自适应控制在供水系统中的应用研究
|
摘要
摘 要
本文针对变频恒压供水系统中控制对象模型难于精确建立以及水泵电机驱
动电源切换控制中的问题，提出了自适应逆控制和自适应模糊控制策略以及锁相
环同步切换最优控制方案，并进行了基于 DSP 控制器的模拟供水系统调试以及基
于 Matlab 环境下的锁相同步控制仿真，结果证明了控制策略的有效性，明显改
善了供水系统的控制性能。
由于供水系统的数学模型具有严重的非线性和时变性，无法用线性化方法近
似，传统 PID 控制器的参数很难调整、适应性差。针对此问题，本文设计了自适
应逆控制器。自适应逆控制通过在线辨识对象模型以获取对象的运行状态，然后
以此运行状态为信息，用自适应算法修改控制器参数，从而使控制器实时跟踪被
控对象的变化，实施正确的控制。论文中给出了在模拟供水实验装置上的调试波
形，可看出显著提高了供水系统水压的控制精度。由于非线性自适应逆控制算法
采用基于递归神经网络的 NARX 滤波器实施建模与控制，计算量较大，需要运算
速度较高的微处理器，使得硬件成本较高。为此又设计了自适应模糊控制器，获
得了较好的控制性能。该方法不需要辨识被控对象的模型，因而计算量小，有利
于系统的实时控制，及降低设备的成本，便于推广应用。
本文针对供水系统中的另一问题，即：在变频供水系统中多台水泵并联运行，
变频电源驱动的水泵电机切换到工频电源时，由于水泵电机的反电动势与工频电
源在频率和相位上的不确定性，可能产生很强的过电流，会导致系统的自动开关
跳闸，并对电器元件产生强烈的冲击。采用锁相同步切换方式，可显著减小冲击
电流。但是一般的锁相环采用 PI 控制，响应速度慢，切换过程长。针对此问题，
本文提出时间最优控制及基于趋近率的变结构控制策略，并对锁相环捕获过程进
行了仿真研究，给出了 Matlab 环境下的仿真波形。结果表明变结构控制在捕获
时间和稳态精度的综合性能上优于 PI 控制。
This paper deals with the problems in water supply system of constant pressure,
where the accurate plant model is difficult to get and the power supplies of the pump
need to switch. Those problems are resolved by adaptive inverse control, adaptive
fuzzy control and synchronous transfer based on phase locking, respectively. The DSP
hardware system of constant pressure control is tested on the analog of water supply
system and the synchronous control of phase locking is simulated in Matlab
environment. The findings prove the validity of the control strategies that improve the
performance of the water supply system.
The model of the water supply system is intrinsic nonlinear and time variant, which
makes it difficult to linearize the plant and tune the PID controller. Thus the
traditional PID controller cannot adapt to the changes. To tackle this problem, we
have designed an adaptive inverse controller and get high precise control, which can
be seen from the pressure waveforms of the experimental equipment. In this case, the
run time states of the plant is achieved by the identification procedure and this
information is used to update the parameters of the controller by adaptive algorithm
that makes the controller can feed right control to the plant as the states changed.
However, the nonlinear adaptive inverse control uses NARX filter as plant model and
controller, which makes the computational complexity of the algorithm increase and
the cost of the hardware become high. So we also designed the adaptive fuzzy
controller that has no need to identify the model of the plant. It makes the
computational complexity decrease, which helps the real time processing, reduce the
cost and simplify the implementation.
Another issue in water supply system is the parallel operation of multiple pumps.
The main problem is, while the pump driven by the VFD is switched to the power
lines, large surge transient currents may be produced by the frequency and phase
difference between the power supply and the counter electromotive force of the motor,
which will result in the failure of the automatic switches and the damage of the
II
摘 要
electric equipments. Synchronous transfer based on phase locking is an advanced
switch mode that can reduce the surge transient currents. However, the general
phase-locked loop adopts PI control, which makes the response become slow and the
switch process long. To avoid this problem, this paper employs time optimal control
and variable structure control by the reaching low method. The simulation of the
capture process, by Matlab, shows that the variable structure control is superior to PI
control both in capture time and steady state performance.
