Evolutionary Optimization of Convolutional Extreme Learning Machine for Remaining Useful Life Prediction

The main objective of this works is to study and improve the performances of the Single hidden Layer Feedforward Neural network (SLFN) for the application of Remaining Useful Life (RUL) prediction of aircraft engines. The most common problems in SLFNs based old training algorithms such as backpropagation are time consuming, over-fitting and the appropriate network architecture identification. In this paper a new incremental constructive learning algorithm based on Extreme Learning Machine algorithm is proposed for founding the appropriate architecture of a neural network under less computational costs. The aim of the proposed training approach is to study its maximum capabilities during RUL prediction by reducing over-fitting and human intervention. The performances of the proposed approach which are evaluated on C-MAPPS dataset and compared with its original variant from the literature. Experimental results proved that the new algorithm outperforms the old one in many metrics evaluations.

Given the difficulty of developing physics-based degradation process models in practice, data-driven prognostics approaches are preferred in several industrial applications. Among data-driven approaches, one can distinguish between (i) degradation-based approaches that predict the future evolution of the equipment degradation and (ii) direct Remaining Useful Life (RUL) prediction approaches which directly predict the equipment RUL. In this work, we compare two direct RUL prediction approaches one based on Back Propagation-Artificial Neural Networks (BP-ANN) and the other one on Extreme Learning Machines (ELM). The two approaches are compared on data from turbofan engines. We consider different prognostic metrics such as RMSE, Accuracy Index, Steadiness Index, a-I metric and the time necessary to train and execute the model. The obtained results show that the ELM-based model is performing only slightly worse than the BP-ANN-based model in terms of accuracy and stability, but it requires a considerably shorter training time.

Remaining useful life (RUL) prediction is a cornerstone of Prognostic and Health Management (PHM), playing a crucial role in ensuring the reliability and safety of equipment. In recent years, deep learning techniques have shown great promise in RUL prediction, providing more reliable and accurate outcomes. However, many current deep learning models still fall short in comprehensive feature extraction. To tackle this limitation, we propose a multi-feature fusion model leveraging a dual-attention mechanism. Initially, convolutional neural networks (CNN) and channel attention mechanisms are employed to preliminarily extract spatial features. Subsequently, a layer combining Long Short-Term Memory (LSTM) and self-attention mechanisms is used to further extract and integrate temporal features. Concurrently, domain knowledge-based handcrafted features are selected for multi-feature fusion. Finally, RUL values are predicted via regression. The effectiveness of the proposed method is validated on widely-used academic datasets, and its superior performance in RUL prediction is demonstrated through comparative analysis with other methods.

Remaining useful life (RUL) prediction of bearings has extraordinary significance for prognostics and health management (PHM) of rotating machinery. RUL prediction approaches based on deep learning have been dedicated to finding a nonlinear mapping relationship between non-stationary monitoring data and RUL. However, most existing approaches pay little attention to the degradation trend of diverse health stages of bearing and lack the discriminative power of crucial degradation features, resulting in the loss of some important information associated with RUL. To address this challenge, this article proposes a novel RUL prediction framework based on degradation-trend-aware deep neural network with attention mechanism (DTADAN). Firstly, the multi-direction features with evident degradation trend are extracted via the analysis of bearing vibration signal from both time domain and time-frequency domain. Next, the deep neural network architecture with attention mechanism is utilized to adaptively learn the critical degradation features beneficial for RUL prediction. Distinct from the existing approaches, the proposed framework is able to dynamically extract key degradation features of the bearing including degradation trend information and effectively fuse multi-direction information to improve RUL prediction accuracy. The performance of the proposed approach is evaluated via case studies on XJTU-SY bearing dataset and PRONOSTIA bearing dataset. Compared with other state-of-the-art approaches, the proposed framework has better predictive accuracy and robustness. Additionally, interpretable analysis is provided to reveal the process of model learning and data characteristics, and the analysis results are helpful in guiding model learning.

