Zero-point fault detection of load cells in truck scale based on recursive principal component analysis and comprehensive evaluation method

Abstract

A zero-point fault of load cells in truck scale is a typical small fault and it is difficult to be detected online. This paper proposes a new method for detecting zero-point fault online by combining a recursive principal component analysis (RPCA) with a comprehensive evaluation method (CEM). In this new method, firstly, the principal component model is updated online and the estimation of load cells is obtained by the principal recursive algorithm based on rank-1 modification, and then the four statistics, i.e., the Hotelling’s T2 statistic, the squared prediction error (SPE) statistic, the Hawkins TH2 statistic, and the principal component related variable residual (PVR) statistic, are used to construct a comprehensive evaluation method (CEM) for fault detection. This proposed method is applied to detecting the zero-point fault of load cells in truck scale, and the experimental results show that the accuracy of the zero-point fault detection is far higher than that of the traditional method (i.e., it is detected by only using any one of the four statistics, such as SPE, T2, TH2, PVR), which proves the effectiveness of this proposed method.