基于虚拟传感器的AGV故障诊断方法

Virtual-sensor-based fault diagnosis method for automated guided vehicles

为实现自动导引车(automated guided vehicle,AGV)的低成本有效故障诊断,提出一种基于虚拟传感器(virtual sensor,VS)的AGV故障检测方法。首先,基于AGV的结构特性,建立紧凑动力学模型;基于所建立的AGV模型,结合H_(∞)控制理论,设计全新的VS,使得能够在无需了解AGV轮胎模型且不进行线性化处理的条件下,对车轮横向力、纵向力及转矩进行有效估计,通过相应的定理证明所提VS的可解性与收敛性;进一步,通过比较AGV模型与VS产生的输出响应构造用于故障诊断的残差信号,并给出基于“IF-THEN”规则的故障诊断策略;最后,通过实验验证所提故障诊断方法的有效性。结果表明,所提方法能够对不同运行情况下的AGV状态进行有效检测,从而可为后续复杂AGV系统的全面状态监测提供有益参考。

In order to realize low-cost and effective fault diagnosis for automated guided vehicles(AGV),a fault detection method based on virtual sensor(VS)is proposed.Firstly,based on the structural characteristics of AGV,a compact dynamic model is established.Then,based on the compact AGV model and H_(∞)control theory,a new adaptive state estimator is designed,which can effectively estimate the lateral force,longitudinal force and torque acting on the wheels without understanding the AGV tire model and linearization.Corresponding theorems are proposed to prove the solvability and convergence of the proposed VS.Furthermore,by comparing the output response generated by the compact AGV model and the proposed VS,the residual signals for fault diagnosis are constructed,and the fault diagnosis strategies based on"IF-THEN"rules are given.Finally,the effectiveness of the proposed fault diagnosis method is verified by experiments The results show that the proposed method can effectively detect the state of AGV under different operating conditions,which can provide a useful reference for the future comprehensive state monitoring of complex AGV systems.