct.2004. Liao Changchu (2005) „Application Technology of S7-300/400PLC? Beijing: China Machine Press, Aug. 2005. Sun Zengqi etc (2004) „Theory and Technique of Intelligent Control? Beijing: Tsinghua University Press & Guangxi: Guangxi University Press ,Jan. 2004. Zhang Guoliang, Zeng Jing, Ke Xizheng (2002) „Fuzzy Control and Its Application in MATLAB? Xi’an: Xi'an Jiaotong University Press.2002. 基于的模糊控制器调节因子S7 - 300PLCPID 摘要:本文涉及了模糊控制器的可调因子。测量因素的自我调节将会实现。模糊控制算法在PID 软件完成，然后在里下载。软件将被用来控制进行实时变化的趋势。STEP7S7- 300 PLC WinCC 和之间通过MPI通信实现。研究 表 关于同志近三年现实表现材料材料类招标技术评分表图表与交易pdf视力表打印pdf用图表说话 pdf 明，这种模糊控制器具有更好的鲁棒S7- 300PLCWinCCPID 性和稳定性。这是一个对控制复杂的长时变滞后系统有效的方法。 关键词:模糊控制，调节因子，温度控制， PIDMPI 简介1. 温度控制是非常重要的工业成果。在现代工业中最常见的温度控制对象的锅炉、电炉、蒸汽厂和蒸馏塔的控制系统(F.G.Hinskey，2004)。温度控制系统一般具有大惯量和延迟的特性，所以现在还很难准确的建立数学模型。在工业生产过程中，有些控制方法已被应用如PID控制(Bolat, E.D., Erkan, K.,Postalcioglu, S., 2005)，史密斯预测控制(He, S.-Z.,Xu, F.-L., Tan, S., 1992)，模型预测控制，模糊控制(庄家峰，陈荣诚，黄郝荣, 2004)，鲁棒控制(Ingram, J.E., Hodel,A.S., Kirkici, H., 1997)神经网络(Khalid, M., Omatu,S. etc 1992)。 PID控制器的许多优点仍然被广泛应用于生产控制领域。但是，时变的工艺大时滞，传统的PID算法有许多不足之处:控制精度较低，结构难以稳定和算法的更敏感的匹配度的车型。因此，工业过程控制，具有较大的时间延迟仍然是一个公认的难题，目前。而对于大滞后，随时间变化的过程，其对象作为参数变化工作条件和环境的改变，它更难以控制。而对于大滞后，时变进程，其工作对象的参数变化条件和环境的变化，更难以控制它。 模糊控制没有被控对象模型和较强的鲁棒性，但传统的模糊控制不能很好地克服大滞后引起的负面影响。本文中，我们给出了一种混合型模糊控制器的设计。 备注:该项目是由计算机辅助重点科技攻关项目(工业部分)江苏省()，科学和江苏科技大学( BE2006090 1293000240创新基金和高校自然科学基金和江苏省高校()。 05KDJ470048 2.控制方法的选择和实施 一般常用的二维模糊控制系统的系统误差e和错误率et作为输入变量。这种控制系统是可以分为两类:模糊PD控制和模糊PI控制。模糊PD控制采用u作为输出，而模糊PI控制以Δu作为输出(Han-Xiong Li, Gatland. H.B., 1996)。本文中，我们选择模糊PI控制器，如图1所示(Chu Jing etc., 1999)。 Ke ΔUt u模糊 t Ku ect 推理 + -1 ECt 1-Z -1 Kc Zu(t-1) 图1模糊PI控制的框图 在这种模糊控制器，是控制变量，是控制变量，是参考输入，模糊控制器的输入是错uPVSVt 误信号和偏差信号，输出是。分别是误差和误差率各自的因素。是模糊和K KEECΔu KPIec tttu控制器的比例因子。模糊控制算法已在生效(廖长春，年)并能在中STEP7 2005S7 – 300PLC下载，显示器画面和趋势图已制定了监控软件(浙昆，年)，并用于监视被控对象的WinCC2004 变化趋势，和两者间的数据通信通过网络建立。