As optimization of parameters affects prediction accuracy and generalization ability of support vector
regression (SVR) greatly and the predictive model often mismatches nonlinear system model predictive control,
a multi-step model predictive control based on online SVR (OSVR) optimized by multi-agent particle swarm
optimization algorithm (MAPSO) is put forward. By integrating the online learning ability of OSVR, the predictive
model can self-correct and adapt to the dynamic changes in nonlinear process well.
TANG Xianlun (唐贤伦), LIU Nianci (刘念慈), WAN Yali (万亚利), GUO Fei (郭飞)
. Multi-Step Model Predictive Control Based on Online Support Vector Regression Optimized by Multi-Agent Particle Swarm Optimization Algorithm[J]. Journal of Shanghai Jiaotong University(Science), 2018
, 23(5)
: 607
-612
.
DOI: 10.1007/s12204-018-1990-1
[1] XI Y G, LI D W, LIN S. Model predictive control:Status and challenges [J]. ACTA Automation Sinica,2013, 39(3): 222-234.
[2] MELANIE N Z, COLIN N J, MANFRED M. Realtimesuboptimal model predictive control using a combinationof explicit MPC and online optimization[J]. IEEE Transactions on Automatic Control, 2011,56(7): 1524-1534.
[3] SHU D Q. Predictive control system and its application[M]. Beijing, China: China Machine Press, 1996(in Chinese).
[4] CHEN J D, PAN F. Online support vector regressionbasednonlinear model predictive control [J]. Controland Decision, 2014, 29(3): 460-464 (in Chinese).
[5] SHIN J, KIM H J, KIM Y. Adaptive support vectorregression for UAV flight control [J]. Neural Networks,2011, 24(1): 109-120.
[6] GAO L F, ZHAO S J, GAO J. Application of artificialfish-swarm algorithm in SVM parameter optimizationselection [J]. Computer Engineering and Applications,2013, 49(23): 86-90 (in Chinese).
[7] DENG N Y, TIAN Y J. New method of data mining:Support vector machine [M]. Beijing, China: SciencePress, 2004 (in Chinese).
[8] ZHONG W M, HE G L, PI D Y, et al. SVM withquadratic polynomial kernel function based nonlinearmodel one-step-ahead predictive control [J]. ChineseJournal of Chemical Engineering, 2005, 13(3): 373-379.
[9] TANG X L, ZHUANG L, HU X D. The support vectorregression based on the chaos particle swarm optimizationalgorithm for the prediction of silicon content inhot metal [J]. Control Theory & Applications, 2009,26(8): 838-842 (in Chinese).
[10] MU C X, ZHANG R M, SUN C M. LS-SVM predictivecontrol based on PSO for nonlinear systems [J]. ControlTheory & Applications, 2010, 27(2): 164-168.
[11] ZENG L Q, LUO F B, DING J M. Application of particleswarm optimization algorithm integrated with tabusearch in reactive power optimization [J]. Power SystemTechnology, 2011, 35(7): 129-133 (in Chinese).
[12] TANG X L, ZHANG H, CUI Y Q, et al. A novelreactive power optimization solution using improvedchaos PSO based on multi-agent architecture [J]. InternationalTransactions on Electrical Energy Systems,2014, 24(5): 609-622.
[13] WANG Z, WANG Y J, SHANG A L. Applicationof multi-agent and particle swarm optimizationin network reconfiguration of ship power system[C]//Proceedings of 4th International Conference onModelling, Identification and Control. Wuhan, China:IEEE, 2012: 1218-1223.
[14] LI L Y, LI D R. Fuzzy entropy image segmentationbased on particle swarm optimization [J]. Progress inNatural Science, 2008, 18: 1167-1171.
[15] MA J S, THEILER J, PERKINS S. Accurate onlinesupport vector regression [J]. Neural Computation,2003, 15(11): 2683-2704.
