基于改进PSO算法的MAP图标定点选择新方法

为提高基于MAP图的控制系统驱动效果，并有效减小控制系统内的存储量，提出了一种基于改进粒子群算法的MAP图中标定点择优选取新方法。以液压机械无级变速传动比控制系统中采用的MAP图为例，将其横坐标的两个变量在其定义域内等分，并采用改进粒子群算法选取等分后每段内的坐标点数量和位置。选取过程采用多目标优化原理结合了随机产生100个点的实际值与MAP图线性插值的平均误差以及选定的标定点数量。为提高算法执行效率，对粒子群算法的迭代准则、惯性权重和学习因子进行改进。结果表明，改进后的粒子群算法收敛速度快，寻优精度高，仅需较少的标定数据即可制作控制效果较佳的MAP图。

New method of MAP fixed-point selection based on improved PSO algorithm

In order to improve the driving effect of the control system based on MAP and reduce the storage capacity of the control system effectively, this paper proposed a new method for selecting the fixed points in the MAP based on the improved particle swarm optimization algorithm. Taking the MAP of hydraulic mechanical continuously variable transmission ratio control system as an example, it divided 2 variables of the abscissa coordinate in its definition domain, and used the improved particle swarm optimization algorithm to select the number and location of coordinate points in each section after equal division. The selection process was based on the principle of multi-objective optimization, which combined the actual values of 100 randomly generated points with the average error of linear interpolation of MAP and the number of selected scalar points. In order to improve the efficiency of the algorithm, it improved the iterative criteria, the inertia weight and the learning factor of the particle swarm optimization. The results show that the improved particle swarm optimization algorithm has the advantages of fast convergence speed and high optimization accuracy, and the MAP with better control effect can be made with less calibration data.