Prognostics And Health Management Research Articles

Joint learning strategy of multi-scale multi-task convolutional neural network for aero-engine prognosis

Remaining useful life (RUL) prediction and health status (HS) assessment are two key tasks in aero-engine prognostics and health management (PHM) system. However, existing deep learning-based prognostic models perform RUL prediction and HS assessment tasks separately, without considering the correlation between these two tasks. Secondly, traditional deep learning can only extract single-scale features, which limits the ability to extract complex degradation features from high-dimensional condition monitoring data. Therefore, this work proposes a multi-scale and multi-task convolutional neural network for joint learning of aero-engine RUL prediction and HS assessment. Firstly, multi-sensor data with multiple cycles are converted into image samples to integrate more condition monitoring information that is beneficial to prognosis. Then, the multi-scale feature fusion block is designed as the shared network for multi-task, utilizing convolutional layers with filters of different sizes to enhance the ability to extract complex degradation features from high-dimensional condition monitoring data. And a multi-layer concatenation block is constructed to integrate multi-scale features at different levels to fully utilize the important information at different levels. On this basis, a multi-task joint learning block is constructed and a joint loss function is developed for joint learning of RUL prediction and HS assessment. Finally, experiments on two engine degradation datasets, CMAPSS and N-CMAPSS, demonstrate that the proposed network has excellent RUL prediction and HS assessment performance, and outperforms other state-of-the-art methods.

Prognostics and Health Management of Unmanned Surface Vessels: Past, Present, and Future

Abstract With the increasing popularity and deployment of unmanned surface vessels (USVs) all over the world, prognostics and health management (PHM) has become an indispensable tool for health monitoring, fault diagnosis, health prognosis, and maintenance of marine equipment on USVs. USVs are designed to undertake critical and extended missions, often in extreme conditions, without human intervention. This makes the USVs susceptible to equipment malfunction, which increases the probability of system failure during mission execution. In fact, in the absence of any crew onboard, system failure during a mission can create a great inconvenience for the concerned stakeholders, which compels them to design highly reliable USVs that must have integrated intelligent PHM systems onboard. To improve mission reliability and health management of USVs, researchers have been investigating and proposing PHM-based tools or frameworks that are claimed to operate in real time. This paper presents a comprehensive review of the existing literature on recent developments in PHM-related studies in the context of USVs. It covers a broad perspective of PHM on USVs, including system simulation, sensor data, data assimilation, data fusion, advancements in diagnosis and prognosis studies, and health management. After reviewing the literature, this study summarizes the lessons learned, identifies current gaps, and proposes a new system-level framework for developing a hybrid (offline-online) optimization-based PHM system for USVs in order to overcome some of the existing challenges.

A hybrid prognostic & health management framework across multi-level engineering systems with scalable convolution neural networks and adjustable functional regression models

With the rapid development of sensor technology and mechatronics integration, there are emerging demands for achieving prognostic and health management (PHM) within a general framework across multi-level engineering systems. However, current PHM frameworks are mainly supported either by field data or degradation models, which are always limited in inadequate robustness to systematically increase operational availability and reduce maintenance costs. For extending the scalability and flexibility of PHM services, this paper proposes a novel hybrid framework that integrates the scalable convolution neural network (SCNN) with the adjustable functional regression model (AFRM) through a modified linear filter. This proposed framework first establishes a data-driven SCNN with scalable time sequences and kernel sizes to enhance the remaining useful life (RUL) generalization. Then, enabled by an adjustable semi-parametric degradation process model, AFRM with flexible lifetime distributions is designed to evaluate time-to-failure (TTF) in real time. Finally, a modified Kalman filter has been explored to coordinate data-driven RUL results with model-based TTF values to schedule predictive maintenance (PdM). High-accuracy prediction, multi-level systems, high-flexibility evaluation are thus encompassed within the hybrid progressive modification. The proposed framework has been practically validated through multi-level studies including the component-level bearings, subsystem-level engines, and system-level machines. Results demonstrate that our hybrid PHM framework achieves significant improvements in prediction accuracy, evaluation adaptability, and scheduling effectiveness.

