Search

Rolling element bearing degradation prediction is an important issue in rotating machinery. With the rapid development of artificial intelligence (AI), AI-based bearing degradation prediction has aroused extensive attention. However, current methods rely on whole life cycle data, which is quite difficult to acquire in real industrial scenarios. To solve this problem, a rotor-bearing dynamic model is built to generate simulation signals for a range of spall sizes, and an improved domain adversarial neural network is proposed to transfer degradation knowledge from simulation data to experimental data. To be specific, complete simulation data is used to pre-train a network for learning comprehensive degradation knowledge, and guides the extracted high-level features in the adversarial domain adaptation stage to align with it as an additional optimization item. The proposed approach is verified on bearing degradation datasets under different working conditions, and results show that the proposed approach can successfully predict bearing degradation progress only with some early stage experimental data.

In the era of Industry 4.0, digital transformation has spurred the swift advancement of technologies, including intelligent predictive maintenance scheduling, prognostics and health management. The accurate prediction of remaining useful life plays a crucial role in these technologies as it extends power equipment’s safe operational duration and decreases the maintenance costs associated with unforeseen shutdowns. Also, the increased accessibility of data for monitoring system conditions has paved the way for the more immense adoption of machine learning models in prognostics and health management for power transformers. At the moment, with the ever-increasing demand for electricity, there is a corresponding increase in the degradation processes of power transformers. Transformers insulation system and more importantly, the paper insulation happens to be the principal part where the degradation is prominent. Therefore, an accurate prediction of the insulating paper condition through its degree of polymerization is required to guarantee the reliability of power transformers. In this regard, the predictions, reliability, and health monitoring of this power equipment can be actualized by modeling the degradation of transformer insulation paper through several machine learning frameworks. In this view, this review paper has been drafted not only to serve as a guide for researchers interested in the fields of transformer insulation system fault prognosis but also to offer insights into potential research directions as existing literature in modeling and evaluating transformer paper insulation is presented.

Data-driven methods have been shown to be suitable for the diagnosis and prognosis of the health of engineering systems. However, the training of data-driven methods usually requires a large amount of data, which is rarely available in industry. In addition, the prediction accuracy often degrades when operating or environmental conditions change or when there are similar systems with different technical characteristics. Transfer learning offers the possibility to transfer knowledge about the degradation behavior between such systems. However, there is a risk that the degradation behavior differs too much, which leads to a so-called negative transfer. Therefore, the authors argue that a similarity assessment of degradation behavior is essential. An assessment based on the operational data of systems seems particularly appropriate. In this paper, the suitability of time series similarity measures for such data-based similarity assessments is investigated. The current state of research is presented. Thereby, no studies on the similarity comparison of degradation curves of engineering systems using time series similarity measures are found. Furthermore, measures for assessing the similarity of degradation curves are identified and categorized. In a case study on filter clogging curves, similarity tests are performed using these measures to find the most similar time series. Various approaches are proposed for evaluation, two of which are used in this paper. The results show that mostly a good selection is made, with some measures performing particularly well. The work presented in this paper represents valuable groundwork for the use of time series similarity measures in transfer learning.

This paper explores nonlinear cumulative damage models applied to Prognostics and Health Management in electronic systems, with a focus on solder joint reliability under variable vibrations. Several cumulative damage models, including the total strain energy density model, are investigated to address the limitation of Miner’s rule. The State-of-Damage (SoD) concept is introduced as a pivotal metric for evaluating Prognostics and Health Management performance within each model. A novel SoD definition with a scale factor is proposed to balance linearity and exponential behavior, meeting the demands of real-time monitoring. Furthermore, by calculating Remaining Useful Life (RUL) using the SoD metric, the RUL predictions from Miner’s rule and total strain energy density model with the scale factor are compared. The results reveal that the total strain energy density model, enhanced with the scale factor, outperforms Miner’s rule in accurately predicting SoD and RUL. This study advances Prognostics and Health Management methodologies by introducing an innovative SoD concept with a scale factor, resulting in enhanced accuracy for both SoD and RUL predictions. This innovation holds the potential for more reliable electronic systems under variable loading conditions.

Little research has been done for artificial intelligence applications of semiconductor backend. This study aims to develop a deep learning based fault diagnosis framework as prognostics and health management (PHM) solutions with a comprehensive analytics process. A linear encoder sensor is employed to measure position, while data preparation is conducted to convert position information into a signal. Then, signal processing is used to extract the features, in which high pass filter is applied to normalize the signal and emphasize the features. High-dimensional features including time domain statistical features, time-frequency domain features obtained by continuous wavelet transform (CWT), and frequency domain features converted by fast Fourier transform (FFT) are extracted to prepare the dataset. An ensemble neural network model that integrates deep neural network (DNN) and convolutional neural network (CNN) is employed to recognize the pattern and diagnose the positioning errors. The accuracy and false alarm rate are considered to support maintenance decisions. An empirical study was conducted in a world leading IC assembly and testing company for validation. The results have shown that the proposed approach can effectively detect inappropriate installation of the bond head in wire bonding equipment for predictive maintenance and cost reduction. [ABSTRACT FROM AUTHOR]

A decomposed dynamic dispatching and preventive maintenance (D-DDPM) framework is presented to overcome the computational challenges of solving large DDPM problems. Stochastic deterioration of machine health is explicitly considered. By efficient allocation of individual machine’s capacity under different machine health, the D-DDPM framework decomposes the global DDPM problems into single-machine DDPM problems while ensuring throughput optimality. Intelligent machine agents are then designed to solve the single-machine DDPM problems efficiently and to perform real-time decisions for individual machines. After the decomposition, the overall computational complexity grows only linearly with the number of machines and the proposed framework could be adopted in large systems. In the numerical study, the D-DDPM framework is near-optimal in small systems that have optimal DDPM solutions. In which, the computation time is reduced by 90.72% and the average cycle time is within 6% of optimal. For large systems without optimal DDPM solutions, numerical results show that the D-DDPM maintains throughput optimality and reduces average cycle time by up to 90% against other control policies. [ABSTRACT FROM AUTHOR]

Approaches for diagnosis and prognosis of the health of engineering systems are divided into data-driven, model-based, and hybrid methods. Data-driven methods depend on the availability of data. Model-based methods require knowledge of the degradation process. A great effort of data generation along with the high complexity of degradation processes often limits both approaches. To mitigate these limitations, the combination of data and knowledge through hybrid methods is examined in this paper. This approach is compared to the alternative approach of reducing the effort of generating training data, as both are gaining importance in diagnostics and prognostics. A new categorization of hybrid prognostic methods for combining data-driven and physics-based models is presented, along with references to existing realizations of these methods. Based on the categorization, a case study on the hybrid remaining useful life prediction of a filtration process is conducted. Several hybrid methods are implemented and tested in this study. Through the combination of models, an improvement in predictive accuracy is achieved. In addition, the paper examines systematic attributes of the individual hybrid methods. Statements on the influence of data scarcity on the predictive accuracy, data-driven models with high variance, and the computational efficiency of the hybrid methods are made. It is shown that these statements are supported by the case study’s results.

