Dissertation > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > Automation systems > Automatic control,automatic control system > Fuzzy control, fuzzy control system

A new auto-tuning PID type fuzzy logic controller

Author LeiTao
Tutor JiaLiMin
School Academy of Railway Sciences
Course Traffic Information Engineering \u0026 Control
Keywords PID type fuzzy controller Self-tuning PID controller In band Low frequency band Incremental Response speed Lead compensation Frequency domain theory Frequency characteristics
CLC TP273.4
Type Master's thesis
Year 2001
Downloads 854
Quotes 0
Download Dissertation

In engineering, PIO controller is the most commonly used controller. Because of its simple structure, and can adapt to the requirements of most industrial processes. PID controller epitomized the system in the past, the present and future state of information. In Japan, around 84.5% of the controllers are PID controllers. However, with the rapid development of modern industrial production, when better static and dynamic performance and robustness of the control system, PID controller often fail to achieve satisfactory results. This time by the setting of the traditional engineering approach (such as Z-N tuning formulas) already can not meet the control requirements, especially the control object with strong coupling, large delay, nonlinear characteristics when. As a result, people began to study a variety of improved PID controller SMITH predictive compensation PID controller, the optimal PID and adaptive plant P'D, and so on. In recent years, with the continuous development of intelligent control technology and computer applications, the use of a combination of intelligent control and PID control technology started to become a popular means of control. Which, FUZZY 10-P workers D is a commonly used method. Generally, PID and fuzzy control technology to combine two common ways, one is a traditional PID controller plus a fuzzy parameter adjustment link, fuzzy inference engine online real-time adjustment of the three parameters of the PID controller However, this method requires a lot of expertise, and very complex set of rules of inference machine. Another way is to direct the formation of a PID type fuzzy controller. General fuzzy controller can be divided into two types: location, type and increment type. Position Controller is equivalent to a PD controller, Incremental equivalent to a PI controller. Easy to achieve because of computer programming, under normal circumstances, are incremental fuzzy controller. A standard incremental fuzzy controller has the structure shown in Figure 1: 'under Chen Maputo renogram! The growth star standard fuzzy controller analytical expression: △ u two, especially [in particular: e ten in particular: (l a) △ e] by adjusting the weighting factor. Can be adjusted for the deviation e and the deviation of the rate of change of △ e different weighting the degree. When simulation can know. Is relatively large, i.e. on the deviation e weighted higher degree of the object! Eve of China Academy of Railway Sciences Master 12 thesis step response overshoot, fast response: When. Small, the step response of the object does not overshoot, but the response is slow. Therefore, adjust the appropriate weighting factor, Q, can be a good to change the performance of the controller. This is a simple way to improve controller. The control system of direct application of differential equations can be accurately solved motion of the system function, but from an engineering point of view, this approach not only significantly less, but too redundant. The physical meaning of the frequency domain method is distinctive and easy object model, and can get the stability margin of the system is that the robustness of the system. Therefore in this paper the application of the theory of frequency domain analysis of FLC, based on the frequency domain of the linear system theory, system's crossover frequency do not. . The frequency characteristics of the system is divided into high, medium and low frequency bands. General in the band, do not. / No 2-2, but lower than the mountain. / 2 for the low frequency band, higher than 2 field 'for the high frequency band. _ Body from open-loop frequency characteristics of the system can analyze the dynamic and steady-state performance. The system low frequency band of the nature of the steady-state performance. The nature of the system high frequency dynamic nature of the major impact of the system to do a quick movement, and also affects the system's anti-interference ability. Band in the nature of the dynamic transition process. Therefore, the nature of the middle band is very important. System band to guarantee the stability of the system, requiring a larger stability margin. Convergence point of the band and the high-and low-frequency band o,. Hill, also have a larger impact on the dynamic performance of the system. The Needlessly * more away from the Do not. , The overshoot of the system is smaller, and Tian. The closer to the field, the faster the response speed of the system. Should therefore ensure that the system is stable under the premise of the mouth, away from Tian, ??mouth. The closer the fields. This system's fast response and small overshoot. Link from the above analysis that you want to lead compensation to the low-band mobile. Seen by the analytical expression of FLC lead compensator link to move to a low-frequency band to increase K, K:. However, when K, K: △ e is too large, since the FLc limiter so K3K: down, so that the deterioration of system performance. Further, when a step signal to the FLC, △ e will produce a relatively large step value, due to the limiting effect of the FLC so △ e the gain will drop a lot. This is also the FLC initial rise slowly. ', In order to solve the above problems

Related Dissertations
More Dissertations