基于思维进化算法的单相电机矢量控制系统
|
摘要
随着电力事业的发展,人们生活水平的提高,对家用电器如冰箱、洗衣机、空调等中的电动机的应用提出了越来越高的要求——要求其向着节能化、智能化的方向发展,于是,变频的概念越来越多地出现在普通百姓的日常生活中。变频方式的出现将避免电动机压缩电机的频繁启动,减小对电网的冲击并延长压缩机的使用寿命。然而,在目前的情况下,对家用电器中较多采用的单相异步电机的相关控制研究还较少。
并且,随着电力电子技术和数字控制技术的发展,交流传动取代直流传动已经成为不可逆转的趋势。与此同时,异步电机系统还存在一定的问题,它是一个多变量、非线性强耦合的时变参数系统,虽然矢量控制技术在一定程度上能使异步电动机得以解耦控制,但是这并不能改变其非线性且具有时变参数的特点。况且,在交流调速系统实际运行过程中不可预测的干扰很多,所以交流调速系统中调节器的设计一直是人们研究的焦点。虽然,到目前为止,对于三相交流异步电机的变频控制已经较为成熟,但是对于单相电机,转速的调节主要采用调节端电压和改变电机极对数的方法。在这两种方法中,前者只适用于小功率低力矩场合例如电风扇等,后者则不能实现无级调速。事实上,单相异步电动机的结构与三相异步电动机较为相似,因而与三相电机一样,使得变频成为这种电机转速调节最为有效的方法。论文基于以上原因,在查阅相关文献和借鉴前人工作经验的基础上,以将单相异步电机在模型上等效为直流电动机为目的,仔细研究了单相电机物理模型和数学模型,通过两相静止坐标系/两相旋转坐标系的变换和直角坐标系/极坐标系的变换,整理出单相电机的运动方程、转矩方程、电压方程等,作为下一步系统设计的基础。
另外,涉及到系统中转矩调节器和转速调节器的设计时的参数,常见的应用在三相电机上的有工程整定法、试凑法和经验值法等。在本文中,综合考虑了工程整定法和经验值等方面的不足,首先用工程整定的方法算出参数的初始值,然后,在详细学习和考量了各种优化算法的特点和优缺点的基础上,由于思维进化算法收敛速度快、定向机制、不受函数条件的限制等良好特点,最终选用它作为调节器参数优化的方法。在给出α=0.6的解空间范围内,即在初始值±60%的范围内,采用思维进化算法的趋同开采和异化勘探策略,实现了设计中参数优化的功能。
在实验仿真阶段,分两个步骤进行,首先,关于调节器的思维进化算法优化部分是在VC平台,用C语言来实现,得出优化好的参数结果;其次,在simulink仿真阶段,搭建系统框图,将各设计参变量代入到系统中,运行得到scope图。
With the development of power industry and the improvement of living standards of human being, more and more requirements about motors for home-appliances such as refrigerators, washing machines, air- conditionings and so on are put forward , which are developed towards energy-saving and intelligent direction. This method will avoid compressor of motor from starting frequently,reduce strike to power and extend compressor life. Then, the concept of varial-frequency appears in the daily life of ordinary people.In present, the research on single-phase motor control is less .
Moreover, with the development of power-electronic-technology and digital control technology, the trend of motor applications is that AC drive will replace the DC drive. At the same time, there are some problems with asynchronous motor. It is a nonlinear and time-change system. It has multiple variables and the personality of strong gearing. Although vector control can solve the problem of gearing to some degree, the system is still nonlinear and time-change. What’s more, during the real running course, there are many disturbance factors. So, the designing of regulators in AC adjusting speed system is the focus of researches so far. So far, although variable frequency control of three-phase AC induction motor has been more mature, for the single-phase motors,main methods of speed regulation are change voltage and the number of poles. The former is only applicable to low-power and low-torque circumstances such as fans, and the latter can not stepless speed regulation.In fact, the structure of single-phase asynchronous motor and three-phase asynchronous motor is similar. Therefore, the variable-frequency becomes the most effective method just as three-phase motor. Because of above reasons , on the base of studying relative documents and referring the experience of ancestors, I studyed carefully the single-phase motor physical and mathematical models for the purpose of equivalent to the DC motor in the model.Through the two-phase static coordinate system / two-phase rotating coordinate system and the cartesian coordinate system transformation / polar coordinates transformat- ion , the equation of motion, torque equation, voltage equation and so on was sorted out.It is the basis of the next step in the system design. In addition, there are several methods to decide parameters of ASR and ATR .