Prognosis is a challenging technology that aims to accurately predict and estimate the remaining useful life of a component or system in order to enhance its reliability and performance. Although prognosis research for predictive maintenance is a well-researched topic, practical examples of successful prognostic applications remain scarce. This is due to the lack of available run-to-failure data to build the prediction model as maintenance is usually conducted regularly to avoid significant defects. This paper proposes a novel prognosis method that can be applied to real-world railway maintenance planning without employing run-to-failure data. The key idea is that the fault severity assessment and approximate remaining time prediction are often all that is needed in order to plan maintenance. Firstly, using motor current signals, a degradation indicator on railway door systems is generated based on the dynamic time warping method to measure similarity between typical normal and faulty behaviour. Then, the K-means algorithm is applied to assess fault severity, followed by the representative time estimation for each level of fault severity. This estimation thus allows the remaining time prediction until reaching the critical fault severity level without using run-to-failure data. As a result, the proposed method enables predictive maintenance planning for railway door systems. In addition, the fault severity threshold can be updated by additional operational data, enabling the remaining time prediction to be more reliable. Furthermore, the proposed method can be applied to conventional railway assets and other electro-mechanical actuators as motor current signals are primarily available from the controller or motor drive without additional sensors.

In this article, we describe the iterative participatory design of SOPHIE, an online virtual patient for feedback-based practice of sensitive patient-physician conversations, and discuss an initial qualitative evaluation of the system by professional end users. The design of SOPHIE was motivated from a computational linguistic analysis of the transcripts of 383 patient-physician conversations from an essential office visit of late stage cancer patients with their oncologists. We developed methods for the automatic detection of two behavioral paradigms, lecturing and positive language usage patterns (sentiment trajectory of conversation), that are shown to be significantly associated with patient prognosis understanding. These automated metrics associated with effective communication were incorporated into SOPHIE, and a pilot user study identified that SOPHIE was favorably reviewed by a user group of practicing physicians. [ABSTRACT FROM AUTHOR]

Machine degradation is a complex, dynamic and irreversible process and its modeling is a leading-edge technology in prognostics and health management (PHM). In recent years, machine learning algorithms have been widely used to model machine degradation. However, these degradation models are not physically interpreted so that their extended uses are reduced and weakened. Aiming at solving this problem and visualizing informative features learned from degradation data, in this paper, a generalized machine degradation modeling methodology is proposed by integrating multiple-source fusion with genetic programming (GP). A composite fitness function of GP tailored for machine degradation modeling is innovatively designed. Afterward, multiple process sensor data, such as temperature, pressure, currents, etc., and non-process sensor data, such as vibration and acoustic signals, can be respectively modeled and fused into structurally interpreted health indicators from the time domain and the frequency domain. Moreover, the proposed methodology can automatically select informative frequency components and sensors, and provide transparent modeling architecture for early fault detection and subsequent monotonic degradation assessment. Another benefit of the proposed methodology is that complicated data preprocessing and manual feature extraction are not required anymore. Hence, the proposed methodology would have many potential applications and it is easy to implement for online machine degradation modeling. A gearbox run-to-failure dataset (non-process data) and an aircraft engine degradation dataset (process data) are studied to verify the effectiveness of the proposed methodology. Comparisons show that structurally interpreted health indicators constructed from the proposed methodology are superior to state-of-the-art works.

Prognostics and health management (PHM) of systems usually depends on appropriate prior knowledge and sufficient condition monitoring (CM) data on critical components’ degradation process to appropriately estimate the remaining useful life (RUL). A failure of complex or critical systems such as heating, ventilation, and air conditioning (HVAC) systems installed in a passenger train carriage may adversely affect people or the environment. Critical systems must meet restrictive regulations and standards, and this usually results in an early replacement of components. Therefore, the CM datasets lack data on advanced stages of degradation, and this has a significant impact on developing robust diagnostics and prognostics processes; therefore, it is difficult to find PHM implemented in HVAC systems. This paper proposes a methodology for implementing a hybrid model-based approach (HyMA) to overcome the limited representativeness of the training dataset for developing a prognostic model. The proposed methodology is evaluated building an HyMA which fuses information from a physics-based model with a deep learning algorithm to implement a prognostics process for a complex and critical system. The physics-based model of the HVAC system is used to generate run-to-failure data. This model is built and validated using information and data on the real asset; the failures are modelled according to expert knowledge and an experimental test to evaluate the behaviour of the HVAC system while working, with the air filter at different levels of degradation. In addition to using the sensors located in the real system, we model virtual sensors to observe parameters related to system components’ health. The run-to-failure datasets generated are normalized and directly used as inputs to a deep convolutional neural network (CNN) for RUL estimation. The effectiveness of the proposed methodology and approach is evaluated on datasets containing the air filter’s run-to-failure data. The experimental results show remarkable accuracy in the RUL estimation, thereby suggesting the proposed HyMA and methodology offer a promising approach for PHM.

