基于多控制器切换的过热汽温控制系统设计
|
摘要
火电厂汽温控制是一个成熟而常新的研究领域。一方面，基于常规PID控制的DCS控制系统已经具有一套规范成熟的体系；另一方面，市场对电厂的供电质量、安全性能及价格体系提出了更高的要求，而且电厂中亚临界、起临界等大型机组的比重越来越大，因此具有大容量，多参数特点的大型机组带来更加复杂多变的热工特性，使得现有的控制策略难以满足新形势下的控制要求。因此，新的控制理论及控制策略的研究和应用对火电厂汽温控制具有迫切的现实意义。
火电厂锅炉是一个非线性、强耦合、大滞后、多变量、工况范围广的复杂对象，本文以某火电厂#2锅炉600MW机组为研究对象，针对其控制品质差、易超温等问题，提出一种基于多控制器切换的过热汽温控制系统设计方法。由于过热汽温控制系统处于不同工况下的模型存在较大差异，本文针对每种不同的模型分别设计出与之相对应的控制器，并通过实时的监测其偏差值判断是否执行控制器的切换，当时滞在一定范围之内时，采用传统的数字PID对其控制；当时滞过大时，采用广义预测控制器进行控制。实验仿真结果表明，通过多控制的切换控制，锅炉过热汽温控制的动态特性明显得到了明显的改善。
Thermal power plant steam temperature control is a well-developed and frequentlyinnovated researching field．On the one hand, the DCS(distributed control system)based oncascade PID control strategy for power plant has been a normative and well-developedsystem． On the other hand，the more high criterions have been required for power plantspower supply quality, security performances and economy because of the market. Andsince the proportion of large units such as sub-vertical and super-critical growing unitshave been applied. Thermal complicated and variable thermal characteristics brought by alarge-capacity, multi-parameter characteristics have been made the existing controlstrategies not easy to meet the control request under the new situation. Therefore, a newresearch and application of control theory and strategies for thermal power plant steamtemperature control have urgent practical significance.
The boiler of thermal power plants is a complex object of nonlinear, strong coupling,large delay, multi-variable, and a wide range of operating conditions. Based on the NO.2steam temperature of600MW a certain power plant which had the problems of bad controlquality and surpassing temperature easily, design method of super-heated steamtemperature control system is proposed based on the switching multi-controller.Correspondingly, the controller has been designed for each different model. And throughreal-time monitoring of its deviation determine, it is to switch the implementation of thecontroller or not. It is time to choose traditional digital PID control scheme when the timelag is in certain range. If the time lag is too large, the generalized predictive controller issuitable. Simulation results show that the dynamic characteristics of the superheated steamtemperature control of boiler have been significantly improved through the multi-controllerswitching method.
火电厂锅炉是一个非线性、强耦合、大滞后、多变量、工况范围广的复杂对象，本文以某火电厂#2锅炉600MW机组为研究对象，针对其控制品质差、易超温等问题，提出一种基于多控制器切换的过热汽温控制系统设计方法。由于过热汽温控制系统处于不同工况下的模型存在较大差异，本文针对每种不同的模型分别设计出与之相对应的控制器，并通过实时的监测其偏差值判断是否执行控制器的切换，当时滞在一定范围之内时，采用传统的数字PID对其控制；当时滞过大时，采用广义预测控制器进行控制。实验仿真结果表明，通过多控制的切换控制，锅炉过热汽温控制的动态特性明显得到了明显的改善。
Thermal power plant steam temperature control is a well-developed and frequentlyinnovated researching field．On the one hand, the DCS(distributed control system)based oncascade PID control strategy for power plant has been a normative and well-developedsystem． On the other hand，the more high criterions have been required for power plantspower supply quality, security performances and economy because of the market. Andsince the proportion of large units such as sub-vertical and super-critical growing unitshave been applied. Thermal complicated and variable thermal characteristics brought by alarge-capacity, multi-parameter characteristics have been made the existing controlstrategies not easy to meet the control request under the new situation. Therefore, a newresearch and application of control theory and strategies for thermal power plant steamtemperature control have urgent practical significance.