In this paper, considering the faults of project-method and experience values, first of all, used the project-method to initialize the parameters, then, chose MEA to optimize them due to its fast speed, targeted mechanisms and not subject to conditions, after a detaile study and consideration of a variety of optimization algorithms based on their advantages and disadvantages . Solution space is the given asα=0.6. With the mining and explorating strategies of MEA parameters was optimized.
In the simulation phase of the experiment,first , parameters of ASR and ATR was optimized in VC with C language using MEA. Then run simulink to get figures.
并且,随着电力电子技术和数字控制技术的发展,交流传动取代直流传动已经成为不可逆转的趋势。与此同时,异步电机系统还存在一定的问题,它是一个多变量、非线性强耦合的时变参数系统,虽然矢量控制技术在一定程度上能使异步电动机得以解耦控制,但是这并不能改变其非线性且具有时变参数的特点。况且,在交流调速系统实际运行过程中不可预测的干扰很多,所以交流调速系统中调节器的设计一直是人们研究的焦点。虽然,到目前为止,对于三相交流异步电机的变频控制已经较为成熟,但是对于单相电机,转速的调节主要采用调节端电压和改变电机极对数的方法。在这两种方法中,前者只适用于小功率低力矩场合例如电风扇等,后者则不能实现无级调速。事实上,单相异步电动机的结构与三相异步电动机较为相似,因而与三相电机一样,使得变频成为这种电机转速调节最为有效的方法。论文基于以上原因,在查阅相关文献和借鉴前人工作经验的基础上,以将单相异步电机在模型上等效为直流电动机为目的,仔细研究了单相电机物理模型和数学模型,通过两相静止坐标系/两相旋转坐标系的变换和直角坐标系/极坐标系的变换,整理出单相电机的运动方程、转矩方程、电压方程等,作为下一步系统设计的基础。
另外,涉及到系统中转矩调节器和转速调节器的设计时的参数,常见的应用在三相电机上的有工程整定法、试凑法和经验值法等。在本文中,综合考虑了工程整定法和经验值等方面的不足,首先用工程整定的方法算出参数的初始值,然后,在详细学习和考量了各种优化算法的特点和优缺点的基础上,由于思维进化算法收敛速度快、定向机制、不受函数条件的限制等良好特点,最终选用它作为调节器参数优化的方法。在给出α=0.6的解空间范围内,即在初始值±60%的范围内,采用思维进化算法的趋同开采和异化勘探策略,实现了设计中参数优化的功能。
在实验仿真阶段,分两个步骤进行,首先,关于调节器的思维进化算法优化部分是在VC平台,用C语言来实现,得出优化好的参数结果;其次,在simulink仿真阶段,搭建系统框图,将各设计参变量代入到系统中,运行得到scope图。
With the development of power industry and the improvement of living standards of human being, more and more requirements about motors for home-appliances such as refrigerators, washing machines, air- conditionings and so on are put forward , which are developed towards energy-saving and intelligent direction. This method will avoid compressor of motor from starting frequently,reduce strike to power and extend compressor life. Then, the concept of varial-frequency appears in the daily life of ordinary people.In present, the research on single-phase motor control is less .
Moreover, with the development of power-electronic-technology and digital control technology, the trend of motor applications is that AC drive will replace the DC drive. At the same time, there are some problems with asynchronous motor. It is a nonlinear and time-change system. It has multiple variables and the personality of strong gearing. Although vector control can solve the problem of gearing to some degree, the system is still nonlinear and time-change. What’s more, during the real running course, there are many disturbance factors. So, the designing of regulators in AC adjusting speed system is the focus of researches so far. So far, although variable frequency control of three-phase AC induction motor has been more mature, for the single-phase motors,main methods of speed regulation are change voltage and the number of poles. The former is only applicable to low-power and low-torque circumstances such as fans, and the latter can not stepless speed regulation.In fact, the structure of single-phase asynchronous motor and three-phase asynchronous motor is similar. Therefore, the variable-frequency becomes the most effective method just as three-phase motor. Because of above reasons , on the base of studying relative documents and referring the experience of ancestors, I studyed carefully the single-phase motor physical and mathematical models for the purpose of equivalent to the DC motor in the model.Through the two-phase static coordinate system / two-phase rotating coordinate system and the cartesian coordinate system transformation / polar coordinates transformat- ion , the equation of motion, torque equation, voltage equation and so on was sorted out.It is the basis of the next step in the system design. In addition, there are several methods to decide parameters of ASR and ATR .
