基于GA-BPNN的PID调节器实现实时调速的研究
|
摘要
遗传算法(GA)和误差反向传播算法的多层前馈网络(BPNN)都具有很强的寻优能力。遗传算法具有全局寻优能力。神经网络在局部寻优效果较好，但容易陷入局部极点。于是本文提出两者相结合来对PID调节器参数寻优。使得PID调节器稳定性、响应速度以及精度都得到提高。
由于GA和BPNN的寻优时收敛速度慢，使得基于GA-BPNN的PID调节器在实际控制系统中的运用受到了限制。为了克服这个局限性，根据无刷直流电机在调速系统中的特点，本文提出一种新型的控制策略：“离线粗调”和“在线细调”相结合。让大部分工作放到离线状态来做，使得已训练好BPNN-PID调节器来在线对无刷直流电动机调速。由于BPNN具有自适应能力和记忆能力，使得该控制系统具有自适应能力和响应速度快的特点。
对于智能算法：GA和BPNN分别寻优PID参数，采用GA优化BPNN的权值，并进行了仿真试验。对于新型控制策略也进行了仿真验证。并在北京闻亭科技发展有限责任公司(WINTECH)所开发的TMS320F240评估板上进行了部分硬件仿真，都得到了很好的控制效果。表明了这种智能算法、新型控制策略是可行的、可靠的。
Genetic algorithm (GA) and Back Propagation neural network (BPNN) have a strong capacity of optimizing parameters. GA is able to optimize parameters in wide range, BPNN have good converge effect in part, but it is prone to run into apices in part, so, In this paper, combining them to optimize the parameters of PID Adjustor. In order to make PID Adjustor become better stability, more quick rate and higher precision.
Because the convergent rate of GA and BPNN is very slow, PID Adjustor based on GA-BPNN can't be widely used in real-time control system. In order to conquer the limitation, in the paper, the late-model control strategy is presented, which combines Off-line Roughly Adjust and On-line Precisely Adjust. In real-time adjust speed system, The Brush -less Direct Current Motor (BLDCM) is relatively fastness, so the mostly work is done in Off-line state, and PID Adjustor Based on trained BPNN is used to real-time control BLDCM. Because BPNN have capabilities of adaptive and memory, the control system have characters of adaptive and fast respond.
For the intelligent algorithm: GA and BPNN optimize parameters of PID Adjustor respectively, which is simulated and validated. And in the paper, it use GA to optimize the initial values of the power of BPNN. The late-model control strategy is also simulated and validated. Based on TMS320F240 EVM board tapped by WINTECH Co. The intelligent algorithm and the late-model control strategy is partly hardware simulation. These simulations get very good control effects, which shows that the intelligent algorithm and the late-model control strategy are feasible and reliable.
由于GA和BPNN的寻优时收敛速度慢，使得基于GA-BPNN的PID调节器在实际控制系统中的运用受到了限制。为了克服这个局限性，根据无刷直流电机在调速系统中的特点，本文提出一种新型的控制策略：“离线粗调”和“在线细调”相结合。让大部分工作放到离线状态来做，使得已训练好BPNN-PID调节器来在线对无刷直流电动机调速。由于BPNN具有自适应能力和记忆能力，使得该控制系统具有自适应能力和响应速度快的特点。
对于智能算法：GA和BPNN分别寻优PID参数，采用GA优化BPNN的权值，并进行了仿真试验。对于新型控制策略也进行了仿真验证。并在北京闻亭科技发展有限责任公司(WINTECH)所开发的TMS320F240评估板上进行了部分硬件仿真，都得到了很好的控制效果。表明了这种智能算法、新型控制策略是可行的、可靠的。
Genetic algorithm (GA) and Back Propagation neural network (BPNN) have a strong capacity of optimizing parameters. GA is able to optimize parameters in wide range, BPNN have good converge effect in part, but it is prone to run into apices in part, so, In this paper, combining them to optimize the parameters of PID Adjustor. In order to make PID Adjustor become better stability, more quick rate and higher precision.
Because the convergent rate of GA and BPNN is very slow, PID Adjustor based on GA-BPNN can't be widely used in real-time control system. In order to conquer the limitation, in the paper, the late-model control strategy is presented, which combines Off-line Roughly Adjust and On-line Precisely Adjust. In real-time adjust speed system, The Brush -less Direct Current Motor (BLDCM) is relatively fastness, so the mostly work is done in Off-line state, and PID Adjustor Based on trained BPNN is used to real-time control BLDCM. Because BPNN have capabilities of adaptive and memory, the control system have characters of adaptive and fast respond.
