For Fault Detection And Diagnosis Research Articles

A systematic approach for data generation for intelligent fault detection and diagnosis in District Heating

This study introduces a novel systematic approach to address the challenge of labeled data scarcity for fault detection and diagnosis (FDD) in District Heating (DH) systems. To replicate real-world DH fault scenarios, we have created a controlled laboratory emulation of a generic DH substation integrated with a climate chamber. Furthermore, we present an FDD pipeline using an isolation forest and a one-class support vector machine for fault detection alongside a random forest and a support vector machine for fault diagnosis. Our research analyzed the impact of data sampling frequencies on the FDD models, revealing that shorter intervals, such as 1-min and 5-min, significantly improve FDD performance. We provide detailed information on six scenarios, including normal operation, a minor valve leak, a valve leak, a stuck valve, a high heat curve, and a temperature sensor deviation. For each scenario, we present their signature, quantifying their unique behavior and providing deeper insights into the operational implications. The signatures suggest that, while variable, faults have a consistent pattern seen in the generic DH substation. While this work contributes directly to the DH field, our methodology also extends its applicability to a broader context where labeled data is scarce.

Residual squeeze-and-excitation convolutional auto-encoder for fault detection and diagnosis in complex industrial processes

Recently, deep learning has attracted increasing attention in process monitoring. However, it is still a big challenge to extract effective features of process data for fault detection and diagnosis (FDD) by using deep learning-based methods. In this study, a novel deep learning-based model, residual squeeze-and-excitation convolutional auto-encoder (RSECAE) is proposed for FDD in complex industrial processes. A hybrid structure that integrates convolution calculation and auto-encoder is proposed to construct multiple convolution layers and deconvolution layers in the encoder and decoder respectively. Thus, RSECAE can learn effective features from the process signals in an unsupervised manner. At the same time, multiple residual squeeze-and-excitation networks (RSENets) are embedded in the deep network, which effectively preserves the critical feature representations of the convolution layer and transmits them to the decoder in the form of residuals. Finally, maximum mean discrepancy (MMD) term is utilized in RSECAE to reduce the distribution difference between the process signals and the learned features. This enables RSECAE to extract effective features with prior distribution information. RSECAE provides a new solution for process FDD for complicated industrial processes. Two industrial cases are adopted to validate the effectiveness of the RSECAE in process FDD. The experimental results demonstrate the effectiveness and superiority of RSECAE in process fault detection and fault diagnosis.

A new integrated framework to fault detection and diagnosis of air handling unit: Emphasizing the impact of symptoms

This study proposes a comprehensive framework for Fault Detection and Diagnosis (FDD) in Air Handling Units (AHU), emphasizing the impact of symptoms associated with various faults. The aim is to address an important limitation in FDD model development where detection is based simply on fault intensity without considering symptom impact or subjective severity criteria across faults. Instead, a new approach is introduced that factors in fault impact, utilizing impact analysis to enhance FDD accuracy and efficiency. The methodology involves three interconnected components: First, a fault impact analysis categorizes 18 fault types into stages per intensity and assesses resulting symptom impacts. Second, symptom-based intensity thresholds are established to categorize fault severity into three levels based on symptom severity. Finally, the goal is to integrate these findings into the FDD process for optimized fault detection and diagnosis. The study successfully categorized 18 faults into three severity levels, with thresholds identified for each. Furthermore, Tree-based models demonstrated effective performance when conducting FDD based on these levels. The approach provides a clear framework through three linked processes. This research combines comprehensive impact analysis with sophisticated classification methods, presenting a more defined, systematic, and efficient FDD approach. This novel methodology substantially advances fault detection and diagnosis, especially for building energy systems.

Applying OPTICS with and without PCA for fault detection of fan coil units using building automation system data

In building operations, Heating, Ventilation, and Air Conditioning (HVAC) system faults lead to substantial energy waste. Due to the rapid growth in the availability of sensing data and advancements in computational technology, AI-based methods have emerged as a powerful tool for Fault Detection and Diagnosis (FDD) of HVAC systems, significantly enhancing the efficiency of building operations. Several studies support supervised learning methods for FDD; however, these are limited in practice due to their reliance on labeled datasets, which are generally unavailable. In this context, unsupervised methods, notably those using cluster analysis, show great promise, particularly for analysing Building Automation System (BAS) data. This paper presents a methodology using OPTICS – an ordered clustering algorithm – for fault detection and explores the impact of PCA on its accuracy for fault detection, using both published and full-scale historical building data. While beneficial for reducing computational costs and enhancing noise reduction — thereby generally improving the clarity of cluster differentiation — PCA’s dimensionality reduction may result in the loss of critical information, leading to some clusters being less discernible or entirely undetected. These overlooked clusters could be indicative of underlying faults, and their obscurity represents a significant limitation of PCA when identifying potential faults in complex data.

