An Immune Genetic Extended Kalman Particle Filter approach on state of charge estimation for lithium-ion battery

Intl J of Energy Research

Fan

et al. 2022

Accurate state of charge (SOC) for the lithium-ion battery is not only related to user experience but also the top target to avoid overcharge and overdischarge and to use it safely. The back propagation (BP) neural network is widely used in SOC estimation, but there exist some issues, such as easily falling local extreme value, converging slowly, or even unable to converge and even overfitting. The Drosophila algorithm has a simple algorithm and strong global optimization ability, but there is also a problem of direct inheritance to reduce the optimization ability. To solve these problems, an individual migration dynamic step Drosophila (Improved Drosophila) algorithm combined with the BP neural network is proposed to estimate the SOC of lithium-ion batteries and improve estimation accuracy. In addition, the performance of the proposed method is compared with that of its traditional algorithms and other commonly used functions. The experiments are carried out to verify the ternary lithium-ion battery under DST and BBDST conditions., the mean absolute error is less than 0.8%, and the root mean square error is less than 1.4%. The SOC estimation is carried out when the current data under the DST condition are missing, which also has good estimation performance, which shows the robustness of the algorithm. Compared to other algorithms, there is good estimation accuracy.

Section: Introductionmentioning

confidence: 99%

A novel Drosophila‐back propagation method for the lithium‐ion battery state of charge estimation adaptive to complex working conditions

Intl J of Energy Research

Fan

et al. 2022

“…17 When Kalman filter cannot solve the nonlinear problem, extended Kalman filter algorithm based on Kalman filter algorithm to solve the nonlinear system, 18 by linearizing the nonlinear state space equation, Kalman filter algorithm to achieve state of charge estimation. 19 The authors propose a cubic Kalman filtering algorithm combining fuzzy adaptive and singular value decomposition in order to solve the problem of slow convergence time of cubic Kalman algorithm. 20 Four kind of algorithms based on Kalman filter extension are introduced in detail (extended Kalman filter, unscented Kalman filter, cubature Kalman filter, and ensemble Kalman filter), the advantages and disadvantages, estimation accuracy and anti-interference ability of these four algorithms were compared in detail.…”

Section: Introductionmentioning

confidence: 99%

“…However, Kalman filter algorithm can only solve linear problems, and the estimation of state of charge is usually a nonlinear process, so it needs to linearize the nonlinear system 17 . When Kalman filter cannot solve the nonlinear problem, extended Kalman filter algorithm based on Kalman filter algorithm to solve the nonlinear system, 18 by linearizing the nonlinear state space equation, Kalman filter algorithm to achieve state of charge estimation 19 . The authors propose a cubic Kalman filtering algorithm combining fuzzy adaptive and singular value decomposition in order to solve the problem of slow convergence time of cubic Kalman algorithm 20 …”

Section: Introductionmentioning

confidence: 99%

Collaborative state estimation of lithium‐ion battery based on multi‐time scale low‐pass filter forgetting factor recursive least squares ‐ double extended Kalman filtering algorithm

Long

Cao

et al. 2022

Circuit Theory & Apps

For the lithium battery management system and real-time safety monitoring, two issues are of great significance, namely, the ability to accurately update the model parameters in real time and to accurately estimate the state of charge and health. In this context, this thesis adopts the second-order RC equivalent circuit model and the forgetting factor recursive least squares -double extended Kalman filtering (FFRLS-DEKF) algorithm with multi-time scales and low-pass filter. Forgetting factor recursive least squares is applied to conduct online parameter identification, and the traditional double extended Kalman filtering algorithm is optimized to evaluate the state of charge and model parameters in the micro-scale and macro-scale. In this way, the error caused by two different characteristics is reduced, and a low-pass filter is added to optimize the fluctuation problem of the estimated value of the model parameters. According to the experiment results, the maximum error between the model simulation value and the actual value of the terminal voltage is 0.0459 V. If the initial value of the state of charge deviates from the actual value, the maximum errors under BBDST and HPPC conditions record 0.0235 and 0.0048, respectively, the forgetting factor recursive least squares -double extended Kalman filtering algorithm with multi-time scales and low-pass filter is able to track the true value within 40 s. Furthermore, the lithium-ion battery state of health both reaches 98% under the two conditions. In summary, the experimental analysis shows that the algorithm helps reduce the influence of initial values on the results, thereby reducing error accumulation and improving the robustness.

“…The scaling approach is realized for improved SOC estimation as one of the most significant factors for performance optimization. [22][23][24][25] The SOC estimation is conducted with nonelectrical parameters and the uniform fiber Bragg grating (FBG). The estimation methods are used to determine the critical battery state and polynomial augmented model construction, including the immune genetic extended Kalman filtering, particle filtering, low-frequency EIS, double adaptive extended Kalman filter, adaptive H-infinity filter, and adaptive correction-unscented Kalman filter.…”

Section: Introductionmentioning

confidence: 99%

Improved Compound Correction‐Electrical Equivalent Circuit Modeling and Double Transform‐Unscented Kalman Filtering for the High‐Accuracy Closed‐Circuit voltage and State‐of‐Charge Co‐Estimation of Whole‐Life‐Cycle Lithium‐Ion batteries

Takyi‐Aninakwa

et al. 2022

Energy Tech

For complex energy storage conditions, it is necessary to monitor the state‐of‐charge (SOC) and closed‐circuit voltage (CCV) status accurately for the reliable power supply application of lithium‐ion batteries. Herein, an improved compound correction‐electrical equivalent circuit modeling (CC‐EECM) method is proposed by considering the influencing effects of ambient temperature and charge–discharge current rate variations to estimate the CCV. Then, an improved adaptive double transform‐unscented Kalman filtering (ADT‐UKF) method is constructed with recursive sampling data correction to estimate the nonlinear SOC. A dynamic window function filtering strategy is constructed to obtain the new sigma point set for the online weighting coefficient correction. For a temperature range of 5–45 °C, the CCV for the improved CC‐EECM responds well with a maximum error of 0.008608 V, and the maximum SOC estimation error is 6.317%. The proposed ADT‐UKF method improves the CCV and SOC estimation reliability and adaptability to the time‐varying current rate, temperature, and aging factors.