Sensor Fault Detection Method Research Articles

Advancing Sensor Data Integrity with Deep Learning-Based Fault Detection

In the realm of modern data-driven systems, accurate and reliable sensor measurements are imperative for informed decision-making and system integrity. The objective of this project is to develop a robust sensor fault detection methodology leveraging the power of deep learning techniques. The ubiquity of sensors such as temperature and humidity sensors has led to a critical need for discerning between accurate readings and erroneous data, thereby enhancing the reliability of these measurements. This study addresses the prevailing challenge of sensor reliability by introducing a data-driven approach that harnesses deep learning algorithms to detect sensor faults promptly and accurately. A comprehensive dataset comprising sensor readings under diverse conditions forms the foundation of this investigation. Multiple cutting-edge deep learning architectures and techniques are systematically explored and evaluated against this dataset to identify the most efficient and precise sensor fault detection method. The outcomes of this research contribute significantly to advancing the state-of-the-art in sensor fault detection, with implications for a wide array of applications reliant on sensor measurements. By harnessing the capabilities of deep learning, this project presents a tangible solution to the challenge of accurately identifying faulty sensor data in real-time, thereby bolstering the dependability and efficacy of sensor-driven systems. Ultimately, this work underscores the potential of integrating advanced machine learning techniques in ensuring the reliability and precision of sensor data, heralding a new era of robust and trustworthy sensor-enabled environments.

Two-Level Integrity-Monitoring Method for Multi-Source Information Fusion Navigation

To address the issue of integrity monitoring for a multi-source information fusion navigation system, a theoretical framework of two-level integrity monitoring is proposed. Firstly, at the system level, a system-integrity-monitoring method based on the Kalman filter weighted least-squares form is established to detect and isolate faulty navigation sources. Secondly, at the sensor level, considering the redundancy of the faulty navigation sources, this paper presents the design of a multi-mode comprehensive fault-detection method for non-redundant navigation sources. Additionally, an extended-dimension matrix optimized sensor-fault detection and verification method for redundant navigation sources is proposed. Finally, integrity risk allocation criteria are established based on the effectiveness of navigation sources to calculate the system protection level and trigger integrity alarms. The two-level integrity-monitoring method was tested on a multi-source information fusion navigation system integrated with an inertial navigation system (INS), global positioning system (GPS), BeiDou satellite navigation system (BDS), Doppler velocity log (DVL), barometric altimeter (BA), and terrain-aided navigation (TAN). Test results demonstrate that the proposed method can effectively isolate the faulty navigation source within 10 s. Furthermore, it can detect the faulty sensors within the faulty navigation sources, thereby enhancing the reliability and robustness of the multi-source information fusion navigation system.

A sensor fault detection, isolation, and estimation method for intelligent vehicles

Accurate and reliable sensor information is crucial for intelligent vehicles to ensure safety and reliability. It is significant to devise effective solutions to diagnose possible sensor faults. In this paper, a sensor fault detection, isolation, and estimation approach is developed for intelligent vehicle-integrated motion systems. The Luenberger observer is designed and embedded into the fault diagnosis framework to real-time estimate the system states. Interval observers are constructed for vehicle subsystems to estimate the interval bounds of the corresponding observation errors. The estimation errors enter into the corresponding intervals generated by the adaptive laws when the sensor fault is free. In the event of a sensor fault, it can be detected efficiently and the faulty sensor can be isolated. Moreover, occurred sensor fault can be estimated by modifying the interval observer structure. The simulation results in the standard J-Turn test scenario are provided to evaluate and verify the effectiveness of the proposed method.

Sensor abnormality detection in multistage compressor units: A “white box” approach using tree-based genetic programming

Sensors are crucial in detecting equipment problems in various plant systems. In particular, detecting sensor abnormality is challenging in the case of utilizing the data which are acquired and stored offline (data logs). These data are normally acquired using Internet of Things (IoT) system and stored in a dedicated server. This situation presents both opportunities and challenges for exploration in sensor abnormality detection task. In this paper, we propose a multistage compressor sensor fault detection method using data logs. In the proposed method, the compressor sensor output is modeled as a function of other sensors using static approach. Subsequently, the model output is used for detecting abnormality by observing the residuals. It has been shown that the histogram of residuals offers rich information to predict abnormality of the targeted sensor. In particular, we explore the concept using Genetic Programming (GP) to generate regression model which offers more “white box” solution to the operators. There are various advantages in this approach. Firstly, the conventional “black box” approach lacks model transparency and, thus, is highly undesirable in critical systems. Secondly, equations are more easily applied in Programmable Logic Controller (PLC) if autonomous flagging is required. We also compare the proposed model with Multiple Linear Regression (MLR) and Neural Network Regression (ANN). Results show that the best generated models are comparable with the latter but with more crisp “white box” mathematical equations utilizing lesser feature inputs (four features only). This model yields R2 of 0.991 and RMSE of 2.1×10−2.