Using transformer and a reweighting technique to develop a remaining useful life estimation method for turbofan engines

Estimating the Remaining Useful Life (RUL) of industrial machinery is an important task in Prognostics and Health Management (PHM). Accurate RUL prediction based on current health status and real-time sensor measurements can help establish efficient maintenance strategies in advance, thus improving the productivity and efficiency of machinery operations. However, considering the nature of industrial machinery, a data imbalance problem, which exists in current RUL estimation datasets like the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset, hinders data-driven approaches. To address the detrimental data imbalance problem and improve estimation performance, this research proposes an Adaptive RUL-wise Reweighting (ARR) technique. ARR uses kernel smoothing to maintain the inherent continuity of RUL and a reweighting method to rebalance the effects of samples with different ground-truth RULs. In addition, this research proposes a novel RUL estimation method that uses a transformer architecture, which has not yet been widely applied for PHM tasks. Using a transformer-based estimation model with high expressive power capable of handling multivariate time-series data has proven effective in RUL estimation. Comprehensive experiments using the C-MAPSS dataset based on cross validation are conducted to find optimal hyperparameter configurations and to validate the proposed method’s effectiveness in RUL estimation. For subsets FD001 and FD003 of the C-MAPSS dataset, the proposed method provides state-of-the-art estimation performance with the lowest Root Mean Squared Error (RMSE) values of 11.36 and 11.28 and score values of 192.22 and 133.41, respectively. The proposed method also shows highly competitive performance on subsets FD002 and FD004 of the C-MAPSS dataset.

A CNN-BiLSTM-Attention approach for EHA degradation prediction based on time-series generative adversarial network

As a representative integrated system for power-by-wire (PBW) systems, Electro-hydrostatic actuator (EHA) has series of advantages such as high power density, compactness, and high efficiency, which is one of the important development directions of future hydraulic system field. However, due to its high integration and high reliability requirements, it is challenging to conduct degradation studies in short period of time with limited data samples. For this type of high integrated mechatronics system, Prognostics and Health Management (PHM) is one of the key works to ensure its safety and reliability, especially the performance degradation prediction presented in this paper. To deal with the small size of EHA data, a time-based data enhancement method for expanding the performance data set is proposed based on Time Generative Adversarial Network (TimeGAN). Considering the complex of working state and system performance, the relationship between the EHA operation data and its health indicator is then analyzed using the CNN-BiLSTM-Attention model, so as to generate the health indicator combine with TimeGAN synthesis data. Finally, CNN-BiLSTM-Attention model with multi-input channels is developed, and EHA data as well as TimeGAN synthesized EHA data are incorporated into the model. The results show that this method can greatly improve the prediction accuracy of EHA performance, and provide a novel method for performance degradation prediction of integrated mechatronic system.

Dirichlet probability navigated fault detection via key-group memory auto-encoder under non-stationary working conditions

Real-time condition monitoring is the foundation of prognostics and health management (PHM) for mechanical systems. Constraint by extravagant cost and insurmountable obstacles to simulate real faults in actual working conditions, establishing fault detection models with historical normal data is the most promising way. Meanwhile, non-stationary fault detection is rarely studied in the literature but fits most in real working conditions. Thus, an innovative method named KGMem-DirAE is proposed to conduct real-time fault detection under non-stationary working conditions using normal data only. In the training stage, normal patterns with different semantics are recorded in the developed key-group memory module (KGMem) combined with auto-encoder. Considering the matching probabilities of encoded latent features and recorded memory units (normal patterns), a statistical anomaly score defined by aggregated negative log-likelihood of Dirichlet distributions is proposed to detect incipient failures. In order to deal with time-varying working conditions and capture fault degradation trends concurrently, normalized time-frequency maps containing fault-evolving information are obtained from vibration data. Both widely studied non-stationary and stationary run-to-failure data sets are deeply researched using the proposed KGMem-DirAE and the experimental results proved its superiority against other comparative anomaly detection methods.

