Research On Current Decoupling Control Of Built-In Permanent Magnet Synchronous Motor

As the oil resources are depleted day by day,the traditional internal combustion engine vehicles will be phased out in the future,and new energy vehicles will enter the market one after another.As the key parts of new energy vehicles,motor and electronic control still have room for optimization in control strategy.As one of the core parts of motor control strategy,decoupling control is necessary.Due to the inherent characteristics of the motor,the current control of the quadrature axis and the straight axis is easily coupled,which is especially serious at high speed,thus affecting the dynamic regulation performance of the motor.Therefore,decoupling control is needed to solve the current coupling problem at high speed in order to improve its control effect.However,the traditional decoupling control has the problem of dependence on motor parameters,and some decoupling methods proposed by academia tend to be theoretical and computationally complex,which is not conducive to engineering applications.Therefore,this paper summarizes several existing decoupling control strategies,and finally chooses feedforward decoupling which has more practical application value in engineering as the research content.The traditional feed-forward decoupling has dependence on motor parameters.In order to meet the requirement of decoupling control,the motor must be able to run stably at high speed.Therefore,the PI parameters of the current regulator are determined by the principle of zero-pole cancellation,and then the stator current distribution below the base speed is optimized by the maximum torque-current ratio control.At the same time,the gradient descent method is used to adjust the flux weakening so that the motor can transit to the constant power region steadily.Through theoretical analysis,it is proved that the method avoids the current integral saturation existing in the traditional flux weakening process.The problem of sum is solved,which realizes the weak magnetic acceleration.In this paper,a seven-segment voltage vector pulse width modulation(SVPWM)method is adopted,and the voltage inverters are over-modulated by linear output of fundamental voltage amplitude,which reduces the demand for bus voltage at high speed.In order to solve the problem that the decoupling effect of the traditional feed-forward decoupling control strategy is not good due to the change of motorparameters when the built-in permanent magnet synchronous motor operates at high speed,the recursive least squares method with forgetting factor is used to identify the parameters of the motor,and the estimated values of the identified motor parameters are substituted into the feed-forward decoupling to improve its robustness to parameter changes.Because of the restriction of the chip's computing ability in practical engineering,the calculation amount of the algorithm needs to be reduced.Therefore,this paper simplifies the parameter identification model.The default resistance and flux linkage remain unchanged,and only the parameters of the cross-axis inductance are identified.It is theoretically proved that the feed-forward decoupling strategy based on the above identification model has no effect on the decoupling effect.Resistance error has no effect on the decoupling effect when the motor runs at high speed.The effect of coupling is minimal.Then considering that the parameters of the motor change slowly and the speed of the motor changes abruptly in real time,the forgetting factor is optimized under the constraints of recognition speed and anti-interference ability.Finally,the simulation results show that the decoupling control method has faster current regulation speed than the traditional feed-forward decoupling method,and thus improves the stability of the motor at high speed.Finally,the sliding mode variable structure control(SMVSC)based on the general exponential reaching rate is used to design the control rate of the outer loop of the speed.Then,under the motor speed mode,the simulation results show that the improved feed forward decoupling control has better dynamic regulation performance than the traditional feed forward decoupling control in the current,speed and torque response.