Data-Driven Sensor Fault Diagnosis Under Closed-Loop Control With Slow Feature Analysis

Recently, data-driven fault diagnosis techniques, especially multivariate statistical process monitoring methods, have been extensively investigated and widely applied to industries. Whether the monitored system is under closed- or open-loop control has a strong impact on the development of fault diagnosis strategies, but this point has not been taken seriously into consideration. This work aims to provide an effective data-driven method for sensor fault diagnosis under closed-loop control with modified slow feature analysis (SFA). The SFA is first revisited and compared with the well-known principal component analysis approach. Through theoretical analysis and an intuitive example, the influence of feedback control on sensor fault diagnosis is demonstrated, and to achieve successful detection, a strategy of incorporating more variables for monitoring is suggested. Then, based on SFA, a new sensor fault detection and classification method is developed. Its detectability analysis is carried out, based on which improved methods are proposed for efficient detection of incipient sensor faults. Finally, case studies on the continuous stirred tank reactor benchmark process demonstrate the effectiveness of the proposed method.

State Estimation in Spacecraft Power Distribution Systems Using Compound Current Sensors

The autonomous operation of the electric power system of a spacecraft is a key enabling functionality for deep-space missions. This functionality is informed by the real-time physical status of its power system, which is determined by processing the inputs obtained from various voltage and current sensors distributed throughout the electric network. In particular, it is important to have an accurate estimate of the current in each branch of the power system to detect overload conditions. However, voltage and current measurements can have an error due to noise and sensor faults. In this work, the focus is on ensuring accurate estimation of the voltages and currents in the presence of such faults. To this end, a method is proposed to optimally design a network of compound current sensors that enhances the ability to reliably identify faulty or biased current sensors. The proposed sensor fault detection method is validated using simulation results.

Single current sensor based fault tolerant control of interior permanent magnet synchronous machine for drive applications

This paper presents an integrated method for current sensor fault detection and fault tolerant control (FTC) for traction interior permanent magnet synchronous motor (IPMSM). The proposed current sensor fault detection method is based on detecting changes in the d-q axis current. The FTC is based on d-q axis current estimation from the reference d-q axis current and the phase current measured by the surviving current sensor. The current estimation process is independent of machine parameters. Hence the estimation is robust and requires less computational cost. The effectiveness of the FTC method is verified by the transient analysis. Such FTC is suitable for electric vehicle traction applications to ensure non-stop control operation of the drive in the entire range of speed. The efficacy of the proposed FTC method is tested through extensive simulations in MATLAB/Simulink environment. The real-time applicability of the proposed FTC method using the cost-effective digital signal processor (DSP) is verified on Texas Instruments© TMSF28379D through the processor in loop (PIL) simulation model.

Fault detection and anti-icing technologies in wind energy conversion systems: A review

Over the past few decades, wind energy has emerged to be one of the most promising and popular sources of renewable energy in different continents. The efficiency and robustness of wind energy conversion systems are highly dependent on the safe operational status of the system and the capability to achieve cost-effective operation. In order to prolong the systems’ operational time and increase the reliability of the wind turbines, predictive and preventive maintenances are required. This is important in reducing the operational and maintenance (O&M) costs of the wind energy conversion systems. This review paper acts as a platform for readers to comprehend the most recent, up-to-date and state-of-the-art approaches in detecting and diagnosing faults in wind energy conversion systems, including cepstral editing procedure, sensor fault detection and isolation method, induction abrasive particle sensor, novel model-based approach and machine learning based data-driven approach. Current innovations in anti-icing technologies, including resistive heaters, application of hybrid electro-heating and superhydrophobic coating, ultrasonic deicing, research on materials of ultrasonic transducers for anti-icing and a novel embedded anti-icing system as a part of the O&M measure to reduce the levelized cost of energy of the wind energy conversion systems are also presented and discussed in this paper.