For the intelligent algorithm: GA and BPNN optimize parameters of PID Adjustor respectively, which is simulated and validated. And in the paper, it use GA to optimize the initial values of the power of BPNN. The late-model control strategy is also simulated and validated. Based on TMS320F240 EVM board tapped by WINTECH Co. The intelligent algorithm and the late-model control strategy is partly hardware simulation. These simulations get very good control effects, which shows that the intelligent algorithm and the late-model control strategy are feasible and reliable.
引文
[1] 王晓明．电动机的单片机控制．北京航空航天大学出版社，2002，5
[2] 李士勇．模糊控制·神经控制和智能控制．哈尔滨工业大学出版社，1996，4
[3] 袁曾仁．人工神经网络及其应用．清华大学出版社，1999，10
[4] P D J Lisboa. Neural networks current applications. International Thomson Computer Press, 1992
[5] 杨智明，王旭，庄显义．遗传算法在自动控制领域中的应用综述．信息与控制，2003，29(4)：329～339
[6] Robert H．Bishop．现代控制系统分析与设计——应用MATLAB和Simulink．清华大学出版社，2003，12
[7] 黄立培．电动机控制．清华大学出版社，2003，9
[8] 深圳市依托普林电力电子有限公司，www.powersave.com.cn
[9] 王强，李剑春，陆永平．无刷直流电动机系统数学模型及局部线性化传递函数．哈尔滨工业大学学报，1998，30(4)：84～88
[10] 韩力群．人工神经网络理论、设计及应用．化学工业出版社，2002，9
[11] Nielson, HR. Theory of the Backpropagation Neural network. IEEE UCNN, 1989, 1: 693～696.
[12] 王小明，曹立明．遗传算法—理论应用与软件实现．西安交通大学出版社，2002，1
[13] Lothar M. Schmit, Fundamental Study Theory of genetic algorithms, Theoretical Computer Science, 2001(259): 1～61
[14] 陶永华．新型PID控制及其应用系列讲座(第一讲——第六讲)．工业仪表与自动化装置，1997(4)：60～64，1997(5)：50～53，1997(6)：57～61，1998(1)：57～62，1998(2)：59～61，1998(3)：55～59
[15] S O Orero, M R Irving. A genetic algorithm for generator scheduling in power systems, Electrical Power & Energy Systems, 1996 (18): 19～26
[16] Goldberg D E. Genetic Algorithms in Search, Optimization, and Machine Learning. Reading, MA, Addison-Wisely, 1989
[17] 刘娜，韩璞．基于遗传算法的PID参数寻优．计算机仿真，2002(19)：70～73
[18] 康赐荣．用遗传算法(GA)实现PID控制．广东自动化与信息工程，2001(1)
[19] A.Blanco, M.Delgado, M.C.Pegalajar. A real-coded genetic algorithm for training recurrent neural networks, Neural Networks, 2001(14): 93～105
[20] 张彤，张华．浮点数编码的遗传算法及应用．哈尔滨工业大学学报，2002 (32)：59～61
[21] 辛香非，朱鳌鑫．遗传算法的适应度函数研究．系统工程与电子技术，1998(11)：58～62
[22] 侯宏霞，王涛，杨国清等．影响遗传算法PID参数优化性能的主要因素，2002，18(4)
[23] Manderick, de Weger M. The Genetic Algorithm and the Structure of the Fitness Landscape, In: Proceedings of the Fourth International Conference on Genentic Algorithms, CA, Morgan Kaufman, 1991: 143～150
[24] 殷铭，张兴华．基于MATLAB的遗传算法实现．电子技术应用2000(1)：9～11
[25] 刘国华，包宏．用MATLAB实现遗传算法程序．计算机应用研究，2001(18)：80～82
[26] Satyadas, Krishnakumar K. Genetic Algorithm Modules in MATLAB: Design and Implementation Using Software Engineering Pratice. Genetic Algorinthm in Engineerning and Computer Science, Winter G(ed).Wiley, 1995: 321～344
[27] 武志云．基于神经网络的PID控制器研究．内蒙古工业大学学报，2000(3)：234～237
[28] 宋滨．基于神经网络的PID控制．青岛大学学报，2001 16)：90～91
