Remaining useful life prediction of lithium-ion battery using a novel particle filter framework with grey neural network

Chen, Lin; Ding, Yunhui; Liu, Bohao; Wu, Shuxiao; Wang, Yaodong; Pan, Haihong

doi:10.1016/j.energy.2021.122581

Cited by 53 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cui et al [13] proposed a comprehensive prediction model based on time-varying particle filter, this method can adaptively select the optimal state-space model according to the characteristics of the data, which improves the prediction accuracy for bearing life; Ge et al [14] proposed a data-driven improved particle filter prediction method, applying a quantum genetic algorithm to solve the particle deterioration problem, and used LSTM neural network to predict the trend of model coefficients achieving good results in predicting the remaining useful life of aero-engines; Wu et al [15] put forward a prediction method combining bat algorithm optimized particle filter and neural network, which is used to predict the state of lithium ion battery, and concluded that the neural network prediction model with two hidden neurons is better than that with three hidden neurons; Zhang et al [16] proposed a prediction method using levy flight optimization particle filter, and used LSTM to learn the deterioration state curve of lithium ions, serving as the state-space model of particle filter algorithm, achieving good prediction results; Chen et al [17] proposed a prediction method combining a new particle filter algorithm with GNN to predict the state of lithium batteries. These scholars have conducted a lot of research on the prediction method combining neural network and improved particle filter, and made some contributions, but these contributions are basically applied to the prediction of lithium-ion battery life or the RUL of aero-engines.…”

Section: Introductionmentioning

confidence: 99%

Remaining Useful Life Prediction for Rolling Bearings With a Novel Entropy-Based Health Indicator and Improved Particle Filter Algorithm

et al. 2023

View full text Add to dashboard Cite

Due to the importance of bearings in modern machinery, the prediction of the remaining useful life (RUL) of rolling bearings has been widely studied. When predicting the RUL of rolling bearings in engineering practice, the RUL is usually predicted based on historical data, and as the historical data increases, the prediction results should be more accurate. However, the existing methods usually have the shortcomings of low prediction accuracy, large cumulative error and failure to dynamically give prediction results with the increase of historical data, which are not suitable for engineering practice.To address the above problems, a novel RUL prediction method is proposed. The proposed method consists of 3 parts: First, the multi-scale entropy-based feature -namely "average multi-scale morphological gradient power spectral information entropy (AMMGPSIE)"from the rolling bearings as the Health indicator (HI) is extracted to ensure all the fault-related information is well-included; Then, the HI is processed with the enhanced Hodrick Prescott trend-filtering with boundary lines (HPTF-BL) to ensure good performance and small fluctuation on the HI; Finally, the deterioration curve is predicted using an LSTM neural network and the improved Particle Filter algorithm that we proposed. The proposed method is validated using the experimental bearing degradation dataset and the casing data of a centrifugal pump bearing from an actual industrial site. Comparing the results with other recent RUL prediction methods, the proposed method achieved state-ofthe-art feasibility and effectiveness, conform to the needs of practical application of the project.

show abstract

Section: Introductionmentioning

confidence: 99%

Remaining Useful Life Prediction for Rolling Bearings With a Novel Entropy-Based Health Indicator and Improved Particle Filter Algorithm

et al. 2023

View full text Add to dashboard Cite

show abstract

“…The predicted output from UPF is iterated back to the training set of data-driven method, hence the online training of LSSVM is also realized. Moreover, the grey neural network (GNN) with sliding windows is proposed to track the battery capacity degradation trend, 19 and the estimated values of battery capacity are used as observation values in PF-based EOL prediction framework. Recently, the future residuals required by the UPF algorithm are predicted by the optimized relevance vector machine (RVM), 20 then the predicted residuals are used in UPF’s EOL prediction phase to update the model’s parameters.…”

Section: Introductionmentioning

confidence: 99%

Early-Stage end-of-Life prediction of lithium-Ion battery using empirical mode decomposition and particle filter

