State of Charge Estimation for Lithium-Ion Batteries Based on TCN-LSTM Neural Networks

Hu, Chunsheng; Cheng, Fangjuan; Ma, Lvyi; Li, Bohao

doi:10.1149/1945-7111/ac5cf2

Cited by 43 publications

(11 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Experimental findings revealed that the proposed TCN-based SOH estimation model exhibited high accuracy in estimation and demonstrated good adaptability across different battery types. 18 Wen proposed a model based on incremental capacity analysis and BP neural network to predict battery SOH at different ambient temperatures. By analyzing the correlation between IC curve characteristics and SOH, the mapping relationship between temperature and IC curve characteristics was established by the least square method, and the SOH prediction model at different temperatures was obtained.…”

Section: Introductionmentioning

confidence: 99%

SOH estimation and RUL prediction of lithium batteries based on multidomain feature fusion and CatBoost model

Zhang,

Yin,

Chen

2023

Energy Science & Engineering

View full text Add to dashboard Cite

In this paper, a lithium‐ion battery State of Health (SOH) estimation algorithm is proposed based on the fusion of multidomain features and the application of a CatBoost model. The aim is to address the issue of low prediction accuracy in SOH caused by the utilization of single‐feature extraction techniques. The algorithm encompasses the extraction of various features from the original charge–discharge data, including time‐domain, frequency‐domain, entropy, and time‐series features. Following the evaluation of feature importance, a feature selection process is conducted to eliminate redundant features that provide a limited contribution to the predictive results. Subsequently, a multiple‐set discriminative correlation analysis is employed to integrate high‐dimensional features. To attain accurate predictions, the CatBoost model is further optimized through the utilization of a sparrow search algorithm. Experimental results demonstrate that the proposed algorithm achieves accurate SOH estimations within individual batteries, as evidenced by mean square error values consistently below 4e−4 and goodness‐of‐fit values exceeding or equal to 0.98. Additionally, the algorithm exhibits reliable prediction capabilities across different batteries operating under the same charge/discharge strategy. Comparative analysis indicates that the adoption of the multidomain feature fusion approach yields improved prediction accuracy in contrast to the utilization of a single feature extraction method.

show abstract

Section: Introductionmentioning

confidence: 99%

SOH estimation and RUL prediction of lithium batteries based on multidomain feature fusion and CatBoost model

Zhang,

Yin,

Chen

2023

Energy Science & Engineering

View full text Add to dashboard Cite

show abstract

“…Using a Genetic Algorithm (GA) to search for the smoothing time interval for the optimal ERTSS, various dynamic unit tests were performed. Reference (Hu et al, 2022) extracts higher-level spatial features between multi variables into the current SOC and historical input, to achieve state assessment. However, the above-mentioned methods have low accuracy and large errors, so they cannot be accurately evaluated.…”

Section: Introductionmentioning

confidence: 99%

State-of-charge estimation method for lithium-ion batteries based on competitive SIR model

2022

Front. Energy Res.

View full text Add to dashboard Cite

In order to meet the needs of life and production and ensure the battery is stable when using the battery, a scheme for reckoning the state of charge of lithium-ion batteries derived from the competitive SIR model is proposed. During the charging process of the battery, the electrolyte and the diaphragm reach the negative electrode of the battery, and the electrolyte escapes from the graphite of the negative electrode to the positive electrode in the case of discharge. The analysis shows that the SIR model belongs to the internal information evolution process, which can infect the surrounding data and evaluate the state of charge better. Through experiments, it is substantiated that the scheme is able to better estimate the state of lithium-ion batteries, the error value is 0.0189, the accuracy is good, and the battery usage can be predicted in time.

show abstract

“…during battery operation as inputs. [18][19][20][21][22] Ref. [23] proposed an improved feedforward-long short-term memory (FF-LSTM) modeling method to realize an accurate whole-life-cycle SOC prediction by effectively considering the current, voltage, and temperature variations.…”

mentioning

confidence: 99%

Battery Modeling Considering Hysteresis Effect and Temperature Adaptability

et al. 2023

View full text Add to dashboard Cite

With the increasing severity of energy crisis and the gradual increase of environmental awareness, countries around the world have devoted themselves to the new energy vehicle industry. [1] As the foundation and core component of electric vehicles, the power battery pack and its management system play a decisive role in the performance and development of electric vehicles. The power battery pack often uses lithium-ion batteries with higher energy density, lower self-discharge rate, and long cycle life. [2][3][4][5][6] Accurate stateof-charge (SOC) estimation is an important factor to control the stable operation of batteries, which directly affects the life and safety of the battery. [7][8][9] At present, SOC estimation can be roughly divided into two categories: principle-based and model-based methods. [10,11] The modelbased SOC estimation method is widely used because of its high robustness and accuracy under variable battery operating conditions. [12,13] Battery models, as an important part of model-based SOC estimation methods, commonly include electrochemical models, blackbox models and equivalent circuit models (ECMs). The electrochemical model describes the diffusion process and charge transfer process of lithium ions in the cell based on porous electrodes and concentrated solution theory and uses a set of coupled partial differential equations to achieve real-time estimation of SOC. [14][15][16] Ref.[17] designed a nonlinear observer with terminal-voltage feedback injection based on the electrochemical single-particle model, which improved the accuracy of SOC estimation. Although the electrochemical model can reflect the actual electrochemical reaction process inside the battery more accurately, the structure is complicated and not conducive to practical simulation and calculation. The blackbox model establishes an abstract mapping relationship between inputs and outputs, reflecting a generalized direct causal relationship between the factors involved. Real-time estimation of SOC can be achieved by taking real-time state quantities (e.g., voltage, current, etc.) during battery operation as inputs. [18][19][20][21][22] Ref. [23] proposed an improved feedforward-long short-term memory (FF-LSTM) modeling method to realize an accurate whole-life-cycle SOC prediction by effectively considering the current, voltage, and temperature variations. Although the calculation of the blackbox model is simple, the accuracy of the estimation depends to a large extent on the quality and quantity of the training data, and the training process may take a long time. From the perspective of external electrical characteristics, ECM uses circuit elements such as resistance, capacitance, and voltage source to form a circuit network to express the relationship between voltage and current in the cell. [24] It cleverly avoids the real structure and complex electrochemical reactions inside the battery and can estimate SOC in real-time by combining it with the filtering algorithm. [25][26][27][28][29] Ref. [30] proposed a novel feedback

show abstract

State of Charge Estimation for Lithium-Ion Batteries Based on TCN-LSTM Neural Networks

Cited by 43 publications

References 59 publications

SOH estimation and RUL prediction of lithium batteries based on multidomain feature fusion and CatBoost model

SOH estimation and RUL prediction of lithium batteries based on multidomain feature fusion and CatBoost model

State-of-charge estimation method for lithium-ion batteries based on competitive SIR model

Battery Modeling Considering Hysteresis Effect and Temperature Adaptability

Contact Info

Product

Resources

About