NASA Datasets Research Articles

Lightweight state-of-health estimation of lithium-ion batteries based on statistical feature optimization

To enhance the model estimation performance and minimize the possibility of overfitting, we propose a statistical optimization strategy of healthy features (HFs) for estimating the state-of-health (SOH) of lithium-ion batteries (LIBs). Firstly, a series of initial HFs were extracted from the voltage, current, temperature, incremental capacity (IC) curves, and differential thermal voltammetry (DTV) curves according to the battery characteristics. Secondly, the six statistical features (i.e., mean, median, lower quartile, range, upper quartile, and standard deviation) of the initial HFs for each charge cycle were calculated. Thirdly, to minimize the impact of redundant features and noise, the optimal HF set was identified by a comparative analysis among different combinations of the six statistical features. Thus, the battery SOH can be estimated using the dual-kernel Gaussian process regression (GPR), which improves the estimation accuracy and generalization ability of the single-kernel GPR. To prevent estimation errors due to manual adjustments, the GPR hyperparameters were optimized via the northern goshawk optimization (NGO) algorithm. Experiments were conducted on the NASA and Oxford datasets, with most estimation errors below 1%. Experimental results demonstrate the lightweight NGO-Dualkernel-GPR model, by incorporating the statistical feature optimization strategy, achieves exceptional SOH estimation performance for different LIBs.

Engine remaining useful life prediction based on PSO optimized multi-layer long short-term memory and multi-source information fusion

Engine as the core component of mechanical equipment, its operating state directly affects whether the equipment can operate normally. Predicting the engine remaining useful life (RUL) can monitor the health of the engine in real time and formulate a timely and reasonable maintenance plan. Aiming at the engine monitoring data with various and long time span, we propose a direct prediction method of engine RUL based on particle swarm optimization (PSO) optimized multi-layer Long Short-Term Memory (LSTM) in this paper. Firstly, the monitoring data that can well reflect the engine degradation trend is screened out, and the samples are constructed through a sliding time window. Then, a multi-layer LSTM model is constructed to mine the deep-seated features of the samples for predicting the engine RUL. Finally, the hyperparameters of the multi-layer LSTM model are optimized automatically by the PSO algorithm to optimize the performance of the model. The effectiveness of this method is verified by NASA data set. RMSE, MAE and the scoring function are used as evaluation indexes. RMSE and score of the prediction results are 12.35 and 284.1, respectively. It has higher prediction accuracy compared with traditional deep learning and machine learning methods.

A novel health indicator by dominant invariant subspace on Grassmann manifold for state of health assessment of lithium-ion battery

The precise estimation of the state of health (SoH) in Lithium-ion batteries (LiBs) relies heavily on a reliable health indicator (HI). Conventional indicators are often constructed by directly concatenating features from multiple sources. It overlooks significant non-linear and correlative information inherent in raw signals. To address this limitation, this paper introduces an innovative approach for SoH estimation in LiBs. Deep features extracted from signals of various sensors are obtained using denoising auto-encoders (DAEs). Then the dominant invariant subspaces (DIS) are calculated through the non-linear transformation of multi-source features on the Grassmann manifold. It can preserve essential and robust characteristics. The health indicator quantifies the geodesic distance of DIS using a projection metric. It provides a more comprehensive inclusion of nonlinear and correlation information. Consequently, this indicator offers heightened precision in discerning differences in health states. Validation of the proposed method is conducted using the NASA dataset. The result demonstrates its effectiveness on the SoH assessment and superiority to the state-of-the-art method.

