Actuator faults in unmanned aerial vehicles (UAVs) can have significant and potentially adverse effects on their safety and performance, highlighting the critical importance of fault diagnosis in UAV design. Ensuring the reliability of these systems in various applications often requires the use of advanced diagnostic algorithms. Artificial intelligence methods, such as deep learning and machine learning techniques, enable fault diagnosis through sample-based learning without the need for prior knowledge of fault mechanisms or physics-based models. However, UAV fault datasets are typically small due to stringent safety standards, which presents challenges for achieving high-performance fault diagnosis. To address this, deep reinforcement learning (DRL) algorithms offer a unique advantage by combining deep learning’s automatic feature extraction with reinforcement learning’s interactive learning approach, improving both learning capabilities and robustness. In this study, we propose and evaluate two DRL-based fault diagnosis models, which demonstrate remarkable accuracy in fault diagnosis, consistently exceeding $99{\rm{\% }}$. Notably, under small sample scenarios, the proposed models significantly outperform traditional classifiers such as decision trees, support vector machines, and multilayer perceptron neural networks. These findings suggest that the integration of DRL enhances fault diagnosis performance, particularly in data-limited environments.

With the widespread application of proton exchange membrane fuel cells (PEMFCs), ensuring the safe and reliable operation of the PEMFCs is becoming more and more important. Timely diagnosis of fault types and the implementation of targeted interventions are crucial for addressing these challenges. In this study, a simulated PEMFC model is firstly introduced by using Fluent, and the effectiveness is validated through experiments involving membrane dry faults, water flooding faults, normal states, and unknown states. Then, a data-driven deep learning convolutional neural network, YOLOv5-CG-AS, is developed, which employs the EfficientViT network as the backbone, incorporating lightweight improvements through the proposed CG-AS attention layer. The results demonstrate that YOLOv5-CG-AS can automatically extract fault features from images for offline fault diagnosis and can perform real-time online diagnosis of multiple parameter curves of PEMFCs. Moreover, the experimental results have validated the feasibility and effectiveness of proposed method and shown the average precision mean Average Precision (mAP) of the trained model reaches 99.50%, superior than other conventional strategies. This has significant implications for advancing fault diagnosis methods, enhancing the reliability and durability of PEMFC systems, and promoting further development in the field.

The performance and confidence in fault detection and diagnostic systems can be undermined by data pipelines that feature multiple compounding sources of uncertainty. These issues further inhibit the deployment of data-based analytics in industry, where variable data quality and lack of confidence in model outputs are already barriers to their adoption. The methodology proposed in this paper supports trustworthy data pipeline design and leverages knowledge gained from one fully-observed data pipeline to a similar, under-observed case. The transfer of uncertainties provides insight into uncertainty drivers without repeating the computational or cost overhead of fully redesigning the pipeline. A SHAP-based human-readable explainable AI (XAI) framework was used to rank and explain the impact of each choice in a data pipeline, allowing the decoupling of positive and negative performance drivers to facilitate the successful selection of highly-performing pipelines. This empirical approach is demonstrated in bearing fault classification case studies using well-understood open-source data.

Current fault diagnosis (FD) methods for heating, ventilation, and air conditioning (HVAC) systems do not accommodate for system reconfigurations throughout the systems’ lifetime. However, system reconfiguration can change the causal relationship between faults and symptoms, which leads to a drop in FD accuracy. In this paper, we present Fault-Symptom Brick (FSBrick), an extension to the Brick metadata schema intended to represent information necessary to propagate system configuration changes onto FD algorithms, and ultimately revise FSRs. We motivate the need to represent FSRs by illustrating their changes when the system reconfigures. Then, we survey FD methods’ representation needs and compare them against existing information modeling efforts within and outside of the HVAC sector. We introduce the FSBrick architecture and discuss which extensions are added to represent FSRs. To evaluate the coverage of FSBrick, we implement FSBrick on (i) the motivational case study scenario, (ii) Building Automation Systems’ representation of FSRs from 3 HVACs, and (iii) FSRs from 12 FD method papers, and find that FSBrick can represent 88.2% of fault behaviors, 92.8% of fault severities, 67.9% of symptoms, and 100% of grouped symptoms, FSRs, and probabilities associated with FSRs. The analyses show that both Brick and FSBrick should be expanded further to cover HVAC component information and mathematical and logical statements to formulate FSRs in real life. As there is currently no generic and extensible information model to represent FSRs in commercial buildings, FSBrick paves the way to future extensions that would aid the automated revision of FSRs upon system reconfiguration.

