基于Gamma退化过程的直升机主减速器行星架剩余寿命预测模型

鉴于Gamma过程具有平稳、独立增量等退化建模所需的属性,将其用于描述设备退化过程,并针对缺乏故障数据时难以进行剩余寿命预测的问题,利用设备运行中采集的表征其退化状态的大量间接状态参数和少量直接状态参数,建立了基于Gamma退化过程的剩余寿命预测模型;针对经验最大化算法中似然函数难以解析求解的问题,引入粒子滤波算法实现了模型参数估计;最后将模型应用于直升机主减速器行星架的剩余寿命预测,得到了不同时刻的预测结果及95%置信区间,验证了预测模型的有效性和准确性。     

融合多源数据的非线性退化建模与剩余寿命预测

针对目前基于单个传感器剩余寿命预测方法存在预测精度不高的问题,该文提出一种融合多源传感器数据的非线性退化建模与剩余寿命预测方法。该方法包括复合健康指标的构建、模型参数的估计和传感器融合系数的确定,在确定融合系数后,结合设备历史寿命数据与实时监测数据,利用Bayesian参数更新公式推导出设备的剩余寿命概率分布,实现设备的剩余寿命在线预测。最后通过由商用模块化航空推进系统仿真生成的发动机退化数据集进行仿真实验,结果表明该文所提方法能够有效提高设备剩余寿命预测的准确性。     

Remaining useful life prediction for hard failures using joint model with extended hazard

In this paper, we investigate a joint modeling method for hard failures where both degradation signals and time‐to‐event data are available. The mixed‐effects model is used to model degradation signals, and extended hazard model is used for the time‐to‐event data. The extended hazard is a general model which includes two well‐known hazard rate models, the Cox proportional hazards model and accelerated failure time model, as special cases. A two‐stage estimation approach is used to obtain model parameters, based on which remaining useful life for the in‐service unit can be predicted. The performance of the method is demonstrated through both simulation studies and a real case study.     

Remaining useful life prediction via long-short time memory neural network with novel partial least squares and genetic algorithm

Advancements in information technology have made various industrial equipment increasingly sophisticated in recent years. The remaining useful life (RUL) of equipment plays a crucial important role in the industrial process. It is difficult to establish a functional RUL model as it requires the fusion of time-series data across different scales. This paper proposes a long-short term memory neural network, which integrates a novel partial least square based on a genetic algorithm (GAPLS-LSTM). The parameters are first analyzed by PLS to obtain the parameter fusion function of the health index (HI). The GA then searches the optimal coefficients of the function; the expected HI values can be calculated with the fusion function. Finally, the RUL of the equipment is predicted with the LSTM method. The proposed GAPLS-LSTM was applied to RUL prediction of a marine auxiliary engine to validate it by comparison against GAPLS-BP and GAPLS-RNN methods. The results show that the proposed method is capable of effective RUL prediction.     

An ensemble model considering health index based classification for remaining useful life prediction

Accurate prediction of remaining useful life (RUL) plays an important role in the formulation of maintenance strategies. However, due to the diversity of the failure mode of equipment, there are significant differences between the degradation data, which greatly affects the accuracy of RUL prediction. In this case, an ensemble prediction model considering health index-based (HI-based) classification is proposed in this paper. Firstly, the stacked autoencoder (SAE) is employed to construct the HI. Then, the time window is used to sequentially process the HI sequence, so that many data segments with the same length can be achieved. To differentiate the data with the similar degradation process, K-means and Xgboost are selected to construct offline and online data classification models respectively. Finally, according to the results of the data classification, the ensemble model that integrates multiple machine learning methods are separately trained and then adaptively used for RUL prediction. In addition, integrating multiple methods helps to improve the generalization ability of the model. The NASA C-MAPSS dataset is applied to verify the effectiveness of the proposed method, and the results show that the proposed method achieves a higher prediction accuracy and shorter computational time than other existing prediction models.     

Remaining useful life prediction for lithium-ion batteries using a quantum particle swarm optimization-based particle filter

ABSTRACT

A novel RUL prediction approach for lithium-ion batteries using quantum particle swarm optimization (QPSO)-based particle filter (PF) is proposed. Compared to particle swarm optimization (PSO)-based PF, QPSO-based PF is proved to have a better performance in global searching and has fewer parameters to control, which makes QPSO-PF easier for applications. Moreover, fewer particles are required by QPSO-PF to accurately track the battery's health status, leading to a reduction of computation complexity. RUL prediction results using real data provided by NASA and compared with benchmark approaches demonstrates the superiority of the proposed approach.     