Investigation of smart thermostat fault detection and diagnosis potential for air-conditioning systems using a Modelica/EnergyPlus co-simulation approach

Smart thermostats offer low-cost alternative for fault detection and diagnosis (FDD) in residential air-conditioning (AC) systems. However, smart thermostats do not directly measure AC performance, but measures indoor air conditions which reflect the indoor air responses to both AC operations and other factors like weather and building gains. Estimating uncontrollable building gain disturbances is essential to differentiate them from AC impacts. Since the uncontrollable disturbances are difficult to monitor using smart thermostat data, there is need to investigate the actual capability of smart thermostats for FDD. In this study, an integrated building and Vapor Compression cycle-based AC model was developed and simulated using EnergyPlus/Spawn and Dymola. The simulation was performed under ten scenarios covering relevant gains, low charge, and low indoor airflow faults. The sensitivity of the time-to-cool feature under these scenarios was studied. The results show about 10% time-to-cool increase from varied internal gains and infiltration. Meanwhile, low charge at 10%, 20% and 30% severity increased time-to-cool by about 3%, 6%, and 20%, respectively, suggesting that if these gains are neglected in an FDD method, it might not be possible to detect low charge below 30% using smart thermostat data since even without a fault, the combined impact of those gains already exceeded that for 20% low charge. For low indoor airflow at 10%, 20% and 30% severity, the increase in time-to-cool was about 3%, 9% and 12%, meaning that only low indoor airflow fault above 30% can be confidently detected if those gains are not captured. Thus, this study demonstrates the need to consider some of these gains in developing smart thermostat based FDD algorithms.

An intelligent fault detection and diagnosis model for refrigeration systems with a comprehensive feature selection method

Feature selection and model establishment are two essential steps for fault detection and diagnosis (FDD) of refrigeration systems. A robust and powerful FDD model combined with a suitable feature selection method can exhibit excellent performance in FDD tasks for refrigeration systems. In this study, a novel FDD method that integrates a comprehensive feature selection method and a deep learning-based intelligent FDD model is proposed. Including three steps, the comprehensive feature selection method combines filter methods and wrapper methods. It can optimize the features and the model jointly by using the multi-objective optimization algorithm to achieve a better performance. In addition, a novel FDD model that combines one-dimensional convolutional neural network (1D-CNN) and self-attention (SA) mechanism is proposed based on the deep learning technology. To evaluate the proposed method, experiments are performed on a miniature refrigeration system under 4 situations with multiple working conditions, forming a dataset for the FDD study. The proposed three-step feature selection method is utilized to obtain the best feature subset. The 1D-CNN and SA FDD model is constructed and the model is jointly optimized with the features. Several comparisons are carried out to demonstrate the effectiveness and superiority of the proposed feature selection method and the FDD model. The results demonstrate that the presented integrated optimization achieved a test accuracy of around 99.66 %, surpassing other popular FDD models including MLP, CNN, and LSTM.

Fault detection and diagnosis using the dynamic network framework

A local model-based method for fault detection and diagnosis (FDD) in large-scale interconnected network systems is introduced, using models in a dynamic network framework. To this end, model validation methods are developed for validating single modules in a dynamic network, which are generalized from the classical auto- and cross-correlation tests for open- and closed-loop systems. Invalidation of the model can indicate the detection of a fault in the system. A fault diagnosis algorithm is developed that includes fault isolation and optimal placement of external excitation signals. Numerical illustrations demonstrate the method’s capability to detect a fault in a local module and isolate it within the entire network system.

Shared Parameter Network: An efficient process monitoring model

The use of deep learning methods for fault detection and diagnosis (FDD) has continued to gain research attention due to the continuous availability of data and the need for reliable FDD methods. In our previous work, we proposed a probabilistic bidirectional recurrent network (PBRN) for process prediction and detection of faults. A limitation of PBRN is the model size. In this paper, we reduce the computational requirements of the original PBRN by introducing a shared parameter network (SPN). The SPN uses a shared parameter space to learn relevant features for process prediction and fault detection. This shared parameter setting reduces the model size and training time significantly. Using the SPN, we achieved a 76 percent reduction in model size and a 48 percent reduction in training time. The training validation loss profile and fault detection performances of both models also demonstrate the superiority of the SPN.

