提出在航空关节轴承试验机加载控制系统中应用极限学习机(ELM)对电液力PID控制参数进行在线调节的方法。解决了BP神经网络梯度下降法速度太慢导致在伺服系统中PID控制实时性太差的问题。实验结果表明,该方法泛化性能好、学习速度快、模型精度高,具有很强的实践指导意义。 A method to adjust the electrohydraulic PID control parameters on-line by using extreme learning machine (ELM) in the loading control system of aviation joint bearing test machine is proposed. Solve the BP neural network gradient descent method is too slow, resulting in poor real-time PID control in the servo system problem. The experimental results show that this method has good generalization performance, fast learning speed and high precision of the model, which has a strong practical guiding significance.