Remaining useful life (RUL) prediction of machinery plays a significant role for predictive maintenance, thus attracting more and more attentions in recent years. Stochastic process model-based methods are widely used in the RUL prediction of machinery. One of the major issues in the stochastic process model-based methods is that how to deal with the unit-to-unit variability during the RUL prediction process. Traditional methods generally handle this issue by introducing a unit-to-unit variability parameter into the model expression and estimate the parameter using the maximum likelihood estimation (MLE) algorithm. There exist two major limitations in the traditional methods. 1) The degradation processes are assumed to be dependent on only the age, which restricts their implementation in the cases of the state-dependent degradation processes. 2) They do not discuss the influence of the unit-to-unit variability in the RUL prediction processes systematically. To deal with these two limitations, a new RUL prediction method based on age- and state-dependent stochastic process models is proposed in this paper. In the proposed method, a generalized expression of the age- and stage-dependent stochastic process models is generated. An enhanced MLE algorithm is developed to estimate the model parameters according to the measurements of the available training units. And the unit-to-unit variability parameter is updated according to the real-time measurements of the testing unit. The effectiveness of the proposed method is demonstrated using a numerical simulation dataset of fatigue crack-growth.

Remaining useful life (RUL) prediction is crucial for the predictive maintenance and health management of mechanical transmission systems. Although data-driven RUL prediction methods have advanced significantly, three key challenges remain: (i) constructing a physically interpretable HI that is invariant to operating conditions yet sensitive to incipient degradation; (ii) achieving adaptive time to start prediction (TSP) detection that accommodates non-stationarity and controls false and missed alarms; (iii) accurate and generalizable RUL prediction under cross-domain scenarios. To address these issues, a method combining unsupervised anomaly detection and deep feature transfer for TSP determination and RUL prediction is proposed. A residual health indicator is constructed from an expected trajectory learned by Gaussian process regression (GPR) on the healthy segment and a Mahalanobis distance (MD) deviation measure. For the first time extreme value theory is used to create dynamic thresholds for performance degradation analysis and TSP determination. During the RUL prediction phase, transfer component analysis mitigates feature distribution discrepancies between source and target domains. A dual long short-term memory (dual-LSTM) network with L2 regularization is then employed for RUL prediction across two transfer tasks. Experimental validation on a public benchmark dataset and a laboratory run-to-failure bearing dataset demonstrates the superior predictive accuracy and generalization capability of the proposed approach compared to existing methods.

The remaining useful life (RUL) prediction is critical for the safe and reliable operation of lithium-ion battery (LIB) systems, which characterizes the aging status of the battery and provides early warning for battery replacement. Most existing RUL prediction methods rely on empirical aging models, and the role of the battery mechanism is not considered in the subsequent algorithm settings. The accuracy and stability of data-driven algorithms are severely limited by battery aging data. A new electrochemical-model-based particle filter (PF) framework for LIB RUL prediction is proposed in this paper. Parameters of a simplified electrochemical model (SEM) are used as state variables of the PF algorithm and these parameters can be identified by applying specially designed current excitations to the battery. The SEM-based capacity simulation process is taken as the observation equation in the PF algorithm framework. Therefore, the mechanism of the battery is fully considered when making the RUL prediction. The proposed method is validated through cyclic aging experiment of a cylindrical LFP/graphite LIB of 45Ah. The accuracy of the method is compared with a data-driven-based PF framework for RUL prediction and shows better accuracy and stability, which provides a choice for achieving high-quality RUL prediction.

In industry, forecast prediction and health management (PHM) is used to improve system reliability and efficiency. In PHM, remaining useful life (RUL) prediction plays a key role in preventing machine failures and reducing operating costs, especially for reliability requirements such as critical components in aviation as well as for costly equipment. With the development of deep learning techniques, many RUL prediction methods employ convolutional neural network (CNN) and long short-term memory (LSTM) networks and demonstrate superior performance. In this paper, a novel two-stream network based on a bidirectional long short-term memory neural network (BiLSTM) is proposed to establish a two-stage residual life prediction model for mechanical devices using CNN as the feature extractor and BiLSTM as the timing processor, and finally, a particle swarm optimization (PSO) algorithm is used to adjust and optimize the network structural parameters for the initial data. Under the condition of lack of professional knowledge, the adaptive extraction of the features of the data accumulated by the enterprise and the effective processing of a large amount of timing data are achieved. Comparing the prediction results with other models through examples, it shows that the model established in this paper significantly improves the accuracy and efficiency of equipment remaining life prediction.