在本文中，我们选择 S7 - 300 PLCWinCCMPI 过程控制设备的锅炉温度作为被控对象。AE2000A 模糊控制算法通过与二维表联机来实现。这个过程可以分为以下三个步骤: 第1步:根据采样信号和控制电路中的给定值计算系统的误差和错误率。然后，模糊化的误差和错误率根据这两个方程计算:K = n/e and K=m/c emaxcmax 第步:根据模糊化误差和错误率查询二维表格。在中，没有查询二维表的特殊指2 STEP7 令。正如我们所知道的，微处理器的数据结构是线性的，所以我们通过这些特点写了这个二维表查询程序。这个二维表阵列有个元素，元素的物理地址是:(首地址)。根据绝n×mα[i][j]+ i×n + j对的物理地址和指令的特性，我们可以的得到元素的值。STEP7α[i][j] 第3步:为了控制被控对象，我们应该defuzzy模糊控制变量Δu ，我们可以从第二步开始。模糊化的公式是:K =Δu/h。 umax 3.自调整模糊控制器的设计 模糊控制器在以下四个要素组成: 1.一个以规则为基础(IF-THEN规则集)，其中包含了专家的语言描述如何实现良好的控制，模糊逻辑量化。 2.推理机制(也被称为“推理引擎”或“模糊推理”模块)，模拟专家的决定作出解释和适用关于如何最好地控制植物的知识。 3.一个模糊化接口，控制器输入转换成信息，推理机制，可以很容易地使用激活和适用规则。 4.一个模糊化接口，转换到过程中的实际投入的推理机制的结论。 模糊控制规则是:如果E=A那么如果EC=B然后Δu=C，可以通过模糊关系来描述，也就j,iij 是R = ? A B C，当误差和错误率从模糊子集A和B中分离开来，我们就可以通过模糊推理规则得到输出变量Δu = (A×B?R)。“质心”去模糊(孙增圻等，2004)是: 1M Z,,(z)z/,(z)0cttct,t,1 我们可以通过在MATLAB的模糊工具箱中的模糊控制器设计得到一个从查询表(张国良，曾京，柯熙政，2002年)，如表1所示。 误差率 ec Δu -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 -6 -5.8 -5.5 -5.45 -5.5 -5.48 -5.09 -4.22 -3.73 -2.8 -2.33 -19. -1.00 0.35 -5 -5.5 -5.59 -5.46 -5.59 -5.5 -5.09 -4.2 -3.6 -2.67 -2.29 -1.61 0.12 0.61 误差 -4 -5.45 -5.46 -5.22 -5.19 -5.17 -4.26 -4.25 -3.02 -2.33 -1.1 0.15 1.06 1.19 e -3 -5.5 -5.59 -5.19 -5.09 -4.88 -3.68 -3.11 -2.26 -1.39 0.30 1.1 2.15 2.33 -2 -5.48 -5.5 -5.17 -4.88 -4.72 -3.47 -2.7 -1.96 0.18 0.15 1.19 2.32 2.8 -1 -5.09 -5.09 -4.43 -3.68 -3.47 -2.28 -1.09 0.30 0.97 1.29 2.79 3.52 3.73 0 -4.33 -4.14 -2.83 -2.24 -2.00 -1.13 0 1.13 1.89 2.24 3.43 4.14 4.33 1 -3.73 -3.52 -1.57 -2.09 -0.93 0.3 1.16 3.03 4.00 4.11 4.15 5.09 5.09 2 -2.8 -2.33 -1.18 -0.30 0.56 1.53 2.23 3.05 4.20 4.22 5.17 5.5 5.48 3 -2.33 -2.1 -0.8 0.3 1.39 2.34 3.11 3.11 4.22 4.84 5.19 5.59 5.5 4 -1.19 -0.80 0.3 1.56 1.74 3.02 3.25 4.26 4.47 5.15 5.22 5.46 5.45 5 -0.78 0.12 1.1 2.19 2.70 3.48 4.14 4.92 4.92 5.5 5.46 5.59 5.5 6 0.45 0.78 1.9 2.37 2.8 3.73 4.22 5.09 5.48 5.5 5.45 5.5 5.8 模糊控制器的控制量查询表表PIDΔu 1 3.1 温度模糊控制器的查询表 的改良 由于实验装置的特殊性，我们需要调整温度模糊控制器的查询表。