CAJ | 학술논문

神经元PID控制器虽然具有自学习和自调整能力，但其增益K不具备这种能力。而增益K对多电机同步控制系统又有着十分显著的影响。为解决这一问题，将自适应PSD控制算法应用于神经元PID控制器中，形成增益K的自适应算法，解决了传统神经元PID控制器增益K无法实时在线调整的缺点，提高了传统神经元PID控制器的自学习和自调整能力。应用MATLAB7.0分别搭建常规PID、神经元PID和基于PSD算法的神经元PID的3种多电机同步控制系统模型，进行仿真比较。仿真结果表明：基于PSD算法的神经元PID多电机同步控制系统具有更好的自适应性和鲁棒性，实现了提高多电机同步控制系统控制精度的目的。
신경원PID공제기수연구유자학습화자조정능력，단기증익K불구비저충능력。이증익K대다전궤동보공제계통우유착십분현저적영향。위해결저일문제，장자괄응PSD공제산법응용우신경원PID공제기중，형성증익K적자괄응산법，해결료전통신경원PID공제기증익K무법실시재선조정적결점，제고료전통신경원PID공제기적자학습화자조정능력。응용MATLAB7.0분별탑건상규PID、신경원PID화기우PSD산법적신경원PID적3충다전궤동보공제계통모형，진행방진비교。방진결과표명：기우PSD산법적신경원PID다전궤동보공제계통구유경호적자괄응성화로봉성，실현료제고다전궤동보공제계통공제정도적목적。
Neuron propotional intergal and differential (PID)controller had self-learning and self-adjusting ability,but whose gain K did not,moreover gain K had an very significant effect on multi-motor synchronous control. In order to solve the problem,by applying self-adjusting PSD control algorithm to neuron PID controller and forming self-adjusting algorithm of gain K,the shortcoming was overcomed that gain K for tranditonal neuron PID controller could not be adjusted in real time on line,and its self-learning and self-adjusting ability were improved. By using MATLAB7.0,three motors synchronous control system model of general PID,neuron PID and neuron PID based on PSD algorithm were respectively set up,and simulated to compare. The simulation result shows that neuron PID multi-motor synchronous control system based on PSD algorithm has better self-adjusting and robustness,which realizes the pur-pose to improve multi-motor synchronous control accuracy.