Figure - available from: Nature Communications
This content is subject to copyright. Terms and conditions apply.
An economic evaluation of retired battery recycling a Comparison of the Pyro- (pyrometallurgical), Hydro-(hydrometallurgical), and ML-direct (machine learning aided direct) recycling methods. b Cost analysis of Lithium Iron Phosphate (LFP) and Nickel Manganese Cobalt Oxide (NMC) batteries using different recycling methods in individual modes. c Cost analysis of LFP and NMC batteries using ML-direct recycling in individual mode. d Cost, revenue, and profit comparison of the individual battery type using different recycling methods in individual mode. e Cost, revenue, and profit comparison using Wasserstein distance voting (WDV), majority voting (MV), and independent learning (IL) methods in hybrid mode. The ratio is the amount of LFP battery to that of NMC battery. f Sensitivity analysis of the profit of WDV, MV, and IL methods in a hybrid model towards sorting accuracy in hybrid mode. The ratio is the amount of LFP to that of the NMC battery. g Comprehensive comparison of different battery recycling technologies in hybrid mode. Source data are provided as a Source Data file. The graphics in panel a were created using icons from Flaticon.com.
Source publication
Unsorted retired batteries with varied cathode materials hinder the adoption of direct recycling due to their cathode-specific nature. The surge in retired batteries necessitates precise sorting for effective direct recycling, but challenges arise from varying operational histories, diverse manufacturers, and data privacy concerns of recycling coll...
Similar publications
Privacy by design is nowadays recognized as essential in bringing data privacy into software systems. However, developers still face many challenges in reconciling privacy and software requirements and implementing privacy protections in software systems. One emerging trend is the adoption of microservices architectures—they bring in some qualities...
Citations
... Ineffective management of this surplus could lead to significant resource wastage and environmental harm. Therefore, the development of advanced battery recycling technologies is crucial 3 . The imperative for recycling is further underscored by the necessity of reclaiming valuable materials. ...
In recent years, deep learning techniques have been extensively used for the identification and classification of lithium-ion batteries. However, these models typically require a costly and labor-intensive labeling process, often influenced by commercial or proprietary concerns. In this study, we introduce RecyBat24, a publicly accessible image dataset for the detection and classification of three battery types: Pouch, Prismatic, and Cylindrical. Our dataset is designed to support both academic research and industrial applications, closely replicating real-world scenarios during the acquisition process and employing data augmentation techniques to simulate various external conditions. Additionally, we demonstrate how the RecyBat24’s detection-oriented annotations can be used to create a second version of RecyBat24for instance-segmentation tasks. Finally, we demonstrate that recent lightweight machine learning models achieve high accuracy, highlighting their potential for classification and segmentation applications where computational resources are constrained.
... Numerous battery diagnostic and prognostic methods have been proposed [10][11][12][13][14][15], including model-based methods, data-driven methods, and hybrid methods. Recently, thanks to the rapid development of artificial intelligence [16,17] and the open source of numerous battery datasets [18][19][20][21][22], data-driven methods combining machine learning (ML) algorithms and feature engineering have achieved superior performance in battery diagnosis and prognosis, showing great application potential in battery material development [23], manufacturing processes improvement [24,25], aging mechanism revelation [26], fast-charging protocol selection [27], management strategy optimization [28,29], recycling [30,31], etc. However, most data-driven methods and their feature engineering rely on laboratory data at specific stages under specific conditions, such as charge/discharge data [32,33], impedance data [34,35], and relaxation data [36]. ...
... Note that, to improve the prediction accuracy of 1 and 2 , we first divided the field SOC into multiple independent intervals. Specifically, the value of is divided into ten independent intervals, namely Specifically, in Dataset 1, according to the numerical range of 1 corresponding to 1 =10 Hz in the training set, the value of 1 is divided into four independent intervals, namely [14,18), [18,22), [22,26), and [26,30]. Therefore, a total of forty ML models are trained for the prediction of 1 corresponding to 1 =10 Hz. ...