Electromagnetic Relays (Electromagnetic Relay (EMR)s) are omnipresent in electrical systems, ranging from mass-produced consumer products to highly specialised, safety-critical industrial systems. Our detailed literature review focused on EMR reliability highlighting the methods used to estimate the State of Health or the Remaining Useful Life emphasises the limited analysis and understanding of expressive EMR degradation indicators, as well as accessibility and use of EMR life cycle data sets. Prioritising these open challenges, a deep learning pipeline is presented in a prognostic context termed Electromagnetic Relay Useful Actuation Pipeline (EMRUA). Leveraging the attributes of causal convolution, a Temporal Convolutional Network (TCN) based architecture integrates an arbitrary long sequence of multiple features to produce a remaining useful switching actuations forecast. These features are extracted from raw, high volume life cycle data sets, namely EMR switching data (Contact-Voltage, Contact-Current). Monte-Carlo Dropout is utilised to estimate uncertainty during inference. The TCN hyperparameter space, as well as various methods to select and analyse long sequences of multivariate time series data are investigated. Subsequently, our results demonstrate improvements using the developed statistical feature-set over traditional, time-based features, commonly found in literature. EMRUA achieves an average forecasting mean absolute percentage error of ±12 % over the course of the entire EMR life.

This brief investigates the problem of fault prognosis in decentralized structure, where given systems are monitored by $r$ local agents, and assumes that only $z$ ($1\leq z\leq r$) local agents’ information can be obtained. In this case, under a state-estimate-based protocol, a notion called $z$ -reliable SE-coprognosability is proposed. Compared with existing results, $z$ -reliable SE-coprognosability is a more universal concept. Besides, with the help of semi-tensor product, an efficient algorithm for state estimation is presented, and based on which, the algorithm on verification of $z$ -reliable SE-coprognosablity is also proposed. A dominating breakthrough is that the complexity of given algorithms is polynomial with respect to the number of agents. [ABSTRACT FROM AUTHOR]

In flight control actuation systems, electromechanical actuators (EMAs) as energy-efficient components play critical roles in next-generation aircraft. Reliable degradation estimation models based on sensing data from sensors installed in EMA can help to improve its availability and safety, which need to be studied. However, the dynamic operating conditions, which will lead to the mismatch issues of the sensing data-based degradation estimation models, reduce the overall accuracy of the degradation estimation. To address this issue, an improved adaptive Wiener process model for EMA degradation estimation based on sensing data with dynamic operating conditions is proposed. First, a health indicator (HI) is extracted from sensing data through physical analysis or extracted using data processing techniques. Then, a novel model for EMA degradation estimation with the improved strong tracking filter and the Wiener process model (i.e., ISTF-WPM) is formulated based on the obtained HI. By introducing a saturation function to the improved strong tracking filter (STF), the performance of sensing data-based degradation estimation with dynamic operating conditions can be improved. The proposed model presents a novel routine for EMA degradation estimation based on sensing data with dynamic operating conditions. In the experiments of this study, the effectiveness of the ISTF-WPM is validated using simulation data as well as practical data collected by sensors of EMA. Experimental results indicate that EMA sensing data-based degradation estimation with the proposed model has a high overall accuracy. [ABSTRACT FROM AUTHOR]

Accurate remaining useful life (RUL) of batteries plays an imperative role in ensuring safe operations and avoiding catastrophic accidents. However, in practice, complicated working conditions bring challenges to accurate battery prognostics. In this article, an accelerated stress factors-based nonlinear Wiener process model is proposed to enrich inadequate battery prognostic works at various operating conditions. To realize online individual battery prognostics, once a new measurement is available, the parameters of a state-space model constructed by the proposed model are posteriorly updated. Then, based on the Peukert law and the Arrhenius equation, two specific accelerated stress-relevant drift functions and their associated degradation models at different discharge rates and temperatures are respectively designed. Subsequently, RUL predictions are conducted using the proposed method. RUL predictions at different discharge rates and different temperatures demonstrate the accuracy and robustness of the proposed prognostic models. According to some general prognostic metrics, the proposed method is proved to be superior to four existing RUL prediction approaches. [ABSTRACT FROM AUTHOR]

To avoid unexpected failures of engineering systems, sensors have been widely used to monitor the degradation process of the systems. A number of studies have been conducted to analyze the collected sensor signals and predict the failure time. However, the existing studies are usually restricted and cannot be adapted to different practical situations. In this paper, we propose a systematic method for degradation modeling and prognosis that can be widely applied in different scenarios. In particular, the proposed method is capable to handle one or multiple sensors, powerful to capture the nonlinear relations between sensor signals and the degradation process with few assumptions, generic to consider multiple failure modes, flexible to deal with unequally spaced sensor measurements or asynchronous signals, and easily understandable with little preprocessing required. The main idea is to predict the failure time of an in-service unit based on a subset of the nearest historical units, where features are extracted from each sensor to describe the progression of sensor signals and local linear regression models are constructed to establish the relation between failure time and the extracted features. The prediction variance is then used as the goodness-of-fit measure, based on which decision-level fusion and feature-level fusion are proposed to combine multiple sensors. A case study with two datasets on the degradation modeling of aircraft engines is conducted which shows satisfactory performance of the proposed method. Note to Practitioners—This paper aims at modeling the collected sensor signals to understand the degradation process of the monitored engineering systems and predict the failure time. The main idea is to measure the similarity of units and predict the failure time of an in-service unit based on a subset of the nearest historical units. The developed method is widely applicable in different practical situations such as multiple sensors, multiple failure modes, asynchronous signals, and missing data. Furthermore, the method requires little preprocessing. There are several steps involved for implementing the proposed method: 1) collecting the sensor signals for historical units and the in-service unit; 2) extracting features from each sensor signal; 3) constructing a local linear model to predict the failure time based on the extracted features, and obtaining the prediction variance on the in-service unit; and 4) combining the information of different sensors using the decision-level fusion or feature-level fusion, if each unit is monitored by multiple sensors. [ABSTRACT FROM AUTHOR]

Recently, the study of fault diagnosis and prognosis of discrete-event systems (DESs) has received considerable attention. This article aims to investigate the decentralized fault prognosis of stochastic DESs (SDESs). The notion of $m$ -step stochastic-coprognosability, called $S_{m} $ -coprognosability, is formalized to capture the capability of stochastic systems to predict a fault at least $m$ steps in advance under the decentralized framework. The verifier is constructed from the stochastic decentralized system, in which $n$ local prognosers are deployed to send the local prognostic decisions to a coordinator for calculating the final prognostic decision. In particular, a necessary and sufficient condition of $S_{m}$ -coprognosability of stochastic systems is derived, and an algorithm for testing $S_{m} $ -coprognosability is given, which is polynomial to the number of states and events but exponential to the number of local sites. Furthermore, the maximum number of $m$ for making the given SDES to be $S_{m} $ -coprognosable is discussed. Note to Practitioners—This article extends the fault prognosis methods of SDESs from centralized framework to decentralized framework, such that for many technologically complex systems whose information collection is scattered in physically separated different sites, every local site makes decision based on information collected by own sensors, without communicating with other local sites, which is more appropriate than the centralized method. [ABSTRACT FROM AUTHOR]