[29] Nan Jiang, Zhiye Zhao, Liqun Ren, Design of structural modular neural networks with genetic algorithm, Advances in Engineering Software, 2003(34): 17～24
[30] 卢洵，基于遗传算法的神经网络优化的研究，信息工程学院学报，1999(18)，18～20
[31] 周春光，张冰等，遗传算法及其在训练前向神经网络中的应用，小型微型计算机
系统，1996(17)，54～58
[32] Tammy Holter, Xiaoqiang Yao, Luis Carlos Rabelo, Albert Jones, Yuehwern Yih. Integration of Neural Networks and Genetic Algorithms for an Intelligent Manufacturing Controller. Computers and Engng. 1995 (29). 211～215
[33] Pavel Hajda, Vladimir Novotny, Xin Feng, Ruoli Yang. SIMPLE FEEDBACK LOGIC, GENETIC ALGORITHMS AND ARTIFICIAL NEURAL NERWORKS FOR REAL-TIME CONTROL OF A COLLECTION SYSTEM, Wat. Sci. Tech. 1998 (38), 187～195
[34] Jahangir Morshed & Jagath J. Kaluarachchi, Application of artificial neural network and genetic algorithm in flow and transport simulations, Advances in Water Resources, 1998(22), 145～158
[35] Fong-Chin Su, Wen-Lan Wu, Design and testing of a genetic algorithm neural network in the assessment of gait patterns, Medical Engineering & Physica, 2000 (22), 67～74
[36] Yas Abbs Alsultanny, Musbah M. Aqel, Pattern recognition using multiplayer neural-genetic algorithm, Neurocomputing, 2003(51), 237～247
[37] A. DUYGU ARBATLI and H. LEVENT AKIN, RULE EXTRACTION FROM TRAINED NEURAL NETWORKS USING GENETIC ALGORITHMS, Nonlinear Analysis, Theory, Methods&Application, 1997(30), 1639～1648
[38] 张光溢，郭前岗，电机学，重庆大学出版社，2002，1
[39] 邓吉，王学着，无刷电机控制技术综述，微特电机，2003(3)，11～13
[40] M. NAGATA, A. IWATA, PWM SIGNAL PROCESSING ARCHITECTURE FOR INTELLIGENT SYSTEMS, Computers Elect. Engng, 1977(23), 393～405
[41] 刘卫国，马瑞卿等，无刷直流电动机控制系统的单神经元控制器，西北工业大学学报，1999(17)
[42] Texas Instruments, TMS320C2xx User's Guide, Printed in U.S.A. by Custom Printing Company Owensville, Missouri, 1977
[43] 刘和平等，TMS320LF240x DSP结构、原理及应用，北京航空航天大学出版社，2002
[44] TMS320F24X高速数学信号处理器原理与应用，北京闻亭科技发展有限公司，2000
[45] Texas Instruments, TMS320C2xx C Source Debugger User's Guide, Printed in U.S.A. by Champion Press Houston, Texas, 1995
[46] Texas Instruments, TMS320C1x /C2x/C2xx/C5x Assembly Language Tools User's Guide, Printed in U.S.A. by Custom Printing Company Owensville, Missour, 1995
[47] 白雪石，杨华，基于DSP的无刷直流电动机控制系统，电气传动自动化，2002(24)，12～143
[48] 欧阳文等，基于DSP的无刷直流电动机的控制设计，微电机，2002(35)，29～31
[49] 郑乐，张晓峰等，基于DSP全数字多相无刷直流电动机控制系统，船电技术，2002 (2)，8～11
[50] 李川等，电机控制专用DSP讲座，微特电机，2001(5)，40～43
[51] 刘卫国，马瑞卿等，无刷直流电动机控制系统的单神经元控制器，西北工业大学学报，1999(17)，124～129
[52] Onuki Jun, Maenosono Tala-aki. ANN accelerator by parallel processor based on DSP. Proceedings of the International Joint Conference on Neural Networks V2 1993. IEEE 1913～1916
[53] 姜阳．基于数字信号处理器的神经元遗传算法自适应PID控制：[硕士学位论文]，2002
[54] Texas Instruments. Implementation of a Speed Controlled Brushless DC Drive Using TMS320F240.