Joint evaluation and prediction of SOH and RUL for lithium batteries based on a GBLS booster multi-task model

From the perspective of safe electric vehicle operation, accurately assessing the state of health (SOH) and remaining useful life (RUL) of lithium batteries holds paramount importance. This paper introduces a novel multi-task learning data-driven model named GBLS Booster, focusing on the joint estimation and prediction of SOH and RUL. GBLS Booster integrates the strengths of GBLS, offering reduced computation, swift computing speed, and harnesses the powerful feature extraction capabilities of the CNN-Transformers algorithm-based Booster. Additionally, the Tree-structured Parzen Estimator (TPE) algorithm is applied to optimize the model.In this study, 10 healthy indicators (HIs) are devised to capture variations in battery SOH. These HIs are derived from readily available sensor data, encompassing current, voltage, and temperature information. The random forest method (RF) is employed to further refine features and minimize data dimensions. Concerning the RUL prediction, the capacity data is often plagued by significant noise. To address this challenge, the complete empirical mode decomposition (CEEMDAN) method is employed for noise reduction decomposition, followed by the utilization of the Pearson correlation coefficient to eliminate noisy data points.The proposed method is rigorously evaluated using the NASA dataset and CLACE dataset for modeling simulation and verification. Comparative analysis with other algorithms is conducted. The results demonstrate the superior performance of the proposed model, showcasing exceptional accuracy (with a minimum Mean Absolute Percentage Error (MAPE) of 0.3348 % for SOH and a minimum Relative Error (RE) of 0.01 % for RUL), robustness, and generalization capabilities.

State-of-health and remaining-useful-life estimations of lithium-ion battery based on temporal convolutional network-long short-term memory

Accurate estimations in state of health (SOH) and remaining useful life (RUL) are significant for safe and efficient operation of batteries. With the development of big data and deep learning technology, the neural network method has been widely used for SOH and RUL estimations because of its excellent nonlinear mapping performance, adaptive performance and self-learning performance. In this paper, a novel hybrid model based on temporal convolutional network-long short-term memory (TCN-LSTM) for SOH and RUL estimations is proposed. The hyperparameters of each layer in the model are optimized using Bayesian optimization algorithm. Three different models, including convolutional neural network-long short-term memory (CNN-LSTM) model, temporal convolutional network (TCN) model and long short-term memory (LSTM) model, are adopted as comparisons to evaluate the performance of the proposed model. All the models are tested using two public battery datasets from National Aeronautics and Space Administration (NASA dataset) and Oxford University (OX dataset). In SOH task, the TCN-LSTM model achieves an accuracy improvement of >16 % and 14 % in NASA and OX datasets, respectively. In RUL task, the accuracies of the TCN-LSTM model and the CNN-LSTM model are superior to other models in NASA dataset; while the LSTM model and the CNN-LSTM model have better performance in OX dataset.

Self-supervised Health Representation Decomposition based on contrast learning

Accurately predicting the Remaining Useful Life (RUL) of equipment and diagnosing faults (FD) in Prognostics and Health Management (PHM) applications requires effective feature engineering. However, the large amount of time series data now available in industry is often unlabeled and contaminated by variable working conditions and noise, making it challenging for traditional feature engineering methods to extract meaningful system state representations from raw data. To address this issue, this paper presents a Self-supervised Health Representation Decomposition Learning(SHRDL) framework that is based on contrast learning. To extract effective representations from raw data with variable working conditions and noise, SHRDL incorporates an Attention-based Decomposition Network (ADN) as its encoder. During the contrast learning process, we incorporate cycle information as a priori and define a new loss function, the Cycle Information Modified Contrastive loss (CIMCL), which helps the model focus more on the contrast between hard samples. We evaluated SHRDL on three popular PHM datasets (N-CMAPPS engine dataset, NASA, and CALCE battery datasets) and found that it significantly improved RUL prediction and FD performance. Experimental results demonstrate that SHRDL can learn health representations from unlabeled data under variable working conditions and is robust to noise interference.

State of health estimation for lithium-ion batteries based on two-stage features extraction and gradient boosting decision tree

To ensure the stable, reliable, and safe operation of lithium-ion batteries, the state of health (SOH) serves as a crucial basis for battery safety monitoring and operational maintenance. However, the complex working environment of batteries leads to the presence of a significant amount of noise data, abnormal data, and data discontinuities, posing challenges for battery feature extraction and SOH estimation. To address this issue, this paper proposes a two-stage transformation-based feature extraction method combined with the gradient boosting decision tree (GBDT) algorithm. The two-stage transformation feature extraction method utilizes singular spectrum analysis (SSA) and fast Fourier transform (FFT) to extract time-domain and frequency-domain health features from the battery data, respectively. The GBDT ensemble learning algorithm is employed to estimate the SOH of lithium-ion batteries, leveraging the capability to effectively uncover and learn relevant features by combining multiple weak learners into a strong model. Furthermore, correlation analysis and ablation experiments are conducted to validate the effectiveness of the two-stage feature extraction method. Additionally, four sets of comparative experiments are designed to demonstrate the superior performance of the GBDT method compared to four other machine learning methods. The experimental results show that the mean absolute error (MAE) and root mean square error (RMSE) of SOH estimation using the proposed method do not exceed 3 % on the NASA and CALCE datasets.