Remaining useful life prediction of lithium-ion battery using a novel health indicator

Remaining useful life (RUL) prediction plays a significant role in the health prognostic of lithium-ion batteries (LIBs). The capacity or internal resistance is commonly used to quantify degradation process and predict RUL of LIB, but those two indicators are difficult to be obtained due to complex operational conditions and high costs, respectively. To address this issue, we extract a novel health indicator (HI) from the battery current profiles that can be directly measured online. Furthermore, the indicator is optimized by Box-Cox transformation and evaluated by correlation analysis for degradation modeling accurately. Finally, relevance vector machine (RVM) algorithm is utilized to make a probabilistic prediction for battery RUL based on the extracted HI. The correlation analysis verifies the effectiveness of the novel HI, and comparative experiments demonstrate the proposed method can predict RUL of LIB more accurately.     

A scheduling model for systems with task and health dependent remaining useful life prognostics

When we schedule a system to perform a task, a factor that should be taken into account is the remaining useful life prognostics of the system. This prognostics of the system may depend not only on the health state of the system, but also on the characteristics of the task to be performed. Assuming such prognostics is available at the time of system scheduling, the problem is to find a method to schedule the system, which can improve the expected profit rate. Two system life models were proposed for the case considered in this paper. Due to the dynamic nature of the problem, a global optimal policy is hard to find, we proposed an approach based on the approximated expected profit rate to schedule the systems. The approach is validated through simulations compared with a number of other task scheduling rules to show the advantage of the proposed approach. We also find the optimal global stationary result by exhaustive search of small scheduling problems of few systems and tasks to compare with the proposed approximate one. Further numerical analyses are presented to demonstrate the process of determining a decision variable and the sensitivity analysis in terms of a cost parameter.     

线性FBM过程随机退化设备剩余寿命自适应预测

现有文献主要通过马尔科夫过程来描述设备的退化轨迹,鲜有考虑历史状态对未来退化轨迹的影响.鉴于此,为了反映退化轨迹的记忆效应对剩余寿命预测的影响,首先基于分数布朗运动建立一种线性随机退化模型,并通过引入随机效应来反映不同样本间的退化差异性;其次,基于弱收敛性理论与分数布朗运动的特性推导剩余寿命分布的近似解析解;再次,基于...     

Hybrid approach for remaining useful life prediction of ball bearings

Remaining useful life (RUL) prediction plays an important role in predictive maintenance systems to support decision‐makers for arranging maintenance tasks and related resources. We propose a hybrid approach that is combined an exponential weighted moving average (EWMA) control chart for anomaly detection and machine learning models such as support vector regression (SVR) and random forest regression (RFR) with differential evolution (DE) algorithm to predict the RULs of ball bearings. Here, DE algorithm is used to find the optimal hyperparameters of SVR model. The datasets of ball bearings from the Prognostics Data Repository of NASA are used to compare the prediction performance of different methods. The degradation behavior of training data from the anomaly time to the end of life is used to transfer learning for the testing data in the SVR and RFR models. The results indicate that the proposed methods outperform the other four existing methods in terms of score. Therefore, the proposed hybrid approach is a reliable tool for the RUL prediction of ball bearings.     

Remaining service life prediction based on gray model and empirical Bayesian with applications to compressors and pumps

In this study, a three-step remaining service life (RSL) prediction method, which involves feature extraction, feature selection, and fusion and prognostics, is proposed for large-scale rotating machinery in the presence of scarce failure data. In the feature extraction step, eight time-domain degradation features are extracted from the faulty variables. A fitness function as a weighted linear combination of the monotonicity, robustness, correlation, and trendability metrics is defined and used to evaluate the suitability of the features for RSL prediction. The selected features are merged using a canonical variate residuals-based method. In the prognostic step, gray model is used in combination with empirical Bayesian algorithm for RSL prediction in the presence of scarce failure data. The proposed approach is validated on failure data collected from an operational industrial centrifugal pump and a compressor.     