The boiler of thermal power plants is a complex object of nonlinear, strong coupling,large delay, multi-variable, and a wide range of operating conditions. Based on the NO.2steam temperature of600MW a certain power plant which had the problems of bad controlquality and surpassing temperature easily, design method of super-heated steamtemperature control system is proposed based on the switching multi-controller.Correspondingly, the controller has been designed for each different model. And throughreal-time monitoring of its deviation determine, it is to switch the implementation of thecontroller or not. It is time to choose traditional digital PID control scheme when the timelag is in certain range. If the time lag is too large, the generalized predictive controller issuitable. Simulation results show that the dynamic characteristics of the superheated steamtemperature control of boiler have been significantly improved through the multi-controllerswitching method.
引文
[1]曾德良,赵征,陈彦桥,刘吉臻.500MW机组锅炉模型及实验分析[J].中国电机工程学报,2003,23(5):149-152.
[2]王广军,何祖威,陈红.基于神经网络和过程机理的锅炉过热系统动态仿真[J].中国电机工程学报,2001,21(12):38-40.
[3]张智焕,王树青.基于多模型的PID控制器[J].仪器仪表学报,2010（23）:122-123.
[4] L.J. Long, J. Zhao. Control of switched nonlinear systems in p-normal form usingmultiple lyapunov functions[J]. IEEE Transactions on Automatic Control,2012,57(5):1285-1291.
[5] L.I. Allerhand, U. Shaked. Robust stability and stabilization of linear switchedsystems with dwell time[J]. IEEE Transactions on Automatic Control,2011,56(2):381-386.
[6]孙泽勇,王庸贵,任德军,张燕.基于PID的温度自动控制系统的实现[J].中国测试技术,2003(2):18-19.
[7]袁佑新,卢立殊,王萍,王华斌.基于GHI神经元的温度控制系统研究[J].武汉理工大学学报,2002,24(3):22-24.
[8]赵为光,杨莹.电锅炉温度控制系统PID参数整定算法的改进[J].工业仪表与自动化装置,2009(1):85-87.
[9]胡寿松.自动控制原理[M].北京:科学出版社,2007.
[10]余雷,费树岷,李勋,刘一福.单神经自适应PSD预测控制在过热汽温切换系统中的应用[J].控制与决策2010,25(12):1906-1908.
[11]盖绍婷.在高速网络下时滞系统的最优扰动抑制研究[D].中国海洋大学硕士学位论文,2010,4.
[12]李建军,姜东娇,马志杰.模糊PID控制在主汽温控制系统中的应用[J].东北电力技术,2008(10):9-11.
[13]刘金琨.先进PID控制MATLAB仿真[M].电子工业出版社,2004.
[14]王伟.广义预测控制理论及其应用[M].北京:科学出版社,1998.
[15]吕俊霞.基于模糊PID的温度控制系统设计与分析[J].精密制造与自动化,2010(2):23-25.
[16]王庆东,冯增健,孙优贤.锅炉热工过程先进控制策略研究综述[J].电力喜用及其自动化学报.2004,16(5):75-80.
[17]张培建,施广仁.工业锅炉的燃烧控制策略研究与应用[J].电力自动化设备,2010,30(1):102-104.
[18]翟军勇,费树岷.基于多模型切换的过热汽温智能控制研究[J].自动化仪表.2005,26(6):10-12.
[19]张云广,沈炯,李益国.基于多模型切换的过热汽温广义预测控制[J].华东电力.2009,37(1):164-168.
[20] Jing Ke，Yizheng Qiao，Jixin Qian．Identification of time-varying delay systemsusing particle swarm optimization[A].Proceedings of the5th WCICA [C].Hangzhou，China，June15-19，2004:330-334.
[21] J. Marsik, V. Strejc. Application of identification-free algorithms for adaptive Control[J].Automatica,1989,25(2):273-277.
[22] J. Marsik. A new conception of digital adaptive PSD control [J].Problems of Controland Information Theory,1983(12):267-277.
[23] Weiming Xu, Zhipei Zu. The Application of Single-Neuron Apaptive Controller inLiquid Level Control. Journal of University of Shanghai for Science and Technology,2001(2):181-182.