锅炉的电热丝是三相电阻丝并且三相电加热管的电流由可控硅的导角的控制。通过实验我们知道，当PLC的模拟量输出模块的最大值是27648时，电加热管的电流表值4.2A。开始时没有电流显示直到该PLC的模拟量输出模块价值约为12500时，然后开始加热电阻丝。在测试开始时，温度值呈上升趋势并且模糊控制器的查询值在[6，-6]和[6,6]之间波动，这只是在表1的最后一行的数据。如果查询值过小，量化的输出值将很少，并且所转化的模拟量输出值太小达不到可控硅的导电值，因此可控硅不能导电， 并且电阻丝不能工作。根据对控制理论为基础的分析，我们知道，在上升阶段需要更大的控制效果以至于能使实际值迅速地达到设定值。所以，我们只是修改了表1的最后一行，新修改的查询表如表2所示。 误差r 误差 ec e 3.0 3.0 3.5 3.5 4.0 4.5 4.5 5.09 5.48 5.5 5.45 5.5 5.8 表 2 改良后的Δu的查询表 在S7 - CPU315 - 2DP的内存中存储此查询表。在实时控制过程中，程序可直接搜索该查询表并根据模糊误差和误差率的值获取控制值Δu，然后再乘以比例因子K，这结果可用于控制作为iju 输出值的被控对象。 调节因子的设计 3.2 比例因子的在线自调整方法被模糊控制器锁使用。由于传统控制影响，模糊控制静态和动态特性之间的矛盾仍然存在。因此，如果我们同时调整三个参数，控制算法将太复杂。从控制器的结构我们可以发现，应更清晰的调整，我们仍然可以达到我们的最终目的调整。为了获得和KK Kuec最佳的控制性能，我们选择了设置上线时设置离线时(楚晶等，)。其原理框图和KK K1999uec 如图显示 Query Table2 vd e+ Control- Ke Quanti-Query v’ led K KuouficationTable1 - Plant d ec Kc dt Rules Fuzzy Set Set 图2 模糊控制其自我调节框图 从测试中我们得到了误差的变化趋势紧接着提取了其变化曲线，如图所示3 e a e b f 0 t c d 图 3 误差变化曲线 在点a处， e(t)，0,太大并且 de/dt，0，为了迅速排除误差我们需要一个有力的控制效果，这样我们可以使K更大。 u 在点处，，并且，，为了避免超过设定值并导致新的波动，我们希望可be(t) 0 de/dt 0e(t)Ku以更小。 在点处，，并且，，为了加快的收敛速度，我们需要使更大。e(t) 0de/dt 0eK cu 在点处，，，，为了避免大的颠覆，控制效果应该减弱，因此应该更小。de(t) 0de/dt 0,K u 同样，我们可以分析其他各点的值。这样一来，我们就可以得到有关值的一组模糊规KKuu 则，根据这些规则就可以建立一个关于值的查询表。这些规则一般表示为:如果并且K E=A ui ’’’’应该满足等式:。在这个等式中表示设置那么，EC=B, K=C (i=0, 1, 2,......n)KK = K KKiuiuuuuouo ’离线表示在模糊控制表中查得的值。K u ’我们可以用同样的方法获得的就如得到模糊控制器的查询表一样，同时在中生KMATLABu 成离线模式，如表所示。3 误差 误差率 ec e -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 -6 6 5.6 5 5 4 4 4 3 2 2 2 2 1 -5 5.6 5.5 5 4 4 3 3 2 2 2 1.5 1 1 -4 5 5 4 4 3 3 2.5 2 2 1.5 1 1 1 -3 5 5 4 4 3 2.5 2 1.5 1 1 1 2 2 -2 5 5 4 3.5 3 2.5 2 1.5 1 1 1.5 2 3 -1 4 4 3.5 3 2.5 2 2 1 1 1.5 2 2 3 0 4 3 3 2.5 2 1 1 1 2 2.5 3 3 4 1 3 2 2 2 1 1 2 2 2.5 2 3.5 4 4 2 3 2 1.5 1 1 1.5 2 2.5 3 3.5 4 5 5 3 2 2 1 1 1 1.5 2 2.5 3 4 4 5 5 4 1 1 1 1.5 2 2 2.5 3 3 4 4 5 5 5 1 1 1.5 2 2 2 3 3 4 4 5 5.5 5.6 6 1 2 2 2 2 3 4 4 4 5 5 5.6 6 ’表 3 K 的查询表 u 4(实验结果及分析 由于温度受外界环境的影响，不同时间段的初始温度是不同的。