Mohamed Faouzi Harkat

Sensor Fault Detection and Isolation by Robust Principal Component Analysis

Sensors are essential components of modern control systems. Any faults in sensors will affect the overall performance of a system because their effects can easily propagate to manipulative variables through feedback control loops and also disturb other process variables. The task for sensor validation is to detect and isolate faulty sensors and estimate fault magnitudes afterwards to provide fault-free values. Model-based methods constitute an important approach to sensor fault detection and isolation (FDI). A model-based approach consists in generating residuals as the difference between the measurements and the estimates provided by the relationships existing between the various variables of the process. The analysis of these residuals may lead to detect and isolate the faulty sensors. Almost all conventional model-based methods presume the knowledge of an accurate model of the system, e.g. transfer function or system matrices in the state space representation. Principal component Analysis (PCA) is a data-driven method which is particularly well adapted to reveal linear relationships among the plant variables without formulating them explicitly and has also been employed for system identification. PCA has some other nice features. It can handle high dimensional and correlated process variables, provides a natural solution to the errors-in-variables problem and includes disturbance decoupling (Li & Qin, 2001). Moreover in the FDI field, Gertler & McAvoy (1997) have shown a close link between PCA and parity space method. Principal component analysis (PCA) has then been applied successfully in the monitoring of complex systems (Chiang & Colegrove, 2007; Harkat et al., 2006; Kano & Nakagawa, 2008). 17

New Adaptive Moving Window PCA for Process Monitoring

Abstract Slow and normal process changes often occur in real processes, which lead to false alarms for a fixed-model monitoring approach. In this paper, two recursive PCA algorithms for adaptive process monitoring are studied. the first algorithm is based on Moving Window Principal Component Analysis (MWPCA), and the second is based on Forgetting Factors Principal Component Analysis (Recursive Weighted PCA). Furthermore, by changing the size and the shift of the window, also for the forgetting factor, we will see the influence of these changes on the monitoring performances. Then, adaptive forgetting factors will be used, for increasing the robustness against outliers. Using the same concept of varying forgetting factors, a new recursive algorithm for adaptive process monitoring based on Moving Window is proposed. By using the current model and the updated mean and covariance structures and an Adaptive Moving Window, a new model is derived recursively (AMWPCA). Based on the updated PCA representation the Q-statistic (SPE) (monitoring metric) is calculated and their control limits are updated. The feasibility and advantages of each algorithms is illustrated by application to Tennessee Eastman process.

New hierarchical approach for multiple sensor fault detection and isolation. Application to an air quality monitoring network

Our work is devoted to the problem of multiple sensor fault detection and isolation using principal component analysis. Structured residuals are used for multiple fault isolation. These structured residuals are based on the principle of variable reconstruction. However, multiple fault isolation based on reconstruction approach leads to an explosion of the reconstruction combinations. Therefore instead of considering all the subsets of faulty variables, we determine the isolable multiple faults by removing the subsets of variables that have too high minimum fault amplitudes to ensure fault isolation. Unfortunately, in the case of a large number of variables, this scheme yet leads to an explosion of faulty scenarios to consider. An effective approach is to use multi-block reconstruction approach where the process variables are partitioned into several blocks. In the first step of this hierarchical approach, the goal is to isolate faulty blocks and then in the second step, from the faulty blocks, faulty variables have to be isolated. The proposed approach is successfully applied to multiple sensor fault detection and isolation of an air quality monitoring network.

Multiple sensor fault detection and isolation of an air quality monitoring network using RBF-NLPCA model

This paper presents a data-driven method based on non-linear principal component analysis to detect and isolate multiple sensor faults. The RBF-NLPCA model is obtained by combining a principal curve algorithm and two three-layer radial basis function (RBF) networks. The reconstruction approach for multiple sensors is proposed in the non-linear case and successfully applied for multiple sensor fault detection and isolation of an air quality monitoring network. The proposed approach reduces considerably the number of reconstruction combinations and allows to determine replacement values for the faulty sensors.

Fault detection and isolation in nonlinear systems with partial Reduced Kernel Principal Component Analysis method

In this article, we suggest an extension of our proposed method in fault detection called Reduced Kernel Principal Component Analysis (RKPCA) (Taouali et al., 2015) to fault isolation. To this end, a set of structured residues is generated by using a partial RKPCA model. Furthermore, each partial RKPCA model was performed on a subset of variables to generate structured residues according to a properly designed incidence matrix. The relevance of the proposed algorithm is revealed on Continuous Stirred Tank Reactor.

Online process monitoring based on kernel method

This paper discusses the monitoring of dynamic process. In recent years, Kernel Principal component analysis (KPCA) has gained significant attention as a monitoring method of nonlinear systems. However, the fixed KPCA model limit its application for dynamic systems. For this purpose a new Variable Moving Window Kernel PCA (VMWKPCA) method is introduced to update the KPCA model. The basic idea of this technique is to vary the size of the moving window depending on the normal change of the process. Then the VMWKPCA method is performed for monitoring a Chemical reactor (CSTR). The simulation results proved that the new method is effective.

Fault detection localization and reconstruction in nonlinear system using RKPCA method and RBC

In this paper we exploit the use of the proposed RKPCA method ([1], [2], [3]) for sensor fault detection, localisation and reconstruction. To this end, a set of structured residues is generated by using partial RKPCA technique. Also to identify fault, the Reconstruction Based Contribution RBC approach [4] was used. The relevance of the evaluated techniques partial RKPCA and RBC is revealed on Continuous Stirred Tank Reactor (CSTR).

Broken bars fault diagnosis in induction machine fed by a voltage source inverter

In this paper, we present the modeling of the induction machine, taking into consideration the rotor fault in order to detect and diagnosis rotor broken bars the analysis of stator current in Concordia pattern compared to the healthy mode case. The obtained results show that the considered method can effectively detect abnormal operating conditions in induction motor and clearly show the possibility of extracting signatures to detect fault. These techniques offer reliable and satisfactory results to diagnose such fault.

Diagnosis and Detection of Short-Circuit in a Three-Phase Induction Machine

Due to its simple construction, low cost, manufacture, and its robustness. The use of induction motors is rapidly and increasingly growing in the industry especially in highly important sectors. This leads us to a serious focus on their operation and their availability. The early detection for motor deterioration can increase plant availability and safety in economical way, help to avoid expensive failures, reduce coasts and the fault number. Many publications have investigated the detection and diagnosis of short circuit fault in electrical machines supplied directly online. This paper presents a technique based on spectral analysis of stator currents to detect short-circuit fault in the stator using a mathematical model of three-phase squirrel cage induction motor under stator short-circuit fault. The simulation show the impact and the effectiveness of failure. The tests are validated by numerical simulation and the obtained results clearly show the possibility of extracting signatures to detect and locate faults.