Sensor Fault Identification Research Articles

Sensor fault identification based on time-lagged PCA in dynamic processes

Principal component analysis (PCA) is widely employed as a multivariate statistical method for fault detection, isolation and diagnosis in chemical processes. Previously, PCA has been successfully used to identify faulty sensors under normal static operating conditions. In this paper, we extend the reconstruction-based sensor fault isolation method proposed by Dunia et al. to dynamic processes. We develop a new method for identifying and isolating sensor faults in an inherent dynamic system. First, we describe how to reconstruct noisy or faulty measurements in dynamic processes. The reconstructed measurements are obtained by simple iterative optimization based on the correlation structure of the time-lagged data set. Then, based on the sensor validity index (SVI) approach developed by Dunia et al., we propose an SVI for fault isolation in dynamic processes. The proposed method was applied to sensor fault isolation in two strongly dynamic systems: a simulated 4×4 dynamic process and a simulated wastewater treatment process (WWTP). In these experiments, the proposed sensor fault identification method correctly and rapidly identified the faulty sensor; in contrast, the traditional PCA-based sensor fault isolation approach showed unsatisfactory results when applied to the same systems.

Identification of sensor faults on turbofan engines using pattern recognition techniques

The possibility to identify faults in the readings of the sensors used to monitor the performance of a high by-pass ratio turbofan engine is examined. A novel method is proposed, based on the following principle: if a measurement set is fed to an adaptive performance analysis algorithm, a set of component performance modification factors (fault parameters) is produced. Faults, which may be present in the measurement set, may be recognized from the patterns they produce on the modification factors. The constitution of a method of this type based on pattern recognition techniques is discussed here. Test cases corresponding to different sensor faults, simulating operation in real conditions are examined. Three kinds of pattern recognition techniques with increasing complexity are used, in order to correctly identify the examined sensor faults. It is demonstrated that by choosing an appropriate formulation it is possible to have a 100% success in the identification of the examined sensor faults.

Probabilistic model for sensor fault detection and identification

AbstractA new Bayesian belief network (BBN) model with discretized nodes is proposed for fault detection and identification in a single sensor. The single‐sensor model is used as a building block to develop a BBN model for all sensors in the process considered. A new fault detection index, a fault identification index, and a threshold setting procedure for the multisensor model are introduced. The fault detection index exploits the probabilistic information of the multisensor model, which, along with the proposed threshold setting procedure, leads to effective detection of faulty sensors. The fault identification index uses only the probabilistic information of the faulty sensor to determine in a discretized fashion the size of faults that should be analyzed within a moving time window to identify the fault type. Single‐sensor model design parameters (prior and conditional probability data) are optimized to achieve maximum effectiveness in detection and identification of sensor faults. The single‐sensor model and optimal values of design parameters are used to develop a multisensor BBN model for a polymerization reactor at steady‐state conditions. Its capabilities to detect and identify bias, drift, and noise in sensor readings are illustrated for single and simultaneous multiple faults by several case studies.

Setting Up of a Probabilistic Neural Network for Sensor Fault Detection Including Operation With Component Faults

The diagnostic ability of probabilistic neural networks (PNN) for detecting sensor faults on gas turbines is examined. The structure and the features of a PNN, for sensor fault detection, are presented. It is shown that with the proposed formulation, a powerful tool for sensor fault identification is produced. A particular feature of the PNN produced is the ability to detect sensor faults even in the presence of engine component malfunction, as well as on deteriorated engines. In such situations, the size of bias that can be identified increases. The way to establish the limits of sensor bias that can be detected is presented along with results from application to test cases with realistic noise magnitudes. The diagnostic procedure proposed here is also supported by an engine performance model. The data used for setting up and testing the PNN are generated by such a model.

An intelligent system for multivariate statistical process monitoring and diagnosis

Abstract A knowledge-based system (KBS) was designed for automated system identification, process monitoring, and diagnosis of sensor faults. The real-time KBS consists of a supervisory system using G2 KBS development software linked with external statistical modules for system identification and sensor fault diagnosis. The various statistical techniques were prototyped in MATLAB, converted to ANSI C code, and linked with the G2 Standard Interface. The KBS automatically performs all operations of data collection, identification, monitoring, and sensor fault diagnosis with little or no input from the user. Navigation throughout the KBS is via menu buttons on each user-accessible screen. Selected process variables are displayed on charts showing the history of the variables over a period of time. Multivariate statistical tests and contribution plots are also shown graphically. The KBS was evaluated using simulation studies with a polymerization reactor through a nonlinear dynamic model. Both normal operation conditions as well as conditions of process disturbances were observed to evaluate the KBS performance. Specific user-defined disturbances were added to the simulation, and the KBS correctly diagnosed both process and sensor faults when present.

Robust Dead-Reckoning System with Function on Sensor-Fault Diagnosis for Mobile Robot.

