基于直接判据提取方式的直轴电流补偿型IPMSM最大转矩电流比控制算法

Direct Axis Current Compensated MTPA Control Algorithm of IPMSM Based on Direct Criterion Calculation

为实现内嵌式永磁同步电机(IPMSM)高性能运行,该文提出一种最大转矩电流比(MTPA)决策与控制算法,旨在简化控制结构、加快系统动态响应、增强系统鲁棒性。该算法根据dq轴电流直接计算电磁转矩的微分项,并以此作为MTPA状态的决策判据。根据判据值对d轴电流参考值进行补偿,实现对MTPA状态的在线追踪。所提算法无需实际或虚拟信号注入和解算环节,系统动态性能得以提高。d轴电流补偿控制解决了算法收敛速度对负载条件较敏感的问题,提升了控制系统的鲁棒性。理论分析、仿真计算和实验结果验证了该方法的有效性和优越性,与虚拟信号注入法相比,所提出的算法在不同负载下对MTPA状态的收敛时间降低为原来的1/5~1/2。

Due to the asymmetry of the magnetic circuit of the quadrature axis and direct axis,there are both permanent magnet torque and reluctance torque in the electromagnetic torque of the interior permanent magnet synchronous motor(IPMSM).In order to improve the utilization rate of reluctance torque,the maximum torque per ampere(MTPA)control has become the preferred control strategy of IPMSM.Corresponding to a certain load torque,the MTPA control minimizes the armature current required to maintain the operation of the motor by distributing the d-q axis current,thereby reducing copper consumption and improving efficiency.Some of the existing MTPA control methods have the problems of immense computation,high complexity,and low practicability.An improved MTPA control algorithm to realize the high-performance operation of IPMSM is proposed in this paper,which aims to simplify the control structure,speed up the dynamic response and enhance the robustness of the system.The proposed algorithm directly calculates the differential term of electromagnetic torque according to d-q axis currents,which is taken as the decision criterion of the MTPA state.According to the convergence relationship between the MTPA criterion and the dq/dt term,when the motor is in a non-MTPA state,the∂Te/∂qis not zero,and the current vector angle q needs to be adjusted.Then,an error-driven closed-loop control structure is constructed with the criterion value equal to zero as the control target.Finally,the given value of the current vector angle qref is adjusted in real time based on the initial value of the current vector angle qini.The criterion is utilized to compensate for the reference value of the d-axis current to realize the tracking and maintenance of the MTPA state.The proposed method does not need an actual or virtual signal injection and demodulation,and the dynamic performance of the system is improved.The d-axis current compensation control solves the problem that the convergence speed of the conventional algorithm is sensitive to load conditions,and improves the robustness of the control system.This paper has compared the proposed algorithms,including fixed-gain current vector angle compensation(FGAC),adaptive-gain current vector angle compensation(AGAC),and d-axis current compensation(DCC),with the current virtual signal injection(VSI)and multiple virtual signals injection(MVSI).The 1 N·m load test show that the time required for the d-q axis current corresponding to converge to the MTPA state is about 6 s(FGAC),1.8 s(AGAC),and 1.8 s(DCC).When the load is changed to 4 N·m,the response time of the three methods is about 2.2 s.The convergence speed of the proposed algorithm for the MTPA state is 4~5 times faster than that of the VSI algorithm.Experiments show that the convergence speed of the DCC algorithm is mostly the same(about 2 s)under different working conditions of 2 N·m and 4 N·m.The dynamic convergence speed of VSI is slower than that of the DCC algorithm under light load and heavy load.Compared with MVSI,DCC’s time complexity(online computation)is reduced by about 85%,and the space complexity(memory footprint)is reduced by about 55%.The following conclusions can be drawn from simulation and experimental analysis:(1)The proposed algorithm establishes the MTPA state direct decision criterion,which shortens the solution convergence time.Taking the VSI dynamic convergence process duration as the benchmark value 1(pu),the convergence time of the proposed algorithm is only 0.2(pu)under different working conditions,faster than most existing methods.(2)The proposed algorithm extracts the MTPA state decision criterion using direct calculation without injecting real/virtual signals into the system and the signal demodulation process.Both time complexity and space complexity are reduced.(3)By setting the adaptive integral gain and optimizing the structure of the MTPA control system,the sensitivity of the dynamic convergence speed of the algorithm to the load,speed,and other working conditions is reduced,and the applicability of the algorithm is enhanced.