本文针对变频恒压供水系统中控制对象模型难于精确建立以及水泵电机驱
动电源切换控制中的问题，提出了自适应逆控制和自适应模糊控制策略以及锁相
环同步切换最优控制方案，并进行了基于 DSP 控制器的模拟供水系统调试以及基
于 Matlab 环境下的锁相同步控制仿真，结果证明了控制策略的有效性，明显改
善了供水系统的控制性能。
由于供水系统的数学模型具有严重的非线性和时变性，无法用线性化方法近
似，传统 PID 控制器的参数很难调整、适应性差。针对此问题，本文设计了自适
应逆控制器。自适应逆控制通过在线辨识对象模型以获取对象的运行状态，然后
以此运行状态为信息，用自适应算法修改控制器参数，从而使控制器实时跟踪被
控对象的变化，实施正确的控制。论文中给出了在模拟供水实验装置上的调试波
形，可看出显著提高了供水系统水压的控制精度。由于非线性自适应逆控制算法
采用基于递归神经网络的 NARX 滤波器实施建模与控制，计算量较大，需要运算
速度较高的微处理器，使得硬件成本较高。为此又设计了自适应模糊控制器，获
得了较好的控制性能。该方法不需要辨识被控对象的模型，因而计算量小，有利
于系统的实时控制，及降低设备的成本，便于推广应用。
本文针对供水系统中的另一问题，即：在变频供水系统中多台水泵并联运行，
变频电源驱动的水泵电机切换到工频电源时，由于水泵电机的反电动势与工频电
源在频率和相位上的不确定性，可能产生很强的过电流，会导致系统的自动开关
跳闸，并对电器元件产生强烈的冲击。采用锁相同步切换方式，可显著减小冲击
电流。但是一般的锁相环采用 PI 控制，响应速度慢，切换过程长。针对此问题，
本文提出时间最优控制及基于趋近率的变结构控制策略，并对锁相环捕获过程进
行了仿真研究，给出了 Matlab 环境下的仿真波形。结果表明变结构控制在捕获
时间和稳态精度的综合性能上优于 PI 控制。
This paper deals with the problems in water supply system of constant pressure,
where the accurate plant model is difficult to get and the power supplies of the pump
need to switch. Those problems are resolved by adaptive inverse control, adaptive
fuzzy control and synchronous transfer based on phase locking, respectively. The DSP
hardware system of constant pressure control is tested on the analog of water supply
system and the synchronous control of phase locking is simulated in Matlab
environment. The findings prove the validity of the control strategies that improve the
performance of the water supply system.
The model of the water supply system is intrinsic nonlinear and time variant, which
makes it difficult to linearize the plant and tune the PID controller. Thus the
traditional PID controller cannot adapt to the changes. To tackle this problem, we
have designed an adaptive inverse controller and get high precise control, which can
be seen from the pressure waveforms of the experimental equipment. In this case, the
run time states of the plant is achieved by the identification procedure and this
information is used to update the parameters of the controller by adaptive algorithm
that makes the controller can feed right control to the plant as the states changed.
However, the nonlinear adaptive inverse control uses NARX filter as plant model and
controller, which makes the computational complexity of the algorithm increase and
the cost of the hardware become high. So we also designed the adaptive fuzzy
controller that has no need to identify the model of the plant. It makes the
computational complexity decrease, which helps the real time processing, reduce the
cost and simplify the implementation.
Another issue in water supply system is the parallel operation of multiple pumps.
The main problem is, while the pump driven by the VFD is switched to the power
lines, large surge transient currents may be produced by the frequency and phase
difference between the power supply and the counter electromotive force of the motor,
which will result in the failure of the automatic switches and the damage of the
II
摘 要
electric equipments. Synchronous transfer based on phase locking is an advanced
switch mode that can reduce the surge transient currents. However, the general
phase-locked loop adopts PI control, which makes the response become slow and the
switch process long. To avoid this problem, this paper employs time optimal control
and variable structure control by the reaching low method. The simulation of the
capture process, by Matlab, shows that the variable structure control is superior to PI
control both in capture time and steady state performance.
