Academic Journal of Science and Technology 
ISSN: 2771-3032 | Vol. 9, No. 2, 2024 

 

47 

Fault	Diagnosis	Model	of	New	Energy	Vehicle	Charging	
Equipment	Based	on	Fuzzy	Fault	Tree	Theory	
Wenqing Li1, a 

1Mengxi College, Jiangsu University, Zhenjiang, Jiangsu, 212013, China 
aWinky_green@163.com 

 

Abstract: This paper probes into the new energy automobile charging equipment fault diagnosis problem. Fuzzy fault tree 
theory is introduced as a solution, fuzzy mathematics is used to improve the accuracy of fault diagnosis, the fault diagnosis model 
of charging equipment is established, and the fault diagnosis system based on Internet of things technology is designed, including 
equipment layer, communication layer, server and client. The system aims to realize the real-time monitoring and processing of 
the fault of new energy vehicle charging facilities, improve the user experience and improve the safety index. 

Keywords: Fuzzy fault tree; Internet of things; Charging equipment; New energy vehicles. 
 

1. Introduction 
Under the situation of global environmental problems and 

energy crisis becoming more and more serious, electric 
vehicles, as a kind of green energy transportation tools, have 
attracted the attention of governments and automobile 
enterprises all over the world[1]. However, the development 
of electric vehicles is not mature at the present stage, and the 
endurance problem is an important resistance to the 
development of electric vehicles. As an important device to 
supplement energy for electric vehicles, the stable and reliable 
operation of charging equipment is particularly important. 
However, the circuit of electric vehicle charging equipment is 
complex, and the reasons for failure are complex. In the case 
of continuous upgrading of electric vehicle charging 
technology and continuous improvement of automation, its 
structure is increasingly complex, which brings great trouble 
to the fault diagnosis and safety warning of electric vehicle 
charging equipment[2]. Charging pile is an important device 
for electric vehicle to supplement energy, and it is also an 
important guarantee for endurance. Therefore, it is of great 
significance to identify charging pile failure and make safety 
warning for ensuring users 'charging experience and charging 
safety. 

At present, there have been a series of studies on the fault 
analysis and detection system of electric vehicle charging pile, 
but there are still some problems: (1) More research focuses 
on fault diagnosis. On the one hand, there is relatively little 
research on intelligent charging pile system combining 
resource scheduling, safety warning and other functions. (2) 
The fault analysis of charging piles mostly relies on the 
theoretical research of complex algorithms, which lacks 
applicability for practical applications[3]. 

In this paper, the fuzzy fault tree theory is used to analyze 
the fault of electric vehicle charging equipment, and a set of 
new energy vehicle charging equipment fault diagnosis 
system is designed based on the Internet of things technology. 
The system can find the root cause of the fault, help the staff 
to deal with it quickly, and improve the processing efficiency. 

2. Fuzzy Fault Tree Theory 

2.1. Defects of traditional fault tree theory 
Fault tree analysis (FTA) is a graphical deduction method 

that divides the causes of system faults into branches[4]. 
The fault tree is composed of "events" and "gates". The 

events are represented by failure probabilities, and the gates 
are used to describe the relations between events[5]. In fault 
tree analysis, the least desired fault state of the system is taken 
as the top event, which is the goal of the analysis, and all the 
factors that can cause the fault are found out as the middle 
event. Then all the factors that cause the middle event are 
found out, and in this way all the causes of the system failure 
are traced back to the bottom event[6]. The logical 
relationship between the system failure and the intermediate 
events and the bottom events is connected by logic gates to 
form a tree diagram to represent the relationship between the 
system and the causes. 

However, the existing theories and methods need to regard 
the occurrence probability of the top event and the bottom 
event of the fault tree as a precise value, which has many 
shortcomings in practical applications: (1) The causes of the 
failure of the system components are not only caused by 
objective uncertain factors, but also by human subjective 
reasons, such as human error, design experience and other 
fuzzy uncertain factors[7]. (2) In practical engineering, the 
failure mechanism of the system is not clear in many cases. In 
complex human-machine systems, due to the imprecision of 
system modeling caused by human factors, correlated failures, 
and common cause failures, pure probability methods are 
difficult to work, and the relationship between events is often 
uncertain. (3) In some high-reliability systems, the failure 
frequency is very low, and a large amount of data can not be 
obtained; 

2.2. Fuzzy fault tree theory 
Fuzzy fault tree is a method that combines fault tree 

analysis method with fuzzy mathematics theory. It extends the 
traditional fault tree analysis to deal with uncertain problems, 
so that the application scope of fault tree analysis is wider[8]. 
When applying the fuzzy fault tree theory to analyze, we need 
to calculate the fuzzy probabilities, which are obtained based 



 

48 

on the previous data and analysis experience. On this basis, 
fuzzy mathematics is used to calculate fuzzy probabilities 
instead of the original precise-valued probabilities. 

The fault probability of components needed in fault 
diagnosis can generally be calculated by historical fault data. 
However, due to the continuous change of the use 
environment of the equipment and the replacement of 
components from time to time, the failure probability of each 
component becomes uncertain. The fuzzy fault tree theory is 
to solve this kind of problem, it uses fuzzy logic to express 
the uncertain component failure probability with fuzzy 
numbers, so that the probability becomes fuzzy probability. 
At the same time, the theory also takes into account the 
"human" this factor can not be ignored, and the judgment of 
the experienced personnel on the fault is also integrated into 
the results through calculation, which is more practical. In 
addition, for those concepts such as the degree of failure that 
cannot be quantified by precise values, fuzzy fault tree theory 
can also be flexibly expressed by fuzzy numbers, which has a 
high degree of flexibility[9]. 