[55] Texas Instruments. Implementation of a Sensorless Speed Controlled Brushless DC Drive Using TMS3240F240
[56] Texas Instruments. Creating a Pulse Width Modulated Signal with a Fixed Duty Cylcle Using the TMS320F240 EVM
[57] Texas Instruments. DSP Solutions for BLDC Motors
[58] Astrom, K. J. et al. Towards Intelligent PID control. Automatica, 1992, 28(1) 1～9
[2] 李士勇．模糊控制·神经控制和智能控制．哈尔滨工业大学出版社，1996，4
[3] 袁曾仁．人工神经网络及其应用．清华大学出版社，1999，10
[4] P D J Lisboa. Neural networks current applications. International Thomson Computer Press, 1992
[5] 杨智明，王旭，庄显义．遗传算法在自动控制领域中的应用综述．信息与控制，2003，29(4)：329～339
[6] Robert H．Bishop．现代控制系统分析与设计——应用MATLAB和Simulink．清华大学出版社，2003，12
[7] 黄立培．电动机控制．清华大学出版社，2003，9
[8] 深圳市依托普林电力电子有限公司，www.powersave.com.cn
[9] 王强，李剑春，陆永平．无刷直流电动机系统数学模型及局部线性化传递函数．哈尔滨工业大学学报，1998，30(4)：84～88
[10] 韩力群．人工神经网络理论、设计及应用．化学工业出版社，2002，9
[11] Nielson, HR. Theory of the Backpropagation Neural network. IEEE UCNN, 1989, 1: 693～696.
[12] 王小明，曹立明．遗传算法—理论应用与软件实现．西安交通大学出版社，2002，1
[13] Lothar M. Schmit, Fundamental Study Theory of genetic algorithms, Theoretical Computer Science, 2001(259): 1～61
[14] 陶永华．新型PID控制及其应用系列讲座(第一讲——第六讲)．工业仪表与自动化装置，1997(4)：60～64，1997(5)：50～53，1997(6)：57～61，1998(1)：57～62，1998(2)：59～61，1998(3)：55～59
[15] S O Orero, M R Irving. A genetic algorithm for generator scheduling in power systems, Electrical Power & Energy Systems, 1996 (18): 19～26
[16] Goldberg D E. Genetic Algorithms in Search, Optimization, and Machine Learning. Reading, MA, Addison-Wisely, 1989
[17] 刘娜，韩璞．基于遗传算法的PID参数寻优．计算机仿真，2002(19)：70～73
[18] 康赐荣．用遗传算法(GA)实现PID控制．广东自动化与信息工程，2001(1)
[19] A.Blanco, M.Delgado, M.C.Pegalajar. A real-coded genetic algorithm for training recurrent neural networks, Neural Networks, 2001(14): 93～105
[20] 张彤，张华．浮点数编码的遗传算法及应用．哈尔滨工业大学学报，2002 (32)：59～61
[21] 辛香非，朱鳌鑫．遗传算法的适应度函数研究．系统工程与电子技术，1998(11)：58～62
[22] 侯宏霞，王涛，杨国清等．影响遗传算法PID参数优化性能的主要因素，2002，18(4)
[23] Manderick, de Weger M. The Genetic Algorithm and the Structure of the Fitness Landscape, In: Proceedings of the Fourth International Conference on Genentic Algorithms, CA, Morgan Kaufman, 1991: 143～150
[24] 殷铭，张兴华．基于MATLAB的遗传算法实现．电子技术应用2000(1)：9～11
[25] 刘国华，包宏．用MATLAB实现遗传算法程序．计算机应用研究，2001(18)：80～82
[26] Satyadas, Krishnakumar K. Genetic Algorithm Modules in MATLAB: Design and Implementation Using Software Engineering Pratice. Genetic Algorinthm in Engineerning and Computer Science, Winter G(ed).Wiley, 1995: 321～344
[27] 武志云．基于神经网络的PID控制器研究．内蒙古工业大学学报，2000(3)：234～237
[28] 宋滨．基于神经网络的PID控制．青岛大学学报，2001 16)：90～91
[29] Nan Jiang, Zhiye Zhao, Liqun Ren, Design of structural modular neural networks with genetic algorithm, Advances in Engineering Software, 2003(34): 17～24
[30] 卢洵，基于遗传算法的神经网络优化的研究，信息工程学院学报，1999(18)，18～20
[31] 周春光，张冰等，遗传算法及其在训练前向神经网络中的应用，小型微型计算机
系统，1996(17)，54～58