Voltage sensor fault detection, isolation and estimation for lithium-ion battery used in electric vehicles via a simple and practical method

Accurate measurement information, especially precise voltage, is essential for model-based multi-state estimation algorithms of lithium-ion battery. Regarding the shortcomings in existing diagnosis methods, such as the difficulty in threshold value determination, low voltage sensor fault detection efficiency and the assumption of no multiple faults occur simultaneously, a simple and practical voltage sensor fault detection, isolation and estimation method is proposed in this paper. Firstly, the widely used algorithm by fusion of recursive least square and extend Kalman filter is adopted to obtain the measurement innovation (MI) between measured and estimated terminal voltage during online state-of-charge estimation process. Subsequently, focusing on the MI generated under faultless and faulty voltage sensors, a simple feature point (FP) identification-based voltage sensor detection and isolation method is further designed. Finally, with the identified FP's y-coordinate, a three-step based voltage sensor fault estimation method is developed to determine the fault mode and the corresponding fault value. Moreover, the proposed voltage sensor fault diagnosis method is verified by Federal Urban Driving Schedule test, whose results demonstrate that it can still realize immediate voltage sensor fault detection (at the moment of sensor fault occurrence) and accurate voltage sensor fault estimation even though there exist certain current sensor fault.

A novel dual time-scale voltage sensor fault detection and isolation method for series-connected lithium-ion battery pack

Considering the limitations in existing correlation coefficient-based, entropy-based and big data analysis-based voltage sensor fault diagnosis methods, we develop a novel dual time-scale voltage sensor fault detection and isolation method for series-connected lithium-ion battery pack in this paper. Firstly, a “ohmic resistance”-based selection method is periodically performed to artificially divide all in-pack cells into “representative cell” and non-representative cells. Secondly, during the “representative cell”-based battery pack state-of-charge (SOC) and cell SOC inconsistence estimation process, the measurement innovation (MI) between measured and estimated voltage of the “representative cell” and non-representative cells is generated in micro time-scale and macro time-scale, respectively. Regarding the “representative cell”, the faulty voltage sensor is immediately detected at the moment of the voltage sensor fault occurrence by catching the abnormal MI. As for the non-representative cells, through analyzing the discreteness degree of generated MI under faultless and faulty voltage sensors, an abnormal variance-based voltage sensor fault diagnosis method and an abnormal variance contribution-based fault location method are developed. The validation results through three sophisticated cases demonstrate that this method can rapidly catch the abnormal features for further voltage sensor fault diagnosis with low complexity and satisfactory robustness even though there exist certain faulty current.

An improved stacking ensemble learning-based sensor fault detection method for building energy systems using fault-discrimination information

Sensor fault detection is essential to maintain operations of heating, ventilation, and air conditioning systems (HVACs) in buildings. Data-driven sensor fault detection methods are becoming increasingly popular recently. However, for the relatively limited modelling data available in the heating, ventilation, and air conditioning system, it is a major challenge to extract fault-discrimination information (FDI) from limited data to develop a data-driven model with higher fault detection performance and lower false-alarm rate. Therefore, this study proposed an improved Stacking (IStacking) sensor fault detection method using fault-discrimination information. Fault-discrimination-information is extracted from original samples by subtracting the statistics with the threshold of each single model. Four different single models (i.e., principal component analysis, one-class support vector machine, K-Means clustering, and autoencoder) are employed to develop the ensemble learning detection method with relatively high generalization ability using a Stacking training manner. Sensor bias faults from two different sources are used for validation. Area under the ROC (receiver operating characteristic) curve (AUC) and false positive rate (FPR) are used to evaluate the detection performance and false-alarm rate. Results indicated that IStacking outperforms the four single models and traditional Stacking when the added biases are ranged [-4 °C, −2 °C] and [3.7 °C, 4 °C]. Since single models are stacked and fault-discrimination information is used to replace the original input, IStacking averagely increases Area under the ROC curve by 2.85% and decreases false positive rate by 5.76%.

Chemical static equipment commonly used sensor fault detection and isolation methods

This paper conducts a study on the faults of common sensors involved in chemical static equipment. Firstly, the types and characteristics of commonly used sensors of chemical static equipment are analyzed, and the characteristics of sensor output signal changes are summarized with the working characteristics of chemical equipment. Then the faults of static equipment sensors are classified and a fault model is established. Through the study of sensor fault detection and isolation methods at home and abroad, the overall scheme of sensor system fault detection and isolation combining single sensor fault detection and isolation method and multi-sensor fault detection and isolation method is proposed. According to the characteristics that chemical processes are generally in a dynamic and stable state and there is a certain correlation between the signals of each detection point in the equipment, a sensor system model is established by using the correlation of multiple sensors on the equipment, and when a sensor in the sensor system fails, the system model changes beyond the threshold value, and a different form of residual generation is used to determine which sensor is faulty and achieve the detection and isolation of faulty sensors. The fault detection method is simulated and studied by using relevant software, combined with a support vector machine and neural network toolbox. The results show that the method proposed in this paper can effectively complete the fault detection and isolation of sensors commonly used in chemical static equipment. The accuracy and reliability of the prediction model are high.