Physics-informed machine learning (PIML) has been emerging as a promising tool for applications with domain knowledge and physical models. To uncover its potentials in power electronics, this article proposes a PIML-based parameter estimation method demonstrated by a case study of dc–dc Buck converter. A deep neural network and the dynamic models of the converter are seamlessly coupled. It overcomes the challenges related to training data, accuracy, and robustness which a typical data-driven approach has. This exemplary application envisions to provide a new perspective for tailoring existing machine learning tools for power electronics. [ABSTRACT FROM AUTHOR]

In the context of Industry 4.0, intelligent manufacturing services put forward the requirement of rapid response in the remaining useful life (RUL) of machinery. To achieve fast-responding and highly accurate RUL prediction services while mining the intra-kernel correlations of the sensor monitoring data, a deep learning-based cloud–edge collaboration framework was proposed in this article. We encapsulated a cloud prediction engine (Cloud-PE) with a deep prediction model in the cloud service layer and an edge prediction engine (Edge-PE) with a shallow prediction model in the edge service layer. The Cloud-PE assisted the Edge-PE in achieving fast and highly accurate RUL prediction by sharing depth model parameters. Both prediction models were constructed on the basis of a novel blueprint separable convolution neural network. To continuously improve the performance of Edge-PE in a context-aware manner, we adopted an update method for the Edge-PE with the assistance of the Cloud-PE. The experimental results demonstrated that the proposed framework can provide more accurate RUL prediction than existing data-driven prediction methods, and the training time of the prediction model is also significantly reduced. [ABSTRACT FROM AUTHOR]

Fault prognosis based on single model is generally inaccurate due to the varying working conditions. A multistage fault prognosis methodology combining stage identification with Bayesian networks (BNs) and time series approach with particular emphasis on the autoregressive moving average (ARMA) model is proposed to solve this problem. In the first stage, degradation data are identified, and outliers are marked by the Euclidean distance. Degenerate attributes of outliers are finely identified by BNs and matched to the corresponding model. In the second stage, the ARMA model is used for prognosis according to the results of the fine identification. Subsequently, the double-precision identification and ARMA submodel prognosis are carried out alternately throughout the prognosis process. Three degradation types of permanent magnet synchronous motor are simulated to verify the applicability of the method. Result shows that it can track the changes in the degradation in time and obtains better results. [ABSTRACT FROM AUTHOR]

Classical machine learning approachesare sensitive to nonstationarity. Transfer learning can address nonstationarity by sharing knowledge from one system to another, however, in areas like machine prognostics and defense, data are fundamentally limited. Therefore, transfer learning algorithms have little, if any, examples from which to learn. Herein, the authors suggest that these constraints on algorithmic learning can be addressed by systems engineering. We formally define transfer distance in general terms and demonstrate its use in empirically quantifying the transferability of models. We consider the use of transfer distance in the design of machine rebuild procedures to allow for transferable prognostic models. We also consider the use of transfer distance in predicting operational performance in computer vision. Practitioners can use the presented methodology to design and operate systems with consideration for the learning theoretic challenges faced by component learning systems. [ABSTRACT FROM AUTHOR]

An in-depth exploration of gene prognosis using different methodologies aids in understanding various biological regulations of genes in disease pathobiology and molecular functions. Interpreting gene functions at biological and molecular levels remains a daunting yet crucial task in domains such as drug design, personalized medicine, and next-generation diagnostics. Recent advancements in omics technologies have produced diverse heterogeneous genomic datasets like micro-array gene expression, miRNA expression, DNA sequence, 3D structures, which are significant resources for understanding the gene functions. In this paper, we propose a novel self-attention based deep multi-modal model, named DeePROG, for the prognosis of disease affected genes based on heterogeneous omics data. We use three NCBI datasets covering three modalities, namely gene expression profile, the underlying DNA sequence, and the 3D protein structures. To extract useful features from each modality, we develop several context-specific deep learning models. Besides, we develop three attention-based deep bi-modal architectures along with DeePROG to leverage the prognosis of the underlying biomedical data. We assess the performance of the models’ in terms of computational assessment of function annotation (CAFA2) metrics. Moreover, we analyze the results in terms of receiver operating characteristics (ROC) curve in high-class imbalance data setting and perform statistical significance tests in terms of Welch's t-test. Experiment results show that DeePROG significantly outperforms baseline models across in terms of performance metrics. The source code and all preprocessed datasets used in this study are available at https://github.com/duttaprat/DeePROG. [ABSTRACT FROM AUTHOR]

Fault diagnosis and prognosis (FDP) is critical for ensuring system reliability and reducing operation and maintenance (O&M) costs. Lebesgue sampling based FDP (LS-FDP) is an event-based approach with the advantages of cost-efficiency, uncertainty management, and less computation. In previous works, LS-FDP approaches are mainly model-based. However, fault dynamic modeling is difficult and time consuming for some complex systems and this severely hinders the applications of LS-FDP. To address this problem, this article presents a data-driven based LS-FDP framework in which deep belief networks (DBN) and particle filter (PF) are integrated to achieve fault state estimation and remaining useful life prediction. In the proposed approach, DBN learns the state evolution model and the Lebesgue time transition model, which are used as diagnostic and prognostic models in PF for FDP. A series of offline and online experiments are conducted on lithium-ion batteries to verify the proposed method. Experimental results and comparison studies show that the proposed approach has higher efficiency in terms of computation and better performance in terms of FDP accuracy and precision. [ABSTRACT FROM AUTHOR]