Transfer learning methodology for machine learning based fault detection and diagnostics applied to building services

Machine Learning (ML) models for Fault Detection and Diagnosis (FDD) can automatically detect anomalies in the operation in large facilities or district heating networks and can help tackling energy wastes. Nevertheless, the development of ML-models is a costly and tedious task requiring large amounts of labelled data. Setting up ML-models for a high number of systems is effort and know-how intensive. However, assets like commercial buildings and district heating networks are constituted of systems with similar topologies. Transferring a ML model initially trained on a source system to a multitude of similar target systems, can help reducing the training costs and facilitating the scalability of ML-based FDD in those assets. To enable this, we have developed a methodology that assesses the potential for Transfer Learning (TL) from a source system to target systems by determining the covariate and concept shifts between the source and target domains and integrating the source model into the target system if the TL assessment is positive. We used a patented method for the model development, that combines two ML-models, that are initially trained on a source system by means of a feedback system. We implemented this methodology on district heating (DH) substations, as DH systems typically contain this kind of subsystems with similar topologies and have thus a high scalability potential for TL. Initial findings showed the effectiveness of TL in adapting the source model to the target domain, resulting in enhanced FDD capabilities with significantly reduced training efforts.

Interpret what a Convolutional Neural Network learns for fault detection and diagnosis in process systems

The focus of this work is on an interpretation strategy on what a Convolutional Neural Network (CNN) has learned for fault detection and diagnosis (FDD) in process systems. Frequency spectra of process variables obtained by Continuous Wavelet Transform (CWT) are adopted as input features. Then, a CNN structure is designed to represent the mappings from input frequency features to different operation conditions. The Layer-wise Relevance Propagation (LRP) strategy is utilized to gain the relevance of each frequency feature to the classification performance. The formulations of relevance propagation for 4 types of CNN layers are presented in detail. The relevance scores are then depicted in heatmaps, where the pixels’ colors denote the contribution degrees and the most significant frequency features are considered as the major bases that the CNN discriminates different operation situations. The proposed interpretation strategy is experimented on the Tennessee Eastman process benchmark. The testing results demonstrate the efficiency of the strategy in interpreting what the CNN has learned to distinguish normal or faulty conditions in the FDD task.

Similarity learning-based fault detection and diagnosis in building HVAC systems with limited labeled data

Machine learning has been widely adopted for fault detection and diagnosis (FDD) in heating, ventilation and air conditioning (HVAC) systems over the past decade due to the ever-increasing availability of massive building operational data. Machine learning-based FDD is flexible and accurate but heavily relies on the availability of sufficient labeled data to develop supervised or unsupervised models. However, collecting labeled data is usually labor-intensive for various types of faulty conditions, significantly limiting the practical implementation of machine learning-based FDD. Therefore, this study proposes a similarity learning-based method using Siamese networks to improve the performance of machine learning-based FDD in applications with limited labeled data. Unlike the conventional supervised approach, the proposed Siamese networks contain two identical long short-term memory subnetworks which take a pair of multivariate time-series samples from the building energy management system as input. The number of training samples can be significantly augmented by generating pairs randomly. In this way, the generalization ability of the machine learning-based FDD is significantly improved in practical applications. Two case studies were designed and conducted using experimental data when labeled data were limited and imbalanced to validate the proposed similarity learning-based method. In case 1, the proposed method improves the fault diagnostic accuracy by at most 45.7% compared with the baseline model when the number of labeled data is limited. In case 2, the proposed method demonstrated better generalization ability when the labeled data is imbalanced.

Generation and evaluation of a synthetic dataset to improve fault detection in district heating and cooling systems

This paper investigates various types of faults in District Heating & Cooling (DHC) systems. Many authors point out that the lack of data hinders the development of good data-driven models for fault detection and diagnosis (FDD). In this work, we design a reference dataset based on simulation and use it to evaluate Machine Learning (ML) models for fault detection.The dataset itself covers six types of DHC system components, covering production, distribution and storage. It is provided as Open Data with corresponding documentation. Most of the models used for generating the dataset are provided as Open Source code.To assess the usefulness of the dataset, we evaluated five ML models on five fault detection tasks. The results highlight varying level of performance on the considered tasks, with faults related to global energy efficiency being easier to handle than those related specifically to thermal losses. Three of the investigated models (Logistic Regression, Support Vector Machine, and XGBoost) provide consistent performance on the considered tasks, achieving accuracy scores up to 99% on the easier tasks and above 66% on one of the more difficult tasks. We also illustrate possibilities for transferring models to real systems with different characteristics, with encouraging results.