A data-driven prediction model for the remaining useful life prediction of lithium-ion batteries

Accurate prediction of remaining useful life (RUL) can ensure the safety and reliability of power batteries during operation, reduce the failure rate and operating costs, and enhance user experience. However, battery degradation is a complex, nonlinear dynamic process that is difficult to fully comprehend and predicting RUL remains a significant challenge. To address this issue, the hybrid data-driven prediction model PCA-CNN-BiLSTM was proposed in this paper, which combines principal component analysis (PCA), convolutional neural network (CNN), and bi-directional long short-term memory (Bi-LSTM) network. PCA was applied to downscale and whiten the health factor (HF) to maximize the extraction of important features of lifespan decay, while reducing the correlation between features. The convolution kernel of the CNN was used to explore the local region feature information of the input information and search for the common patterns among the neighboring data. Additionally, the model parameters and computational efforts were reduced through pooling. Finally, battery RUL prediction was achieved using Bi-LSTM, which has the advantages of effectively enhancing model accuracy and reducing the risk of over-fitting by taking into account both past and future data. The performance of the proposed model was evaluated utilizing NASA and CALCE's battery datasets, and the results suggest that it exhibits a high level of accuracy across various datasets. Compared to other methods, the PCA-CNN-BiLSTM method has the best performance indicators for predicting battery RUL, including RMSE, MAE, MAPE, RULe and DOL. This indicates that the proposed model has better fitting performance, accuracy, robustness, and generalization ability.

Optimized data-driven approach for remaining useful life prediction of Lithium-ion batteries based on sliding window and systematic sampling

The prediction of remaining useful life (RUL) in lithium-ion batteries (LIB) serves as a critical health index for evaluating battery parameters, including efficiency, robustness, and accuracy. Existing RUL prediction techniques encounter challenges related to data extraction based on battery parameters and data samples, as well as the selection of appropriate model hyperparameters to achieve accurate outcomes. To address these issues, this article presents a novel framework for RUL prediction using a hybrid optimized data-driven approach, which combines a cascaded forward neural network (CFNN) with the innovative jellyfish optimization (JFO) technique. Initially, a comprehensive 46-data feature-based framework is developed for training the proposed hybrid JFO-CFNN model, employing a mathematical systematic sampling (SS) method. The SS method facilitates the selection of 15 relevant data features, from the LIB parameters such as temperature, current, and voltage, from each charging cycle. Additionally, the developed framework mitigates the capacity regeneration effect present with the capacity degradation profile by employing the overlapping sliding window (OSW) technique. Moreover, the CFNN model parameters, such as hidden neurons and learning rate, are optimized using the newly introduced meta-heuristic JFO technique. To validate the performance of the proposed hybrid JFO-CFNN model, comparisons are made with other models such as the JFO-optimized artificial neural network (ANN) and deep neural network (DNN) model as well as with another battery dataset obtained from the Massachusetts Institute of Technology (MIT) Stanford battery database. The outcomes based on the NASA battery suggested the average RMSE of the proposed JFO-CFNN model below 0.1 and 0.15 for NASA and MIT Stanford battery datasets. The experimental results demonstrate the high prediction accuracy and applicability of the proposed hybrid JFO-CFNN model.