Meng

Azib

Yue

2023

Proceedings of the Institution of Mechanical Engineers, Part A:

View full text Add to dashboard Cite

The predictive maintenance is a major challenge for improving battery safety without compromising performance. Its main objective is to predict the end-of-life (EOL) and assess the associated prediction uncertainty. In this paper, we consider this issue and propose a hybrid method combining empirical mode decomposition (EMD) and particle filter (PF) to perform the battery early EOL prediction and its uncertainty assessment. The proposed approach is tested on the two most widely accessible public lithium-ion battery degradation datasets from the Prognostics Center of Excellence at NASA Ames and the Center for Advanced Life Cycle Engineering at the University of Maryland. To avoid the overfitting, a major constraint in the prediction phase, only the residual sequence obtained after EMD decomposition of the raw data is used as the measurement input for PF. Besides, the sum of standard deviation of all intrinsic mode functions (IMFs) is used to determine the noise characteristic for the PF algorithm. Extensive experiments are conducted to show the importance of uncertainty assessment for the early EOL prediction. Although the distance between the true EOL and the mean predicted EOL has no obvious decrease when more operation data is available, the results show a clear decreasing trend of EOL prediction uncertainty when the prediction starts from later operation cycles. Particularly, a strong experimental support is provided for filtering-based prediction methods in the early EOL prediction.

show abstract

“…It is mainly divided into Kalman filters and particle filters (PFs): Kalman filters assume that the noise obeys Gaussian distribution, based on the minimum variance estimation criterion, and mainly solves the assumption of a linear system and Gaussian probability model, including the extended Kalman filter, 5 the unscented Kalman filter, 6 the adaptive unscented Kalman filter, 7 and the constrained Kalman filter 8 . PFs assume that, compared with the Kalman filter, the state space model of the PF can be nonlinear and the noise distribution can be of any type; 9 they are often used in combination with other methods, for example, Mo et al 10 combined the Kalman filter and particle swarm optimization algorithm to propose a PF‐based SOH estimation method for the estimation of SOH in LiBs, Li et al 11 proposed an online remaining useful life (RUL) prediction method based on unscented PF and least‐squares support vector machine, and Chen et al 12 used a novel PF framework with gray neural network to predict lithium battery SOH and RUL.…”

Section: Introductionmentioning

confidence: 99%

Cross‐domain state‐of‐health estimation of Li‐ion batteries based on transfer neural network with soft‐dynamic time warping

Zhang

Lin

et al. 2023

Energy Science & Engineering

View full text Add to dashboard Cite

The success of deep learning in the field of state‐of‐health (SOH) estimation relies on a large amount of battery data and the fact that all data possess the same probability distribution. While in real situations, a model based on one working condition data set may not be valid for another working condition data set due to distribution differences. Therefore, this article proposes a transfer learning method using soft‐dynamic time warping (soft‐DTW) as the statistical feature in the feature transfer method, called soft‐DTW domain adaptation network (SDDAN). By combining the prediction error with the time‐series gap in the model training process, the feature transformation can make the obtained prediction results more similar to the source domain results, which can help us to obtain better prediction results in the target domain. Experimental results show that SDDAN can effectively predict the SOH of Li‐ion batteries and significantly improve the performance of feature learning and knowledge transfer.

show abstract

Remaining useful life prediction of lithium-ion battery using a novel particle filter framework with grey neural network

Cited by 53 publications

References 19 publications

Remaining Useful Life Prediction for Rolling Bearings With a Novel Entropy-Based Health Indicator and Improved Particle Filter Algorithm

Remaining Useful Life Prediction for Rolling Bearings With a Novel Entropy-Based Health Indicator and Improved Particle Filter Algorithm

Early-Stage end-of-Life prediction of lithium-Ion battery using empirical mode decomposition and particle filter

Cross‐domain state‐of‐health estimation of Li‐ion batteries based on transfer neural network with soft‐dynamic time warping

Contact Info

Product

Resources

About