本文選題：伺服電機 + PID控制�。� 參考：《哈爾濱理工大學》2014年碩士論文

【摘要】：伺服電機在工業(yè)領(lǐng)域主要用于控制機械元件的運轉(zhuǎn)，可實現(xiàn)準確的速度、位置控制。目前交流伺服電機已經(jīng)取代直流電機，成為了伺服系統(tǒng)的主流，而永磁同步伺服電機(PMSM)因為其優(yōu)越的性能得到了愈加廣泛的應用。 永磁同步伺服電機的控制策略直接影響了整個伺服系統(tǒng)的性能，對PMSM的控制一直以來都是研究的重點。但是由于實際的伺服系統(tǒng)中存在著不確定性以及外部干擾、內(nèi)部擾動，一般的PID控制難以滿足控制要求，近年來神經(jīng)網(wǎng)絡控制、滑模變結(jié)構(gòu)控制等控制策略的研究成為熱點，以更好的提高系統(tǒng)的工作性能。 本文建立了PMSM的數(shù)學模型，運用增量式PID與積分可分離結(jié)合起來的方法對PMSM的常規(guī)PID控制策略進行了仿真分析。隨后借助神經(jīng)網(wǎng)絡的學習能力，去調(diào)節(jié)PID控制器的參數(shù)，進一步改善了PID控制器的性能。滑模變結(jié)構(gòu)控制在控制性能上優(yōu)于常規(guī)PID控制，該控制策略能夠使系統(tǒng)穩(wěn)定的工作在滑模面，進一步提高系統(tǒng)的穩(wěn)定性。由于小腦模型關(guān)節(jié)控制器(CMAC)是神經(jīng)網(wǎng)絡的一個分支，其具有快速的學習能力，為了使系統(tǒng)更加穩(wěn)定，構(gòu)造了基于CMAC網(wǎng)絡的滑�？刂破�，使用CMAC網(wǎng)絡去補償滑模控制，達到加快收斂速度的目的。經(jīng)過仿真分析，分別得到上述各種控制策略的控制結(jié)果，可知神經(jīng)網(wǎng)絡能夠提高提高PID控制器以及滑模控制器的性能。并且把基于CMAC網(wǎng)絡的滑模控制與常規(guī)的PID控制算法、神經(jīng)網(wǎng)絡控制PID以及滑模變結(jié)構(gòu)控制進行仿真對比分析，可以清楚的看到基于CMAC網(wǎng)絡滑�？刂频南到y(tǒng)無論是在收斂速度還是在跟蹤精度方面均能得到了明顯的改善和加強。
[Abstract]:Servo motor is mainly used to control the operation of mechanical components in the industrial field, which can achieve accurate speed and position control. At present AC servo motor has replaced DC motor and become the mainstream of servo system. Permanent magnet synchronous servo motor (PMSM) has been more and more widely used because of its superior performance. The control strategy of PMSM has a direct impact on the performance of the whole servo system. The control of PMSM has always been the focus of research. However, due to the uncertainty and external disturbance and internal disturbance in the actual servo system, the general pid control is difficult to meet the control requirements. In recent years, neural network control, sliding mode variable structure control and other control strategies have become a hot topic. To better improve the performance of the system. In this paper, the mathematical model of PMSM is established, and the conventional pid control strategy of PMSM is simulated and analyzed by combining incremental pid and integral. Then the parameters of pid controller are adjusted by the learning ability of neural network, and the performance of pid controller is further improved. The sliding mode variable structure control is superior to the conventional pid control in control performance. The control strategy can make the system work stably on the sliding mode surface and further improve the stability of the system. Because the cerebellar model joint controller (CMAC) is a branch of neural network, it has fast learning ability. In order to make the system more stable, a sliding mode controller based on CMAC network is constructed, and CMAC network is used to compensate sliding mode control. To speed up convergence. The simulation results show that the neural network can improve the performance of pid controller and sliding mode controller. And the sliding mode control based on CMAC network is compared with the conventional pid control algorithm, neural network control pid and sliding mode variable structure control. It can be clearly seen that the system based on CMAC network sliding-mode control can be improved and strengthened obviously in terms of convergence speed and tracking accuracy.
【學位授予單位】：哈爾濱理工大學
【學位級別】：碩士
【學位授予年份】：2014
【分類號】：TM383.4

【引證文獻】

相關(guān)碩士學位論文 前1條

1 張建新;天線測試轉(zhuǎn)臺的結(jié)構(gòu)設計及對準誤差分析[D];哈爾濱工業(yè)大學;2015年

，

本文編號：2077752