The fault diagnosis of complex timed discrete event systems remains a relevant and challenging problem in industry. As systems degrade, the synchronization required for normal operation associated with timed event sequences deteriorate. However, existing discrete event approaches cannot handle complexity well as there may be potentially hundreds of event sequences that still result in normal system behavior, and they do not scale well due to the state explosion problem. To circumvent these issues, a new data-driven augmented intelligence framework called EveSyncIAI is proposed that uses a Timed Petri Net of the normal process to enable selective extraction of discriminating time delay features, which are then used alongside machine learning for fault diagnosis. The developed methodology is applied to a case study of an operational discrete manufacturing system in the semiconductor industry. Multiclass machine learning approaches as well as binary anomaly detection algorithms with varying levels of supervision are directly compared with and without EveSyncIAI-extracted features. With precision, recall, and F1-score values above 0.90 being achieved for multi-fault classification and AUC scores exceeding 0.98 for semi-supervised and unsupervised anomaly detection approaches, the framework presented is promising for the fault diagnosis of timed sequence-sensitive systems. [ABSTRACT FROM AUTHOR]

In many real-world prognostics and health management tasks, where the available training samples are insufficient, deep neural networks are highly vulnerable to overfitting. To address this problem, in this article, we propose a novel health representation learning method based on a Siamese network. This method prevents overfitting by utilizing a constraint by which the differences between samples in the embedding space of the Siamese network should follow the differences in the remaining useful life (RUL) values via the introduction of a multitask learning scheme. In addition, since the learned embedding space reflects the dynamics of degradation, each training sample can be used as a reference to estimate the RUL of a test sample. By combining the estimates for all training samples, the proposed method enables robust RUL prediction. Experimental results show that the proposed learning and estimation method contributes to improving not only RUL prediction performance but also robustness to data insufficiency. [ABSTRACT FROM AUTHOR]

This article focuses on individualized degradation modeling and prognostics for a heterogeneous group, where each individual unit shows a distinct degradation process. Existing degradation models usually treat each unit separately and do not fully utilize the distinct characteristics of each individual. In this study, we propose a generic framework to handle the heterogeneity across units by effectively leveraging the intrinsic covariate information, which is closely related to the unit’s degradation process. Specifically, we employ a multivariate Gaussian process (MGP) to nonparametrically establish the relation between the covariate information and degradation process. Through modeling the unit similarities based on the covariates, efficient information transfer among units is enabled for better degradation modeling and prognostics, as the collected degradation signals from one unit can be shared with the entire heterogeneous group. A theoretical justification for the proposed model is also investigated. Simulation studies are presented to evaluate the parameter estimation accuracy and the sensitivity of the proposed method. A case study on the Alzheimer’s disease (AD) neuroimaging initiative data set is further conducted, which demonstrates the advantage of the proposed method over existing benchmark approaches. Note to Practitioners—This article is motivated by the practical issue of degradation modeling and prognostics for a heterogeneous group, where all units in a group share some similarities and each unit has its own distinct individual-level characteristics (covariates). The covariates in this study refer to static intrinsic characteristics instead of dynamic external environmental conditions. Several practical examples are explained in with more details. There are two fundamental questions involved: 1) how to quantify the distinct individual characteristics of each unit while representing the group-level commonalities among all units and 2) how to perform degradation modeling and prognostics of a newly launched unit with few degradation signals available. The novelty of this article lies in encoding the available knowledge about individual covariates and group-level commonalities into the degradation modeling and prognostics. There are three main steps involved when implementing the proposed method: 1) collecting degradation signals, failure time, and covariate information of heterogeneous units; 2) constructing an MGP-based degradation model; and 3) predicting the degradation status and remaining useful life of the in-service units based on their covariates and signals. The proposed method is particularly useful when we collect various intrinsic covariates which can effectively represent the individual-level characteristics and when only sparse or no data are available for the units of interest. [ABSTRACT FROM AUTHOR]

Fault prognosis of discrete-event systems (DESs) aims to predict the occurrence of fault beforehand such that certain protective measures may be adopted before the fault occurs. This article investigates the reliable coprognosability issue for decentralized stochastic DESs (SDESs) facing the possible unavailability of some local agents. The main contributions are as follows. First, we formalize the notion of $r$ -reliable coprognosability for SDESs. In general, an $r$ -reliably coprognosable SDES with $n$ local sites $(1 \leq r \leq n)$ can predict the occurrences of faults even though $n-r$ local agents are invalid. Second, we construct a reliable coprognoser from the given stochastic system and present a necessary and sufficient condition for testing $r$ -reliable coprognosability by the reliable coprognoser. Third, due to the exponential complexity of testing $r$ -reliable coprognosability by reliable coprognoser, a reliable coverifier is constructed and an alternate necessary and sufficient condition for verifying $r$ -reliable coprognosability of SDESs by the reliable coverifier is proposed, which is polynomial time. [ABSTRACT FROM AUTHOR]

Patients with Disorder of Consciousness (DoC) entering Intensive Rehabilitation Units after a severe Acquired Brain Injury have a highly variable evolution of the state of consciousness which is a complex aspect to predict. Besides clinical factors, electroencephalography has clearly shown its potential into the identification of prognostic biomarkers of consciousness recovery. In this retrospective study, with a dataset of 271 patients with DoC, we proposed three different Elastic-Net regressors trained on different datasets to predict the Coma Recovery Scale-Revised value at discharge based on data collected at admission. One dataset was completely EEG-based, one solely clinical data-based and the last was composed by the union of the two. Each model was optimized, validated and tested with a robust nested cross-validation pipeline. The best models resulted in a median absolute test error of 4.54 [IQR = 4.56], 3.39 [IQR = 4.36], 3.16 [IQR = 4.13] for respectively the EEG, clinical and hybrid model. Furthermore, the hybrid model for what concerns overcoming an unresponsive wakefulness state and exiting a DoC results in an AUC of 0.91 and 0.88 respectively. Small but useful improvements are added by the EEG dataset to the clinical model for what concerns overcoming an unresponsive wakefulness state. Data-driven techniques and namely, machine learning models are hereby shown to be capable of supporting the complex decision-making process the practitioners must face. [ABSTRACT FROM AUTHOR]

Synergetic recovery of both somatosensory and motor functions is highly desired by limb amputees to fully regain their lost limb abilities. The commercially available prostheses can restore the lost motor function in amputees but lack intuitive sensory feedback. The previous studies showed that electrical stimulation on the arm stump would be a promising approach to induce sensory information into the nervous system, enabling the possibility of realizing sensory feedback in limb prostheses. However, there are currently limited studies on the effective evaluation of the sensations evoked by transcutaneous electrical nerve stimulation (TENS). In this paper, a multichannel TENS platform was developed and the different stimulus patterns were designed to evoke stable finger sensations for a transradial amputee. Electroencephalogram (EEG) was recorded simultaneously during TENS on the arm stump, which was utilized to evaluate the evoked sensations. The experimental results revealed that different types of sensations on three phantom fingers could be stably evoked for the amputee by properly selecting TENS patterns. The analysis of the event-related potential (ERP) of EEG recordings further confirmed the evoked sensations, and ERP latencies and curve characteristics for different phantom fingers showed significant differences. This work may provide insight for an in-depth understanding of how somatosensation could be restored in limb amputees and offer technical support for the applications of non-invasive sensory feedback systems. [ABSTRACT FROM AUTHOR]

