基于RBF神经网络的液黏调速离合器 活塞位移控制器设计

针对液黏调速离合器存在非线性和控制精度较低，难以满足工业领域较高的传动特性需求等问题，提出了基于RBF (radial basis function, 径向基函数)神经网络的液黏调速离合器活塞位移滑模控制策略。对液黏调速离合器局部结构进行改进，增设位移传感器和导电滑环以采集位移信号；建立了电液比例溢流阀和液黏调速离合器的数学模型，设计并分析了基于RBF神经网络的液黏调速离合器活塞位移滑模控制器；搭建了液黏调速离合器AMESim-MATLAB联合仿真模型。仿真结果表明：基于RBF神经网络的液黏调速离合器活塞位移滑模控制可以有效地适应液黏调速离合器的非线性，并解决滑模控制的抖振问题，能够提高控制精度，使液黏调速离合器控制器具有很好的鲁棒性，可以满足较高的工业需求。

Displacement controller design for piston of hydro-viscous clutch based on RBF neural network

The hydro-viscous clutch has relatively serious nonlinearity and lower control accuracy, which is difficult to meet the requirement of high transmission characteristics in the industrial field. Based on these issues, sliding mode control strategy of piston displacement based on RBF (radial basis function) neural network was proposed. The local structure of the hydro-viscous clutch was improved. The displacement sensor and the conductive slip ring were added to acquire the displacement signal. The mathematical models of the electro-hydraulic proportional relief valve and the hydro-viscous clutch were derived. The sliding mode controller based on RBF neural network of piston displacement of hydro-viscous clutch was designed and analyzed. An AMESim-MATLAB co-simulation model of hydro-viscous clutch was built. The simulation results showed that the sliding mode control of piston displacement based on RBF neural network (RBFSMC) could effectively cope with the serious nonlinearity of hydro-viscous clutch, and solved the chattering phenomenon of sliding mode control. The piston displacement controller can improve the control precision, make the controller of hydro-viscous clutch have good robustness and meet the high industrial demand.