Linear Induction Motor Model Predictive Control Strategy Research

The degree of urban traffic congestion has increasingly become an important factor restricting the development of urban economy.The development of rail transit is one of the important means to solve this problem.Linear Induction Motor(LIM)has been widely used in rail transit.Linear induction motor in operation will be affected by the side effect,the motor parameters will change with the motor running speed,will directly affect the performance of the control system.In this paper,the mathematical model of LIM considering the edge effect is combined with the model predictive control method.The design process of weight coefficient of torque control predicted by traditional model needs a lot of experiments.In order to solve the difficult problem of weight coefficient setting,the model prediction flux control method was adopted,and the mathematical relationship between electromagnetic thrust and primary flux amplitude was established to convert the cost function into primary flux vector,thus avoiding the tedious weight coefficient design process.Traditional vector allocation methods usually use the SVPWM method,which is usually a high switching frequency.Based on the model prediction flux control,this paper adopts the three-vector voltage vector distribution method to reduce the switching frequency of the inverter under the condition that the current harmonics and thrust fluctuations are basically consistent with the effect of the vector control method.Moreover,the model prediction flux control method also has good dynamic response performance and good parameter robustness.The traditional vector control is highly dependent on the accuracy of the parameters of the current loop PI controller.In this paper,the multi-step model predictive control method is adopted to reduce the dependence on the controller parameters.The selection range of weight coefficient in this method is large.Choosing suitable predictive length and weight coefficient can also obtain good dynamic response performance and decoupling performance.The method also has good parameter robustness.In addition,this paper also proposes an improved method of three-vector allocation to solve the problem of mismatch of vector synthesis when the three-vector allocation method is used in sector switching,which effectively solves the problem of incorrect voltage vector synthesis and improves the steady-state performance.Because the experimental conditions are difficult to satisfy the experimental verification of the motion control strategy of LIM,a LIM model is built on the d SPACE semi-physical real-time simulation platform.The platform can debug the parameters of the control system in real time,which can greatly improve the development efficiency and reduce some risks in the experiment.Finally,the above control methods are verified on the d SPACE semi-physical platform,and the correctness and effectiveness of the d SPACE semi-physical platform and the proposed control methods are verified.