基于ARMAX模型的挖掘机液压系统故障检测

Fault Detection Approach of Excavator's Hydraulic System Based on ARMAX Model

根据挖掘机的可靠性水平和智能化程度,提出了一种基于ARMAX(Auto-Regressive Moving Average Exogenous)模型的挖掘机液压系统故障检测方法.使用系统正常状态下的信号样本建立系统的ARMAX辨识模型,通过休哈特控制图分析AR-MAX辨识模型的输出残差,以获取故障检测的阈值;然后,将系统故障状态下的信号样本代入ARMAX辨识模型,当模型残差超过阈值,即能判定系统发生故障.实验分析表明,基于ARMAX模型的故障检测方法够有效地检测挖掘机液压系统产生的故障.

In order to improve rehabihty and intelligence of excavator,a fault detection approach of the excavator＇s hydrauhc system based on auto-regressive moving average exogenous （ARMAX） model was proposed. Firstly, an ARMAX model was estabhshed with sample data under normal condition, and the residuals of ARMAX model were analyzed by Shewart control charts to obtain the thresholds of fault detection. Secondly, sample data under fault condition was imported to the ARMAX model, if the model residuals exceed the thresholds, the system faults could be detected. Experimental results show that the proposed fault detection approach can be effectively apphed to the excavator＇s hydraulic system. 3figs., 10refs.