Optimization Of Field Oriented Control For Permanent Magnet Synchronous Motor In Electric Vehicle Based On Torque-speed-current MAP

With the rapid advancement of global automotive electrification,electric vehicles are becoming increasingly important modes of transportation.The electric drive system is the core component of an electric vehicle,and its performance will determine the driving quality of the vehicle.The motor drive system mainly includes the motor and the inverter.Although China has made great progress improving these elements in recent years,there is still much work to be done to match other countries' level of expertise.The areas which are most in need of development include control techniques and the solutions to extreme special conditions.This article is a major projects of the Ministry of Science and Technology.The research object is the permanent magnet synchronous motor(PMSM)which takes advantage of rare earth resources in China.This research divides vector control into three processes based on optimizing the control strategy of the PMSM: target torque to target current process,target current to target voltage process,and target voltage to the power device switching process.Also,a detailed optimization plan is proposed.In addition,two unique fault conditions of the PMSM,fault prevention and treatment scheme,are analyzed.In terms of converting the target torque into the target current,this paper proposes a target current distribution vector control strategy based on torque-speed-current MAP.The central idea is to replace the feedback field weakening adjustment in the traditional vector control by calibrating the data to the area where the motor can run through the dynamometer.Firstly,the paper derives the basic maximum current torque ratio and the maximum voltage torque ratio from the mathematical model of the motor,and uses these two curves and current limiting conditions to solve the torque-speed-current MAP of the control region.Secondly,according to the variation law of motor parameters,the influence of the change of motor sensitive parameters on the distribution of MAP data is analyzed accordingly,which provides compensation ideas for the actual control of the motor.To adapt to the inverter voltage fluctuations,the voltage fluctuation correction factor is introduced.Therefore,the control algorithm can be applied to other voltage conditions only if it is successfully calibrated within one voltage range.Finally,the paper proposes a set of experimental methods for calibrating the torque-speed-current MAP of the motor using an electric dynamometer.The calibration process is clearly divided into several steps to achieve data measurement,extraction and integration,and finally an interpolation tool is used to form an isometric table which clearly identifies the working ideas for calibration engineers and speeds up the calibration of the motor MAP.In terms of converting the target current into the target voltage,this paper proposes an optimization scheme based on the original PI current control.For this reason,two algorithms,feedforward decoupling control and internal model decoupling control,are compared.The decoupling degree of the two decoupling methods under the condition of accurate parameters is analyzed,and a current control transfer function is constructed to analyze the robustness of the two decoupling methods to the parameter changes.It is found that the internal model decoupling can better achieve independent control of the d-q axis current.Finally,the internal model decoupling and the anti-saturation function PI control were combined to carry out simulations,benches,and vehicle experiments.In terms of converting the target voltage to the power switching device,this paper proposes a control vector duty cycle over modulation technology.The optimization of the algorithm makes the motor power increase.At the same time,the voltage and current harmonic components caused by over modulation and the influence of harmonic components on the motor control performance are analyzed.Accordingly,a variable-bandwidth low-pass filter is designed to eliminate the influence of harmonics on the entire control system and simulations.In addition,benches and vehicle experiments were also carried out.Regarding the handling of special conditions,this paper focuses on solving the two intrinsic shortcomings of the PMSM compared to AC asynchronous motors: overvoltage and overcurrent which are easily caused by position encoder failure at high speed,and the high potential for permanent magnets to demagnetize easily under high temperature or high current conditions.As for the conditions of high speed and high back EMF,the causes of inverter damage caused by the failure of the PMSM rotary encoder are analyzed.The protection strategy of estimating the rotor flux linkage angle by using the position sensorless algorithm is proposed.Considering both hardware and software costs,detection accuracy,control influence and other factors,the position estimation safety redundancy strategy based on back-EMF observer technology is adopted after analyzing the advantages and disadvantages of multiple estimation methods.In this case,the failure of the position sensor at high speed can be prevented,and parking safely can be ensured when the above situation occurs.As for demagnetization of high temperature permanent magnets,after analyzing the performance of the magnetic material,a rotor temperature estimation strategy is proposed based on the PMSM thermal network model,which can estimate the temperature of the rotor permanent magnet and magnetic steel in real time.Compared with the traditional method of measuring only the stator temperature,the large performance redundancy of the motor is not necessary any more.This puts the motor intrinsic performance on full display.In this paper,a set of power reduction strategies is proposed based on the measured temperature of the stator and the estimated temperature of the rotor.Thus,the real-time protection of the permanent magnet is realized,which greatly reduces the risk of demagnetization of the permanent magnet and improves the ability of the system to continuously maintain the maximum power.Lastly,to verify this control strategy,an inverter with a peak power of 40 kW and a rated voltage of 144 V was designed.The main control unit of the inverter uses an Infineon automotive-grade MCU XMC4500,and the power part uses Infineon's discrete MOSFET IRF200F223 10-tube parallel connection scheme.The inverter's hardware uses circuit-board parasitic inductance drive technology,which reduces switching losses.The software uses a MOSFET die temperature estimation algorithm that enables the power electronics capability to perform well.At the same time,a dynamometer bench platform was set up for power testing to accurately measure steady and dynamic data from the inverter.In the end,an A00 pure electric vehicle was used as the vehicle's testing platform,and the performance of the entire vehicle was greatly improved.