【作者】 韩光信； 赵聚乐； 胡云峰；

【Author】 HAN Guang-xin;ZHAO Ju-le;Hu Yun-feng;Jilin institute of chemical technology;Jilin university;

【机构】 吉林化工学院； 吉林大学；

【摘要】 针对欠驱动非线性系统的输入受限问题,本文以球平衡器的轨迹跟踪控制为研究对象,提出了一种基于滑模控制的神经网络抗饱和补偿控制。选取合适的滑模趋近律来保证系统的快速跟踪和稳定性能,使用径向基神经网络对控制器饱和误差进行逼近补偿;为保证补偿系统稳定,神经网络权值由Lyapunov稳定理论推得的自适应律确定。为提高系统轨迹跟踪精度和防止滑模抖振,使用粒子群算法对滑模控制器参数以及神经网络权值进行优化。仿真实验表明,该算法具有良好的鲁棒性能与抗饱和性能。更多还原

【Abstract】 In order to solve the problem of input limitation in underactuated nonlinear systems.The trajectory tracking control of ball balancer was taken as the research object,a neural network anti-saturation compensation control based on sliding mode control was proposed. Appropriate sliding mode approach law was selected to ensure the system’s fast tracking and stable performance. The radial basis neural network is used to approximate and compensate the saturation error of the controller.In order to improve the tracking accuracy and prevent the chattering of sliding mode, the sliding mode controller parameters and neural network weights are optimized by particle swarm optimization(pso). Simulation results show that the algorithm has good robustness and anti-saturation performance.更多还原