[24] A. B. Rad, W. L. Lo, K. M. Tsang. Simultaneous online identification of rationaldynamics and time delay: a correlation-based approach [J]. IEEE Transactions onControl Systems Technology (S1063-6536),2003,11(6):957-959.
[25] Haralambos Sarimveis，Alex Alexandridis，Stefanos Mazarakis，George Bafas．A newalgorithm for developing dynamic radial basis function neural network models basedon genetic algorithms[J]． Computers and Chemical Engineering，2004,28(I-2):209-2l7.
[26]幸星.工业加热炉温度Smith模糊自适应PID控制算法研究[J].中国科技信息,2009(12):51-53.
[27] Dongqiang Qiu,Yaofa Tu. The Status and Protect of Neural Control. Automation&Instrumentation,2001(5):1-7.
[28] Xiaodong Wang. Direct Drive System of Single-neuron Based on S Funtion. ElectricAutomation,1996(5):4-7
[29] X. M. Ren, A. B. Rad, P T Chan, W L Lo. Online identification of continuous-timesystems with unknown time delay [J]. IEEE Transactions on Automatic Control(S0018-9286),2005,50(9):1418-1422.
[30]林辉,常继彬.基于PID控制的温度大滞后系统算法研究[J].甘肃科学学报,2011,2(1):119-121.
[31] S. J. Kelly, M. D. Rogars, D. W. Hoffman. Quadratic Dynamic Matrix Control ofHydrocracking Reactors[C]. Proceeding of the1998American Control Conferece,1998(1):295-300.
[32]韦庆志,李正明,孙俊.基于模糊自适应PID控制的锅炉过热蒸汽温度控制系统[J].机械设计与制造,2010(7):173-175.
[33] D. H. Nguyen, B. Widrow. Neural Networks for Self-learning Control System.INTJControl,1995,4(6):1439-1451.
[34] M. Clerc．The swarm and the queen：towards a deterministic and adaptive Particleswarm optimization[A]. Proceedings of the1999Congress on EvolutionaryComputation[C]，Washington，USA，July6-9，1999，V3：1951-1957
[35]席裕庚李德伟林姝.模型预测控制——现状与挑战.自动化学报[J]，2013,39(3):222-236..
[36] L. Yu, M.Q. Zhang, S.M. Fei. Nonlinear adaptive sliding mode switching control withaverage dwell-time[J]. International Journal of System Science,2013,44(3):471-478.
[2]王广军,何祖威,陈红.基于神经网络和过程机理的锅炉过热系统动态仿真[J].中国电机工程学报,2001,21(12):38-40.
[3]张智焕,王树青.基于多模型的PID控制器[J].仪器仪表学报,2010（23）:122-123.
[4] L.J. Long, J. Zhao. Control of switched nonlinear systems in p-normal form usingmultiple lyapunov functions[J]. IEEE Transactions on Automatic Control,2012,57(5):1285-1291.
[5] L.I. Allerhand, U. Shaked. Robust stability and stabilization of linear switchedsystems with dwell time[J]. IEEE Transactions on Automatic Control,2011,56(2):381-386.
[6]孙泽勇,王庸贵,任德军,张燕.基于PID的温度自动控制系统的实现[J].中国测试技术,2003(2):18-19.
[7]袁佑新,卢立殊,王萍,王华斌.基于GHI神经元的温度控制系统研究[J].武汉理工大学学报,2002,24(3):22-24.
[8]赵为光,杨莹.电锅炉温度控制系统PID参数整定算法的改进[J].工业仪表与自动化装置,2009(1):85-87.
[9]胡寿松.自动控制原理[M].北京:科学出版社,2007.
[10]余雷,费树岷,李勋,刘一福.单神经自适应PSD预测控制在过热汽温切换系统中的应用[J].控制与决策2010,25(12):1906-1908.
[11]盖绍婷.在高速网络下时滞系统的最优扰动抑制研究[D].中国海洋大学硕士学位论文,2010,4.
