The Chaotic Motion Control Of Permanent Magnet Linear Synchronous Motor Based On Neural Network

The operating of permanent magnet synchronous linear motor show the characteristic of obvious time variation. When the objective factors like the operating temperature of system change, the parameters of the model will change too. When the operating parameters are higher than the critical point as the chaotic attractor occured, the chaos begin. The chaotic motion has a big influence on the operating quality and reliability of the motor, it will make a big loss to the production. As a matter of fact, it is very important to the efficient operation of the entire power system, that we analyze and research the chaotic characteristics of permanent magnet linear synchronous motor, and that we research the control measures of insuring the motor operating stably. research work of this paper is as follows:Firstly, this article achieved the chaotic model of permanent magnet synchronous linear motor by using the Coordinate transformation, which like the Lorenz chaotic system. Than we proved that the permanent magnet synchronous linear motor is very sensitive to the dependence of the operating parameters. When the parameters go into a certain range, the motor system will begin to have a chaotic motion. We used the simulation platform of MATLAB in the whole article. Taking into account the actual operation, the operating parameters of permanent magnet synchronous linear motor because of the interference of external or internal factors and make random changes, but also taking into account through human action, its operating parameters reaches a certain value after stabilizing at a chaotic state, with fixed parameters of the model runs. Based on this, the paper analyzes the permanent magnet linear synchronous motor parameters fixed and dynamic model parameter uncertainty, for each of the parameters of the model are proposed two optimal control strategy.Secondly, When operating parameters fixed, in order to suppress the chaotic state of permanent magnetlinear synchronous motor, this paper presented a feedback control method of radial basis function neural network, which is based on the theory of chaos control. The control variables of this method are the stator q-axis current、the direct axis current and the speed of permanent magnet linear synchronous motor. Firstly, chaotic dynamic characteristics were learned by RBFNN which is reflected by the three variables. Than, this paper give a feedback to the control chaotic system by using the trained radial basis function neural network model. The simulation shows that the chaotic state of permanent magnet linear synchronous can restore to a stable equilibrium point quickly. The limitation is that, only in the situation of motor parameters are fixed, can this control tragedy reflect a good performance of control.Tertiary, The parameters based on the motor operating characteristics are easily changed with the external factors changed, and the dynamic model is changeable. Therefore, this paper presented a control method based on RBF neural network for single neuron, which is integrated by auto disturbances rejection controller. This method absorbs the advantages of RBF neural network, control and auto disturbance rejection control. And it can use the control approximation ability and learning speed of RBF neural network, can also use the advantages of auto disturbance rejection control not relying on the model and real time compensation.Research results show that the proposed control method has high control precision, strong robustness and fast response speed, has the extremely important application value for the safe and efficient operation of the motor.