Data-driven state of health estimation for lithium-ion battery based on voltage variation curves

The state of health (SOH) estimation of lithium-ion batteries (LIBs) is crucial for battery management system, but the accuracy and generalizability of the widely used data-driven methods are strongly dependent on the selection of LIBs health features (HFs). In this paper, four types of LIBs with different anode types from four datasets, including NASA dataset, CALCE dataset, Oxford dataset and UL-PUR dataset, were selected to extract the area of constant current charging and discharging voltage curves as two sets of HFs, and then the high correlation between the HFs and SOH is verified by their Pearson coefficient. Secondly, with the two sets of HFs, the SOH of selected batteries in the four datasets are evaluated under Gaussian Process Regression, Long and Short-Term Memory neural network and Back Propagation neural network respectively. With a training/test set ratio model of 50/50 and cross-validation method, all algorithms obtain accurate SOH estimation results. Finally, the estimation results are compared with reference data under the same dataset and training mode, and it is found that the proposed method shows better estimation accuracy and robustness than other evaluation methods by multiple HFs or even complex algorithms.

A New Improved Prediction of Software Defects Using Machine Learning-based Boosting Techniques with NASA Dataset

Predicting when and where bugs will appear in software may assist improve quality and save on software testing expenses. Predicting bugs in individual modules of software by utilizing machine learning methods. There are, however, two major problems with the software defect prediction dataset: Social stratification (there are many fewer faulty modules than non-defective ones), and noisy characteristics (a result of irrelevant features) that make accurate predictions difficult. The performance of the machine learning model will suffer greatly if these two issues arise. Overfitting will occur, and biassed classification findings will be the end consequence. In this research, we suggest using machine learning approaches to enhance the usefulness of the CatBoost and Gradient Boost classifiers while predicting software flaws. Both the Random Over Sampler and Mutual info classification methods address the class imbalance and feature selection issues inherent in software fault prediction. Eleven datasets from NASA's data repository, "Promise," were utilised in this study. Using 10-fold cross-validation, we classified these 11 datasets and found that our suggested technique outperformed the baseline by a significant margin. The proposed methods have been evaluated based on their abilities to anticipate software defects using the most important indices available: Accuracy, Precision, Recall, F1 score, ROC values, RMSE, MSE, and MAE parameters. For all 11 datasets evaluated, the suggested methods outperform baseline classifiers by a significant margin. We tested our model to other methods of flaw identification and found that it outperformed them all. The computational detection rate of the suggested model is higher than that of conventional models, as shown by the experiments..

Performance evaluation of software defect prediction with NASA dataset using machine learning techniques

Performance evaluation of software defect prediction with NASA dataset using machine learning techniques

A Rapid Test Method Based on Prediction with Swarm Intelligent Optimization and Monte Carlo Expansion for Lithium-Ion Battery Cycle Life

The cycle life test offers significant sustainment for utilization and maintenance of lithium-ion batteries. The traditional way is continuous charge-discharge testing without interruption, which often takes one year or even longer. Therefore, this paper proposes a rapid cycle life test method based on intelligent prediction to replace the continuous test, which shortens the test period and accelerates product replacement. The original capacity data is decoupled into the short-term regeneration trajectory and the long-term degradation trajectory, which are predicted by the long-short term memory model optimized by swarm intelligent algorithm. The data expansion technique based on Monte Carlo sampling is used to increase the diversity of training data to improve the prediction accuracy. The feasibility and effectiveness are proved by NASA data sets. The results show that the cycle life test time reduced by at least 90% with the error less than 3 cycles.

Enhanced Whale Optimization Algorithm with Wavelet Decomposition for Lithium Battery Health Estimation in Deep Extreme Learning Machines

Lithium battery health state estimation can help optimize battery usage and management strategies. In response to the challenges faced by traditional battery management systems in accurately estimating the State of Health of lithium-ion batteries and addressing issues such as capacity recovery and noise interference, this paper proposes a method based on wavelet decomposition and an improved whale optimization algorithm optimized deep extreme learning machine for estimating the SOH of lithium-ion batteries. Firstly, the lithium-ion battery capacity degradation sequence is extracted, and the wavelet decomposition method is used to decompose the battery capacity into global and local degradation trends. Next, the non-linear convergence factor and the whale optimization algorithm with adaptive weights are employed to optimize the deep extreme learning machine for predicting each trend component. Finally, the prediction results are effectively integrated to obtain the lithium-ion battery SOH. This experimental method is validated using NASA and CALCE datasets, and the results indicate that the root mean square error and mean absolute percentage error are both below 0.95%, with relative accuracy and absolute correlation coefficients exceeding 98%. This demonstrates the method’s excellent accuracy and robustness.