Challenges on prognostics and health management for wind turbine components

This paper explores the applicability of prognostics and health management (PHM) for wind turbines (WTs), presenting the PHM approach along with challenges and opportunities in the context of WT components. First, the PHM framework is introduced, consisting of three blocks: observation, analysis, and action. Critical components and failure modes for WTs are identified, and data acquisition strategies using supervisory control and data adquisition (SCADA) and condition monitoring (CM) data are discussed. Prognostics, specifically remaining useful life (RUL) estimation, employs physics model-based, data-driven, and hybrid models. Finally, challenges and opportunities related to data, analysis and CM, and developing RUL prediction models have been found. Data challenges include data standardization, limited public datasets, and data quality issues. Analysis and CM challenges address new sensorless and non-intrusive techniques, as well as the fusion of data sources. Prognostics model challenges involve uncertainty management, interpretability issues, and the need for online updates. Addressing challenges requires incorporating physical knowledge, utilizing transfer learning, and improving online RUL prediction methods.

An adaptive model with dual-dimensional attention for remaining useful life prediction of aero-engine

As an indispensable part in prognostics and health management (PHM) of mechanical systems, the remaining useful life (RUL) estimation has been studied widely and intensively. Recently, with the boom in deep learning, the traditional data-driven prognostic methods like the temporal convolutional neural networks (TCNs), have made remarkable progress in the RUL estimation. However, the conventional TCNs methods do not have a well-established mechanism to parallel weigh the dual-dimensional features of multi-sensory data. Moreover, the conventional TCNs have a fixed network structure which is not flexible to learn deep temporal representations. Here, we propose an adaptive model with dual-dimensional attention for aero-engine RUL prognostics. Specifically, the dual-dimensional multi-head attention (DMHA) module is constructed to assign weights to features in the spatial and temporal dimensions, and then preforms the feature fusion. The mathematical derivations of the DMHA module parameter iterations are performed, which enhances the interpretability of the model. Next, the adaptive TCN (ATCN) is designed to further learn the time correlations of the fused features. Compared with the traditional TCN, the ATCN can adjust the network structure adaptively and extract the deep temporal representations better. Last, the prediction ability of the DMHA-ATCN is verified by conducting experiments on the C-MAPSS dataset.

Structured Data Ontology for AI in Industrial Asset Condition Monitoring

This paper proposes an ontology for prognostics and health management (PHM) applications involving sensor networks monitoring industrial machinery. Deep learning methods show promise for the development of autonomous PHM systems but require vast quantities of structured and representative data to realize their potential. PHM systems involve unique and specialized data characterized by time and context, and thus benefit from tailored data management systems. Furthermore, the use of dissimilar standards and practices with respect to database structure and data organization is a hinderance to interoperability. To address this, this paper presents a robust, structured data ontology and schema that is designed to accommodate a wide breadth of PHM applications. The inclusion of contextual and temporal data increases its value for developing and deploying enhanced ML-driven PHM systems. Challenges around balancing the competing priorities of structure and flexibility are discussed. The proposed schema provides the benefits of a relational schema with some provisions for noSQL-like flexibility in areas where PMH applications demand it. The selection of a database engine for implementation is also discussed, and the proposed ontology is demonstrated using a Postgres database. An instance of the database was loaded with large auto-generated fictitious data via multiple Python scripts. CRUD (create, read, update, delete) operations are demonstrated with several queries that answer common PHM questions.

Remaining useful life estimation of bearing via temporal convolutional networks enhanced by a gated convolutional unit

In the field of prognostics and health management (PHM) for industrial equipment and systems, the estimation of remaining useful life (RUL) constitutes a fundamental task. A reliable and accurate method for estimating the RUL is therefore essential. This paper proposes a dynamic self-adaptive ensemble model, aimed at improving the rolling bearing RUL prediction. This model integrates an adaptive multi-scale feature extractor, a gated convolutional unit (GCU) and temporal convolutional networks (TCN). Through a redesign of the data flow, this model directly incorporates multi-scale comprehensive feature evaluation indicators into the neural network data flow, significantly enhancing the model's feature extraction capabilities. Subsequently, the study extends the traditional TCN by incorporating the GCU module and its gating mechanisms, further strengthening the model's capacity to capture long-term dependencies in sequence tasks. Experimental results demonstrate that, compared to existing state-of-the-art (SOTA) models, our method achieves at least a 10% increase in the prediction accuracy on two public run-to-failure bearing datasets. Beyond the tested datasets, the architecture that directly maps multi-scale evaluation indicators into the structure of neural network data flows also holds potential for broader application across diverse PHM tasks, promising significant improvements in the industrial safety and efficiency.