In this paper, considering the faults of project-method and experience values, first of all, used the project-method to initialize the parameters, then, chose MEA to optimize them due to its fast speed, targeted mechanisms and not subject to conditions, after a detaile study and consideration of a variety of optimization algorithms based on their advantages and disadvantages . Solution space is the given asα=0.6. With the mining and explorating strategies of MEA parameters was optimized.
In the simulation phase of the experiment,first , parameters of ASR and ATR was optimized in VC with C language using MEA. Then run simulink to get figures.
引文
[1]陈光东,郝晓田.单相电容电机变频调速的逆变结构[J].电气传动,1997年第3期:15～17
[2]陈伯时.电力拖动自动控制系统[M],第2版.北京:机械工业出版社,1999.10
[3]陈伯时.电力拖动自动控制系统-运动控制系统[M],第3版.北京:机械工业出版社, 2004.4
[4]潘晓晟.郝世勇.MATLAB电机仿真精华50例,北京:电子工业出版社,2007.7
[5]王俊丽.思维进化计算——搜索算法的开发和算法性能的分析[学位论文].太原:太原理工大学.2003.4
[6]周旭东.采用思维进化计算求解最大团问题[学位论文].太原:太原理工大学.2004.4
[7]介婧.曾建潮.单群体思维进化算法[J].2000中国控制与决策学术年会论文集:245~249
[8]谢克明.杜永贵.孙承意.基于思维进化机器学习算法在水泥生料配比中的应用[J].Proceeding of the 3rd World Congress on Intelligent Control and Automation.132~134
[9]方崇智,萧德云.过程辨识[M].北京:清华大学出版社,1988.8.
[10]孙承意.思维进化计算的产生与进展[J], Hefei:China Academic Journal Electronic Publishing House,1994-2008.
[11] Holland J. H., Adaptation in natural and artificial systems[J], Ann Arbor: University of Michigan Press, 1975.
[12] Bagley J. D., The behavior of adaptive systems which employ genetic and correlation algorithms (Doctoral dissertation, University of Michigan) [J], Dissertation Abstracts International, 28 (12), 5106B, 1967.
[13] Wei Lijun, Sun Yan and Sun Chengyi, A more efficient similartaxis strategy of MEBML, in Proc. Of IASTED Inter. Conf. on Modeling and Simulation (IASTED MS'99) [J], pp.l-5, May 5-8, 1999.
[14]刘金琨.先进PID控制及其MATLAB仿真[M].北京:电子工业出版社,2003.1
[15]孙承意,王皖贞,贾鸿雁,思维进化计算的产生与进展[J],昆明:2001年中国智能自动化会议(CIAC2001). 80-86, 2001年8月13-16.
[16]曾建潮,查凯.基于思维进化机器学习的异化策略研[J], Hefei:Proceeding of the 3rd World Congress on Intelligent Control and Automation.129-131.