Objective: Diagnosis and prognosis of patients with disorders of consciousness (DOC) is a challenge for neuroscience and clinical practice. Transcranial magnetic stimulation combined with electroencephalography (TMS-EEG) is an effective tool to measure the level of consciousness. However, a scientific and accurate method to quantify TMS-evoked activity is still lacking. This study applied fast perturbational complexity index (PCIst) to the diagnosis and prognosis of DOC patients. Methods: TMS-EEG data of 30 normal healthy participants (NOR) and 181 DOC patients were collected. The PCIst was used to assess the time-space complexity of TMS-evoked potentials (TEP). We selected parameters of PCIst in terms of data length, data delay, sampling rate and frequency band. In addition, we collected Coma Recovery Scale–Revised (CRS-R) values for 114 DOC patients after one year. Finally, we trained the classification and regression model. Results: 1) PCIst shows the differences among NOR, minimally consciousness state (MCS) and unresponsive wakefulness syndrome (UWS) and has low computational cost. 2) Optimal parameters of data length and delay after TMS are 300 ms and 101–300 ms. Significant differences of PCIst at 5–8 Hz and 9–12 Hz bands are found among NOR, MCS and UWS groups. PCIst still works when TEP is down-sampled to 250 Hz. 3) PCIst at 9–12 Hz shows the highest performance in diagnosis and prognosis of DOC. Conclusions: This study confirms that PCIst can quantify the level of consciousness. PCIst is a potential measure for the diagnosis and prognosis of DOC patients. [ABSTRACT FROM AUTHOR]

In today's digital world, we are equipped with modern computer-based data collection sources and feature extraction methods. It enhances the availability of the multi-view data and corresponding researches. Multi-view prediction models form a mainstream research direction in the healthcare and bioinformatics domain. While these models are designed with the assumption that there is no missing data for any views, in the real world, certain views of the data are often not having the same number of samples, resulting in the incomplete multi-view dataset. The studies performed over these datasets are termed incomplete multi-view clustering or prediction. Here, we develop a two-stage generative incomplete multi-view prediction model named GIMPP to address the missing view problem of breast cancer prognosis prediction by explicitly generating the missing data. The first stage incorporates the multi-view encoder networks and the bi-modal attention scheme to learn common latent space representations by leveraging complementary knowledge between different views. The second stage generates missing view data using view-specific generative adversarial networks conditioned on the shared representations and encoded features given by other views. Experimental results on TCGA-BRCA and METABRIC datasets proves the usefulness of the developed method over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Recently, digital pathology image-based prognosis prediction has become a hot topic in healthcare research to make early decisions on therapy and improve the treatment quality of patients. Therefore, there has been a recent surge of interest in designing deep learning method solving the problem of prognosis prediction with digital pathology images. However, whole slide histopathological images (WSIs) based prognosis prediction is still a challenge due to the large size of pathological images, the heterogeneity of tumors and the high cost of region of interests (ROIs) labeling. In this study, we design a novel two-stage deep learning framework for prognosis prediction (TSDLPP) based on WSIs. Our proposed framework consists of two-stage paradigms: 1) training tissue decomposition network (TDNet) to divide WSIs into cancerous and non-cancerous regions, 2) integrating general prognosis-related densely connected CNN (GPR-DCCNN) and morphology-specific prognosis-related densely connected CNNs (MSPR-DCCNNs) to extract different level features of pathological images. In the end, we apply TSDLPP to the prognosis prediction of breast cancer using The Cancer Genome Atlas (TCGA) datasets. Experiment results demonstrate that TSDLPP obtains superior performance of prognosis prediction compared with the existing state-of-arts methods. [ABSTRACT FROM AUTHOR]

Identifying biomarkers of heterogeneous complex diseases has always been one of the focuses in medical research. In previous studies, the powerful network propagation methods have been applied to finding marker genes related to specific diseases, but existing methods are mostly based on a single network, which may be greatly affected by the incompleteness of the network and the ignorance of a large amount of information about physical and functional interactions between biological components. Other methods that directly integrate multiple types of interactions into an aggregate network have the risks that different types of data may conflict with each other and the characteristics and topologies of each individual network are lost. Meanwhile, biomarkers used in clinical trials should have the characteristics of small quantity and strong discriminate ability. In this study, we developed a multiplex network-based dual ranking framework (DualRank) for heterogeneous complex disease analysis. We applied the proposed method to heterogeneous complex diseases for diagnosis, prognosis, and classification. The results showed that DualRank outperformed competing methods and could identify biomarkers with the small quantity, great prediction performance (average AUC = 0.818) and biological interpretability. [ABSTRACT FROM AUTHOR]

Deep learning techniques have recently brought many improvements in the field of neural network training, especially for prognosis and health management. The success of such an intelligent health assessment model depends not only on the availability of labeled historical data but also on the careful samples selection. However, in real operating systems such as induction machines, which generally have a long reliable life, storing the entire operation history, including deterioration (i.e., bearings), will be very expensive and difficult to feed accurately into the training model. Other alternatives sequentially store samples that hold degradation patterns similar to real ones in damage behavior by imposing an accelerated deterioration. Labels lack and differences in distributions caused by the imposed deterioration will ultimately discriminate the training model and limit its knowledge capacity. In an attempt to overcome these drawbacks, a novel sequence-by-sequence deep learning algorithm able to expand the generalization capacity by transferring obtained knowledge from life cycles of similar systems is proposed. The new algorithm aims to determine health status by involving long short-term memory neural network as a primary component of adaptive learning to extract both health stage and health index inferences. Experimental validation performed using the PRONOSTIA induction machine bearing degradation datasets clearly proves the capacity and higher performance of the proposed deep learning knowledge transfer-based prognosis approach. [ABSTRACT FROM AUTHOR]