Integrated generative networks embedded with ensemble classifiers for fault detection and diagnosis under small and imbalanced data of building air condition system

Faults in building Heating, Ventilation, and Air-condition (HVAC) system create an uncomfortable indoor environment and cause energy waste. The data-driven method has been widely applied for Fault Detection and Diagnosis (FDD) in the complex building HVAC system. This method relies on the availability of many fault data which is difficult to collect. This makes it quite challenging to apply the data-driven methods for the FDD of the HVAC system. Thus, a novel data-driven FDD method that only utilizes small fault data collected from a Variable Refrigerant Flow air condition system has been proposed. Under different conditions, the fault and normal data are collected in an enthalpy difference laboratory to create small and imbalanced data. A generative network is developed by combining Wasserstein Generative Adversarial Network with Gradient Penalty and Variational Auto-Encoder. To improve the FDD classifier’s accuracy and to train an end-to-end network model using small and imbalanced data, two ensemble classifiers are embedded into the generative network. The dataset includes normal and fault data have been applied to train the modified generative network, and two ensemble classifiers are used to detect and diagnose the fault, respectively. The performance indexes show that the proposed method is much better than the SMOTE-based methods in almost all training groups. Besides, the comparison between the proposed method and generative network with a single classifier indicates that the ensemble classifiers can improve the F1-score of fault detection and the accuracy of fault diagnosis.

A novel image-based transfer learning framework for cross-domain HVAC fault diagnosis: From multi-source data integration to knowledge sharing strategies

Data-driven classification models have gained increasing popularity for fault detection and diagnosis (FDD) tasks considering their advantages in implementation flexibility and modeling accuracies. To tackle the wide existence of data shortage challenges for individual buildings, transfer learning can be adopted to enhance the applicability of data-driven approaches. At present, limited studies have been conducted to explore the potentials of transfer learning in HVAC FDD tasks, leaving the following two key questions unanswered, i.e., (1) whether the tabular data collected from different building systems can be effectively integrated and utilized as the source data for transfer learning, and (2) whether the operational patterns learnt from a specific building system can be interchangeably applied for FDD tasks of other systems. This study proposes a novel image-based transfer learning framework to tackle the multi-source data compatibility challenge in the building field, while investigating the value of transfer learning in cross-domain FDD tasks. Data experiments have been designed to quantify the value of transfer learning given different data amounts, imbalance ratios, and transfer learning strategies. The research results validate the usefulness of image-based transfer learning for HVAC FDD tasks. The insights obtained are valuable for multi-source building operational data integration and cross-domain knowledge sharing.

Machine learning and deep learning based methods toward industry 4.0 predictive maintenance in induction motors: State of the art survey

Purpose: Developments in Industry 4.0 technologies and Artificial Intelligence (AI) have enabled data-driven manufacturing. Predictive maintenance (PdM) has therefore become the prominent approach for fault detection and diagnosis (FD/D) of induction motors (IMs). The maintenance and early FD/D of IMs are critical processes, considering that they constitute the main power source in the industrial production environment. Machine learning (ML) methods have enhanced the performance and reliability of PdM. Various deep learning (DL) based FD/D methods have emerged in recent years, providing automatic feature engineering and learning and thereby alleviating drawbacks of traditional ML based methods. This paper presents a comprehensive survey of ML and DL based FD/D methods of IMs that have emerged since 2015. An overview of the main DL architectures used for this purpose is also presented. A discussion of the recent trends is given as well as future directions for research.Design/methodology/approach: A comprehensive survey has been carried out through all available publication databases using related keywords. Classification of the reviewed works has been done according to the main ML and DL techniques and algorithmsFindings: DL based PdM methods have been mainly introduced and implemented for IM fault diagnosis in recent years. Novel DL FD/D methods are based on single DL techniques as well as hybrid techniques. DL methods have also been used for signal preprocessing and moreover, have been combined with traditional ML algorithms to enhance the FD/D performance in feature engineering. Publicly available datasets have been mostly used to test the performance of the developed methods, however industrial datasets should become available as well. Multi-agent system (MAS) based PdM employing ML classifiers has been explored. Several methods have investigated multiple IM faults, however, the presence of multiple faults occurring simultaneously has rarely been investigated.Originality/value: The paper presents a comprehensive review of the recent advances in PdM of IMs based on ML and DL methods that have emerged since 2015.