Automatic software bug prediction using adaptive golden eagle optimizer with deep learning

In the software maintenance and development process, the software bug detection is an essential problem because it related with the complete software successes. So, the earlier software bug detection is essential to enhance the software efficiency, reliability, software quality and software cost. Moreover, the efficient software bug prediction is a critical as well as challenging operation. Hence, the efficient software bug prediction model is developed in this article. To achieve this objective, optimized long short-term memory is developed. The important stages of the proposed model is preprocessing, feature selection and bug detection. At first the input bug dataset is preprocessed. In preprocessing, the duplicate data instances are removed from the dataset. After the preprocessing, the feature selection is done by Adaptive Golden Eagle Optimizer (AGEO). Here the traditional GEO algorithm is altered by means of opposition-based learning (OBL). Finally, the proposed approach utilizes a long short-term memory (LSTM) based recurrent neural network (RNN) for bug prediction. Long Short-Term Memory (LSTM) network is a type of recurrent neural network. The promise and NASA dataset are considered as the input for bug prediction. the performance of proposed approach is analysed based on various metrics namely, accuracy, F- measure, G-measure and Matthews Correlation Coefficient (MCC).

State of health estimation of lithium-ion battery using energy accumulation-based feature extraction and improved relevance vector regression

The state of health (SOH) of lithium-ion battery is an important component of intelligent battery management system. Precise SOH estimation provides feasible and safe guidance for the energy system driven by lithium-ion battery. A novel SOH estimation method using energy accumulation of equal discharge voltage difference (EAEDVD) features extraction and improved relevance vector regression (RVR) is developed. Firstly, the EAEDVD based health feature is extracted from the discharge process as the parameters to describe battery aging, which considers the limitation of incomplete discharge greatly hindering the possibility of extracting traditional aging feature from the whole cycle. Then, the health features with high correlation are selected via integrated correlation analysis from EAEDVD curve smoothed by filter method, which eliminate redundant information. Second, this paper constructs a RVR model to estimate SOH, aiming at the problem of RVR model parameter selection, sparrow search algorithm (SSA) is developed to optimize the model parameters for improving the performance via finding the optimal solution. Finally, the feasibility is verified based on two battery datasets from NASA and laboratory. For the four batteries in the NASA dataset, the error indicators are all within 1 %, and for the two batteries from the laboratory, the mean error indicators are 0.75 %, 1.05 % and 1.05 % respectively, which indicates that proposed method has high accuracy, strong robustness and applicability.

Remaining Useful Life Prediction for Lithium-Ion Batteries With a Hybrid Model Based on TCN-GRU-DNN and Dual Attention Mechanism

The instability of lithium-ion batteries may result in system operation failure and cause safety accidents, thus predicting the remaining useful life (RUL) accurately is helpful for reducing the risk of battery failure and extending its useful life. In this paper, a hybrid model based on TCN-GRU-DNN and dual attention mechanism is proposed for enhancing the RUL prediction accuracy of lithium-ion batteries. Firstly, the Temporal Convolutional Network (TCN) with a feature attention mechanism is applied to form an encoder module to capture the battery capacity regeneration phenomenon, and then a Gated Recurrent Unit (GRU) with a temporal attention mechanism is denoted as a decoder module for better characterizing the decay trend of the capacity series. Finally, the final prediction results are output through a Deep Neural Network (DNN). We conducted experiments on two data sets NASA and CALCE. The prediction errors are presented in subsequent experiments under different evaluation standards such as absolute error (AE), root mean square error (RMSE), mean absolute error (MAE), and R-squared error (R²). The experimental results demonstrate that the proposed model can achieve a more accurate prediction for RUL on lithium-ion batteries, with RMSE does not exceed 2.407% in the NASA dataset and does not exceed 0.897% in the CALCE lithium-ion battery dataset respectively.