A Design for Availability Approach for Use with PHM


 
 
 Prognostics and Health Management (PHM) methods are incorporated into systems for the purpose of avoiding unanticipated failures that can impact system safety, result in additional life cycle cost, and/or adversely affect the availability of the system. Availability is the probability that a system will be able to function when called upon to do so. A vailability depends on the system’ s reliability (how often it fails) and its maintainability (how efficiently and frequently it is pro-actively maintained, and how quickly it can be repaired and restored to operation when it does fail). Availability is directly impacted by the success of PHM. Increasingly, customers of critical systems are entering into “availability contracts” in which the customer either buys the availability of the system (rather than actually purchasing the system itself) or the amount that the system developer/manufacturer/supplier is paid for the system is a function of the availability achieved by the customer. Predicting availability based on known or predicted system reliability, operational parameters, logistics, etc., is relatively straightforward and can be accomplished using existing methods. However, while determining the availability that results from a set of events is straightforward, determining the events that result in a desired availability is not, and prediction of a system’s attributes to meet an availability requirement can only be accomplished using “brute force” search-based methods that are not general and become quickly impractical for real systems and when uncertainties are introduced. This paper presents a “design for availability” approach that starts with an availability requirement and uses it to predict the required logistics, design and operation parameters. The method is general and can be applied when the inputs to the problem are uncertain (even the availability requirement can be a probability distribution). The method is demonstrated on several examples with and without PHM.
 
 

Deep Learning-Based Prognostics and Health Management Model for Pilot-Operated Cryogenic Safety Valves.

This paper highlights the significance of safety and reliability in modern industries, particularly in sectors like petroleum and LNG, where safety valves play a critical role in ensuring system safety under extreme conditions. To enhance the reliability of these valves, this study aims to develop a deep learning-based prognostics and health management (PHM) model. Past empirical methods have limitations, driving the need for data-driven prediction models. The proposed model monitors safety valve performance, detects anomalies in real time, and prevents accidents caused by system failures. The research focuses on collecting sensor data, analyzing trends for lifespan prediction and normal operation, and integrating data for anomaly detection. This study compares related research and existing models, presents detailed results, and discusses future research directions. Ultimately, this research contributes to the safe operation and anomaly detection of pilot-operated cryogenic safety valves in industrial settings.

Transformer Based Long-Term Prognostics for Dynamic Operating PEM Fuel Cells

Proton Exchange Membrane Fuel Cell (PEMFC) is considered as a clean alternative energy that has been widely used in many fields. However, the large-scale commercialization of PEMFC is still limited by its durability performance. Prognostics and Health Management (PHM) is considered as a helpful solution to improve the durability of PEMFC with degradation prediction and health-based control and maintenance methods. The long-term PEMFC prognostic task, as a primary component in PHM, has received extensive attention from both academia and industry. Recently, the combination of deep neural networks and PEMFC prognostics has expressed a broad research perspective. The deep-learning-based methods, such as Convolutional Neural Network (CNN), Echo State Network (ESN), and Recurrent Neural Network (RNN) family methods, have been well-studied by many researchers and have shown satisfactory performance in degradation prediction. However, the long-term prediction performance is still limited by the model structure and accumulative errors. This paper focuses on the degradation of the PEMFC and a Transformer-based PEMFC prognostic framework is proposed to predict the long-term degradation of the PEMFC system. The Transformer model is applied for the first time to predict the degradation of the PEMFC, and a series-attention mechanism is proposed to replace the self-attention mechanism in the Vanilla Transformer which could improve the health indicator (HI) prediction performance. Finally, the PEMFC dynamic durability test data is utilized to evaluate the performance of the proposed framework in both multi-step-ahead prediction and long-term prognostic conditions, and the experimental results illustrate the feasibility and effectiveness of the proposed method.

