Self-Tuning of PI Speed Controller Gains Using Fuzzy Logic Controller

Abstract

The role of proportional-integral (PI) controller and proportional-integral-derivative (PID) controller as a speed controller for a Permanent Magnet Synchronous Motor (PMSM) in high performance drive system is still vital although new control techniques such as vector control theory that is more effective -but complex-is available. However, PI controller is slow in adapting to speed changes, load disturbances and parameters variations without continuous tuning of its gains. Conventional approach to these issues is to tune the gains manually by observing the output of the system. The tuning must be made on-line and automatic in order to avoid tedious task in manual control. Hence, an on-line self-tuning scheme using fuzzy logic controller (FLC) is proposed in this paper. The performance of the developed proposed controller is tested through a wide range of speeds as well as with load and parameters variations through simulation using MATLAB/SIMULINK. It is found that the proposed 25 rules FLC with adaptive input and output scaling factors enhances the performance of the system especially at high load inertia. The simulation results show that the developed controller can well adapt to speed changes as well as sudden speed reduction besides fast recovery from load torque and parameters variation and these show remarkable improvement compared to conventional PI controller performance.

A Multi-Wavelet Neural Network Particle Swarm Optimization Based MRPID of Interior Permanent Magnet Synchronous Motor Drives

“…Therefore the comparison analysis of the existing and the proposed illustrates that the proposed controller has zero steady state error and shows 86 % of improvement than that of the existing controller. (1) and (11). At time t=0 the error occurred will be of about 50% for the existing controller which is highly reduced by using the proposed controller and it attains the steady state at time t=0.004 sec.…”

“…Although they are uncomplicated, these controllers are very sensitive to parameter differences, change in load, changes in command speed and other uncertainties [10]. The design of PI or PID controller is uncomplicated as the computations of the proportional gain, integral gain and derivative gain are prepared using secondorder method or Ziegler-Nichols method [11]. The uncertainty problem in the PI and PID controller can be accomplished by the fuzzy neural network, in order to develop the speed tracking precision of the conventional controllers, an adaptive and artificial intelligent controllers have been proposed [12].…”

Kararsız Koşullarda SMDA Motorun Dayanıklı Hız Kontrolü İçin Bulanık-Tabanlı Akıllı Denetleyici Tasarımı

“…Eğer bir denetleyici değişen çevresel koşullara rağmen denetleme işlemini yerine getirebiliyorsa böyle denetleyicilere uyarlamalı denetleyiciler (adaptive) denir. Uyarlamalı BMD'lerin kendini ayarlayan (self-tuning) ve kendini organize eden (self-organizing) denetleyiciler gibi farklı tipleri vardır [35], [36]. Bu çalışmada geleneksel BMD'de sabit olan çıkışın ölçekleme faktörlerini denetlenmekte olan sistemin performansına bağlı olarak geliştirilmiş ikinci bir kural tabanına göre değiştirerek bulanık mantık denetleyiciler için kendini ayarlama tekniği ortaya çıkmaktadır [37].…”

Fuzzy Logic Based Module-Level Power Electronics for Mitigation of Rapid Cloud Shading in Photovoltaic Systems

“…Chapter 3 describes the improved power optimizer design using four subsections: improved MPPT, mode selection and active power curtailment, ramp rate reduction using fuzzy logic control, and a self-tuning gain controller for the DC regulation stage. The design of the MPPT algorithm is adapted from [4], the standard buck boost converter presented by Matlab/Simulink was customized for this model, and the work presented by [7] was adapted for use within this application; however, the remaining subsystems are designed by the author. The variable step MPPT Chapter 4 introduces the test cases used to obtain the results found in this chapter.…”

“…A fuzzy logic gain controller was implemented to manage load side variations to enhance conventional PI controller functionality [7]. The proportional and integral gains, K p and K i respectively, are the two parameters which will adaptively change with the varying load.…”