An ensemble approach to sensor fault detection and signal reconstruction for nuclear system control

Abstract

To efficiently control a process, accurate sensor measurements must be provided of the signals used by the controller to decide which actions to actuate in order to maintain the system in the desired conditions. Noisy or faulty sensors must, then, be promptly detected and their signals corrected in order to avoid wrong control decisions. In this work, sensor diagnostics is tackled within an ensemble of Principal Component Analysis (PCA) models whose outcomes are aggregated by means of a local fusion (LF) strategy. The aggregated model thereby obtained is used for both the early detection and identification of faulty sensors, and for correcting their measured values. The fault detection decision logic is based on the Sequential Probability Ratio Test (SPRT). The proposed approach is demonstrated on a simulated case study concerning the pressure and level control in the pressurizer of a Pressurized Water Reactor (PWR). The obtained results show the possibility to achieve an adequate control of the process even when a sensor failure occurs

Unsupervised Clustering for Fault Diagnosis in Nuclear Power Plant Components

“…Figure 9 is a schematic representation of the pressurizer system for which a Matlab SIMULINK model has been developed, based on the application of the mass and energy conservation equations to the two regions of vapor and liquid; exchanges between the two regions, due to evaporation of liquid and condensation of steam, are taken into account [Kuridan et al, 1998;Todreas et al, 1990]. The system of nonlinear differential equations describing the model is detailed in [Baraldi et al, 2010]. In order to represent a realistic situation, the simulations have been carried out based on the operational parameters of a standard PWR pressurizer (Table 1).…”

“…The control of the level L and the pressure P in the pressurizer is achieved through a feedback control scheme which reproduces that used in a standard PWR pressurizer. According to the control scheme illustrated in [Baraldi et al, 2010], the pressure f P and level f L are the controlled signals as well as the controller input signals; the sprayers mass flow rate sp m , and the heaters power Q are the controller outputs. The present case study focuses on some faults which can occur to the pressurizer control system and can lead to undesired behaviors of the pressurizer.…”

Variance sensitive adaptive threshold-based PCA method for fault detection with experimental application

“…One of the most common multivariate statistical process control (MSPC) methods used for this purpose is principal component analysis. PCA has been used for various multivariate data analysis techniques such as process monitoring, quality control, sensor and process fault diagnosis [22][23][24][25][26][27][28][29][30][31][32][33][34]. In this section, the general principle of using PCA for fault detection is presented.…”

“…Most processes are well equipped with the sensors to realize automatic monitoring and control. However, most of the research studies are based on simulation results [23,27,29].…”

“…In both situations, multivariate statistical techniques such as principal component analysis and partial least squares (PLS) can play a major role. These methods have been successfully applied to solve a wide range of multivariate problems in the chemical processes [22][23][24][25], semiconductor processing [26], machining processes [27], waste water treatment [28], nuclear power systems [29], air-conditioning processes [30,31], building central chilling systems [32][33][34]. Most processes are well equipped with the sensors to realize automatic monitoring and control.…”