为了增加实验的可比性，在每个实验中我们选取同样范围内变化的温度。温度控制器的采样时间为t= 2秒。 图 4 PID控制器在4?时的温度响应曲线 图5 模糊控制器在4?时的温度响应曲线 图 6 模糊PID控制器在4?时的温度响应曲线 对上面个控制器的分析如下: (一)PID控制器。温度变化范围为14?-18?，虚线表示的设定值。通过图4中我们可以看到上升时间为t=180s,超调量σ%= 15.25,，稳定范围为? 0.45。 γ (二)模糊控制器。温度变化范围为18.1?-22.1?。通过图5中，我们可以看到上升时间 为t = 126s,超调量σ%= 10,，稳定范围为? 0.16。该系统运行稳定和上升时间较短。 (三)模糊PID控制器。温度变化范围为22?-26?。通过图6我们可以知道该系统运行稳定和上升时间也很短，更何况，它比传统的模糊控制器稳态精度更好。该上升时间是t= 106s,超调量γσ%= 8.6,，稳定范围为? 0.06。 5(结论 在本文中，我们选择过程控制系统的锅炉温度作为被控对象，基于PID控制器、模AE2000A 糊控制器和模糊PID控制器作出了分析和控制效果比较。我们发现，模糊PID控制器具有可调的因素比其他两个明显的优势。它有一个更好的动态，静态响应特性和较强的鲁棒性，所以它可以消除系统的残余误差。我们可以得到这样的结论:模糊PID控制是处理变时过程控制并伴随大的时滞控制问题的有效方法。、 参考文献 : ()Bolat, E.D., Erkan, K., Postalcioglu, S.2005。微控制器温度控制的实验自整定。用计算机做PID 工具，。EUROCO 2005，国际会议第1卷，2005年11月21-24号，页码:266 - 269 2005 庄家峰，陈荣诚，黄郝荣(2004) ‘通过实施经常性的硬件控制的温度模糊控制器'模糊系统，2004。诉讼。 2004年电机及电子学工程师联合会国际会议第2卷，2004年7月25-29日。页码:795 - 799第2卷 楚晶等(1999)‘模糊控制中的应用和原则’。北京:中国机械工业出版社， 1999年5月 F.G.Hinskey(2004)，过程控制系统应用、设计和调优(第三版)，中文版。翻译萧德昀，吕波明。北京:清华大学出版社 李韩雄，Gatland.(1996) 传统模糊控制及其自身强化的系统，人与控制论，B部分，H.B 光学学报26卷，第5期，1996年10月 。页码:791 - 797 He，S.-Z.，Xu,F.-L.，(1992)。 新的自适应Smith预估控制器,TENCON 92。未来Tan, S 可实现技术:计算机，迈向21世纪的通信与自动化。 1992年电机及电子学工程师联合会国际会议第10卷 。 1992年11月11-13，页码:1038-1042年第2卷 (1997)，鲁棒温度控制在高温蒸气压的测量，工业电Ingram, J.E., Hodel, A.S., Kirkici, H.子，控制与仪表，1997年。 IECON 97。第23届国际会议第1卷， 1997年11月9-14号，页码:149 - 154第1卷 (1992)，神经网络控制器的温度控制系统，控制系统杂志，电机及电Khalid, M., Omatu, S. 子学工程师联合会第12卷，第3期，1992年6月，页码:58 – 64 ，数字对象标识符10.1109/37.165518 哲坤(2004)，WinCC解释了对西门子WinCC用简单语言的深化，北京:北京航空航天大学出版社，2004年10月 廖常初(2005)，北京科技应用S7-300/400PLC，中国机械工业出版社，2005年8月 孙增圻等(2004)的理论与智能控制技术，北京:清华大学出版社和广西:广西大学出版社， 2004年1月 张国良，曾京，柯熙政(2002年)，模糊控制及其在MATLAB应用，西安:西安交通大学出版社，2002年