Prognostics and health management is an emerging discipline to scientifically manage the health condition of engineering systems and their critical components. It mainly consists of three main ...aspects: construction of health indicators, remaining useful life prediction, and health management. Construction of health indicators aims to evaluate the system's current health condition and its critical components. Given the observations of a health indicator, prediction of the remaining useful life is used to infer the time when an engineering systems or a critical component will no longer perform its intended function. Health management involves planning the optimal maintenance schedule according to the system's current and future health condition, its critical components and the replacement costs. Construction of health indicators is the key to predicting the remaining useful life. Bearings and gears are the most common mechanical components in rotating machines, and their health conditions are of great concern in practice. Because it is difficult to measure and quantify the health conditions of bearings and gears in many cases, numerous vibration-based methods have been proposed to construct bearing and gear health indicators. This paper presents a thorough review of vibration-based bearing and gear health indicators constructed from mechanical signal processing, modeling, and machine learning. This review paper will be helpful for designing further advanced bearing and gear health indicators and provides a basis for predicting the remaining useful life of bearings and gears. Most of the bearing and gear health indicators reviewed in this paper are highly relevant to simulated and experimental run-to-failure data rather than artificially seeded bearing and gear fault data. Finally, some problems in the literature are highlighted and areas for future study are identified.
Prognostics of the remaining useful life (RUL) has emerged as a critical technique for ensuring the safety, availability, and efficiency of a complex system. To gain a better prognostic result, ...degradation information is quite useful because it can reflect the health status of a system. However, due to the lack of accurate information about the plants' degradation, the prognostic model is usually not well established. To solve this problem, this paper proposes a two-stage strategy that is in the context of data-driven modeling to predict the future health status of a bearing, where the degradation information was estimated by calculating the deviation of multiple statistics of vibration signals of a bearing from a known healthy state. Then, a prediction stage based on an enhanced Kalman filter and an expectation-maximization algorithm were used to estimate the RUL of the bearing adaptively. To verify the effectiveness of the proposed approach, a real-bearing degradation problem was implemented.
State-of-charge (SOC), which indicates the remaining capacity at the current cycle, is the key to the driving range prediction of electric vehicles and optimal charge control of rechargeable ...batteries. In this paper, we propose a combined convolutional neural network (CNN) - long short-term memory (LSTM) network to infer battery SOC from measurable data, such as current, voltage, and temperature. The proposed network shares the merits of both CNN and LSTM networks and can extract both spatial and temporal features from input data. The proposed network is trained using data collected from different discharge profiles, including a dynamic stress test, federal urban driving schedule, and US06 test. The performance of the proposed network is evaluated using data collected from a new combined dynamic loading profile in terms of estimation accuracy and robustness against the unknown initial state. The experimental results show that the proposed CNN-LSTM network well captures the nonlinear relationships between SOC and measurable variables and presents better tracking performance than the LSTM and CNN networks. In case of unknown initial SOCs, the proposed network fast converges to true SOC and, then, presents smooth and accurate results, with maximum mean average error under 1% and maximum root mean square error under 2%. Moreover, the proposed network well learns the influence of ambient temperature and can estimate battery SOC under varying temperatures with maximum mean average error under 1.5% and maximum root mean square error under 2%.
Lithium-ion batteries are critical components to provide power sources for commercial products. To ensure a high reliability of lithium-ion batteries, prognostic actions for lithium-ion batteries ...should be prepared. In this paper, a prognostic method is proposed to predict the remaining useful life (RUL) of lithium-ion batteries. A state-space model for the lithium-ion battery capacity is first constructed to assess capacity degradation. Then, a spherical cubature particle filter (SCPF) is introduced to solve the state-space model. The major idea of the SCPF is to adapt a spherical cubature integration-based Kalman filter to provide an importance function of a standard particle filter (PF). Once the state-space model is determined, the extrapolations of the state-space model to a specified failure threshold are performed to infer the RUL of the lithium-ion batteries. Degradation data of 26 lithium-ion battery capacities were analyzed to validate the effectiveness of the proposed prognostic method. The analytical results show that the proposed prognostic method is more effective in the prediction of RUL of lithium-ion batteries, compared with an existing PF-based prognostic method.
Accurate state-of-charge (SOC) estimation is critical for driving range prediction of electric vehicles and optimal charge control of batteries. In this paper, a stacked long short-term memory ...network is proposed to model the complex dynamics of lithium iron phosphate batteries and infer battery SOC from current, voltage, and temperature measurements. The proposed network is trained and tested using data collected from the dynamic stress test, US06 test, and federal urban driving schedule. The performance on SOC estimation is evaluated regarding tracking accuracy, computation time, robustness against unknown initial states, and compared with results from the model-based filtering approach (unscented Kalman filter). Moreover, different training and testing data sets are constructed to test its robustness against varying loading profiles. The experimental results show that the proposed network well captures the nonlinear correlation between SOC and measurable signals and provides better tracking performance than the unscented Kalman filter. In case of inaccurate initial SOCs, the proposed network presents quick convergence to the true SOC, with root mean square errors within 2% and mean average errors within 1%. Moreover, the estimation time at each time step is sub-millisecond, making it appropriate for real-time applications.
