Citation: Gao, H.; Zang, B.; Sun, L.;
Chen, L. Evaluation of Electric
Vehicle Integrated Charging Safety
State Based on Fuzzy Neural
Network. Appl. Sci. 2022, 12, 461.
https://doi.org/10.3390/
app12010461
Academic Editors: Pierluigi Siano,
Hassan Haes Alhelou and
Amer Al-Hinai
Received: 8 November 2021
Accepted: 31 December 2021
Published: 4 January 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Evaluation of Electric Vehicle Integrated Charging Safety State
Based on Fuzzy Neural Network
Hui Gao
1
, Binbin Zang
1,
* , Lei Sun
2
and Liangliang Chen
3
1
College of Automation, College of Artificial Intelligence, Nanjing University of Posts
and Telecommunications, Nanjing 210023, China; gaohui2005@163.com
2
State Grid Jiangsu Electric Power Co., Ltd., Research Institute, Nanjing 211103, China; slsz_2014@126.com
3
NARI Technology Co., Ltd., Nanjing 211106, China; N1220055918@126.com
* Correspondence: nyzangbinbin@163.com
Featured Application: This work is suggested to be applied to the safety assessment of electric
vehicle charging process, which may be embedded into the system to effectively assess the charging
state and provide certain help for reducing charging failures in the future.
Abstract:
Electric vehicles have been promoted worldwide because of their high energy efficiency
and low pollution. However, frequent charging safety accidents have to a certain extent restricted
the development of electric vehicles. Therefore, it is extremely important to accurately evaluate
the safety state of EV charging. The paper presents an integrated safety assessment method for
electric vehicle charging safety based on fuzzy neural network. The integrated fault model was
established by analyzing the correlation between truck–pile–grid. Then the integrated evaluation
index was analyzed and sorted out, and the comprehensive fuzzy evaluation method used to evaluate.
Following this, the improved GA_BP neural network algorithm was used to calculate the weight.
Compared with the evaluation effect before and after the improvement, the simulation results
show that the GA_BP neural network has higher accuracy and smaller error than the ordinary BP
neural network. Finally, the feasibility and effectiveness of the evaluation method was verified by a
case study.
Keywords: electric vehicles; fuzzy neural network; safety assessment method; evaluating indicator
1. Introduction
In order to promote a global carbon balance, countries have formulated their own
“carbon reduction” targets [
1
]. China is striving to achieve the “carbon peak” before
2030 and the “carbon neutral” “dual carbon” goal before 2060 [
2
]. New energy electric
vehicles play an important role in reducing carbon emissions, reducing consumption of
fossil energy, and promoting the development of electrified transportation [
3
]. However, in
the promotion process of electric vehicles, power battery failure and charging equipment
safety problems have become obstacles to the rapid development of electric vehicles [
4
].
Therefore, EV fault diagnosis, safety warnings, and other issues have become the focus of
research in various countries [5].
As the main energy source of the vehicle, the power battery system is not only the
core component of electric vehicles, but also a technical bottleneck restricting their de-
velopment [
6
]. The inevitable performance degradation of the power battery in use will
lead to the decrease of the vehicle driving range, the deterioration of the power perfor-
mance and the shortening of service life, which will cause an increase in the safety risks.
Accurate prediction and diagnosis of power battery faults is an important guarantee to
improve the safety and reliability of electric vehicles [
7
]. In the actual operation of electric
vehicles, many factors such as electromagnetic interference, road conditions, and driving
habits can lead to battery system failures, and complex, nonlinear, or multi-parameter
Appl. Sci. 2022, 12, 461. https://doi.org/10.3390/app12010461 https://www.mdpi.com/journal/applsci