... Therefore, a total of forty ML models are trained for the prediction of 1 corresponding to 1 =10 Hz. According to the numerical range of 2 corresponding to 2 =312.5 Hz in the training set, the value of 2 is divided into four independent intervals, namely [14,18), [18,22), [22,26), and [26,30]. Therefore, a total of forty ML models are trained for the prediction of 2 corresponding to 2 =312.5 Hz. ...
Aiming at the dilemma that most laboratory data-driven diagnostic and prognostic methods cannot be applied to field batteries in passenger cars and energy storage systems, this paper proposes a method to bridge field data and laboratory data using machine learning. Only two field real impedances corresponding to a medium frequency and a high frequency are needed to predict laboratory real impedance curve, laboratory charge/discharge curve, and laboratory relaxation curve. Based on the predicted laboratory data, laboratory data-driven methods can be used for field battery diagnosis and prognosis. Compared with the field data-driven methods based on massive historical field data, the proposed method has the advantages of higher accuracy, lower cost, faster speed, readily available, and no use of private data. The proposed method is tested using two open-source datasets containing 249 NMC cells. For a test set containing 76 cells, the mean absolute percentage errors of laboratory real impedance curve, charge curve, and discharge curve prediction results are 0.85%, 4.72%, and 2.69%, respectively. This work fills the gap between laboratory data-driven diagnostic and prognostic methods and field battery applications, making all laboratory data-driven methods applicable to field battery diagnosis and prognosis. Furthermore, this work overturns the fixed path of developing field battery diagnostic and prognostic methods based on massive field historical data, opening up new research and breakthrough directions for field battery diagnosis and prognosis.
... As the most expensive part of the vehicle, the battery pack plays a key role in performance, range, and on-the-road costs. Therefore, precise estimation of the capacity of the battery is vital for energy management optimization, enhanced longevity, and reliability assurance within the EV [1][2][3]. One of the critical aspects of EV management is the battery capacity estimation, especially under load -this greatly impacts the range, efficiency and performance of the vehicle as a whole. ...
... New ML methods are emerging as strong tools to model these complexities. Dealing with this issue has also been tackled in several other works, such as using ANN, RF and recurrent architectures like LSTM [1,4]. ANNs were inspired by a combination of computational frameworks based on how human brains are structured. ...
... Lithium-ion batteries (LIBs), due to their high energy density, long cycle life, and environmental friendliness, are widely used in EVs power batteries and energy storage applications [6][7][8][9]. The vast market has accelerated the technological development of LIBs, posing stricter demands on their efficiency, health, safety, and environmental friendliness [10][11][12]. ...
The accurate state of health (SOH) estimation of lithium-ion batteries is crucial for efficient, healthy, and safe operation of battery systems. Extracting meaningful aging information from highly stochastic and noisy data segments while designing SOH estimation algorithms that efficiently handle the large-scale computational demands of cloud-based battery management systems presents a substantial challenge. In this work, we propose a quantum convolutional neural network (QCNN) model designed for accurate, robust, and generalizable SOH estimation with minimal data and parameter requirements and is compatible with quantum computing cloud platforms in the Noisy Intermediate-Scale Quantum. First, we utilize data from 4 datasets comprising 272 cells, covering 5 chemical compositions, 4 rated parameters, and 73 operating conditions. We design 5 voltage windows as small as 0.3 V for each cell from incremental capacity peaks for stochastic SOH estimation scenarios generation. We extract 3 effective health indicators (HIs) sequences and develop an automated feature fusion method using quantum rotation gate encoding, achieving an R2 of 96%. Subsequently, we design a QCNN whose convolutional layer, constructed with variational quantum circuits, comprises merely 39 parameters. Additionally, we explore the impact of training set size, using strategies, and battery materials on the model’s accuracy. Finally, the QCNN with quantum convolutional layers reduces root mean squared error by 28% and achieves an R2 exceeding 96% compared to other three commonly used algorithms. This work demonstrates the effectiveness of quantum encoding for automated feature fusion of HIs extracted from limited discharge data. It highlights the potential of QCNN in improving the accuracy, robustness, and generalization of SOH estimation while dealing with stochastic and noisy data with few parameters and simple structure. It also suggests a new paradigm for leveraging quantum computational power in SOH estimation.