[17]杨爱军.交流异步电动机的矢量控制[J].船舶电子对抗, 2004.6, 27(3):36～38 [13]
[18]H.Can, E.Akin. Neural network-based stator voltage compensator for low-frequency operation of a vector-controlled induction motor drive[J].Electrical Engineering, 2002.2, 32(8):287～293
[19]B.karanayil, M.F.Ranhman, C.Granthan. Investigation of an on-line rotor resistance identification with a new stator resistance observer for induction motor drive using Artificial Neural-Networks[J]. IEEE,2003:1883～1888
[20]江建辉.基于矢量控制原理的异步电动机调速系统的研究与设计. [学位论文].西安:西北工业大学,2003.3
[21]李士勇.模糊控制·神经控制和智能控制理论[M],第2版.哈尔滨:哈尔滨工业大学出版社,1998.9
[22]B.karanayil, M.F.Ranhman, C.Granthan. Rotor Resistance Identification using Artificial Neural Networks for a Speed Sensorless Vector Controlled Induction Motor Drive[J]. IEEE, 2003: 419～424 [13]
[23]Chen Zhimei, Zhang Jinggang, Zeng Jingchao. A New Method of Sliding Model Control and Application to AC Servo System[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:759～762
[24]赵立新,丁筱玲,刘双喜.转速、转子电流双闭环控制变频调速器[J].农业机械报, 2002.3, 33(2):135～137
[25]郑朝科,唐顺华.电力拖动基础[M].上海:同济大学出版社,1996.5
[26]王晓英,罗文广.基于MATLAB/SIMULINK的异步电动机矢量控制调速系统仿真[J].东北电力技术,2004(1):14～16
[27]林峰.新型无速度传感器矢量控制交流调速系统[J].浙江大学学报,2001.1,35(1):67～71
[28]王东兴,庄斌,苏剑霞.用多种控制策略实现矢量变频调速系统的速度调节[J].电网技术学报,2002.6, 26(6): 65～67
[29]B.karanayil, M.F.Ranhman, C.Granthan. Rotor Resistance Identification using Artificial Neural Networks for an Indirect Vector Controlled Induction Motor Drive[J]. IEEE, 2001:1315～1320
[30]周熙炜.基于扩展卡尔曼滤波算法的异步电机参数辨识. [学位论文].西安:西安理工大学,2003.3
[31]孙宇新,王纪俊,刘贤兴.交流电机矢量控制中的转子磁链辨识的神经网络实现[J].中小型电机,2003.4,30(4):4～7
[32]刘勇,康立山,陈毓屏.非数值并行算法(第二册)遗传算法[M].北京:科学出版社,1995.1
[33]洪乃刚.矢量控制交流调速系统的研究[J].安徽工业大学学报,2004.4,21(2):124～127
[34]Liang Zhonghua, Hu Qing, Yu Haiyan etc. Application of Genetic Algorithms in Sensorless Speed Vector-Controlled AC Adjusting System[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:1286～1288
[35]周玲玲,侯立军,苏彦民.异步电机调速系统建模与仿真研究[J].微电机,2004,37(1):33～37
[36]Luis Zorzano Martinez, Antonio Zorzano Martinez. IDENTIFICATION OF INDUCTION MACHINES USING ARTIFICIAL NEURAL NETWORKS[C]. ISIE’97:1259～1264
[37]郭瑞.异步电动机自适应矢量控制系统的设计与仿真. [学位论文].辽宁:辽宁工程技术大学,2002.3
[38]Yang Wenqiang, Jia Zhengchun, Xu Qiang. A New Algorithm for Flux and Speed Estimation in Induction Machine[C]. Shenyang University of Technology,ICEMS.Beijing, 2001:698～701
[39]徐小增,李叶松,秦忆.基于参数辨识的磁链定向感应电机自适应控制[J].中山大学学报(自然科学版),2002.11,41(6):26～29。
[40]雷华.感应电机的神经网络速度估计方法研究.[学位论文].重庆:重庆大学,2003.5
[41]Zheng Ping. Study on the Impact of Parameter Variations and Relevant Compensation Based of Sensorless Rotor Flux Oriented Induction Machines[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:1308～1312
[42]B.karanayil, M.F.Ranhman, C.Granthan. On-line Rotor Resistance Identification for Induction Motor Drive with Artificial Neural Networks Supported by a Simple PI Stator Resistance Estimator[J]. IEEE, 2003:433～438
[43]B.karanayil, M.F.Ranhman, C.Granthan. Stator and rotor resistance observer for induction motor drive using Fuzzy Logic and Artificial Neural Networks[J].IEEE,2003:124～131
[44]冬雷.矢量控制中感应电动机转子电阻的自适应辨识[J].电工技术学报,2002.8,17(4):39～44
[45]Abdelgham Zergaoui, Abdelhak Bennia. IDENTIFICATION AND CONTROL OF ANASYNCHRONOUS MACHINE USING NEURAL NETWORKS[J]. IEEE, 1999:1043～1046
[46]Alessandro Goedtel . Load Torque Estimation in Induction Motors Using Artificial Neural Networks[J].IEEE,2002:1379～1384
[47]Chich-Yi Huang, Tien-Chi Chen, Ching-Lien Huang. Robust Control of Induction Motor with A Neural-Network Load Torque Estimator and A Neural-Network Identification[J]. IEEE,1999:990～998
[48] Abdelgham Zergaoui, Abdelhak Bennia. IDENTIFICATION AND CONTROL OF ANASYNCHRONOUS MACHINE USING NEURAL NETWORKS[J]. IEEE, 1999:1043～1046.