Remaining useful life prediction via a variational autoencoder and a time-window-based sequence neural network

The prediction of remaining useful life (RUL) has attracted much attention, and it is also a key section for predictive maintenance. In this study, a novel hybrid deep learning framework is proposed for RUL prediction, where a variational autoencoder (VAE) and time-window-based sequence neural network (twSNN) are integrated. Among it, VAE is used to extract the hidden and low-dimensional features from the raw sensor data, and a loss function is designed to extract useful data features; by using a sliding time window, twSNN can predict RUL dynamically; meanwhile, it can simplify the network architecture in the time dimension. Furthermore, to achieve higher performance on various failure conditions, long short-term memory (LSTM) cell and convolutional LSTM (ConvLSTM) cell are designed for twSNN respectively. A case study is completed with a dataset of aircraft turbine engines. It is found that the proposed frameworks with LSTM cell and ConvLSTM cell have better performance on both single failure mode and multiple failure modes. The results also show that the prediction accuracy is averagely improved by 6.65% for single failure mode and 15.05% for multiple failure modes respectively.     

Degradation Feature Selection for Remaining Useful Life Prediction of Rolling Element Bearings

Rolling element bearings are among the most widely used and also vulnerable components in rotating machinery equipment. Recently, prognostics and health management of rolling element bearings is more and more attractive both in academics and industry. However, many studies have been focusing on the prognostic aspect of bearing prognostics and health management and few efforts have been performed in relation to the optimal degradation feature selection issue. For more effective and efficient remaining useful life predictions, three goodness metrics of correlation, monotonicity and robustness are defined and combined for automatically more relevant degradation feature selection in this paper. Effectiveness of the proposed method is verified by rolling element bearing degradation experiments. Copyright © 2015 John Wiley & Sons, Ltd.     

State estimation of discrete event systems for RUL prediction issue

This paper concerns Remaining Useful Life (RUL) estimation of discrete event systems. For that purpose, physics-based models with partially observed stochastic Petri nets are used to represent the system and its sensors. The advantage of the proposed modelling approach is to provide a realistic representation of the system, including the interaction between the normal behaviours and the failure processes. From the proposed modelling and collected measurements, timed trajectories, which are consistent with the observations, are obtained. Based on the event dates, our approach consists in evaluating the probabilities of the consistent behaviours using probabilistic models. State estimation is obtained as a consequence. The most probable future degradations, from the current state, are then considered and a method for fault prognosis is presented. Finally, the prognosis result is used to estimate the RUL as a time interval. A case study is proposed to show the applicability of the proposed method.     

Transfer learning based deep learning model and control chart for bearing useful life prediction

The remaining useful life (RUL) of the machine is one of the key information for predictive maintenance. If there is a lack of predictive maintenance strategy, it will increase the maintenance and breakdown costs of the machine. We apply transfer learning techniques to develop a new method that predicts the RUL of target data using degradation trends learned from complete bearing test data called source data. The training length of the model plays a crucial role in RUL prediction. First, the exponentially weighted moving average (EWMA) chart is used to identify the abnormal points of the bearing to determine the starting point of the model's training. Secondly, we propose transfer learning based on a bidirectional long and short-term memory with attention mechanism (BiLSTMAM) model to estimate the RUL of the ball bearing. At the same time, the public data set is used to compare the estimation effect of the BiLSTMAM model with some published models. The BiLSTMAM model with the EWMA chart can achieve a score of 0.6702 for 11 target bearings. The accuracy of the RUL estimation ensures a reliable maintenance strategy to reduce unpredictable failures.     

A transformed gamma process for bounded degradation phenomena

Most of the stochastic models adopted to describe the evolution over time of degradation phenomena of technological units assume that their degradation level can increase indeterminately. However, these degradation phenomena are typically subjected to obvious bounds, if only because technological units have finite size. In fact, very often, this inconsistency does not significantly affect the effectiveness of unbounded degradation models, since degrading units are usually assumed to fail when their degradation level exceeds a failure threshold that is much smaller than the obvious bounds. Nevertheless, in some cases, due to the very nature of the underlying degradation mechanism, less obvious bounds could exist, which are not necessarily far from the failure thresholds. The question that arises is whether the use of a bounded degradation model, in this latter type of experimental situations, could be beneficial. For this purpose, since a bounded degradation process should necessarily have dependent increments, in this paper we investigate the potential of a new bounded transformed gamma (TG) process to adequately describe bounded degradation phenomena and predict their future evolution. Differently from other existing gamma process based bounded degradation models, here the upper bound is treated as an unknown parameter that has to be estimated from the available degradation data. A numerical example is presented where the parameters of the proposed model are estimated from simulated data. Then the model is applied to a set of wear measures of cylinder liners that equip a diesel engine for marine propulsion, which have also stimulated this study. Model parameters are estimated by using the maximum likelihood (ML) method. The fitting ability of the proposed new bounded process is compared to that of an unbounded gamma process, which was previously adopted to analyze the same liner wear data. Obtained results are critically discussed in the paper.     