... Rechargeable batteries are ubiquitous in modern industry, including electric vehicles, power grids, and portable devices [12,24,36,38]. Nevertheless, batteries inevitably degrade with cyclic operation due to intrinsic electrochemical mechanisms [4,7,34]. ...
Battery Life Prediction (BLP), which relies on time series data produced by battery degradation tests, is crucial for battery utilization, optimization, and production. Despite impressive advancements, this research area faces three key challenges. Firstly, the limited size of existing datasets impedes insights into modern battery life data. Secondly, most datasets are restricted to small-capacity lithium-ion batteries tested under a narrow range of diversity in labs, raising concerns about the generalizability of findings. Thirdly, inconsistent and limited benchmarks across studies obscure the effectiveness of baselines and leave it unclear if models popular in other time series fields are effective for BLP. To address these challenges, we propose BatteryLife, a comprehensive dataset and benchmark for BLP. BatteryLife integrates 16 datasets, offering a 2.4 times sample size compared to the previous largest dataset, and provides the most diverse battery life resource with batteries from 8 formats, 80 chemical systems, 12 operating temperatures, and 646 charge/discharge protocols, including both laboratory and industrial tests. Notably, BatteryLife is the first to release battery life datasets of zinc-ion batteries, sodium-ion batteries, and industry-tested large-capacity lithium-ion batteries. With the comprehensive dataset, we revisit the effectiveness of baselines popular in this and other time series fields. Furthermore, we propose CyclePatch, a plug-in technique that can be employed in a series of neural networks. Extensive benchmarking of 18 methods reveals that models popular in other time series fields can be unsuitable for BLP, and CyclePatch consistently improves model performance establishing state-of-the-art benchmarks. Moreover, BatteryLife evaluates model performance across aging conditions and domains. BatteryLife is available at https://github.com/Ruifeng-Tan/BatteryLife.
... For instance, in the context of the LIB industry, data related to battery production, including the usage of key materials and consumer usage patterns, can offer valuable information on potential safety risks to participants in the reverse supply chain, such as collectors, remanufacturers, and recyclers of processing used batteries [15]. Integrating data from various stakeholders for training a Gen AI model shared by multiple supply chain entities can help organizations to manage safety more efficiently. ...
Circular supply chains (CSCs) aim to minimize ecological footprints by closing the loop on resource use. However, safety concerns in CSCs, especially in activities like recycling and remanufacturing, present significant barriers. Current safety management systems in CSCs are often reactive and limited in scope, failing to address safety concerns in inter-organizational complexities. Therefore, we explore how generative artificial intelligence (Gen AI) can revolutionize safety management in CSCs by providing proactive monitoring, predictive analysis, and enhanced training. Nevertheless, integrating this technology requires addressing technical, data security, operational, trust, and ethical issues. Future research and practices should focus on overcoming these barriers to harness Gen AI's potential for enhancing safety in CSCs.
... [20][21][22][23][24][25][26][27] Manufacturers find the available signals, such as current, voltage, and capacity, integrated with machine learning, promising for predicting battery lifetime trajectories and internal states under diverse conditions. [28][29][30][31][32][33][34][35] However, these models struggle to predict intermediate degradations, with lifetime trajectory prediction posing a greater challenge due to the need for more sensory data that has not yet been established at the time of prediction. [36][37][38] Despite the extrapolation abilities, the statistical model still relies on the availability of long-term observational data, where necessary data collecting time conflicts with the goals of reducing verification time and costs. ...
The paper proposes a physics-informed model to predict battery lifetime trajectories by computing thermodynamic and kinetic parameters, saving costly data that has not been established for sustainable manufacturing, reuse, and recycling.
... Recycling alternatively uses residual values of retired batteries by materials extraction or structural repair 26,27 , suitable for irreversibly degraded retired batteries. Compared with pyrometallurgy and hydrometallurgy 28,29 , direct recycling stands out for superior profitability, lower energy consumption, and reduced carbon emissions 26,27,30 . Leveraging lithium replenishment and postprocessing of cathode materials, direct recycling achieves efficient material structure repair, and performance restoration 31 . ...