Prognostics Aware Control Design for Extended Remaining Useful Life

As most of the safety critical industrial systems remain sensitive to functional degradation and operate under closed loop, it becomes imperative to take into account the state of health within the control design process. To that end, an effective assessment and extension of the Remaining Useful Life (RUL) of complex systems is a standing challenge that seeks novel solutions at the cross-over of Prognostics and Health Management (PHM) domain as well as automatic control. This paper considers a dynamical system subjected to functional degradation presents a novel control design strategy. Wherein the assessment of state of health of the system is taken into account leading to effective prediction of the RUL as well as its extension. To that end, the degradation model is considered unknown but input-dependent. The control design is formulated as an optimization problem wherein a suitable comprise is reached between the performance and desired RUL of the system. The main contribution of the paper remains in proposal of set-point modulation based approach wherein the control input at a given present time stage is modulated in such way that futuristic health of the system over a long time horizon is extended whilst assuring acceptable performance. The effectiveness of the proposed strategy is assessed in simulation using a numerical example as well as liquid propellant rocket engine case.

Technical language processing for Prognostics and Health Management: applying text similarity and topic modeling to maintenance work orders

AbstractModern manufacturing paradigms have incorporated Prognostics and Health Management (PHM) to implement data-driven methods for fault detection, failure prediction, and assessment of system health. The maintenance operation has similarly benefitted from these advancements, and predictive maintenance is now being used across the industry. Despite these developments, most of the approaches in maintenance rely on numerical data from sensors and field devices for any sort of analysis. Text data from Maintenance Work Orders (MWOs) contain some of the most crucial information pertaining to the functioning of systems and components, but are still regarded as ‘black holes’, i.e., they store valuable data without being used in decision-making. The analysis of this data can help save time and costs in maintenance. While Natural Language Processing (NLP) methods have been very successful in understanding and examining text data from non-technical sources, progress in the analysis of technical text data has been limited. Non-technical text data are usually structured and consist of standardized vocabularies allowing the use of out-of-the-box language processing methods in their analysis. On the other hand, records from MWOs are often semi-structured or unstructured; and consist of complicated terminologies, technical jargon, and industry-specific abbreviations. Deploying traditional NLP to such data can result in an imprecise and flawed analysis which can be very costly. Owing to these challenges, we propose a Technical Language Processing (TLP) framework for PHM. To illustrate its capabilities, we use text data from MWOs of aircraft to address two scenarios. First, we predict corrective actions for new maintenance problems by comparing them with existing problems using syntactic and semantic textual similarity matching and evaluate the results with cosine similarity scores. In the second scenario, we identify and extract the most dominant topics and salient terms from the data using Latent Dirichlet Allocation (LDA). Using the results, we are able to successfully link maintenance problems to standardized maintenance codes used in the aviation industry.

Intelligent fault diagnosis method of rolling bearing based on multi-source domain fast adversarial network

Fault diagnosis of rolling bearings is among the most crucial links in the prognostic and health management of bearings. To solve the problem of single-source domain transfer learning that cannot adapt well to the target domain, a transfer diagnosis method based on multi-source domain fast adversarial network (MSDFAN) is proposed. First, signals from all domains are input into a common subnetwork of fast neural networks to reduce the complexity and network running time of neural networks. Secondly, several adversarial networks are constructed as domain specific feature extractors and then use Higher-order Moment Matching to reduce distribution differences between A and B domains. The two experimental cases of rolling bearing support the effectiveness and superiority of the proposed MSDFAN.

Post-clustering Prioritization Framework for Autonomous Decision Making in the Absence of Ground Truth via Hypothetical Probing

A generic prioritization framework is introduced for addressing the problem of automated prioritization of region of interest or target selection. The framework is based on the assumption that clustering of preliminary data for preidentified regions or targets of interest within an operational area has already occurred, i.e., post-classification, and that the clustering quality can be expressed as an energy/objective function. Region or target of interest prioritization then means to rank regions or targets of interest according to their probability of changing the energy/objective function value upon subsequent hypothetical probing as opposed to actually conducted reexamination, i.e., thorough follow-up or in-situ measurements. The mathematical formalism for calculating these probabilities to contribute to this change of the energy/objective function value is introduced and validated through numerical simulations. Moreover, these probabilities can also be understood as a confidence-check of the classification, i.e., the pre-clustering of the preliminary data. The operation of the prioritization framework is independent of the algorithm used to pre-cluster the preliminary data, and supports autonomous decision-making. It is widely applicable across many scientific disciplines and areas, ranging from the microscopic to the macroscopic scale. Due to its ability to help maximize scientific return while optimizing resource utilization, it is particularly relevant in the context of resource-constrained autonomous robotic planetary exploration as it may extend the Remaining Useful Lifetime (RUL) – a key aspect in Prognostics and Health Management (PHM) – of space missions. On a more general, PHM-relevant level, the prioritization framework may provide an additional mechanism of identifying and correcting the maintenance status of system components to assist predictive maintenance or condition-based maintenance.

