基于改进蚁群算法的单神经元PID控制在数控精校机电液伺服系统中的应用

The Application of the Single Neuron PID Control Based on Improved Ant Colony Algorithm in the Hydraulic-servo System of the Precise Straightening Press

数控精校机电液伺服系统是一个参数变化比较大的非线性系统,常规的PID控制效果难以让人满意。采用单神经元PID控制器,在线调整权值,增强了对非线性系统的调节控制功能,然而参数的选择非常重要。文章提出一种改进的连续蚁群算法,用于优化单神经元PID控制器的参数,并用于精校机电液伺服系统,通过MATLAB仿真结果可得系统的阶跃响应,证明该算法比一般的蚁群算法精度更高,收敛更快,同时体现了该蚁群算法在实际应用中的实用性。

The hydraulic-servo system of the precise straightening press is a nonlinear system whose pa- rameters change largely, the effect of the conventional PID control makes it difficult satisfaction. The single neuron PID controller was taken, the relevance weights was adjusting online, the adjustment con- trol function of the nonlinear system was strengthened, however the selection of parameters is very im- portant. This paper puts forward an improved ant colony algorithm which was used to optimize the pa- rameters of the single neuron PID controller and used in the hydraulic-servo system of the precise straight- ening press. The step response of the system was got by the matlab simulation results, it proved that the improved algorithm has the higher precise and the faster convergence than the average ant algorithm, and it reflected the practicality of the algorithm in the practical application.