... Thus, it is challenging to retrieve fieldavailable SOH data. The lifetime data integrity remains a major challenge, calling for the SOH estimation only using field data, opposite to historical data or under controllable conditions 30,40,41 . One solution is to perform a capacity calibration test at the retired battery collection field, i.e., state of charge (SOC) conditioning, which is straightforward, however, it requires unaffordable test time and extra electricity costs. ...
... It is noted that the astrained generative learning model is used to generate more data samples, especially under, but not limited to, unseen retirement conditions. For instance, Case3 uses the physically tested pulse voltage response data at 5%, 25%, and 50% SOCs to generate data at 10,15,20,30,35,40, and 45% SOCs. All interpolation cases exhibit a low MAPE error below 2%, even if the model is never trained with such data. ...
Rapid and accurate state of health (SOH) estimation of retired batteries is a crucial pretreatment for reuse and recycling. However, data-driven methods require exhaustive data curation under random SOH and state of charge (SOC) retirement conditions. Here, we show that the generative learning-assisted SOH estimation is promising in alleviating data scarcity and heterogeneity challenges, validated through a pulse injection dataset of 2700 retired lithium-ion battery samples, covering 3 cathode material types, 3 physical formats, 4 capacity designs, and 4 historical usages with 10 SOC levels. Using generated data, a regressor realizes accurate SOH estimations, with mean absolute percentage errors below 6% under unseen SOC. We predict that assuming uniform deployment of the proposed technique, this would save 4.9 billion USD in electricity costs and 35.8 billion kg CO2 emissions by mitigating data curation costs for a 2030 worldwide battery retirement scenario. This paper highlights exploiting limited data for exploring extended data space using generative methods, given data can be time-consuming, expensive, and polluting to retrieve for many estimation and predictive tasks.
... Zhu et al. 18 offered a comprehensive overview of second-life batteries (SLBs), highlighting the need for comprehensive evaluations of the remaining service life for proper performance assessment. Recent research has focused on improving inconsistency detection 21,22 , state of health (SOH) estimation 23,24 , and cathode material sorting 25 to allocate these batteries appropriately for reuse and recycling applications. Many studies have investigated the application of ESSs to retired batteries for solar and wind power generation, primarily by examining environmental 13,26 , economic 27 , and comprehensive sustainability assessments [28][29][30] . ...
Reuse and recycling of retired electric vehicle (EV) batteries offer a sustainable waste management approach but face decision-making challenges. Based on the process-based life cycle assessment method, we present a strategy to optimize pathways of retired battery treatments economically and environmentally. The strategy is applied to various reuse scenarios with capacity configurations, including energy storage systems, communication base stations, and low-speed vehicles. Hydrometallurgical, pyrometallurgical, and direct recycling considering battery residual values are evaluated at the end-of-life stage. For the optimized pathway, lithium iron phosphate (LFP) batteries improve profits by 58% and reduce emissions by 18% compared to hydrometallurgical recycling without reuse. Lithium nickel manganese cobalt oxide (NMC) batteries boost profit by 19% and reduce emissions by 18%. Despite NMC batteries exhibiting higher immediate recycling returns, LFP batteries provide superior long-term benefits through reuse before recycling. Our strategy features an accessible evaluation framework for pinpointing optimal pathways of retired EV batteries.
... This ranking indicates that high SOC states and higher average driving speed are important factors accelerating battery aging, providing profound insights for mitigating the degradation of electric vehicle batteries. Utilizing feature importance rankings based on the RF algorithm, Tao et al. [60] successfully identified useful mechanistic features to ensure sorting accuracies of battery electrode materials. Furthermore, numerous studies have considered RFR-based SOH estimation methods as powerful baselines [7,[61][62][63][64][65][66][67][68], highlighting the immense appeal of RFR as a typical ensemble learning algorithm for battery aging estimation. ...
... These local models are aggregated into a global model through ensemble learning to make better predictions. Recently, Tao et al. [60] proposed an innovative Wasserstein-distance voting strategy to aggregate local models, achieving accurate classification of heterogeneous retired batteries. The success of this study is attributed to federated learning-based multi-source data sharing and ensemble learning-based fusion of multiple local models. ...