[2]陈伯时.电力拖动自动控制系统[M],第2版.北京:机械工业出版社,1999.10
[3]陈伯时.电力拖动自动控制系统-运动控制系统[M],第3版.北京:机械工业出版社, 2004.4
[4]潘晓晟.郝世勇.MATLAB电机仿真精华50例,北京:电子工业出版社,2007.7
[5]王俊丽.思维进化计算——搜索算法的开发和算法性能的分析[学位论文].太原:太原理工大学.2003.4
[6]周旭东.采用思维进化计算求解最大团问题[学位论文].太原:太原理工大学.2004.4
[7]介婧.曾建潮.单群体思维进化算法[J].2000中国控制与决策学术年会论文集:245~249
[8]谢克明.杜永贵.孙承意.基于思维进化机器学习算法在水泥生料配比中的应用[J].Proceeding of the 3rd World Congress on Intelligent Control and Automation.132~134
[9]方崇智,萧德云.过程辨识[M].北京:清华大学出版社,1988.8.
[10]孙承意.思维进化计算的产生与进展[J], Hefei:China Academic Journal Electronic Publishing House,1994-2008.
[11] Holland J. H., Adaptation in natural and artificial systems[J], Ann Arbor: University of Michigan Press, 1975.
[12] Bagley J. D., The behavior of adaptive systems which employ genetic and correlation algorithms (Doctoral dissertation, University of Michigan) [J], Dissertation Abstracts International, 28 (12), 5106B, 1967.
[13] Wei Lijun, Sun Yan and Sun Chengyi, A more efficient similartaxis strategy of MEBML, in Proc. Of IASTED Inter. Conf. on Modeling and Simulation (IASTED MS'99) [J], pp.l-5, May 5-8, 1999.
[14]刘金琨.先进PID控制及其MATLAB仿真[M].北京:电子工业出版社,2003.1
[15]孙承意,王皖贞,贾鸿雁,思维进化计算的产生与进展[J],昆明:2001年中国智能自动化会议(CIAC2001). 80-86, 2001年8月13-16.
[16]曾建潮,查凯.基于思维进化机器学习的异化策略研[J], Hefei:Proceeding of the 3rd World Congress on Intelligent Control and Automation.129-131.
[17]杨爱军.交流异步电动机的矢量控制[J].船舶电子对抗, 2004.6, 27(3):36～38 [13]
[18]H.Can, E.Akin. Neural network-based stator voltage compensator for low-frequency operation of a vector-controlled induction motor drive[J].Electrical Engineering, 2002.2, 32(8):287～293
[19]B.karanayil, M.F.Ranhman, C.Granthan. Investigation of an on-line rotor resistance identification with a new stator resistance observer for induction motor drive using Artificial Neural-Networks[J]. IEEE,2003:1883～1888
[20]江建辉.基于矢量控制原理的异步电动机调速系统的研究与设计. [学位论文].西安:西北工业大学,2003.3
[21]李士勇.模糊控制·神经控制和智能控制理论[M],第2版.哈尔滨:哈尔滨工业大学出版社,1998.9
[22]B.karanayil, M.F.Ranhman, C.Granthan. Rotor Resistance Identification using Artificial Neural Networks for a Speed Sensorless Vector Controlled Induction Motor Drive[J]. IEEE, 2003: 419～424 [13]
[23]Chen Zhimei, Zhang Jinggang, Zeng Jingchao. A New Method of Sliding Model Control and Application to AC Servo System[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:759～762
[24]赵立新,丁筱玲,刘双喜.转速、转子电流双闭环控制变频调速器[J].农业机械报, 2002.3, 33(2):135～137
[25]郑朝科,唐顺华.电力拖动基础[M].上海:同济大学出版社,1996.5
[26]王晓英,罗文广.基于MATLAB/SIMULINK的异步电动机矢量控制调速系统仿真[J].东北电力技术,2004(1):14～16
[27]林峰.新型无速度传感器矢量控制交流调速系统[J].浙江大学学报,2001.1,35(1):67～71