引文
1 金传伟，毛宗源. 变频调速技术在水泵控制系统中的应用. 电子技术应用. 2000，（9）:
38-39，44
2 胡纲衡，唐瑞球. 交流变频调速的切换控制技术. 电工技术杂志. 2001，（6）:43-45
3 孙 斌，张宏建. 模糊变频恒压供水系统设计. 电气自动化. 2002，（3）:23-25
4 徐国忠，诸静. 单片机模糊控制恒压供水系统的研究. 微电子学与计算机. 1998，（4）:8-12
5 戴广平. 电动机变频器与电力拖动. 中国石化出版社. 1999: 89-106
6 Marek Hartman, Eugeniusz Loeiec, Piotr Boguslawski. Water Pumping Stations Remote Control
System in Depression Areas Based GSM-900 Cellular Communications System. TCSET2002,
Lviv-Slavsko,Ukraine, 2002
7 Hiromitsu Kato,Hiromitsu Kurisu, Teruji Sekozawa.Interactive Multiple Objective Decision
Method for Water Supply Scheduling in Hybrid Network Models. JAPAN
8 Frank A.Dewinter, The Application of a 3500 HP Variable Frequency Drive for Pipeline Pump
Control. IEEE Transactions on Industry Applications. 1989,25(6):1019-1024
9 M. Elbelkacemi, A. Lachhab, M. Limouri, B. Dahhou, A. Essaid. Adaptive Control of a Water
Supply System. Control Engineering Practice. 2001, 9(3):343-349
10 Bryan Coulbeck, Bogumil Ulanicki, Vladimir V.Deviatkov, Sergei Kosov, Iosiff Glukhovsky.
Pressure Control of a Moscow Water Supply System using Expert System Technology.
Proceedings of the 35th IEEE Conference on Decision and Control, Kobe, Japan, December
1996, (4):4498 - 4499
11 B.Widrow, E.Walach 著，自适应逆控制.刘树堂，韩崇昭 译. 西安交通大学出版社，1999
12 Bernard Widrow, Gregory L.Pett. Adaptive Inverse Control Based on Linear and Nonlinear
Adaptive Filtering. Neural Networks for Identification, Control, Robotics, and Signal/Image
1996. Proceedings. IEEE International Workshop on , 1996:30 - 38
13 Gregory L.Pett. Adaptive Inverse Control of Linear and Nonlinear Systems using Dynamic
Neural Networks. IEEE Transactions on Neural Networks. 2003.14(2):360-376
14 Kevin M.Passino ,Stephen Yurkovich. Fuzzy Control. Tsinghua University Press, Prentice Hall,
2001:50-100
- 58 -
参考文献
15 J.R.Layne, K.M.Passino.Fuzzy Model Reference Learning Control for Cargo Ship Steering.
IEEE Control Systems Magazine. 1993, 13(6):23 - 34
16 LiXin Wang. Universal Approximation by Hierarchical Fuzzy Systems. Fuzzy Sets and
Systems. 1998,93:223-230
17 A.J. van der Wal. Application of Fuzzy Logic Control in Industry, Fuzzy Sets and Systems,
1995, 74:33-41
18 谭延良，郭怡倩. 一种新型模糊 PID 控制的变频恒压供水系统. 排灌机械，2001，19（5）:
35-38
19 方大寿，李纪扣. 模糊一 PI 复合控制恒压供水系统. 微小型计算机开发与应用，1996，
（4）:32-35
20 张承慧，程兆林. 发电厂循环水泵变频调速自整定 PID 模糊控制. 电力系统自动化，2002，
26（14）: 59-62
21 Karl Johan ?str?m, Bj?rn Wittenmark. Adaptive Control. 2nded. Science Press, Pearson
Eduation North Asia Limited, 2003:20-24,390-398
22 L.Ljung. System Identification,Theory for the user. Second Edition. Tsinghua University Press,
Prentice Hall, 2001:509-519
23 曾光，徐艳平，宁耀斌. 变频电源与工频电源间的同步切换控制. 西安理工大学学报.