Automatic Software Bug Prediction Using Adaptive Artificial Jelly Optimization With Long Short-Term Memory

AbstractIn the software maintenance and development process, software bug detection is an essential problem because it is related to complete software success. It is recommended to begin anticipating defects at the early stages of creation rather than during the assessment process due to the high expense of fixing the found bugs. The early stage software bug detection is used to enhance software efficiency, reliability, and software quality. Nevertheless, creating a reliable bug-forecasting system is a difficult challenge. Therefore, in this paper, an efficient, software bug forecast is developed. The presented technique consists of three stages namely, pre-processing, feature selection, and bug prediction. At first, the input datasets are pre-processed to eliminate the identical data from the dataset. After the pre-processing, the important features are selected using an adaptive artificial jelly optimization algorithm (A2JO) to eliminate the possibility of overfitting and reduce the complexity. Finally, the selected features are given to the long short-term memory (LSTM) classifier to predict whether the given data is defective or non-defective. In this paper, investigations are shown on visibly obtainable bug prediction datasets namely, promise and NASA which is a repository for most open-source software. The efficiency of the presented approach is discussed based on various metrics namely, accuracy, F- measure, G-measure, and Matthews Correlation Coefficient (MCC). The experimental result shows our proposed method achieved the extreme accuracy of 93.41% for the Promise dataset and 92.8% for the NASA dataset.

An efficient state-of-health estimation method for lithium-ion batteries based on feature-importance ranking strategy and PSO-GRNN algorithm

As a critical index in ensuring the safe and reliable deployment of lithium-ion batteries (LIBs), the estimation performance of state-of-health (SOH) seriously affects the popularization of electric vehicles (EVs). However, for the existing SOH estimation methods based on machine learning algorithms, the health features are manually extracted mainly based on personal experience, without algorithm-based automatic preferential selection. Thus, this study proposes a battery SOH estimation method based on the feature-importance ranking strategy and generalized regression neural network optimized by particle swarm optimization (PSO-GRNN). Firstly, 19 vital features were extracted from charging voltage, charging current, temperature, and incremental capacity (IC) curves. Secondly, the importance indexes for these features were calculated and optimally ranked by fusing the machine learning algorithm (random forest) and statistical analysis (Pearson correlation coefficient). Accordingly, different combinations of high-ranking features were compared to determine the best number of the final feature set. Thirdly, the SOH estimation model is built by the PSO-GRNN algorithm with optimal 7 high-ranking features. Based on the publicly available NASA datasets, the proposed SOH estimation model was validated on the data from a single-cell battery. Furthermore, the datasets from three cells with different cycling test conditions were utilized, where the data from two cells and the other cell were alternately selected as the training and test sets, respectively. The SOH estimation errors (squared absolute error and mean squared error) for single-cell and multi-cell experiments were less than 1 % and 3 %, respectively. Experimental results show that the proposed method is robust and can achieve accurate and lightweight SOH estimation.

State of health confidence estimation for lithium-ion battery based on probabilistic ensemble learning

Uncertainties in a battery would result in unreliable state of health (SOH) estimation. Considering the greater risk after reaching the end of life (EOL), designing a suitable ensemble learning to provide early warning before reaching EOL with uncertainty measurement is desirable for confidence estimation. In this paper, a novel probabilistic ensemble learning method-Gaussian process-based neural networks is proposed for the SOH confidence estimation by describing the uncertainties in probabilistic form. First, different neural networks are built based on health features. Second, battery data are classified under the recovery of capacity and normal operation conditions to characterize the uncertainties of the data under different operation conditions. Besides, the Gaussian process-based neural networks method is constructed based on the data from different conditions for neural networks weighted ensemble with the probabilistic form of Gaussian distribution. Therefore, the uncertainties are measured in the probabilistic form considering different operation conditions which is different from other methods. With the probabilistic form, the confidence interval could be determined to ensure the real SOH within the confidence interval, which improves the estimation performance of the proposed method because of the early warning near the EOL. Finally, the effectiveness is validated by NASA data sets and our experiment with the commercial 18650 lithium-ion battery. From the results, the mean error is less than 1% and real SOH is within the confidence interval.