An approach to sensor fault detection and identification (FDI) in the dead reckoning of mobile robots is proposed. It is based on the interacting multiple-model (IMM) algorithm. Changes of sensor normal/failure modes are explicitly modeled as switching from one mode to another in a probabilistic manner, and mode probabilities and estimates from single-model-based filters, each of which is based on a model matching to a particular system mode, interact with each other effectively. To provide better fault decision, a simple decision rule is incorporated into the IMM based algorithm. The proposed FDI algorithm is implemented on our mobile robot. Sixteen system modes (one normal mode and fifteen 'hard' sensor-failure modes) of four internal sensors (two wheel-encoders, a steering potentiometer and a yaw-rate gyro) are handled. Experimental results illustrate that the dead reckoning with the proposed FDI system allows robust motions of the robot subject to sensor failures.

Robust Dead Reckoning System with Interacting Multiple Model Based Sensor Fault Diagnosis for Mobile Robot

An interacting multiple-model (IMM) approach to sensor fault detection and identification (FDI) in the dead reckoning of mobile robots is proposed. Changes of sensor normal/failure modes are explicitly modeled as switching from one mode to another in a probabilistic manner, and the sensor fault diagnosis and the robot pose estimation are achieved via a bank of parallel Kalman filters. The dead reckoning algorithm is implemented on a skid-steered type of robot, where the thirty-two system modes (one normal mode and thirty-one hard sensor-failure modes) of five sensors (four wheel-encoders and one yaw-rate gyro) are handled. Experimental results validate the effectiveness of the dead reckoning system.

Multiple Sensor Disturbance Identification through Principal Component Analysis

Previous publications have focused upon the application of principal components analysis (PCA) for sensor fault identification through data reconstruction. These reconstruction based methods do not address the problem of fault propagation to other sensor measurements and as a consequence misleading fault identification can result. A novel method for the identification of multiple sensor disturbances during process monitoring is proposed based upon the T2-statistic. By using the T2-statistic in conjunction with the associated principal component score contribution plot, multiple sensor disturbances can be identified in a sequential manner. The methodology is demonstrated on two industrial plant simulations.

A unified geometric approach to process and sensor fault identification and reconstruction: the unidimensional fault case

Fault detection and process monitoring using principal component analysis (PCA) have been studied intensively and applied to industrial processes. This paper addresses some fundamental issues in detecting and identifying faults. We give conditions for detectability, reconstructability, and identifiability of faults described by fault direction vectors. Such vectors can represent process as well as sensor faults using a unified geometric approach. Measurement reconstruction is used for fault identification, and consists of sliding the sample vector towards the PCA model along the fault direction. An unreconstructed variance is defined and used to determine the number of principal components for best fault identification and reconstruction. The proposed approach is demonstrated with data from a simulated process plant. Future directions on how to incorporate dynamics and multidimensional faults are discussed.

Sensor Fault Detection and Identification in a Pilot Plant Under Process Control

The experimental evaluation of an automatic procedure for sensor fault detection and identification in a real process under closed-loop control is the objective of the present research. The scheme proposed here is very robust to faults in the main sensors of a multiloop control system, thus improving safety and reliability of plant operations. A state variable transformation is carried out in order to derive a model suitable for Recursive Least Squares (RLS) identification valid for all regimes of operation. The fault detection method is based on a moving window statistical analysis of the estimated model parameters. Simultaneously, a state estimation scheme, based on the Extended Kalman Filter (EKF), enables the fault identification, reduces false alarms and provides redundant measurements for alternative control purposes. Experimental runs were carried out in an industrial-scale pilot plant. Despite the large number of uncertainties and nonlinearities in the process, the system exhibited a good performance when faults occurred in the sensors of the control loops.

Joint diagnosis of process and sensor faults using principal component analysis

This paper presents a unified approach to process and sensor fault detection, identification, and reconstruction via principal component analysis. The principal component analysis model partitions the measurement space into a principal component subspace where normal variation occurs, and a residual subspace that faults may occupy. Both process faults and sensor faults are characterized by a direction vector, which describes the behavior of the fault. Fault reconstruction is accomplished by sliding the sample vector as close as possible to the principal component subspace. When the actual fault is assumed, the maximum reduction in the squared prediction error is achieved. A fault-identification index is defined in terms of the reconstructed squared prediction error. Fault detectability, reconstructability, and identifiability conditions are derived and demonstrated with a geometric interpretation. Numerous examples are provided to verify the method and conditions derived in the paper. An unreconstructed variance is defined and used to determine the number of principal components for best reconstruction. The proposed approach is applied to a data set from an industrial boiler process.

Detection, isolation, and identification of sensor faults in nuclear power plants

This paper presents methods to detect, locate, and identify sensor degradations. The first method is based on simple redundancy and consists in generating residuals by comparison of measurements provided by physically redundant sensors. The second uses analytical redundancy. Residuals are generated by comparing each measurement with an estimate computed from models of the process. The efficiency of each method is evaluated first on a simulated case and then on measurements obtained from a real power plant.