[32] Tammy Holter, Xiaoqiang Yao, Luis Carlos Rabelo, Albert Jones, Yuehwern Yih. Integration of Neural Networks and Genetic Algorithms for an Intelligent Manufacturing Controller. Computers and Engng. 1995 (29). 211～215
[33] Pavel Hajda, Vladimir Novotny, Xin Feng, Ruoli Yang. SIMPLE FEEDBACK LOGIC, GENETIC ALGORITHMS AND ARTIFICIAL NEURAL NERWORKS FOR REAL-TIME CONTROL OF A COLLECTION SYSTEM, Wat. Sci. Tech. 1998 (38), 187～195
[34] Jahangir Morshed & Jagath J. Kaluarachchi, Application of artificial neural network and genetic algorithm in flow and transport simulations, Advances in Water Resources, 1998(22), 145～158
[35] Fong-Chin Su, Wen-Lan Wu, Design and testing of a genetic algorithm neural network in the assessment of gait patterns, Medical Engineering & Physica, 2000 (22), 67～74
[36] Yas Abbs Alsultanny, Musbah M. Aqel, Pattern recognition using multiplayer neural-genetic algorithm, Neurocomputing, 2003(51), 237～247
[37] A. DUYGU ARBATLI and H. LEVENT AKIN, RULE EXTRACTION FROM TRAINED NEURAL NETWORKS USING GENETIC ALGORITHMS, Nonlinear Analysis, Theory, Methods&Application, 1997(30), 1639～1648
[38] 张光溢，郭前岗，电机学，重庆大学出版社，2002，1
[39] 邓吉，王学着，无刷电机控制技术综述，微特电机，2003(3)，11～13
[40] M. NAGATA, A. IWATA, PWM SIGNAL PROCESSING ARCHITECTURE FOR INTELLIGENT SYSTEMS, Computers Elect. Engng, 1977(23), 393～405
[41] 刘卫国，马瑞卿等，无刷直流电动机控制系统的单神经元控制器，西北工业大学学报，1999(17)
[42] Texas Instruments, TMS320C2xx User's Guide, Printed in U.S.A. by Custom Printing Company Owensville, Missouri, 1977
[43] 刘和平等，TMS320LF240x DSP结构、原理及应用，北京航空航天大学出版社，2002
[44] TMS320F24X高速数学信号处理器原理与应用，北京闻亭科技发展有限公司，2000
[45] Texas Instruments, TMS320C2xx C Source Debugger User's Guide, Printed in U.S.A. by Champion Press Houston, Texas, 1995
[46] Texas Instruments, TMS320C1x /C2x/C2xx/C5x Assembly Language Tools User's Guide, Printed in U.S.A. by Custom Printing Company Owensville, Missour, 1995
[47] 白雪石，杨华，基于DSP的无刷直流电动机控制系统，电气传动自动化，2002(24)，12～143
[48] 欧阳文等，基于DSP的无刷直流电动机的控制设计，微电机，2002(35)，29～31
[49] 郑乐，张晓峰等，基于DSP全数字多相无刷直流电动机控制系统，船电技术，2002 (2)，8～11
[50] 李川等，电机控制专用DSP讲座，微特电机，2001(5)，40～43
[51] 刘卫国，马瑞卿等，无刷直流电动机控制系统的单神经元控制器，西北工业大学学报，1999(17)，124～129
[52] Onuki Jun, Maenosono Tala-aki. ANN accelerator by parallel processor based on DSP. Proceedings of the International Joint Conference on Neural Networks V2 1993. IEEE 1913～1916
[53] 姜阳．基于数字信号处理器的神经元遗传算法自适应PID控制：[硕士学位论文]，2002
[54] Texas Instruments. Implementation of a Speed Controlled Brushless DC Drive Using TMS320F240.
[55] Texas Instruments. Implementation of a Sensorless Speed Controlled Brushless DC Drive Using TMS3240F240
[56] Texas Instruments. Creating a Pulse Width Modulated Signal with a Fixed Duty Cylcle Using the TMS320F240 EVM
[57] Texas Instruments. DSP Solutions for BLDC Motors
[58] Astrom, K. J. et al. Towards Intelligent PID control. Automatica, 1992, 28(1) 1～9