Loading

The Estimation of Battery State of Charge using Corny Network
Ismail1, Firdaus2, Rakiman3, Daddy Budiman4, Sardani5

1Ismail, Department of Electrical Engineering Department, Politeknik Negeri Padang, Padang, Indonesia.

2Firdaus, Department of Electrical Engineering Department, Negeri Padang, Padang, Padang, Indonesia.

3Rakiman, Department of Mechanical Engineering, Politeknik Negeri Padang, Padang, Indonesia.

4Daddy Budiman, Department of Mechanical Engineering, Politeknik Negeri Padang, Padang, Indonesia. 

5Sardani, Department of Electrical Engineering, Politeknik Negeri Padang, Padang, Indonesia.

Manuscript received on 09 January 2024 | Revised Manuscript received on 19 January 2024 | Manuscript Accepted on 15 March 2024 | Manuscript published on 30 March 2024 | PP: 5-11 | Volume-12 Issue-6, March 2024 | Retrieval Number: 100.1/ijrte.F799912060324 | DOI: 10.35940/ijrte.F7999.12060324

Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS |  Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: State of charge (SOC) estimation of lithium-ion batteries has been extensively studied, and the estimation accuracy was mainly investigated through the development of various battery models and dynamic estimation algorithms. All battery models, however, contain inherent model bias due to the simplifications and assumptions, which cannot be effectively addressed through the development of various conventional and intelligent computational methods. Consequently, some existing techniques for estimating battery state of charge (SOC) have not been very accurate in predicting battery SOC characteristics, using both traditional and intelligent computation methods. There are some drawbacks to employing deep learning to estimate SOC battery, such as the use of complicated algorithms or networks, overfitting, and others. The proposed method, the Corny architecture, has a narrow layer design. This design has low computational cost and prevents overfitting. The result shows that the accuracy of the method is very high. The predicted and targeted values are almost merged in a single line. The RMSE and MAX error indexes are very low. The accuracy of the model is acceptable; the electric vehicle battery can be expected to last longer and perform mobility tasks more reliably. Finally, this method also demonstrates that the accuracy of estimating the state of charge (SOC) of a battery in an electric vehicle can be improved by utilising narrow learning layers.

Keywords: Battery, Corny Architecture, Electric Vehicle, State of Charge.
Scope of the Article: Computer Architecture and VLSI