摘要
为了提升传统模型预测控制(MPC)电压矢量的利用率、减少计算量、降低时间消耗,提出了动态MPC策略。动态MCP由简化MPC和传统MPC分时使用组成。简化MPC备选电压矢量少、计算量小。但当系统处于动态变化状态时,较少的备选电压矢量使得转速超调量高、跟踪速度慢。在此基础上,采用I_(d)=0的控制方法,并以d轴电流偏差为参考,提出了简化电流变化的动态切换控制策略。当电流偏差在电流变化系数范围内时,使用简化MPC,并只提供3个备选电压矢量;在范围外时,使用传统MPC,并使用全部7个电压矢量。试验结果表明,动态MPC策略保留了两者的优点,既有较好的动态响应,又缩短了计算周期。该设计可有效应用于低成本、算力不强的电机应用场景。
Dynamic model predictive control(MPC) strategy is proposed in order to enhance the utilization of voltage vectors of conventional MPC,reduce the computational effort,and decrease the time consumption.Dynamic MCP consists of simplified MPC and conventional MPC for time-shared use.The simplified MPC has fewer alternative voltage vectors and less computation.However,when the system is in dynamic change state,less alternative voltage vector makes the speed overshoot high and the tracking speed slow.On this basis,the control method with I_(d)=0 is adopted,and the dynamic switching control strategy with simplified current change is proposed with the d axis current deviation as a reference.When using the simplified MPC within the current deviation in the range of current variation coefficients,only three alternative voltage vectors are provided;when using the conventional MPC outside the range,all 7 voltage vectors are used.The experimental results show that the dynamic MPC strategy retains the advantages of both,with both a better dynamic response and a shorter computation period.The design can be effectively used in low-cost and less arithmetic motor application scenarios.
作者
陈振刚
贾洪平
CHEN Zhengang;JIA Hongping(School of Electrial Information Engineering,Jiangsu University,Zhenjiang 212013,China)
出处
《自动化仪表》
CAS
2024年第4期45-50,共6页
Process Automation Instrumentation
关键词
永磁同步电机
模型预测控制
简化模型预测控制
电流系数
动态模型预测控制
预测时间补偿
Permanent magnet synchronous motor(PMSM)
Model predictive control(MPC)
Simplified MPC
Current coefficient
Dynamic MPC
Prediction time compensation