Remaining useful life prediction for multi-sensor systems using a novel end-to-end deep-learning method

Accurate Remaining Useful Life (RUL) prediction for complex systems like aero-engines is vital for optimizing maintenance strategies. However, the intrinsic complexity of engine degradation, marked by highly coupled sensor dependencies and multi-scale temporal dynamics, continues to pose a fundamental challenge to data-driven approaches. Existing methods often exhibit incomplete temporal feature modeling and fail to capture the intrinsic spatial coupling among multi-channel sensor signals. Moreover, domain shifts caused by fluctuating flight conditions and hybrid fault modes lead to significant performance degradation in unlabeled target domains.To address these challenges, this paper proposes a deep adversarial domain adaptive transfer learning model named GMSIT-WDADA.First, we design GMSIT, a feature extractor that incorporates gated convolution, a selective state space model, and a sparse inverted Transformer. This architecture achieves the extraction of multi-scale temporal features, including short-term transient local features, long-term trend global features, and cross-temporal correlations, while explicitly capturing the spatial coupling relationships across sensors.Second, a neural network is employed to estimate the Wasserstein distance as a replacement for the conventional domain discriminator, achieving more stable cross-domain feature alignment.Experimental results on the C-MAPSS dataset demonstrate that the proposed model can accurately predict RUL values. Furthermore, RUL prediction experiments on 12 cross-domain scenarios of turbofan engines demonstrate that GMSIT-WDADA performs reliable RUL prediction in unlabeled target domains, achieving 7% lower average RMSE and 39% lower average Score compared with the best methods, further verifying the superiority and robustness of the proposed method.

Lithium-ion battery capacity and remaining useful life prediction using board learning system and long short-term memory neural network

Background Accurate prediction of the remaining useful life (RUL) of assets is fundamental to the development of effective maintenance strategies and overall asset management. Despite significant advancements, there remains a notable gap in integrating fault detection and diagnostics (FDD) with RUL prediction models to create more comprehensive and accurate maintenance systems. One of the key challenges in this field is the limited ability of current models to generalize effectively across different types of equipment and varying operating conditions. This gap emphasizes the need for further research and innovation in developing robust and adaptable RUL prediction methodologies that can be applied broadly across diverse industrial scenarios. Methodology This review systematically evaluates the machine learning (ML) and deep learning (DL) techniques used for anomaly detection and RUL prediction, focusing on their efficacy and practical application. By adhering to the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) criteria, the review identifies and addresses the deficiencies in existing models. It explores a range of machine learning and deep learning methods, including probabilistic approaches, hybrid models that combine multiple machine learning techniques, and neural networks designed to handle large-scale time-series data. The review also examines the potential for synergy between machine learning models and FDD, aiming to enhance the precision of equipment monitoring and the early detection of defects. The challenges of data variability, the irregularity in equipment deterioration, and the interpretability of complex models are highlighted. Results The analysis reveals that while current machine learning and deep learning models have made considerable strides in predicting the RUL of assets, significant challenges remain, particularly in their ability to generalize across various equipment types and operational contexts. Hybrid models and neural networks have shown promise in improving the accuracy of RUL predictions, especially when managing large, complex datasets. However, the irregular nature of equipment wears and tear, coupled with data variability, continues to pose significant challenges. The review highlights the need for more robust and adaptable models that can not only predict RUL more accurately but also integrate seamlessly with FDD systems to provide a more holistic approach to maintenance. Conclusion This comprehensive review focusses on the need for continued research in developing more integrated, generalizable, and efficient predictive maintenance systems. By exploring the application of AI in virtual assistants, the review suggests promising avenues for extending asset longevity and optimizing maintenance schedules. While current models offer valuable insights, they must evolve to address the identified gaps in generalizability and model interpretability.