Simple and Robust Current Sensor Fault Detection and Compensation Method for 3-Phase Inverters

The present paper proposes a genuine, simple to implement, robust and reliable method for current sensors fault detection and compensation used for 3-phase inverters. Its functionality is based on an algorithm programed into a Field Programmable Gate Array (FPGA) as general controller. Usually, 3-phase inverters are controlled using field oriented control (FOC) which is generally triggered to sample the measured currents on the established switching frequency. As the latter is much smaller than the base clock of the FPGA, this allows hundreds of free time samples to perform other calculations in-between two FOC samplings. In this paper, a method that uses these free and unused time samples is presented. This method performs calculations for fault detection and compensation ensuring that at each new FOC sampling, this will receive the correct current data to reach continuous operation despite faults. The interleaving principle of the FOC with the fault detection method, as it will be proven in the paper, ensures high reliability of the complete controller diminishing the possibility of undesired or faulted readings to disturb the FOC’s calculations. The fault detection philosophy is based on continuous comparison of the instantaneous measured currents (sensor’s response) against reference values. The experimental results presented in the paper prove reliable operational performances of the method in both steady state and transient conditions. The added value of the paper consists in its genuine approach to handle the fault detection and compensation in-between two PWM ticks, ensuring that no faulted measurements can reach the control unit. This added value is based on functionality divided on several functions triggered by an internal generated clock synchronizing and handshaking their operations towards one goal: fault detection and compensation.

Sensor Fault Detection and Diagnosis Method for AHU Using 1-D CNN and Clustering Analysis.

This paper presents a fault detection and diagnosis (FDD) method, which uses one-dimensional convolutional neural network (1-D CNN) and WaveCluster clustering analysis to detect and diagnose sensor faults in the supply air temperature (Tsup) control loop of the air handling unit. In this approach, 1-D CNN is employed to extract man-guided features from raw data, and the extracted features are analyzed by WaveCluster clustering. The suspicious sensor faults are indicated and categorized by denoting clusters. Moreover, the Tc acquittal procedure is introduced to further improve the accuracy of FDD. In validation, false alarm ratio and missing diagnosis ratio are mainly used to demonstrate the efficiency of the proposed FDD method. Results show that the abrupt sensor faults in Tsup control loop can be efficiently detected and diagnosed, and the proposed method is equipped with good robustness within the noise range of 6 dBm∼13 dBm.

Sensor fault detection and diagnosis in the presence of outliers

In this paper, a sensor fault detection and diagnosis (FDD) method is proposed for linear state-space models in the presence of outliers. The t-distribution with unknown scale matrix and degrees of freedom (dof) parameter is used to describe the measurement noise. By using the variational Bayesian inference, the states, the scale matrix, and the dof parameter are estimated simultaneously. Since the noise distribution is no longer the Gaussian, a modified residual evaluation is proposed to detect the fault. After that, the cause of fault can be determined by observing the changes on measurement noise covariance. Two continuous stirred tank reactor (CSTR) process is conducted to demonstrate that the proposed method can provide more reliable FDD results than the existing methods when measurements contain outliers.

An enhanced PCA-based chiller sensor fault detection method using ensemble empirical mode decomposition based denoising

In heating, ventilating and air conditioning (HVAC) systems, sensor faults cause improper control strategy resulting in both energy penalty and service costs. As the first and foremost step of an entire on-line fault-tolerant control strategy, sensor fault detection is crucial for maintaining system operation performance. Principal component analysis (PCA) is a widely studied sensor fault detection method in HVAC area. However, the noise information contained in field sensor measurements may confuse the PCA model training process and deteriorate the sensor fault detection performance. To address the problem, this study presented an enhanced PCA-based sensor fault detection method using ensemble empirical mode decomposition (EEMD) denoising. The proposed EEMD-PCA method includes two steps: (1) EEMD is adopted to decompose and denoise the raw measured data, namely eliminating the possible noise information and extracting the possible useful information; (2) PCA is employed to establish the Q-statistic with threshold for sensor fault detection. A case study on a real screw chiller system was conducted to validate the proposed method. 11 sensors were chosen for modeling by qualifying chiller energy performance. Field operating data were used for validation with various magnitudes of added biases. Results revealed that EEMD-PCA showed better detection performance than PCA for 8 critical sensors. The fault detection ratios of the 8 sensors were increased by 22% at average. After pre-processed by the EEMD method, the denoised data were smoother than the raw data. Data distributions in the PCA residual subspace indicated that EEMD-PCA is sensitive to the smaller biases since the denoised data can separate faulty data from the normal data.