2001，（3）:265-268
24 张厥胜，郑继禹，万心平. 锁相技术. 西安电子科技大学出版社，2002:1-12,180-182
25 白同云，吕晓德. 电磁兼容设计. 北京邮电大学出版社，2001:21-38,42-72,92-110
26 蔡弈刚. 流体管道动力学. 浙江大学出版社，1990:1-30
27 Simon Haykin.Nural Networks:A Comprehensive Foundation. Second Edition. Tsinghua
University Press, Prentice Hall, 2001:156-255,664-785
28 彭鞍虹. 通用变频器异步电动机的传递函数. 鞍山钢铁学院学报. 2000，（6）:447-449
29 B.Widrow, S. Stearns. Adaptive Signal Processing. Prentice-Hall, Englewood Cliffs, NJ, 1985:
206-228
30 N. A. Hizal. Improved Adaptive Model Control. Springer-Verlag Heidelberg, 1999, 51(3):181
-190
31 Simon Haykin. Adaptive Filter Theory. Fourth Edition. Tsinghua University Press, Printice Hall,
2001:4-6,203-227,466-501
- 59 -
北京工业大学硕士学位论文
32 Andreas Kirsch. An Introduction to the Mathematical Theory of Inverse Problems. Springer
–Verlag, Beijing World Publishing Corpration, 1999:1-85
33 党映农，韩崇昭. 基于Volterra基函数网络的自适应逆控制方法. 西安交通大学学报.
2000，（9）:8-12
34 Helder J. Cochofel, B.Sc. A Neural Network Environment for Adaptive Inverse Control. Intl.
Joint Conf. on Neural Networks.Anchorage, Alaska,1998
35 Jin-Tsong Jeng. Nonlinear Adaptive Inverse Control for the Magnetic Bearing System. Journal
of Magnetism and Magnetic Materials. 2000,209:186-188
36 诸静. 模糊控制原理与应用. 机械工业出版社，2001
37 高为炳. 变结构控制的理论及设计方法. 科学出版社，1998:278-289
38 高为炳. 离散时间系统的变结构控制. 自动化学报. 1995，（2）:154-161
39 刘培玉. 应用最优控制. 大连理工出版社，1990:100-118
40 Bernard Widrow, M.Lehr. 30 Years of Adaptive Neural Networks:Perceptron,Madaline,and
Backpropagation, Proc IEEE special Issue on Neural Networks. 1990, 78:1415-1442
41 Bernard Widrow, Gregory L. Plett. Nonlinear Adaptive Inverse Control. Proceedings of the
36th Conference on Decision & Control, San Diego, California USA, 1997
42 Plett, G.L, Bottrich.H. DDEKF Learning for Fast Nonlinear Adaptive Inverse Control. 2002.
IJCNN '02. Proceedings of the 2002 International Joint Conference on Neural Networks, IEEE
Trans.on Neural Networks, 2002,3:2092 – 2097
43 Plett, G.L. Adaptive Inverse Control of Plants with Disturbances. Ph.D.diss., Stanford Univer-
sity, May,1998:45-114,125-138
44 Gene F.Franklin, J.David Powell, Michael Workman. Digital Control of Dynamic Systems. 3rd
ed, Tsinghua University Press, Addison-Wesley, 2001:66-68,599-611
45 Mahesh M.Swarny. Harmonic Interaction between 1500 kVA Supply Transformer and VFD
Load at an Industrial Plant. IEEE Transactions on Industry Applications. 1998,.34(5):897-903
46 Fran?oise Beaufays. Two-Layer Linear Structures for Fast Adaptive Filtering. Ph.D.diss.,
Stanford University, June, 1995:5-23,86-104
47 Bernard Widrow, Luo.Fa-Long. Microphone Arrays for Hearing Aids: An Overview. Speech
Communication, 2003, 39(1):139-146
48 杨 芳，张宏伟. 城市供水负荷短期预测方法. 天津大学学报. 2002，35（2）:167- 170
- 60 -
参考文献
49 李 平，孙优贤. 最优模糊控制器的系统设计. 控制理论与应用. 1995，12（1）:46-52
