基于遗传算法的模糊自适应控制在DTC的应用  

Application of Adaptive Fuzzy Control Based on Genetic Algorithm in Direct Torque Control

作　　者：胡俊达[1] 

机构地区：[1]湖南工程学院电气与信息工程系,湘潭411101

出　　处：《电工技术学报》2004年第6期78-84,共7页Transactions of China Electrotechnical Society

基　　金：湖南省自然科学基金资助项目(03JJY6019)。

摘　　要：系统中采用了两个模糊控制器,其一是用模糊控制器取代了常规直接转矩控制系统的磁链和转矩控制器;其二是根据异步电动机的运行特点提出了一种基于遗传算法的模糊自适应PI控制器作为直接转矩控制系统的速度调节器。在这种模糊自适应PI控制器的设计中,用自适应模糊控制器推断出PI控制器的参数,用遗传算法和参数调整算法对模糊控制器的量化因子和比例因子分别进行离线粗调和在线优化细调,解决了常规模糊控制器在电机直接转矩控制过程中特别是起动过程因参数不变带来的超调量大、响应慢、不稳定性等问题。仿真试验结果验证了所研究设计的控制器的有效性,系统在动态响应、抗干扰能力、增强鲁棒性等方面均获得了满意的效果。Two fuzzy controllers were used in the system.One was used to substitute for flux and torque controller of traditional direct torque control(DTC)system,and the other was an adaptive PI-type fuzzy controller based on GA which advanced according to the operating characteristic of induction motor as speed controller of DTC system. In the design of adaptive PI-type fuzzy controller ,the parameters of PI controller were got from the adaptive fuzzy inference and the quantizing factor and the scaling factor of fuzzy controller were adjusted roughly off line with genetic algorithm and precisely on line with self-adjusting parameter algorithm ,which solved the problems such as big overshoot、slow speed corresponding、unsteady and so on brought by fixed parameters in the process of DTC control in motor and especially starting up .The result of simulation demonstrates the validity of controller designed in this paper. The system has got a more satisfactory result of dynamic response、strong anti-interference and strong robustness than the DTC system with traditional fuzzy controller.

关 键 词：遗传算法 自适应模糊控制 直接转矩控制 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP311.52[自动化与计算机技术—控制科学与工程]