[28]王东兴,庄斌,苏剑霞.用多种控制策略实现矢量变频调速系统的速度调节[J].电网技术学报,2002.6, 26(6): 65～67
[29]B.karanayil, M.F.Ranhman, C.Granthan. Rotor Resistance Identification using Artificial Neural Networks for an Indirect Vector Controlled Induction Motor Drive[J]. IEEE, 2001:1315～1320
[30]周熙炜.基于扩展卡尔曼滤波算法的异步电机参数辨识. [学位论文].西安:西安理工大学,2003.3
[31]孙宇新,王纪俊,刘贤兴.交流电机矢量控制中的转子磁链辨识的神经网络实现[J].中小型电机,2003.4,30(4):4～7
[32]刘勇,康立山,陈毓屏.非数值并行算法(第二册)遗传算法[M].北京:科学出版社,1995.1
[33]洪乃刚.矢量控制交流调速系统的研究[J].安徽工业大学学报,2004.4,21(2):124～127
[34]Liang Zhonghua, Hu Qing, Yu Haiyan etc. Application of Genetic Algorithms in Sensorless Speed Vector-Controlled AC Adjusting System[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:1286～1288
[35]周玲玲,侯立军,苏彦民.异步电机调速系统建模与仿真研究[J].微电机,2004,37(1):33～37
[36]Luis Zorzano Martinez, Antonio Zorzano Martinez. IDENTIFICATION OF INDUCTION MACHINES USING ARTIFICIAL NEURAL NETWORKS[C]. ISIE’97:1259～1264
[37]郭瑞.异步电动机自适应矢量控制系统的设计与仿真. [学位论文].辽宁:辽宁工程技术大学,2002.3
[38]Yang Wenqiang, Jia Zhengchun, Xu Qiang. A New Algorithm for Flux and Speed Estimation in Induction Machine[C]. Shenyang University of Technology,ICEMS.Beijing, 2001:698～701
[39]徐小增,李叶松,秦忆.基于参数辨识的磁链定向感应电机自适应控制[J].中山大学学报(自然科学版),2002.11,41(6):26～29。
[40]雷华.感应电机的神经网络速度估计方法研究.[学位论文].重庆:重庆大学,2003.5
[41]Zheng Ping. Study on the Impact of Parameter Variations and Relevant Compensation Based of Sensorless Rotor Flux Oriented Induction Machines[C]. Shenyang University of Technology, ICEMS. Beijing, 2001:1308～1312
[42]B.karanayil, M.F.Ranhman, C.Granthan. On-line Rotor Resistance Identification for Induction Motor Drive with Artificial Neural Networks Supported by a Simple PI Stator Resistance Estimator[J]. IEEE, 2003:433～438
[43]B.karanayil, M.F.Ranhman, C.Granthan. Stator and rotor resistance observer for induction motor drive using Fuzzy Logic and Artificial Neural Networks[J].IEEE,2003:124～131
[44]冬雷.矢量控制中感应电动机转子电阻的自适应辨识[J].电工技术学报,2002.8,17(4):39～44
[45]Abdelgham Zergaoui, Abdelhak Bennia. IDENTIFICATION AND CONTROL OF ANASYNCHRONOUS MACHINE USING NEURAL NETWORKS[J]. IEEE, 1999:1043～1046
[46]Alessandro Goedtel . Load Torque Estimation in Induction Motors Using Artificial Neural Networks[J].IEEE,2002:1379～1384
[47]Chich-Yi Huang, Tien-Chi Chen, Ching-Lien Huang. Robust Control of Induction Motor with A Neural-Network Load Torque Estimator and A Neural-Network Identification[J]. IEEE,1999:990～998
[48] Abdelgham Zergaoui, Abdelhak Bennia. IDENTIFICATION AND CONTROL OF ANASYNCHRONOUS MACHINE USING NEURAL NETWORKS[J]. IEEE, 1999:1043～1046.