The tribological behavior can be informative about the incipient faults of robot manipulators. This study explores the evolution of friction characteristics from cold start to thermal equilibrium through a series of steady-state friction experiments. Based on these experimental observations, a friction-based fault diagnosis framework is proposed. The fault diagnosis process primarily involves defining the healthy state, decomposing friction curves and their features, and anomaly detection. Given the dependence of friction characteristics on different sources of faults, the parameters of steady-state experimental friction model are divided into two categories: one associated with contact interactions and the other related to non-contact regimes. Subsequently, confidence regions corresponding to distinguishable friction characteristics are independently constructed. These regions encapsulate the statistical description of the healthy state, characterized by mean values and the covariance of the friction characteristic parameter vectors during the unloaded state. In addition, we conduct experiments that consider the influence of applied loads on friction behavior. These experiments serve as a test set for comparison against nominal statistics. Leveraging the similarity between the effects of wear and load on friction, we introduce equivalent load thresholds to assess the severity of joint degradation. The results demonstrate the feasibility of employing confidence region views based on friction characteristic classification for fault detection and isolation.

Sensed data from high-value engineering systems is being increasingly exploited to optimise their operation and maintenance. In aerospace, returning all measured data to a central repository is prohibitively expensive, often causing useful, high-value data to be discarded. The ability to detect, prioritise and return useful data on asset and in real-time is vital to move toward more sustainable maintenance activities.

We present a data-driven solution for on-line detection and prioritisation of anomalous data that is centrally processed and used to update individualised digital twins (DT) distributed onto remote machines. The DT is embodied as a convolutional neural network (CNN) optimised for real-time execution on a resource constrained gas turbine monitoring computer. The CNN generates a state prediction with uncertainty, which is used as a metric to select informative data for transfer to a remote fleet monitoring system. The received data is screened for faults before updating the weights on the CNN, which are synchronised between real and virtual asset.

Results show the successful detection of a known in-flight engine fault and the collection of data related to high novelty pre-cursor events that were previously unrecognised. We demonstrate that data related to novel operation are also identified for transfer to the fleet monitoring system, allowing model improvement by retraining. In addition to these industrial dataset results, reproducible examples are provided for a public domain NASA dataset.

The data prioritisation solution is capable of running in real-time on production-standard low-power embedded hardware and is deployed on the Rolls-Royce Pearl 15 engines.

As an important index to quantitatively measure the motion performance of a manipulator, motion reliability is affected by many factors, such as joint clearance. The present research utilized a UR10 manipulator as the research object. A factor mapping model for influencing the motion reliability was established. The link flexibility factor, joint flexibility factor, joint clearance factor, and Denavit–Hartenberg (DH) parameters were comprehensively considered in this model. The coupling relationship among the various factors was concisely expressed. Subsequently, the nonlinear response surface method was used to calculate the reliability and sensitivity of the manipulator, which provided an applicable reference for its trajectory planning and motion control. In addition, a data-driven fault diagnosis method based on the kernel principal component analysis (KPCA) was used to verify the motion accuracy and sensitivity of the manipulator, and joint rotation failure was considered as an example to verify the accuracy of the KPCA method. This study on the motion reliability of the manipulator is of great significance for the current motion performance, adjusting the control strategy and optimizing the completion effect of the motion task of a manipulator.

A data-driven approach for multiclass fault diagnosis of drive fed induction motor (IM) using stator current at steady-state condition is a complex pattern classification problem. The applied DWT-IDWT algorithm in this work is reinforced by a novel selection criterion for mother wavelet application and justifies the originality of the work. This investigation has exploited the built-in feature selection process of Random Forest (RF) classifier to resolve the most challenging issues in this area, including bearing and stator fault detection. RF has shown an outstanding performance without application of any feature selection technique because of its distributive feature model. The robustness of the results backed by the experimental verification shows an encouraging future of RF as a classifier in the area of intelligent fault diagnosis of IM.

