An MEWMA-based segmental multivariate hidden Markov model for degradation assessment and prediction

Degradation is an unavoidable phenomenon in industrial systems. Hidden Markov models (HMMs) have been used for degradation modeling. In particular, segmental HMMs have been developed to model the explicit relationship between degradation signals and hidden states. However, existing segmental HMMs deal only with univariate cases, whereas in real systems, signals from various sensors are collected simultaneously, which makes it necessary to adapt the segmental HMMs to deal with multivariate processes. Also, to make full use of the information from the sensors, it is important to differentiate stable signals from deteriorating ones, but there is no good way for this, especially in multivariate processes. In this paper, the multivariate exponentially weighted moving average (MEWMA) control chart is employed to identify deteriorating multivariate signals. Specifically, the MEWMA statistic is used as a comprehensive indicator for differentiating multivariate observations. Likelihood Maximization is used to estimate the model parameters. To avoid underflow, the forward and backward probabilities are normalized. In order to assess degradation, joint probabilities are defined and derived. Further, the occurrence probability of each degradation state at the current time, as well as in the future, is derived. The Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset of NASA is employed for comparative analysis. In terms of degradation assessment and prediction, the proposed model performs very well in general. By sensitivity analysis, we show that in order to improve further the performance of the method, the weight of the chart should be set relatively small, whereas the method is not sensitive to the change of the in-control average run length (ARL).

Improved Statistical Pattern Analysis Monitoring for Complex Multivariate Processes Using Empirical Likelihood

This article developed an improved statistical pattern analysis (SPA) monitoring strategy for fault detection of complex multivariate processes using empirical likelihood. The technique based on statistical pattern analysis performs fault detection by inspecting change in the statistics of process variables (e.g., mean value, correlation coefficient, variance, kurtosis, etc.). It is capable of monitoring non-Gaussian or even nonlinear processes. However, the original SPA framework explicitly computes all the high-order statistics, which significantly increases the scale and dimensionality of the problem, especially in the case of complex multivariate processes. To alleviate this difficulty, we propose monitoring changes in the statistics with the same order using empirical likelihood, which is a widely used estimation method to construct confidence limits or regions for parameters with similar properties. As a result, changes in statistics of the same order can be translated into a single index; hence more information on the faulty conditions can be observed. Furthermore, by considering statistics of the same order, the scale of the problem is reduced significantly. The improved statistical pattern analysis monitoring strategy is suitable for monitoring complex multivariate processes. The performance of the improved method is illustrated by an application study to fault detection of the Tennessee Eastman (TE) process.