Wind turbines play an increasingly important role in renewable power generation. To ensure the efficient production and financial viability of wind power, it is crucial to maintain wind turbines’ reliability and availability (uptime) through advanced real-time condition monitoring technologies. Given their plurality and evolution, this article provides an updated comprehensive review of the state-of-the-art condition monitoring technologies used for fault diagnosis and lifetime prognosis in wind turbines. Specifically, this article presents the major fault and failure modes observed in wind turbines along with their root causes, and thoroughly reviews the techniques and strategies available for wind turbine condition monitoring from signal-based to model-based perspectives. In total, more than 390 references, mostly selected from recent journal articles, theses, and reports in the open literature, are compiled to assess as exhaustively as possible the past, current, and future research and development trends in this substantial and active investigation area. [ABSTRACT FROM AUTHOR]

Digital electronics has become a standard for controlling electrical systems. This is due to the constant improvement of the digital devices, whether in terms of density, performance, flexibility of use, or cost reduction. This article looks into system-on-chip (SoC) field-programmable gate array (FPGA) for controlling complex electrical energy systems. These devices encompass multicore floating-point microprocessors embedded with standard peripherals and an FPGA fabric that allows the design of custom peripherals and specific hardware (HW) accelerators. Thus, SoC FPGA devices can be regarded as a good compromise between “super” microcontrollers (very fast in terms of computation but with a fixed microarchitecture) and pure FPGAs (ideal for specific concurrent microarchitectures but limited in terms of density). [ABSTRACT FROM AUTHOR]

The performance of fault diagnosis and prognosis (FDP) relies greatly on the model that describes the fault behavior and dynamics under study. In the past decades, many researches focused on establishing reliable, robust, and accurate models to describe fault growth dynamics. However, uncertainties in the modeling process severely affect the accuracy and precision of FDP results. A major challenge in FDP is how to deal with the uncertainties that affect the performance of the system. For commonly used state-space model, uncertainties exist in both state transition equation and the measurement equation. In this article, the influence of uncertainties on FDP results is studied and compared in terms of accuracy and precision. Furthermore, this article proposes a differential model decomposition approach, which can automatically acquire appropriate model parameters and uncertainties for the state-space model. The effectiveness and advantages of the proposed method are verified and demonstrated with three different application scenarios, including comparison studies on public data and online experimental verification. [ABSTRACT FROM AUTHOR]

Condition Monitoring (CM) is an essential element of securing reliable operating conditions of Wind Turbines (WT) in a wind farm. CM helps optimize maintenance by providing Remaining Useful Life (RUL) forecast. However, the expected RUL is not often reliable due to uncertainty associated with the prediction horizon. In this paper, we employ high-level fusion methods to expect the RUL of WT bearings. For this purpose, various features are extracted by vibration signals to capture deterioration paths. Then, a Bayesian algorithm is utilized to determine RUL for each selected feature. Eventually, high-level fusion schemes, including Hurwicz, Choquet integral, Ordered Weighted Averaging operator, are employed to integrate RUL numbers and lessen associated uncertainty in the prediction horizons. Besides, a pessimistic fusion strategy is driven to obtain a bounded uncertainty for the worst RUL prediction. The fusion methods are assessed by ten-year vibration signals of Canadian wind farms. Experimental results confirm accurate results with bounded uncertainty for high-level fusion approaches. [ABSTRACT FROM AUTHOR]

Fast kurtogram (FK) is an effective tool in fault diagnosis, but it still has two defects. Firstly, its indicator is easy to be affected by the random impact. Secondly, its fixed frequency band segmentation rules might lead to over-decomposition or under-decomposition problems. Therefore, by combining a more robust indicator, envelope harmonic-to-noise ratio (EHNR), with an adaptive frequency band segmentation method based on scale-space representation (SSR), a completely parameterless adaptive spectrum analysis technology, EHNR-SSR, is constructed. The EHNR is more robust to random impact and independent of prior parameters, and the SSR-based frequency band segmentation has adaptive adjustment ability. Additionally, the EHNR can characterize the signal with specific periodicity, which makes the proposed method has the capability of compound fault detection. The superiority of EHNR-SSR is verified through simulated signals and experimental tests. The results reveal that EHNR-SSR can identify bearing fault features from vibration mixture and realize compound fault detection. [ABSTRACT FROM AUTHOR]

Feature extraction is essential for accurate knowledge-based prognostics and health management (PHM) of the wind turbine system. As a classic solution of knowledge-based PHM, representation learning faces the challenges of long sequence length, missing values, and insufficient labeling when dealing with multivariate time series wind power data. This paper proposes a weighted representation learning-based feature extraction model to address the above problems simultaneously. Firstly, we design a novel self-attentive convolutional autoencoder (SACAE) model which can automatically extract high-level compressed feature matrix representations from long-length wind power time series, thereby retaining more necessary information. Then a missing-weighted dependent wild bootstrapped maximum mean discrepancy (MDMMD) loss function is proposed to capture the temporal dependence, numerical similarity, and distribution similarity of samples and avoid the influence of these missing values. Finally, we present the wind turbine PHM framework to improve the utilization of unlabeled data by using them to train the MDMMD-SACAE model and then the limited labeled data can get more accurate feature representation through the trained MDMMD-SACAE model. Experimental studies on three PHM cases verify the effectiveness and robustness of the proposed approach in the wind turbine system. [ABSTRACT FROM AUTHOR]

One of the main concerns associated with diagnosis, prognosis, and health management (DPHM) of engineering systems is the accuracy of estimates that are derived from Bayesian tracking methods. Estimating the exiting degradation based on stochastic models and evaluating the remaining useful life (RUL) of the system is inherently associated with variances that characterize the inaccuracy of estimation techniques. Furthermore, there are scenarios where a single measurement does not necessarily generate sufficient information regarding the system states, leading one to require multiple readings (and, hence, multidimensional analysis) to deduce diagnostic and/or prognostic decisions. This article introduces a novel approach for solving complex nonlinear multivariable Bayesian models that are utilized for estimation and prediction problems that would be, otherwise, challenging or impractical to solve through available methods, such as particle filters (PFs). Theoretical derivation and strategies that are developed in this article are verified through numerical case study simulations for electrohydraulic servo valves (EHSVs) that constitute a core component of many hydraulic actuators, such as multifunctional spoilers (MFSs), which are widely utilized in aircraft flight control systems. Our developed results are compared with those that are derived through PF in order to illustrate and demonstrate the advantages, benefits, and improvements that are accomplished by applying our proposed methodologies. [ABSTRACT FROM AUTHOR]

