基于主成分分析马氏距离的支腿控制阀健康评估

支腿控制阀的性能是影响起重机支腿系统伸缩性能的重要因素之一。为了准确的评估支腿控制阀的健康性能,提出一种支腿系统的性能衰退与健康状态的评估方法,该方法基于PCA降维与马氏距离相结合的分析模型,建立不同状态下传感器信号与关键零部件的映射关系,从而达到对起重机支腿系统性能衰退量化评估的目的。该方法应用于起重机支腿控制阀的压力信号,通过对传感器信号内蕴关系及起重机在各年份提取的特征在特征空间的相关性分析,以得到量化性能评估结果。与常见的其他方法相比,该模型能够准确地反映起重机支腿系统历年来的性能衰退趋势,具有更好的鲁棒性与泛化性。

Health evaluation of a crane’s leg control valve based on PCA and Mahalanobis distance

The performance of a leg control valve is one of important factors affecting telescopic performance of a crane’s leg system.Here, in order to accurately evaluate health performance of a leg control valve, an evaluation method for crane leg system’s performance degradation and health state was proposed based on the analysis model using the principal component analysis (PCA) dimension reduction combined with Mahalanobis distance to establish a mapping relation between sensor signals and key components under different states, and achieve the goal to quantitatively evaluate a crane leg system’s performance degradation.The method was used with pressure signals of a crane’s leg control valve.Its quantitative performance evaluation results were obtained using the correlation analysis of sensor signal implication relation and crane’s each year features extracted in the feature space.It was shown that compared with other methods, the proposed model can accurately reflect performance degradation trend of a crane leg system over years with better robustness and generalization.