This paper considers actuator redundancy management for a redundant multirotor Unmanned Aerial Vehicle (UAV) under actuators failures. Different approaches are proposed: using robust control (passive fault tolerance), and reconfigurable control (active fault tolerance). The robust controller is designed using high-order super-twisting sliding mode techniques, and handles the failures without requiring information from a Fault Detection scheme. The Active Fault-Tolerant Control (AFTC) is achieved through redistributing the control signals among the healthy actuators using reconfigurable multiplexing and pseudo-inverse control allocation. The Fault Detection and Isolation problem is also considered by proposing model-based and model-free modules. The proposed techniques are all implemented on a coaxial octorotor UAV. Different experiments with different scenarios were conducted for the validation of the proposed strategies. Finally, advantages, disadvantages, application considerations and limitations of each method are examined through quantitative and qualitative studies.

This work deals with the application of the fuzzy logic to automate diagnosis of bearing defects in rotating machines based on vibration signals. The classification tool used is a fuzzy inference system (FIS) of Mamdani type. The vector form of input contains parameters extracted from the signals collected from the test bench studied. The output vector contains the classes for the different operating modes of the experimental study. The results show that; pretreatment data (filtering, decimation,...), the choice of parameters of fuzzy inference system (input variables and output, types and parameters of membership functions associated with different input and output variables of the system, the generation of fuzzy inference rules,...) are of major importance for the performance of fuzzy inference system used as a tool for fault diagnosis of rotating machinery.

Fault diagnosis plays an important role in the operation of modern robotic systems. A number of researchers have proposed fault diagnosis architectures for robotic manipulators using the model-based analytical redundancy approach. One of the key issues in the design of such fault diagnosis schemes is the effect of modeling uncertainties on their performance. This paper investigates the problem of fault diagnosis in rigid-link robotic manipulators with modeling uncertainties. A learning architecture with sigmoidal neural networks is used to monitor the robotic system for off-nominal behavior due to faults. The robustness, sensitivity, missed detection and stability properties of the fault diagnosis scheme are rigorously established. Simulation examples are presented to illustrate the ability of the neural network based robust fault diagnosis scheme to detect and accommodate faults in a two-link robotic manipulator.

Modern industry is concerned about extending the lifetime of its critical processes and maintaining them only when required. Significant aspects of these trends include the ability to diagnose impending failures, prognosticate the remaining useful lifetime of the process and schedule maintenance operations so that uptime is maximized. Prognosis is probably the most difficult of the three issues leading to condition-based maintenance (CBM). This paper attempts to address this challenging problem with intelligence-oriented techniques, specifically dynamic wavelet neural networks (DWNNs). DWNNs incorporate temporal information and storage capacity into their functionality so that they can predict into the future, carrying out fault prognostic tasks. Such fundamental issues as the network structure, learning algorithms, stability analysis, uncertainty management, and performance assessment are studied in a theoretical framework. An example is presented in which a trained DWNN successfully prognoses a defective bearing with a crack in its inner race.

Traditional expert systems for fault diagnosis have a bottleneck in knowledge acquisition, and have limitations in knowledge representation and reasoning. A new expert system shell for fault diagnosis is presented in this paper to develop multiple knowledge models (object model, rules, neural network, case-base and diagnose models) hierarchically based on multiple knowledge. The structure of the expert system shell and the knowledge representation of multiple models are described. Diagnostic algorithms are presented for automatic modeling and hierarchical reasoning. It will be shown that the expert system shell is very effective in building diagnostic expert systems.

This paper presents a methodology for assessment of diagnosability of mechanical and hydraulic systems. The method is developed on the basis of relationships between system performance parameters and physical objects, that is, components of the system. These relationships are identified by system functional domain and are modeled in terms of a bipartite graph, called Diagnosability Bipartite Graph (DBG). A matrix called Diagnosability Matrix (DM) represents the DBG. Various diagnosability parameters of the system are derived from the DBG and the DM and these are useful in evaluation and comparison of design variants of the system. These parameters: are maximum number of set conflicts (MNS), maximum number of components in a set conflict (MNCS), diagnosability effort and cost (DEC), and average merit of diagnosability (AMD). The design having the lowest value of MNCS, AMD, and DEC; and highest value of MNS has the highest diagnosability. On the basis of these, a best design alternative is selected from diagnosability point of view. Moreover, components, which have poor diagnosability, are also identified. Maximum number of set conflicts (MNS) also guides in system fault diagnosis. The proposed procedure aids in the design and development of maintainable systems from diagnosability consideration. The method can also be used for evaluating and comparing the diagnosability of the systems. This method is illustrated with the help of two examples.