The existing deep learning-based fault prognostic methods require massive labeled condition monitoring (CM) data to train a well-generalized model. However, acquiring massive labeled CM data for real-case machines is infeasible due to time, economic costs, and safety concerns. Fortunately, we can handily obtain labeled CM data from relevant but different machines such as from accelerated degradation experiments in laboratories, which contain partially shared prognosis knowledge correlated to real-case machines. Accordingly, to bridge this practical gap, a novel Bayesian deep dual network with domain adaptation is developed to achieve transfer fault prognosis across different machines with distinct structures, measurement settings, and operating conditions. A deep convolutional neural network (DCNN)-multiple layer perceptron (MLP) dual network is first employed to extract abundant degradation representations from time series-based and time-frequency spectrum-based raw features. Then, domain adaptation regularization is imposed to relieve significant distribution discrepancy issue existing across different machines. Finally, the proposed DCNN-MLP dual network integrated with domain adaptation module is extended into Bayesian dual network through variational-inference (VI)-based method. The experimental verification demonstrates that the proposed method can accurately predict the remaining useful life percentage of testing bearings without any labeled CM data in target domain, and comparisons with other existing methods are also included. [ABSTRACT FROM AUTHOR]

This article investigates the robust fault prognosis of discrete-event systems (DESs) against the loss of observations of events. To solve this problem, a dilated automaton is generated from the plant by adding an unobservable transition attached to every event subject to loss of observations. We prove the equivalence between the robust prognosability of the plant model and the prognosability of the dilated model. In order to deal with unobservable cycles in the dilated model, a new necessary and sufficient condition of robust prognosability based on diagnoser is presented. Moreover, a polynomial algorithm for the verification of robust prognosability is proposed by constructing the verifier to search the existence of some cycle generated by a faulty trace and an unprognosable trace that run synchronously and keep the identical observation before the fault occurs. Our results generalize previous works on fault prognosis of DESs by taking the loss of observations into account. Note to Practitioners—The research in this article is motivated by the problem of fault prognosis in the manufacturing system. If some information used for prediction is lost, can engineers still make correct decisions about the occurrences of fault? This article aims to investigate the approach of assuring the accuracy and reliability of prediction by introducing the notion of robust prognosis. A polynomial-time algorithm of verifying the robust prognosability is proposed by constructing the tester automaton. [ABSTRACT FROM AUTHOR]

The automated conveyor system, as the core component in the modern manufacturing world, has gained lots of attention from researchers. To optimize the operation of the conveyor system, range-inspection control (RIC) has been considered an efficient strategy to bring this conventional technology to an intelligent level. Various algorithms have been put into use to achieve optimal control. However, the current methodologies are only focusing on control optimization, not scaled into the smart manufacturing framework. The schema of alignment and corporation between the physical and virtual spaces for the system remains an important problem. Therefore, the work in this article aims for an effective framework of implementation between the physical and virtual stations in an automated conveyor system. Since increasingly more application scenarios rely on the digital twin (DT) technology to realize the integration of physical and virtual systems, we proposed the DT automated conveyor system (DT-ACS) that constructs the road map to implement the RIC-based conveyor system under the background of a smart factory. Besides, profit-sharing-based deep Q-networks (PDQNs) have been proposed to cope with the RIC optimization problem. The robustness and efficiency of the proposed PDQN were evaluated via sets of experiments. The discussion and conclusion are presented at last accordingly. Note to Practitioners—This article aims to propose a strategy of control optimization for conveyor-based manufacturing systems under the digital twin (DT) framework. The conveyor system can be flexible to control the running flows to avoid overloading workstations. Due to the complex environment in the production line, the range that is able to be inspected and the capacity of the reserve area can be considerably diverse among the workstations. To maximally evaluate our framework, we set a comparatively complex environment for the experiments. Nevertheless, to obtain practically ideal performance under other circumstances, the parameters should be precisely tested and fine-tuned with simulation in advance. [ABSTRACT FROM AUTHOR]

Convolutional neural network (CNN), with the ability of feature learning and nonlinear mapping, has demonstrated its effectiveness in prognostics and health management (PHM). However, an explanation on the physical meaning of a CNN architecture has rarely been studied. In this article, a novel wavelet-driven deep neural network, termed as WaveletKernelNet (WKN), is presented, where a continuous wavelet convolutional (CWConv) layer is designed to replace the first convolutional layer of the standard CNN. This enables the first CWConv layer to discover more meaningful kernels. Furthermore, only the scale parameter and translation parameter are directly learned from raw data at this CWConv layer. This provides a very effective way to obtain a customized kernel bank, specifically tuned for extracting defect-related impact component embedded in the vibration signal. In addition, three experimental studies using data from laboratory environment are carried out to verify the effectiveness of the proposed method for mechanical fault diagnosis. The experimental results show that the accuracy of the WKNs is higher than CNN by more than 10%, which indicate the importance of the designed CWConv layer. Besides, through theoretical analysis and feature map visualization, it is found that the WKNs are interpretable, have fewer parameters, and have the ability to converge faster within the same training epochs. [ABSTRACT FROM AUTHOR]

Advanced mechatronics equipment requires reliable and effective performance degradation assessment to guarantee long-term operations. Current data-driven predictions endow the operation and maintenance of mechatronic equipment flexibly and intelligently. However, the sufficient and labeled data in real industrial scenes may not be satisfied, resulting in negative impacts of overfitting and time-consuming annotations. In this article, we propose a novel prognostic model, namely unsupervised meta gated recurrent unit (UMGRU) containing a dual-cycle learning architecture with the designed clustering assignment module to deal with few-shot prognostics under unlabeled historical data. It integrates the strength of double gradient based optimizations for abstracting general degradation knowledge and offering a sensitive model status for precisely online adaptation with limited on-site data. Besides, mini-batch pseudolabels are automatically assigned within each inner cycle learning and further participate in parameter upgrades. Finally, both experimental and industrial data are used to verify the effectiveness of UMGRU. [ABSTRACT FROM AUTHOR]