Ensemble model for the degradation prediction of proton exchange membrane fuel cell stacks

Proton exchange membrane fuel cell (PEMFC) stacks are widely used in mobile and portable applications due to their clean and efficient model of operation. We propose an ensemble model based on a stacked long short-term memory model that combines three machine-learning models, including long short-term memory with attention mechanism, support vector regression, and random forest regression, to improve the degradation prediction of a PEMFC stack. The prediction intervals can be derived using the dropout technique. The proposed model is compared with some existing models using two PEMFC stacks. The results show that the proposed model outperforms the other models in terms of mean absolute percentage error and root mean square error. Regarding the remaining useful life prediction, the proposed model with the sliding window approach can provide better results.     

An ensemble model for engineered systems prognostics combining health index synthesis approach and particle filtering

Prognostics, in other words, remaining useful life (RUL) estimation is a core task of prognostics and health management (PHM). Reliable RUL predictions can reduce maintenance costs, improve production efficiency, and avoid unexpected downtime. Lots of models for RUL predictions have been proposed; however, noise and the nonlinear nature of degradation phenomena often leads to poor prognostics results, and the acquired engineered system data are usually subject to a high level of uncertainty. This makes the RUL estimation models less than satisfactory. Accurate RUL estimation and prediction not only rely on an accurate model but also depend on the adjustments of model parameters to track the variation. In this paper, an ensemble model combining the health index synthesis (HIS) approach and improved particle filtering (PF) is introduced. HIS approach was used to obtain the synthesized health index (SHI) for an engineered system with multiple sensors, which indicated the system's degradation model, while the improved PF approach was used to adjust the parameters of the degradation model obtained from the HIS approach and optimized the RUL estimation results. The performance of the prognostics approach introduced in this paper was demonstrated by using turbofan engine degradation data sets, which was supplied by NASA Ames, and results were compared with several usually used methods.     

New fusion and selection approaches for estimating the remaining useful life using Gaussian process regression and induced ordered weighted averaging operators

In this paper, we propose new fusion and selection approaches to accurately predict the remaining useful life. The fusion scheme is built upon the combination of outcomes delivered by an ensemble of Gaussian process regression models. Each regressor is characterized by its own covariance function and initial hyperparameters. In this context, we adopt the induced ordered weighted averaging as a fusion tool to achieve such combination. Two additional fusion techniques based on the simple averaging and the ordered weighted averaging operators besides a selection approach are implemented. The differences between adjacent elements of the raw data are used for training instead of the original values. Experimental results conducted on lithium-ion battery data report a significant improvement in the obtained results. This work may provide some insights regarding the development of efficient intelligent fusion alternatives for further prognostic advances.     

Remaining Life Prediction Based on Wiener Processes with ADT Prior Information

Precisely predicting the remaining life for an individual plays an important role in condition‐based maintenance, so Bayesian inference method, which can integrate useful data from several sources to improve the prediction accuracy, has became a research hot. Aiming at the situation that accelerated degradation tests have been widely applied to assess the reliability of products, a remaining life prediction method based on Bayesian inference by taking accelerated degradation data as prior information is proposed. A Wiener process with random drift, diffusion parameters is used to model degradation data, and conjugate prior distributions of random parameters are adopted. To solve the problem that it is hard to estimate the hyper parameters from accelerated degradation data using an Expectation Maximization algorithm, a data extrapolation method is developed. With acceleration factors, degradation data are extrapolated from accelerated stress levels to the normal use stress level. Acceleration factor constant hypothesis is used to deduce the expression of acceleration factor for a Wiener degradation model. Besides, simulation tests are designed to validate the proposed method. The method of constructing the confidence levels for the remaining life predictions is also provided. Finally, a case study is used to illustrate the application of our developed method. Copyright © 2015 John Wiley & Sons, Ltd.