直接转矩控制系统的集成智能控制
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
本文提出集成智能算法实现对异步电动机直接转矩控制系统的控制。集成智能算法是各种智能算法的集大成,它利用各自智能算法的优点互相取长补短,发挥整体优势。本文所采用的集成智能算法是模糊理论,神经网络,遗传算法和免疫算法的集成,并且提出了用免疫遗传算法优化模糊神经网络的新控制算法。
     模糊理论运用模糊集理论将控制知识和经验以模糊量形式表达,并运用模糊推理方法进行决策、实施控制,其数学理论体系完整,算法明确。神经网络不需要精确的数学模型,能够解决许多复杂的、不确定的和非线性的问题,在大信息处理与复杂的实时控制方面显示出巨大的优势和潜力。遗传算法对许多用传统数学难以解决或明显失效的复杂问题,特别是最优化问题提供了一个有效的途径。免疫算法从生物免疫系统中获得启示,对开发设计新的智能优化模型,具有重要意义和发展前景。
     异步电动机直接转矩控制具有控制算法简单、动态响应快,受较少电动机参数影响等特点,是继矢量控制之后出现的另一种高性能的异步电动机控制方法。本课题采用基于免疫系统与遗传算法相结合的免疫遗传算法优化模糊神经网络控制器实现对异步电动机进行直接转矩控制,它进一步解决了直接转矩控制中的参数优化和低速性能不好的问题,提高了该系统的静动态性能。基于免疫遗传算法的模糊神经网络控制器具有比模糊神经网络控制器更好的性能,它使整个控制系统易于稳定,调节速度快,在控制中明显优于其它的控制方式,这将使基于直接转矩控制技术的感应电动机的用途进一步拓展。
In the paper, we brought forward using integrated algorithm to realize the control of the DCT system. Integrated intelligent algorithm is the integration of intelligent algorithm. It learns from strong points of the intelligent algorithm to offset their weakness and exert the superiority of entirety. In the paper, the integrated intelligent algorithm we adapted is the integration of fuzzy theory, neural network, genetic algorithm and immune algorithm. And bringing forward the new algorithm of FNN optimized by immune genetic algorithm.
    Fuzzy theory adapts fuzzy set theory to express control information and experience by fuzzy quantity and manage fuzzy reasoning to decide and control. Its mathematical theory system is integrated and algorithm is clear. Neural network needn't precise mathematical model. It can settle many complex, uncertain and nonlinear problems. In the information processing and complex real time control, it brought out great dominant and potentiality. Genetic algorithm provides an efficient channel to complex problems that tradition mathematic is difficult to settle or obvious failure, especially optimization problems. Immune algorithm obtains pointer from organism immune system. It has important meaning and long term potential to develop and plan new intelligent optimization model.
    Induction motor DTC has the performances of control algorithm easy, dynamic response fast and motor parameters affect less. It is a high performance induction motor control method after vector control. The thesis adapted immune genetic algorithm based on combination of immune system and genetic algorithm to optimize FNN and realize DTC of induction motor. It settled the problems of parameter optimization and performance of low velocity. It improved static property and dynamic performance of system. FNN controller based on immune genetic algorithm has better performance than FNN controller. It can make the control system stability and regulating speed faster. In control, it gets an advantage over other methods. It can make the use of induction motor that based on DTC technique further prolongation.
引文
[1] 李夙编.导步电动机直接转矩控制.北京:机械工业业出版社,1998.
    [2] 冯江华,陈高华,黄松涛.异步电动机直接转矩控制,电工技术学报,1999,6,14(3):29~33.
    [3] 喻宗泉.智能控制技术进展,自动化与仪器仪表,2000(1):2~3.
    [4] 顾毅.智能控制发展综述,信息技术,2000(6):39~40.
    [5] Timmis J, Neal M, Hum J. An artificial immune system for data analysis[J]. Biosystems, 2000, 55(1): 143~150.
    [6] Krishnaku M, Sasaki M, Takahashi K.Adaptive learning method of neural network controller using an immune feedback law[A], 6th Int. Conf. on Neural Information Processing[C]. CA, USA, 1999, 641~646.
    [7] 张伟.无速度传感器异步电机矢量控制系统控制方法的研究:学位论文.浙江:浙江大学,2001.
    [8] Cao Chengzhi, Wang Xin, Lu Muping. Direct Torque Control Based on FNN and Optimization. The Second International Conference on Machine Learning and Cybemetics.Xi'an, China.760~764,2003.
    [9] 王磊,免疫进化计算理论及应用:学位论文.西安:西安科技大学,2001.
    [10] 曹承志编.微型计算机控制新技术.北京:机械工业出版社,2001.
    [11] Buckley JJ. Hayashi Y. Fuzzy neural networks: a survey. Fuzzy sets and System, 1994, 66(1): 1~13.
    [12] 孙笑辉,韩曾晋,张曾科.基于直接转矩控制的感应电动机模糊神经元网络实现,电气传动,2001(6):7~9.
    [13] 王建辉,顾树生.模糊控制技术在异步机直接转矩控制中的应用,东北大学学报,1997,12,18(6):640~643.
    [14] 韦巍,张国宏.人工免疫系统及其在控制系统中的应用,控制理论与应用,2002,4,19(2):157~160.
    
