Research On Fusion Control Method Of Battery Parameters And Motor Characteristics For Pure Electric Vehicle

At the same time to meet the pure electric vehicle dynamic performance and endurance needs,separate battery management or motor control is to use the energy from the individual object efficiency,in order to give full play to the power of the battery energy,the power battery and motor drive system(PBMDS)should work in a collaborative way from the perspective of vehicle efficiency optimization.From the battery energy distribution point of view to see the impact of motor control performance,while the control of the motor should also see the battery performance changes.the PBMDS model which can realize the whole system analysis and the fusion control is established,and fusion control method based on the power battery parameters and the motor characteristic is used to improve the driving performance of the electric vehicle and improve the energy utilization efficiency.In this paper,the research on PBMDS fusion control technology of pure electric vehicle is studied,and the following research results are obtained:1.In order to realize the fusion control and theoretical analysis of PBMDS,the change of battery parameters was observed from the characteristics of motor operation,the influence of motor parameters was analyzed from the perspective of battery parameter variation.In this paper,a joint model is established based on the dynamic parameters V-RC equivalent circuit model of the battery and the motor unified equivalent circuit.Using the variable parameter second-order RC circuit to refine and simulate the polarization characteristics of the battery,the battery model to add a constant voltage source dynamic directional analog voltage hysteresis characteristics.The new model has the full state of charge(SOC)simulation accuracy of the battery,more accurate response to the dynamic characteristics of the battery,and can reduce the model parameter identification error.Based on the double reaction theory and the phasor transformation,the new parameters are introduced to establish the unified circuit model of the motor.The new parameters of the motor can be used to simulate the salient pole effect of the motor accurately.The battery and the motor model are coupled by the inverter to obtain the joint model.2.Aiming at the establishment of an adaptive battery model parameter identification method,the parameter identification of the dynamic parameters of the battery model under full SOC is realized,a method of parameter identification of battery model with multi-frequency AC signal injection is proposed.This method has the characteristics of adapting the number of AC signals according to the number of parameters of the model.Themulti-point identification of the model parameters is realized for different battery SOC,which is the SOC dynamic parameter identification algorithm of the SOC from the point to the line.The multi-parameter circuit is transformed into a single-port network,and the network is analyzed by the pilot function.The multi-parameter equations are established and solved.The calculation is simplified and the real-time online identification is realized.This process is recursively sampled under full SOC.Each parameter of the battery model has a function relationship with SOC,that is,the model parameters change dynamically with the battery SOC.The accuracy of the model parameter identification is greatly improved,and the method can be extended to other battery model parameter identification,the battery model parameter identification plays an important role in the battery parameter SOC estimation.The method is recursively sampled under full SOC,which can determine the function of each parameter of the battery model with SOC,to achieve dynamic parameter identification.The identification accuracy of the battery model parameters is greatly improved and can be extended to a variety of battery model parameter identification,and this method will play an important role in the SOC estimation of the battery parameters.3.Aiming at reducing the SOC estimation error and improving the estimation accuracy of the battery in the fusion control,an adaptive extended Kalman filter algorithm(AEKF)battery SOC estimation method based on the dynamic parameter V-RC model is proposed.AEKF estimated battery SOC depends on battery model accuracy,the method of using the battery dynamic parameter model can reduce the estimation error of the battery SOC compared with the traditional constant parameter model.The method used for the initial condition and the uncertainty of the noise only causes the SOC estimation to fluctuate for a short time.The adaptive filter can carry on the real-time estimation and the correction according to the state of the system,has the good system divergence inhibition,the error in Short time to quickly reduce,and system estimation accuracy is improved.4.In order to improve the driving performance of the electric vehicle through the fusion control,the battery parameters are introduced as relevant quantities in the motor control,and the fusion control method based on the battery parameters and the motor characteristics is proposed.Both the battery parameters SOC and SOH are introduced in the motor start control and cruise control,and the motor control achieves the desired control under limited energy.The method uses the real-time parameter of the power battery as the starting principle of the motor starting judgment,and the closed loop determines the starting current satisfying the starting condition,it avoids the current given too much to cause energy loss compared to conventional motor starting control.The adaptive fuzzy PID control is used to generate thecorrection quantity on-line fuzzy reasoning.The input error is detected during the starting process.The PID parameters can be adjusted on-line,and the reference current can be tracked quickly and steadily.The more efficient motor torque output with strong robustness,energy efficiency is improved under the premise of ensuring the performance of electric vehicle drive.Real-time estimation of battery parameters incorporated into cruise control motor,combined with fuzzy neural network PID control method,for electric vehicle drive motor with different conditions,by training the neural network to memorize the fuzzy control rules,the optimal effect can be achieved by adjusting the on-line self-learning of the controller when the model parameters change.Compared with the simple motor control,this method can predict the condition of the PBMDS and improve drive performance.