Pruning Quantized Unsupervised Meta-Learning DegradingNet Solution for Industrial Equipment and Semiconductor Process Anomaly Detection and Prediction

Machine- and deep-learning methods are used for industrial applications in prognostics and health management (PHM) for semiconductor processing and equipment anomaly detection to achieve proactive equipment maintenance and prevent process interruptions or equipment downtime. This study proposes a Pruning Quantized Unsupervised Meta-learning DegradingNet Solution (PQUM-DNS) for the fast training and retraining of new equipment or processes with limited data for anomaly detection and the prediction of various equipment and process conditions. This study utilizes real data from a factory chiller host motor, the Paderborn current and vibration open dataset, and the SECOM semiconductor open dataset to conduct experimental simulations, calculate the average value, and obtain the results. Compared to conventional deep autoencoders, PQUM-DNS reduces the average data volume required for rapid training and retraining by about 75% with similar AUC. The average RMSE of the predictive degradation degree is 0.037 for Holt–Winters, and the model size is reduced by about 60% through pruning and quantization which can be realized by edge devices, such as Raspberry Pi. This makes the proposed PQUM-DNS very suitable for intelligent equipment management and maintenance in industrial applications.

A bearing remaining life prediction method under variable operating conditions based on cross-transformer fusioning segmented data cleaning

Bearing remaining useful life (RUL) prediction research based on deep learning mostly emphasizes model performance and effective feature vectors, overlooking different densities of outlier distributions in vibration signals at varying degradation stages. Moreover, forecasting models focus on capturing cross-time dependencies, ignoring the dependencies between different variables. To solve these problems, this paper proposes an unsupervised segmented data cleaning algorithm and a RUL prediction framework adaptable to variable operating conditions. The method consists of four steps: (1) Multi-domain feature extraction and selection establish a feature vector space reflecting degradation trends. (2) Segmented data cleaning divides degradation stages, using different penalty factors for outlier cleaning. (3) Cleaned vibration signals undergo a second round of multidomain feature engineering and degradation-stage division. (4) A two-stage Cross-Transformer model is used for RUL prediction. The method proposed has been validated on the prognostics and health management (PHM) bearing degradation dataset. In the constant condition prediction task, the root mean square error (RMSE) and mean absolute error (MAE) were improved to 1.88 and 5.78, respectively. In the variable condition prediction task, the proposed method outperformed existing methods, with an improvement of 59.10 % in RMSE, demonstrating strong generalization performance and practical application value.

Dynamic predictive maintenance strategy for system remaining useful life prediction via deep learning ensemble method

In data-driven prognostics and health management (PHM), most studies focus only on prognostics performance but rarely consider maintenance decision problems. However, simple predictive maintenance decisions are not effective in dealing with the complex operating conditions faced in modern industrial systems. Thus, we propose a complete data-driven dynamic predictive maintenance strategy for system remaining useful life (RUL) prediction via deep learning ensemble method to solve this problem. This deep learning ensemble method is composed of a convolutional neural network (CNN) and a bidirectional long short-term memory network (Bi-LSTM), which aims to effectively predict the system RUL. Then, we consider a dynamic predictive maintenance strategy with uncertain system mission cycles based on the RUL predicted by deep learning ensemble method. Meanwhile, this dynamic predictive maintenance strategy includes order, stock, and maintenance decisions. In addition, the number of missions performed by the system and the reliability of the last performed mission are presented based on the mission cycle and the predicted RUL. Finally, experimental results from the NASA turbofan engine dataset C-MAPSS show the favorable performance of the proposed dynamic predictive maintenance strategy compared to the existing maintenance strategy.