[12]李建军,姜东娇,马志杰.模糊PID控制在主汽温控制系统中的应用[J].东北电力技术,2008(10):9-11.
[13]刘金琨.先进PID控制MATLAB仿真[M].电子工业出版社,2004.
[14]王伟.广义预测控制理论及其应用[M].北京:科学出版社,1998.
[15]吕俊霞.基于模糊PID的温度控制系统设计与分析[J].精密制造与自动化,2010(2):23-25.
[16]王庆东,冯增健,孙优贤.锅炉热工过程先进控制策略研究综述[J].电力喜用及其自动化学报.2004,16(5):75-80.
[17]张培建,施广仁.工业锅炉的燃烧控制策略研究与应用[J].电力自动化设备,2010,30(1):102-104.
[18]翟军勇,费树岷.基于多模型切换的过热汽温智能控制研究[J].自动化仪表.2005,26(6):10-12.
[19]张云广,沈炯,李益国.基于多模型切换的过热汽温广义预测控制[J].华东电力.2009,37(1):164-168.
[20] Jing Ke，Yizheng Qiao，Jixin Qian．Identification of time-varying delay systemsusing particle swarm optimization[A].Proceedings of the5th WCICA [C].Hangzhou，China，June15-19，2004:330-334.
[21] J. Marsik, V. Strejc. Application of identification-free algorithms for adaptive Control[J].Automatica,1989,25(2):273-277.
[22] J. Marsik. A new conception of digital adaptive PSD control [J].Problems of Controland Information Theory,1983(12):267-277.
[23] Weiming Xu, Zhipei Zu. The Application of Single-Neuron Apaptive Controller inLiquid Level Control. Journal of University of Shanghai for Science and Technology,2001(2):181-182.
[24] A. B. Rad, W. L. Lo, K. M. Tsang. Simultaneous online identification of rationaldynamics and time delay: a correlation-based approach [J]. IEEE Transactions onControl Systems Technology (S1063-6536),2003,11(6):957-959.
[25] Haralambos Sarimveis，Alex Alexandridis，Stefanos Mazarakis，George Bafas．A newalgorithm for developing dynamic radial basis function neural network models basedon genetic algorithms[J]． Computers and Chemical Engineering，2004,28(I-2):209-2l7.
[26]幸星.工业加热炉温度Smith模糊自适应PID控制算法研究[J].中国科技信息,2009(12):51-53.
[27] Dongqiang Qiu,Yaofa Tu. The Status and Protect of Neural Control. Automation&Instrumentation,2001(5):1-7.
[28] Xiaodong Wang. Direct Drive System of Single-neuron Based on S Funtion. ElectricAutomation,1996(5):4-7
[29] X. M. Ren, A. B. Rad, P T Chan, W L Lo. Online identification of continuous-timesystems with unknown time delay [J]. IEEE Transactions on Automatic Control(S0018-9286),2005,50(9):1418-1422.
[30]林辉,常继彬.基于PID控制的温度大滞后系统算法研究[J].甘肃科学学报,2011,2(1):119-121.
[31] S. J. Kelly, M. D. Rogars, D. W. Hoffman. Quadratic Dynamic Matrix Control ofHydrocracking Reactors[C]. Proceeding of the1998American Control Conferece,1998(1):295-300.
[32]韦庆志,李正明,孙俊.基于模糊自适应PID控制的锅炉过热蒸汽温度控制系统[J].机械设计与制造,2010(7):173-175.
[33] D. H. Nguyen, B. Widrow. Neural Networks for Self-learning Control System.INTJControl,1995,4(6):1439-1451.
[34] M. Clerc．The swarm and the queen：towards a deterministic and adaptive Particleswarm optimization[A]. Proceedings of the1999Congress on EvolutionaryComputation[C]，Washington，USA，July6-9，1999，V3：1951-1957
[35]席裕庚李德伟林姝.模型预测控制——现状与挑战.自动化学报[J]，2013,39(3):222-236..
[36] L. Yu, M.Q. Zhang, S.M. Fei. Nonlinear adaptive sliding mode switching control withaverage dwell-time[J]. International Journal of System Science,2013,44(3):471-478.