Decoupling Control Of Two Motor Variable Frequency Systems Based On Support Vector Machines Inverse

Multi-motor variable frequency speed-regulating system is an electronic control system,which is widely applied in modern industrial production,such as,textile, metallurgy,machine,paper,steel rolling,and so on.Because the system has some special characteristics as multi-variable,nonlinear,strong coupling,traditional control methods aren't able to control it effectively.How to gain great control performance has become an key issue in modern electrical driver area.With the financial aid of National Nature Science Fund(60874014),China Ministry of Education Fund (20050299009) and Natural Science Fund of Jiangsu Province(BK2007094), focusing on a two-motor variable frequency speed-regulating system composed of two AC induction motors and inverters,the decoupling control of two motor variable frequency systems based on Support Vector Machines(SVM) Inverse was studied in this paper.Firstly,the concept of the inverse system and the theory of SVM is introduced. To deal with the bottlenecks of the traditional inverse system methods,combining the good ability of modeling nonlinear processes set available of SVM,the SVM inverse control was proposed.Secondly,the reversibility of this two-motor system working on vector control mode was testified.On the basis of reversibility analysis of original system,the inverse model approximated by the SVM was cascaded with the original system.Thirdly,based on MATLAB6.5,the model of this two-motor variable frequency system was constructed by S-FUNCTION,and then the effectiveness of the decoupling control based on Support Vector Machines was testified by simulation.Finally,based on S7-300PLC experiment system,the SVM inverse control system was constructed in PLC,and then a mass of experiments was made to study the performance of decoupling,validity and robustness.Both the simulation and experiment results demonstrate that the combined method has good static and dynamic decoupling performance.And the strong robustness to both variation of parameters and load torque disturbance can be achieved by using the proposed method.