Prescribed Performance Neuroadaptive Fault-Tolerant Compensation for MIMO Nonlinear Systems Under Extreme Actuator Failures

This article investigates the issue of neuroadaptive tracking control for a family of unknown multi-input multi-output (MIMO) nonlinear uncertain systems subject to extreme actuation failures. Different from most existing methods that are built upon partial loss of actuation effectiveness, here, in this article, we explicitly consider the situation that some actuators at some particular channel completely fail to work, an issue that has not been well addressed. By integrating the neural network approximation technique with two error transformations, a neuroadaptive fault-tolerant control strategy is developed with two attractive features: 1) it is capable of coping with the scenario that some of the actuators suffer from extreme actuation faults without the need for fault detection and diagnosis (FDD)/fault detection and isolation (FDI) or actuator switching and 2) the tracking error is forced to converge to a prescribed residual (symmetric or asymmetric) boundary at a preassignable decay mode within a prechosen finite settling time despite actuator failures and external disturbances. Numerical simulation studies confirm the effectiveness and benefits of the presented control approach.

Flight evaluation of simultaneous actuator/sensor fault reconstruction on a quadrotor minidrone

In view of the increase in the number of Unmanned Aerial Vehicles (UAVs) in the commercial and private sectors, it is imperative to make sure that such systems are safe, and thus resilient to faults and failures. This paper considers the numerical design and practical implementation of a linear parameter-varying (LPV) sliding mode observer for Fault Detection and Diagnosis (FDD) of a quadrotor minidrone. Starting from a nonlinear model of the minidrone, an LPV model is extracted for design, and the observer synthesis procedure, using Linear Matrix Inequalities (LMI), is detailed. Simulations of the observer FDD show good performance. The observer is then implemented on a Parrot® Rolling Spider minidrone and a series of flight tests is performed to assess the FDD capabilities in real time using its on-board processing power. The flight tests confirm the performance obtained in simulation, and show that the sliding mode observer is able to provide reliable fault reconstruction for quadrotor minidrone systems.

Reduced Gaussian process regression based random forest approach for fault diagnosis of wind energy conversion systems

This paper proposes a novel Reduced Gaussian Process Regression (RGPR)-based Random Forest (RF) technique (RGPR-RF) for fault detection and diagnosis (FDD) of wind energy conversion (WEC) systems. First, two RGPR models are proposed to deal with WEC features extraction and selection. The proposed RGPR models extract the most relevant information from the WEC system data while reducing the computation burden compared to the classical GPR model. The complexity reduction is ensured by the selection of the most effective samples through the dimensionality reduction (DR) metrics including Hierarchical K-means (HKmeans) clustering and Euclidean distance (ED). Next, in order to classify the WEC faults and improve the diagnosis abilities, RF classifier is developed. The proposed RGPR H K m e a n s -RF and RGPR E D -RF techniques boost the classification speed and accuracy using a reduced number of features where only the most relevant and sensitive characteristics are kept in case of redundancy. The open-circuit, wear-out, and short-circuit are the three transistor faults considered in order to illustrate the effectiveness and robustness of the developed techniques. The obtained results show that the proposed RGPR-RF technique is characterized by a low computation time and high diagnosis accuracy (an average accuracy of 99.9%) compared to the conventional RF classifiers.

Data-Driven Fault Diagnosis for Electric Drives: A Review

The need to manufacture more competitive equipment, together with the emergence of the digital technologies from the so-called Industry 4.0, have changed many paradigms of the industrial sector. Presently, the trend has shifted to massively acquire operational data, which can be processed to extract really valuable information with the help of Machine Learning or Deep Learning techniques. As a result, classical Condition Monitoring methodologies, such as model- and signal-based ones are being overcome by data-driven approaches. Therefore, the current paper provides a review of these data-driven active supervision strategies implemented in electric drives for fault detection and diagnosis (FDD). Hence, first, an overview of the main FDD methods is presented. Then, some basic guidelines to implement the Machine Learning workflow on which most data-driven strategies are based, are explained. In addition, finally, the review of scientific articles related to the topic is provided, together with a discussion which tries to identify the main research gaps and opportunities.

A data-driven Bayesian network learning method for process fault diagnosis

This paper presents a data-driven methodology for fault detection and diagnosis (FDD) by integrating the principal component analysis (PCA) with the Bayesian network (BN). Though the integration of PCA-BN for FDD purposes has been studied in the past, the present work makes two contributions for process systems. First, the application of correlation dimension (CD) to select principal components (PCs) automatically. Second, the use of Kullback-Leibler divergence (KLD) and copula theory to develop a data-based BN learning technique. The proposed method uses a combination of vine copula and Bayes’ theorem (BT) to capture nonlinear dependence of high-dimensional process data which eliminates the need for discretization of continuous data. The data-driven integrated PCA-BN framework has been applied to two processing systems. Performance of the proposed methodology is compared with the independent component analysis (ICA), kernel principal component analysis (KPCA), kernel independent component analysis (KICA), and their integrated frameworks with the BN. The comparative study suggests that the proposed framework provides superior performance.