Identification of faulty sensors using principal component analysis

AbstractEven though there has been a recent interest in the use of principal component analysis (PCA) for sensor fault detection and identification, few identification schemes for faulty sensors have considered the possibility of an abnormal operating condition of the plant. This article presents the use of PCA for sensor fault identification via reconstruction. The principal component model captures measurement correlations and reconstructs each variable by using iterative substitution and optimization. The transient behavior of a number of sensor faults in various types of residuals is analyzed. A sensor validity index (SVI) is proposed to determine the status of each sensor. On‐line implementation of the SVI is examined for different types of sensor faults. The way the index is filtered represents an important tuning parameter for sensor fault identification. An example using boiler process data demonstrates attractive features of the SVI.

Sensor Fault Identification and Reconstruction Using Principal Component Analysis

This paper presents the use of Principal Component Analysis (PCA) for sensor fault identification via reconstruction. The principal component model captures the measurement correlations and reconstructs each variable by using iterative substitution and optimization. The effect of different sensor faults on model based residuals is analyzed and a new indicator called the Sensor Validity Index (SVI) is defined to determine the status of each sensor. An example using boiler process data demonstrates the attractive features of the SVI.

Use of principal component analysis for sensor fault identification

This paper make use of PCA for sensor fault identification via reconstruction. The principal component model captures the measurement correlations and reconstructs each variable to define associated residuals and a Sensor Validity Index (SVI). The filter applied to the SVI adds an important feature for sensor fault isolation because reduces the effect of false alarms and allows the identification of different types of sensor faults.

A scheme for fault tolerance in earth sensors

A system is presented that uses dual-redundant Earth sensors to measure pitch and roll errors of a three-axis stabilized space craft, with provision for autonomously detecting and identifying a faulty Earth sensor, and automatically selecting the outputs of the fault-free sensor for closed-loop attitude control, before failures cause major problems. A brief description is given of the system, and various failure modes of Earth sensors and their effects are discussed. Novel techniques and algorithms for automatic fault detection, identification, and reconfiguration (FDIR) of dual-redundant Earth sensors are developed. The algorithms are validated through computer simulations, and the results are presented. The proposed scheme can easily be implemented without much penalty on hardware, power consumption, and processing time.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Mathematical modeling of a steam generator for sensor fault detection

A dynamic model for a nuclear power plant steam generator (vertical, preheated, U-tube recirculation-type) is formulated as a sixth-order nonlinear system. The model integrates nodal mass and energy balances for the primary water, the U-tube metal and the secondary water and steam. The downcomer flow is determined by a static balance of momentum. The mathematical system is solved using transient input data from the Philippsburg 2 (FRG) nuclear power plant. The results of the calculation are compared with actual measured values. The proposed model provides a low-cost tool for the automatic control and simulation of the steam generating process. The “parity-space” algorithm is used to demonstrate the applicability of the mathematical model for sensor fault detection and identification purposes. This technique provides a powerful means of generating temporal analytic redundancy between sensor signals. It demonstrates good detection rates of sensor errors using relatively few steps of scanning time and allows the reconfiguration of faulty signals.

Fault Detection and Isolation in a Nuclear Reactor

A fault detection and identification methodology has been developed for sensor and plant component validation, with special emphasis on applications to nuclear powerplants. The methodology is particularly suitable for on-line fault diagnostics and does not rely on detailed knowledge of sensor and plant noise statistics. The algorithm has been computer coded for real-time applications and validated by on-line demonstration in an operating nuclear reactor. ARIOUS methods for fault detection and identification (FDI) of sensors have been reported in the literature.14 However, current practice in the nuclear industry is restricted to a few rather rudimentary techniques such as like-sensor comparisons, limit checking, auctioneering, etc. Although these techniques generally serve to improve system safety, availability, and operability, some limitations, such as the inability to identify gradual drifts and to detect common mode failures, significantly curtail their effectiveness. (If two or more elements fail identically, due to a common cause, the failure is called common mode.) The above limitations can often be circumvented with the aid of advanced computer-aided diagnostic techniques that have been developed for aerospace systems. In addition to improvement of plant availability and operability, these techniques promise to aid plant operators in making valid and timely decisions, thereby enhancing plant safety. The FDI methodology reported in this paper is developed on the basis of the parity space concept,3 which takes into account inconsistencies among all data sources. Any malfunctioning sensors are isolated by sequential checking until a relative consistency among the remaining (normal) sensors is achieved. This methodology does not require a detailed knowledge of sensor and plant noise statistics. Error bounds that are allowed for normal operation of the sensors are sufficient for making decisions. Real-time computer codes have been developed for detection and identification of failed sensors and plant components. As a proof of concept, these codes were verified by demonstration of on-line detection and identification of sensor failures in the 5 MW(t) nuclear reactor presently in operation at MIT, Cambridge, Mass.