A Multi-mode Incipient Sensor Fault Detection and Diagnosis Method for Electrical Traction Systems

This paper proposes a data-driven sensor fault detection and diagnosis (FDD) method for electrical traction systems. Considering their switched characteristics, electrical traction systems can be regarded as switched systems. A mixture non-Gaussian data set will be formed, which can be firstly divided into six different operation modes, and principal component analysis (PCA) is then used for feature extraction in each mode. For two fault indicators in principal and residual subspaces, their probability density functions (PDFs) are estimated and used to determine reasonable thresholds for FDD. The proposed methodology extends the application of multivariate statistical technology to electrical traction systems. It can be applied easily and effectively without requirements on system parameters, and can deal with incipient sensor faults in traction system. Experiments with several different types of incipient sensor faults are conducted, which can demonstrate the effectiveness of the proposed method.

Chiller sensor fault detection based on empirical mode decomposition threshold denoising and principal component analysis

This paper presented a chiller sensor fault detection method based on Empirical Mode Decomposition (EMD) threshold denoising and Principal Component Analysis (EMD-TD-PCA). EMD-TD-PCA was developed to remove noise in the original data set so as to improve fault detection efficiency for temperature sensor bias faults. First, EMD was used to enhance the data quality by removing the noise contained in the raw data set. A threshold denoising way was employed since it only eliminated noise but maintained the useful information. Second, the data processed with EMD threshold denoising were used to build the PCA model. The Q-statistic was adopted to detect the sensor faults. The proposed method was compared with the traditional PCA method. The operational data of a screw chiller system in an electric factory were used to evaluate this method. Results show that the EMD-TD-PCA method can effectively improve the fault detection efficiency especially for temperature sensor faults with introduced biases ranging from −2°C to +2 °C at the 0.5 °C interval, and also there will be certain differences in fault detection efficiency for different sensors.

Improved sensor fault detection, diagnosis and estimation for screw chillers using density-based clustering and principal component analysis

Since the outdoor and indoor load conditions always change continuously, it may lead to the various operation characteristics in chillers. If a single principal component analysis (PCA) model is applied to detect and diagnose sensor faults under a wide range of chiller operation conditions, the various operation characteristics may confuse the sensor fault analysis process in two aspects. One is that the single PCA model may not be sensitive enough to detect less server faults due to the inadaptable and fixed fault detection boundary; the other is that the single PCA model can hardly distinguish some new normal operation conditions from serious sensor faults since it is impossible for a training PCA model to cover all chiller operation conditions. To overcome the limitations, this study proposed an improved sensor fault detection, diagnosis and estimation (FDD&E) method combining density-based clustering with PCA. Clustering analysis, i.e., the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) can automatically classify operation data into clusters and recognize the corresponding operation conditions. Instead of a single PCA model, using sub-PCA models to describe each normal operation condition improved the sensitivity and reliability of fault detection and diagnosis as well as the accuracy of sensor fault estimation. The proposed method was validated using field operation data of an existing screw chiller plant while various sensor faults of different magnitudes were introduced. Results reveal that the proposed method shows better sensor FDD&E results than conventional PCA-based sensor FDD&E method using a single PCA model.

A decentralized sensor fault detection and self-repair method for HVAC systems

This study proposes a novel decentralized sensor fault detection and self-repair method for heating, ventilation and air-conditioning systems. From the perspective of network structure, sensor fault diagnosis in heating, ventilation and air-conditioning systems is distributed to the updated smart sensors without the monitoring host, which is necessary in the traditional centralized method. A fully distributed flat sensor network is established based on fundamental physical equations. Similar to the structure, mechanism and characteristics of biological communities, a smart sensor needs only to communicate with adjacent nodes and operate collaboratively to complete sensor fault detection and self-repair tasks. These tasks are formulated as a constrained optimization and are solved by a decentralized algorithm with a penalty function executed in all the sensor nodes in parallel. The diagnosis model introduces an exponential function method to determine the precise location and undertake self-repair of a fault node. Simulation results on a chilled water system illustrate the effectiveness of the proposed method. Practical application: The traditional sensor fault detection and diagnosis methods for heating, ventilation and air-conditioning systems are based on a centralized structure with several deficiencies, such as high maintenance and labor costs, link congestion and operational lag. This study presents a decentralized sensor network structure and exponential-function-based method that possess the advantages of plug-and-play, rapid deployment, high flexibility and convenience for engineering implementation without having to build a central monitor. The efficiency and effectiveness of the proposed method are demonstrated via a case study.