混合动力车载电机直接转矩控制算法优化

发布时间：2018-03-01 14:37

本文关键词： 车载感应电机直接转矩控制空间矢量脉宽调制滑模变结构模型参考自适应　出处：《吉林大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着环境污染与能源危机问题的日益显现,汽车工业领域逐渐将混合动力汽车视为行业发展的重要趋势。驱动系统作为混合动力汽车的核心组成部分,对汽车控制性能有重要影响。由于交流感应电机本身具有易于生产、结构简单、成本低、效率高和可靠性好等优点,国内外大部分混合动力汽车选择以交流感应电机为驱动电机。直接转矩控制作为一种新兴交流电机控制技术,具备控制思路新颖、目标明确以及系统结构简单等优点,与其它控制策略相比有明显优势。本文选择以混合动力车载感应电机为研究对象,结合车载感应电机运行条件对直接转矩控制系统进行分析研究,并针对传统直接转矩控制系统中存在的一些缺陷设计优化改进方案。主要研究内容如下:1.在传统直接转矩控制系统中,转矩和定子磁链偏差由滞环比较器获得,再通过开关矢量选择表选择合适的电压矢量。这种控制结构使系统受到滞环宽度的影响,电压矢量只能在有限个数中切换,并且造成逆变器开关频率不固定。本文基于空间矢量脉宽调制技术对控制系统进行改进,SVPWM技术可以根据上一周期磁链和转矩的偏差预测出下一周期的期望空间电压矢量,并且期望电压矢量能够由所在扇区内的两个相邻非零电压矢量和零矢量作用合成。不仅克服电压矢量数量上的限制,也固定了逆变器的开关频率。实现对定子磁链平滑调节,优化电机转矩脉动。2.在直接转矩控制系统中,需要获得准确的电机转速才能实现系统的闭环控制。但是,测速仪器不仅增加经济成本还需要定期维护,在一些工业条件下较难实现。本文根据交流电机数学模型和直接转矩控制原理,结合模型参考自适应原理设计转速估计环节,实现对电机转速的精确估计。3.本文利用滑模变结构控制理论对控制系统进行优化。根据滑模变结构控制理论,设计磁链滑模控制器和转矩滑模控制器,替代原始SVPWM模块中PI调节器生成参考电压。同时设计定子磁链滑模观测器,提高定子磁链观测精度。利用李雅普诺夫稳定性原理验证变结构控制器和观测器的稳定性。通过应用滑模变结构控制,克服了系统在实际运行过程中电机参数变化和外界扰动的影响,有效的增强了系统的鲁棒性。在MATLAB/Simulink环境中建立控制系统仿真模型。通过对比分析仿真实验结果,证明改进后直接转矩控制系统的定子磁链轨迹和电机转矩脉动获得明显改善,同时实现对电机转速精确估计。实验结果验证了优化方案的有效性。
[Abstract]:With the emergence of environmental pollution and energy crisis, hybrid electric vehicles (HEV) are regarded as an important trend in automobile industry. The driving system is the core component of HEV. The AC induction motor has the advantages of easy production, simple structure, low cost, high efficiency and good reliability. Most hybrid electric vehicles at home and abroad choose AC induction motor as the driving motor. As a new AC motor control technology, direct torque control has the advantages of novel control ideas, clear objectives and simple system structure. Compared with other control strategies, this paper chooses the hybrid vehicle loaded induction motor as the research object and analyzes the direct torque control system in combination with the on-board induction motor operating conditions. The main research contents are as follows: 1. In the traditional direct torque control system, the torque and stator flux deviation are obtained by hysteresis comparator. Then the switching vector selection table is used to select the appropriate voltage vector. This control structure makes the system be affected by the hysteresis width, and the voltage vector can only be switched in a limited number. Based on space vector pulse width modulation (SVPWM), an improved SVPWM technique can be used to predict the expected space voltage vector for the next period according to the deviation of flux and torque in the previous period. Moreover, the expected voltage vector can be combined by two adjacent nonzero voltage vectors and zero vectors in the sector. It not only overcomes the limitation of the number of voltage vectors, but also fixes the switching frequency of the inverter. In the direct torque control system, the accurate speed of the motor is needed to realize the closed-loop control of the system. However, the speed measuring instrument not only increases the economic cost, but also needs to be maintained on a regular basis. It is difficult to realize in some industrial conditions. According to the mathematical model of AC motor and the principle of direct torque control, combined with the adaptive principle of model reference, the speed estimation link is designed in this paper. In this paper, the sliding mode variable structure control theory is used to optimize the control system. According to the sliding mode variable structure control theory, the flux chain sliding mode controller and the torque sliding mode controller are designed. Instead of the Pi regulator in the original SVPWM module, the reference voltage is generated. At the same time, the stator flux linkage sliding mode observer is designed. The stability of variable structure controller and observer is verified by using Lyapunov stability principle. It overcomes the influence of motor parameter change and external disturbance during the actual operation of the system, and effectively enhances the robustness of the system. The simulation model of the control system is established in the MATLAB/Simulink environment. The simulation results are compared and analyzed. It is proved that the stator flux locus and the torque ripple of the motor are obviously improved after the improvement of the direct torque control system, and the accurate estimation of the motor speed is also realized. The experimental results verify the effectiveness of the optimized scheme.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U469.7

【参考文献】