50 沈毅力，李大海，李天石. 自适应逆控制及其在电掖伺服加载系统中的应用. 机床与液
压.2003，（1）:192-194
51 熊茂华. 基于80C196单片机的神经网络逆动态控制伺服系统. 机床与液压. 2002，（3）:
110-111
52 G.G.古德温，孙贵生. 自适应滤波、预测与控制. 科学出版社，1992:98-107，178-183
38-39，44
2 胡纲衡，唐瑞球. 交流变频调速的切换控制技术. 电工技术杂志. 2001，（6）:43-45
3 孙 斌，张宏建. 模糊变频恒压供水系统设计. 电气自动化. 2002，（3）:23-25
4 徐国忠，诸静. 单片机模糊控制恒压供水系统的研究. 微电子学与计算机. 1998，（4）:8-12
5 戴广平. 电动机变频器与电力拖动. 中国石化出版社. 1999: 89-106
6 Marek Hartman, Eugeniusz Loeiec, Piotr Boguslawski. Water Pumping Stations Remote Control
System in Depression Areas Based GSM-900 Cellular Communications System. TCSET2002,
Lviv-Slavsko,Ukraine, 2002
7 Hiromitsu Kato,Hiromitsu Kurisu, Teruji Sekozawa.Interactive Multiple Objective Decision
Method for Water Supply Scheduling in Hybrid Network Models. JAPAN
8 Frank A.Dewinter, The Application of a 3500 HP Variable Frequency Drive for Pipeline Pump
Control. IEEE Transactions on Industry Applications. 1989,25(6):1019-1024
9 M. Elbelkacemi, A. Lachhab, M. Limouri, B. Dahhou, A. Essaid. Adaptive Control of a Water
Supply System. Control Engineering Practice. 2001, 9(3):343-349
10 Bryan Coulbeck, Bogumil Ulanicki, Vladimir V.Deviatkov, Sergei Kosov, Iosiff Glukhovsky.
Pressure Control of a Moscow Water Supply System using Expert System Technology.
Proceedings of the 35th IEEE Conference on Decision and Control, Kobe, Japan, December
1996, (4):4498 - 4499
11 B.Widrow, E.Walach 著，自适应逆控制.刘树堂，韩崇昭 译. 西安交通大学出版社，1999
12 Bernard Widrow, Gregory L.Pett. Adaptive Inverse Control Based on Linear and Nonlinear
Adaptive Filtering. Neural Networks for Identification, Control, Robotics, and Signal/Image
1996. Proceedings. IEEE International Workshop on , 1996:30 - 38
13 Gregory L.Pett. Adaptive Inverse Control of Linear and Nonlinear Systems using Dynamic
Neural Networks. IEEE Transactions on Neural Networks. 2003.14(2):360-376
14 Kevin M.Passino ,Stephen Yurkovich. Fuzzy Control. Tsinghua University Press, Prentice Hall,
2001:50-100
- 58 -
参考文献
15 J.R.Layne, K.M.Passino.Fuzzy Model Reference Learning Control for Cargo Ship Steering.
IEEE Control Systems Magazine. 1993, 13(6):23 - 34
16 LiXin Wang. Universal Approximation by Hierarchical Fuzzy Systems. Fuzzy Sets and
Systems. 1998,93:223-230
17 A.J. van der Wal. Application of Fuzzy Logic Control in Industry, Fuzzy Sets and Systems,
1995, 74:33-41
18 谭延良，郭怡倩. 一种新型模糊 PID 控制的变频恒压供水系统. 排灌机械，2001，19（5）:
35-38
19 方大寿，李纪扣. 模糊一 PI 复合控制恒压供水系统. 微小型计算机开发与应用，1996，
（4）:32-35
20 张承慧，程兆林. 发电厂循环水泵变频调速自整定 PID 模糊控制. 电力系统自动化，2002，
26（14）: 59-62
21 Karl Johan ?str?m, Bj?rn Wittenmark. Adaptive Control. 2nded. Science Press, Pearson
Eduation North Asia Limited, 2003:20-24,390-398
22 L.Ljung. System Identification,Theory for the user. Second Edition. Tsinghua University Press,
Prentice Hall, 2001:509-519
23 曾光，徐艳平，宁耀斌. 变频电源与工频电源间的同步切换控制. 西安理工大学学报.