Accurately predicting the lifetime of lithium-ion batteries in early cycles is crucial for ensuring the safety and reliability, and accelerating the battery development cycle. However, most of ...existing studies presented poor prediction results for early prediction, due to the nonlinear battery capacity fade with negligible variation in early cycles. In this paper, to achieve an accurate early-cycle prediction of battery lifetime, a comprehensive machine learning (ML) based framework containing three modules, the feature extraction, feature selection, and machine learning based prediction, is proposed. First, by analysing the evolution pattern of various informative parameters, forty-two features are manually crafted based on the first-100-cycle charge-discharge raw data. Second, to manage feature irrelevancy and redundancy, four typical feature selection methods are adopted to generate an optimal lower-dimensional feature subset. Finally, the selected features are fed into six representative ML models to effectively predict the battery lifetime. Numerical experiments and paired t-test are conducted to statistically evaluate the performance of the proposed framework. Results show that the wrapper-based feature selection method outperforms other methods, and significantly improves the prediction performance of subsequent ML models. Both before and after wrapper feature selection, the elastic net, Gaussian process regression, and support vector machine present better performance than other complex ML prediction models. The support vector machine model combined with wrapper feature selection statistically presents the best result for battery lifetime prediction, with a root-of-mean-square-error of 115 cycles, and a R2 of 0.90. Finally, when compared with an existing work, the root-of-mean-square-error is substantially decreased from 173 to 115 cycles, by using the proposed framework.
•A machine learning based framework is proposed for early prediction of battery lifetime.•Various features are extracted and explored for effective lifetime prediction.•Four typical feature selection methods are explored to manage feature irrelevancy and redundancy.•Six representative machine learning models are investigated for effective lifetime prediction.•Various comparative experiments are conducted to demonstrate the effectiveness of proposed methods.
Previous studies have shown that online data, such as search engine queries, is a new source of data that can be used to forecast tourism demand. In this study, we propose a forecasting framework ...that uses machine learning and internet search indexes to forecast tourist arrivals for popular destinations in China and compared its forecasting performance to the search results generated by Google and Baidu, respectively. This study verifies the Granger causality and co-integration relationship between internet search index and tourist arrivals of Beijing. Our experimental results suggest that compared with benchmark models, the proposed kernel extreme learning machine (KELM) models, which integrate tourist volume series with Baidu Index and Google Index, can improve the forecasting performance significantly in terms of both forecasting accuracy and robustness analysis.
•We propose kernel extreme learning machine with different kernel functions to accurately forecast tourist arrivals.•Co-integration and granger causality relationship between search engine query data and tourist arrivals are verified.•Baidu index and Google trends data are simultaneously regarded as exogenous variables to forecast Beijing tourist arrivals.•Our proposed approach significantly improves the forecasting performance better than other benchmarks used in this study.
Railway inspection has always been a critical task to guarantee the safety of the railway transportation. The development of deep learning technologies brings new breakthroughs in the accuracy and ...speed of image-based railway inspection application. In this work, a series of one-stage deep learning approaches, which are fast and accurate at the same time, are proposed to inspect the key components of railway track including rail, bolt, and clip. The inspection results show that the enhanced model, the second version of you only look once (YOLOv2), presents the best component detection performance with 93% mean average precision (mAP) at 35 image per second (IPS), whereas the feature pyramid network (FPN) based model provides a smaller mAP and much longer inference time. Besides, the detection performances of more deep learning approaches are evaluated under varying input sizes, where larger input size usually improves the detection accuracy but results in a longer inference time. Overall, the YOLO series models could achieve faster speed under the same detection accuracy.
Machinery condition monitoring and fault diagnosis has attracted much attention because it is beneficial to reducing maintenance costs and improving industrial profits. Adaptive fault components ...extraction (AFCE) is the most crucial step for machinery fault diagnosis, and its core is statistical indices. Existing statistical indices including kurtosis and correlated kurtosis are empirical statistical indices (ESIs), and they cannot exactly quantify fault-related information in signals and distinguish fault components from interferential components. Thus, the ESIs might be easily affected by random impulsive noise, low-frequency components, etc. To solve this problem, a new statistical index named optimized weights spectrum based index (OWSI) is proposed in this article. The OWSI satisfies two good properties to guarantee exact quantification of fault components and effectively distinguish interferential components. Moreover, a new OWSI-based methodology is proposed to realize AFCE, and it can be implemented with signal decomposition algorithms such as variational mode decomposition without needing careful parameters tuning. Bearing and gear real-world fault signals are studied to verify the effectiveness of the proposed methodology. Results show that the proposed methodology is superior to ESI-based methods including classic fast kurtogram and newly developed feature mode decomposition.
Accurate battery lifespan prediction is critical for the quality evaluation and long-term planning of battery management systems. As battery degradation process is typically nonlinear, accurate early ...prediction of cycle life with significantly less degradation is extremely challenging. Approaches using machine learning techniques, which are mechanism-agnostic alternatives, to predict battery lifespan are therefore becoming more and more attractive. In this paper, a gradient boosting regression tree (GBRT) is proposed to model complex nonlinear battery dynamics and predict battery lifespan through various extracted battery features. Essentially, the GBRT works by constructing additional trees through minimizing the prediction residues from existing base models. It can identify complex and nonlinear end-to-end relationship and meanwhile provide relative importance for each input feature. Various potential features including voltage-related features, capacity-related features, and temperature-related features are constructed in both discharge time dimension and life cycle dimension and explored for effective lifespan prediction. Key hyper-parameters including learning rate, number of trees, and maximum number of splits are investigated for optimal GBRT prediction. Comparative studies confirm that the proposed method is significantly superior to other machine learning algorithms for battery lifespan prediction with limited data samples and various input features, with mean average percentage error around 7%.
•Various features are extracted and explored for effective battery lifespan prediction.•A GBRT model is proposed to directly predict battery lifespan from extracted features.•The impact of model hyper-parameters on lifespan prediction is investigated.•The relative importance of input features is analyzed and discussed.•Its performance is compared with other state-of-art machine learning methods.