Bearings are critical components in mechanical systems, and their degradation process typically exhibits distinct stages, making stage-based remaining useful life (RUL) prediction highly valuable. This paper presents a model that combines correlation analysis feature extraction with a Graph Neural Network (GNN)-based approach for bearing degradation stage classification and RUL prediction, aiming to achieve accurate bearing life prediction. First, the proposed Pearson–Spearman correlation metric, along with Kernel Principal Component Analysis (KPCA) and autoencoders, is used to group and fuse health indicators (HIs), thereby obtaining a health indicator (HI) that effectively reflects the bearing degradation process. Then, a model combining Graph Convolutional Network (GCN) and Long Short-Term Memory (LSTM) networks is proposed for bearing degradation stage classification. Based on the classification results, the Adaptive Attention GraphSAGE–LSTM (AAGL) model, also introduced in this study, is employed to precisely predict the bearing’s remaining useful life.

Organic electrochemical transistors (OECTs) show abundant potential in biosensors, artificial neuromorphic systems, brain-machine interfaces, etc With the fast development of novel functional materials and new device structures, OECTs with high transconductance (gm > mS) and good cycling stabilities (> 10,000 cycles) have been developed. While stability characterization is always time-consuming, to accelerate the development and commercialization of OECTs, tools for stability prediction are urgently needed. In this paper, OECTs with good cycling stabilities are realized by minimizing the gate voltage amplitude during cycling, while a remaining useful life (RUL) prediction framework for OECTs is proposed. Specifically, OECTs based on p(g2T-T) show tremendously enhanced stability which exhibits only 46.1% on-current (ION) and 33.2% peak gm decreases after 80,000 cycles (53 min). Then, RUL prediction is proposed based on the run-to-failure (RtF) aging tests (cycling stability test of OECTs). By selecting two aging parameters (ION and peak gm) as health indicators (HI), a novel multi-scale feature fusion (MFF) method for RUL prediction is proposed, which consists of a long short-term memory (LSTM) neural network based multi-scale feature generator (MFG) module for feature extraction and an attention-based feature fusion (AFF) module for feature fusion. Consequently, richer effective information is utilized to improve the prediction performance, where the experimental results show the superiority of the proposed framework on multiple OECTs in RUL prediction tasks. Therefore, by introducing such a powerful framework for the evaluation of the lifetime of OECTs, further optimization of materials, devices, and integrated systems relevant to OECTs will be stimulated. Moreover, this tool can also be extended to other relevant bioelectronics.

Remaining Useful Life (RUL) prediction is a critical technology in prognostics and health management. Rotating machinery plays an indispensable role in industrial processes, and accurate RUL prediction for such equipment is essential to ensure production continuity, optimize maintenance strategies and reduce operational costs. In recent years, deep learning has emerged as a powerful tool for the analysis of multi-source monitoring data derived from rotating machinery, owing to its exceptional proficiency in feature extraction and non-linear modelling. This approach can effectively excavate the deeper-level information within the data, thereby improving prediction accuracy. Consequently, deep learning demonstrates significant application value and broad development prospects in RUL prediction for rotating machinery. This paper provides a comprehensive review of the application of deep learning techniques in RUL prediction for rotating machinery. Firstly, it introduces a unified framework for RUL prediction based on deep learning methodologies. The article then elaborates on the theoretical foundations of several key learning methods, including autoencoders, convolutional neural networks, long short-term memory, gated recurrent units, as well as emerging models such as transformers, temporal convolutional networks and graph neural networks. Subsequently, the paper summarizes the applications of these deep learning methods in RUL prediction for rotating machinery, evaluating the specific problems each method addresses based on their distinct characteristics. Finally, it discusses the significant challenges that deep learning faces within this domain and outlines potential directions for future research.