    
    [15] 杨建国,李蓓智,俞蕾.基于免疫遗传算法的优化设计,机械设计,2002,9,14~17.
    [16] 曹承志,曲红梅,路战红,修国一.利用模糊神经网络对定子电阻进行在线检测的研究,仪器仪表学报:2002,6,23(3):291~294.
    [17] 朱湘临,诸德宏,朱莉.采用遗传算法的直接转矩控制系统仿真研究,江苏大学学报,2002,11,23(6):75~78.
    [18] 冯培悌.模糊控制器在异步电动机直接转矩控制中的应用,电工技术学报,1999,6,14(3):29~33.
    [19] 李辉,韩红,韩崇昭,朱洪艳.基于遗传算法的模糊逻辑控制器优化设计,西安交通大学学报,2002,4,36(4):385~389.
    [20] 韩生廉,周文愚.基于免疫更新机制的遗传算法及其在规划问题中的应用,控制与决策,2002,5,17(3):343~345.
    [21] 张其光,王执铨.基于遗传算法的模糊神经网络控制器设计及其稳定性分析,控制理论与应用,1999,10,16(5):767~769.
    [22] 程启明,陈熙源,万德钧.基于遗传算法的模糊神经网络智能控制器及其应用,系统工程与电子技术,1999,21(8),41~44.
    [23] 周志坚,毛宗源.一种最优模糊神经网络控制器,控制与决策,2000,5,15(3):358~364.
    [24] 陈丽安,张培铭,缪希仁.基于免疫遗传算法的智能化电磁电器全局优化设计,电工电能新技术,2003,1,22(1):17~20.
    [25] 王磊,潘进,焦李成.免疫算法,电子学报,2000,7,28(7):74~78.
    [26] 孙笑辉,韩曾晋.异步电动机直接转矩控制启动方法仿真研究,电气传动,2000(2):13~17.
    [27] 王焱,刘景录,孙一康.免疫遗传算法对精轧机组负荷分配的优化,北京科技大学学报,2002,6,24(3):339~341.
    [28] 张军,刘克胜,王煦法.一种基于免疫调节和共生进化的神经网络优化设计方法,计算机研究与发展,2000,8,37(8),924~930.
    
    
    [29] 张桂新.异步电动机直接转矩控制系统的建模与仿真,中小型电机,2001,28(1):25~27.
    [30] 何新军,罗隆福,孙红梅.基于TMS320F240 DSP的直接转矩控制系统研究,电机电器技术,2002(5):5~7.
    [31] 窦曰轩,王洪艳.模糊直接转矩控制系统MATLAB/SIMULINK仿真,计算机仿真,2001,3,18(2):70~72.
    [32] 贾建强,韩如成,左龙.基于MATLAB/SIMULINK的交流电机调速系统建模与仿真,电机与控制学报,2000,6,4(2):91~93.
    [33] 卢和,王惠.一种新型模糊神经网络自适应控制器,电气传动,2001(4):31~32.
    [34] 廖俊,朱世强,林建亚.一种新型的模糊神经网络控制器,仪器仪表学报,1998,4,19(2):163~167.
    [35] 陈非,敬忠良,姚晓东.一种模糊神经网络的快速参数学习算法,控制理论与应用,2000,8,19(4):583~587.
    [36] 曹承志,曲红梅,路战红,杨晓波.MATLAB软件包中SIMULINK环境下直接转矩控制系统的仿真,电机与控制学报,2001,6,5(2):111~114.
    [37] 曲瀛,窦曰轩.采用模糊自适应速度调节器的模糊直接转矩控制系统,自动化与仪表,2001,1(16):23~25.
    [38] 王海仙,陈福民.感应电动机直接转矩控制系统的数字仿真,电工技术学报,1995,8(3):6~10.
    [39] 马宪民,桂佑林.基于DSP的直接转矩控制系统的设计与实现,自动化与仪器仪表,2002(2):44~45.
    [40] 施保华,秦娟英.基于MATLAB/SIMULINK动态仿真模型的异步电动机直接转矩控制的研究,电气传动自动化,2001,12,23(6).
    [41] 莫红苹.基于TMS320F240 DSP异步电动机最佳效率控制的研究:学位论文.西安:西安理工大学,2002.
    
    
    [42] 徐中,基于FNNS的感应电动机直接转矩研究:学位论文.大连:大连理工大学,2002.
    [43] 高建英.基于遗传算法设计模糊RBF神经网络控制器:学位论文.大连:大近理工大学,2000.
    [44] 薛定宇,陈阳泉编.基于MATLAB/SIMULINK的系统仿真技术与应用.北京:清华大学出版社,2002.
    [45] Dasgupta. D, Forrest S. Artificial immune systems in industrial applications[A].Proceedings of the Second Int. cone on Intelligent Processing and Manufacturing of Materials[C], Florida, USA, 1999,10~15,257~267.
    [46] Jun J H, Lee D W, Sim K B. Realization of cooperative strategies and swarm behavior in distributed autonomous robotic systems using artificial immune system[A]. IEEE Int. Conf.systems, Man and Cybernetics[C]. Tokyo, Japan 1999,4:614~619.
    [47] Chen Y C, Teng C C. A model reference control structure using a fuzzy neural network [J]. Fuzzy sets and systems,1995,73(3):291~312.
    [48] James N Nash. Direct Torque Control, Induction Motor Vector Control Without an Encoder. IEEE Trans. On Industry Application, 1997, 33(2): 333~341.
    [49] Buckley JJ, Reilly K D. Back propagation and genetic algorithms for training fuzzy neural nets[A]. Proceedings of the Fifth IEEE In ternational Conference on Fuzzy Systems[C], New Orleans, 1996, 2~6.
    [50] Luis A Cabrera. Learning Techniques to Train Neural Networks as a Selector for Inverter-Fed Induction Machines Using Direct Torque Control. IEEE TRANS. On Power Electronics, 1997, 12(5).

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700