2001，（3）:265-268
24 张厥胜，郑继禹，万心平. 锁相技术. 西安电子科技大学出版社，2002:1-12,180-182
25 白同云，吕晓德. 电磁兼容设计. 北京邮电大学出版社，2001:21-38,42-72,92-110
26 蔡弈刚. 流体管道动力学. 浙江大学出版社，1990:1-30
27 Simon Haykin.Nural Networks:A Comprehensive Foundation. Second Edition. Tsinghua
University Press, Prentice Hall, 2001:156-255,664-785
28 彭鞍虹. 通用变频器异步电动机的传递函数. 鞍山钢铁学院学报. 2000，（6）:447-449
29 B.Widrow, S. Stearns. Adaptive Signal Processing. Prentice-Hall, Englewood Cliffs, NJ, 1985:
206-228
30 N. A. Hizal. Improved Adaptive Model Control. Springer-Verlag Heidelberg, 1999, 51(3):181
-190
31 Simon Haykin. Adaptive Filter Theory. Fourth Edition. Tsinghua University Press, Printice Hall,
2001:4-6,203-227,466-501
- 59 -
北京工业大学硕士学位论文
32 Andreas Kirsch. An Introduction to the Mathematical Theory of Inverse Problems. Springer
–Verlag, Beijing World Publishing Corpration, 1999:1-85
33 党映农，韩崇昭. 基于Volterra基函数网络的自适应逆控制方法. 西安交通大学学报.
2000，（9）:8-12
34 Helder J. Cochofel, B.Sc. A Neural Network Environment for Adaptive Inverse Control. Intl.
Joint Conf. on Neural Networks.Anchorage, Alaska,1998
35 Jin-Tsong Jeng. Nonlinear Adaptive Inverse Control for the Magnetic Bearing System. Journal
of Magnetism and Magnetic Materials. 2000,209:186-188
36 诸静. 模糊控制原理与应用. 机械工业出版社，2001
37 高为炳. 变结构控制的理论及设计方法. 科学出版社，1998:278-289
38 高为炳. 离散时间系统的变结构控制. 自动化学报. 1995，（2）:154-161
39 刘培玉. 应用最优控制. 大连理工出版社，1990:100-118
40 Bernard Widrow, M.Lehr. 30 Years of Adaptive Neural Networks:Perceptron,Madaline,and
Backpropagation, Proc IEEE special Issue on Neural Networks. 1990, 78:1415-1442
41 Bernard Widrow, Gregory L. Plett. Nonlinear Adaptive Inverse Control. Proceedings of the
36th Conference on Decision & Control, San Diego, California USA, 1997
42 Plett, G.L, Bottrich.H. DDEKF Learning for Fast Nonlinear Adaptive Inverse Control. 2002.
IJCNN '02. Proceedings of the 2002 International Joint Conference on Neural Networks, IEEE
Trans.on Neural Networks, 2002,3:2092 – 2097
43 Plett, G.L. Adaptive Inverse Control of Plants with Disturbances. Ph.D.diss., Stanford Univer-
sity, May,1998:45-114,125-138
44 Gene F.Franklin, J.David Powell, Michael Workman. Digital Control of Dynamic Systems. 3rd
ed, Tsinghua University Press, Addison-Wesley, 2001:66-68,599-611
45 Mahesh M.Swarny. Harmonic Interaction between 1500 kVA Supply Transformer and VFD
Load at an Industrial Plant. IEEE Transactions on Industry Applications. 1998,.34(5):897-903
46 Fran?oise Beaufays. Two-Layer Linear Structures for Fast Adaptive Filtering. Ph.D.diss.,
Stanford University, June, 1995:5-23,86-104
47 Bernard Widrow, Luo.Fa-Long. Microphone Arrays for Hearing Aids: An Overview. Speech
Communication, 2003, 39(1):139-146
48 杨 芳，张宏伟. 城市供水负荷短期预测方法. 天津大学学报. 2002，35（2）:167- 170
- 60 -
参考文献
49 李 平，孙优贤. 最优模糊控制器的系统设计. 控制理论与应用. 1995，12（1）:46-52
50 沈毅力，李大海，李天石. 自适应逆控制及其在电掖伺服加载系统中的应用. 机床与液
压.2003，（1）:192-194
51 熊茂华. 基于80C196单片机的神经网络逆动态控制伺服系统. 机床与液压. 2002，（3）:
110-111
52 G.G.古德温，孙贵生. 自适应滤波、预测与控制. 科学出版社，1992:98-107，178-183