It is defined as the set of fault symptoms of the equipment, 
and is the total number of fault symptom types:𝑈𝑚 

 
𝑈 𝑈 , 𝑈 , ⋯ , 𝑈               (1) 

 
Is defined as the fault cause set and is the total number of 

fault cause types:𝑉𝑛 
 

𝑉 𝑉 , 𝑉 , ⋯ , 𝑉               (2) 
 
According to the influence degree of each bottom event 

contained in the fault symptom set on the equipment, the 
fuzzy vector of fault symptom can be defined as follows. 

  
𝑢 𝑢 , 𝑢 , ⋯ , 𝑢               (3) 

  
It is defined that the fault cause causes the equipment 

failure, and the membership degree of each fault cause is, at 
this time, the fuzzy vector of the fault cause is defined 
as:𝑉 𝑉 𝑣  

 
 

𝑣 𝑣 , 𝑣 , ⋯ , 𝑣                (4) 
 

 
The fuzzy decision equation is listed as follows. 
 
 

𝑣 𝑢°𝑅                   (5) 
 
Where is the fuzzy relation matrix:𝑅 
 
 

𝑅 𝑟 ∙

𝑟 𝑟 ⋯ 𝑟
𝑟 𝑟 ⋯ 𝑟

⋮ ⋮ ⋮ ⋮
𝑟 𝑟 ⋯ 𝑟

        (6) 

 
 
And has.0 𝑟 1, 1 𝑖 𝑚, 1 𝑗 𝑛 
The established fuzzy relation matrix of dimensions is 

essentially a mapping. 𝑚 𝑛𝑅The fuzzy relation matrix can 
map the fuzzy fault symptom set to the fuzzy fault cause set, 
so that the bottom events of different fault states are integrated 

to make decisions, and it has high adaptability. The column 
represents the fault cause, the row represents the fault 
symptom, and the matrix element represents the membership 
degree of the first symptom to each cause.𝑟 𝑖𝑉 𝑗𝑈  

The fuzzy synthesis operator "" is needed in the fuzzy 
diagnosis, which means that the fuzzy relation matrix R and 
the fault cause fuzzy vector are operated in a certain operation 
mode, and finally the fuzzy relation equation is solved to 
obtain the fault result.  ° 𝑣 Through this fuzzy combination 
operation, we can effectively deal with the uncertainty and 
fuzziness, and get more comprehensive and accurate fault 
diagnosis results. 

3. Fault Diagnosis Model for New 
Energy Vehicle Charging 
Equipment 

Due to the practical needs of fault diagnosis and the 
particularity of information fusion technology, it is feasible to 
use information fusion technology in fault diagnosis. At 
present, fault localization methods mainly include Bayesian 
network fault localization, fuzzy information fusion 
localization, evidence theory fusion fault localization, etc. In 
this paper, fuzzy fault tree is mainly used to calculate legal 
fault information. 

According to the analysis, the fault components of charging 
equipment include the total fault tree of charging pile, the 
fault tree of charging module, the fault tree of main circuit, 
the fault tree of charging gun, the fault tree of control circuit, 
the fault tree of user terminal equipment and the fault tree of 
charging pile over temperature. These causes are classified as 
intermediate events and bottom events, and the middle events 
are caused by the bottom events. 

Taking the risk of "charging equipment failure" as the top 
event, we first look for the direct causes that can lead to the 
risk, and summarize several intermediate events as the direct 
causes according to the working principle of charging 
equipment. These intermediate events are "charging module 
failure", "main circuit failure", "charging gun failure", 
"control circuit failure", "user terminal failure", "charging pile 
overtemperature failure"[10]. As these intermediate events 
can be further subdivided, each intermediate event can be 
corresponding to one or more events, until subdivided to the 
bottom event, these bottom events constitute the basic 
component leading to system failure. 

The first-level intermediate event "charging module fault" 
can be subdivided into "AC input fault", "communication 
fault", "output short circuit fault", "output overvoltage fault" 
and "output overcurrent fault"[11], in which "AC input fault" 
can be subdivided into "AC power outage" and "AC input 
missing item". "Communication fault" can be further 
subdivided into "communication protocol problem", 
"hardware problem", "insulation detection module 
communication interruption", "AC module communication 
interruption", "meter communication interruption"[12]. The 
first-level intermediate event "main circuit fault" can be 
subdivided into "output fuse fault", "reactor anomaly", 
"transformer secondary coil anomaly", "power switch 
anomaly" and "unipolar switch anomaly". The occurrence of 
any one of these bottom events can lead to the occurrence of 
its corresponding intermediate events. The partition results 
are shown in Table 1 and Table 2. The logic structure diagram 
of the fault tree is shown in Figure. 1. 

 



 

49 

 
Table 1. Intermediate event system code and interpretation 

Event codes Event name 
𝑻 Failure of charging equipment 

𝑬𝟏 Failure of charging module 
𝑬𝟐 Main circuit failure 
𝑬𝟑 Charging gun failure 
𝑬𝟒 Control circuit failure 
𝑬𝟓 User terminal fault 
𝑬𝟔 Overtemperature fault of charging pile 
𝑬𝟕 Ac input fault 
𝑬𝟖 Communication failure 
𝑬𝟗 Output short circuit fault 
𝑬𝟏𝟎 Output overvoltage fault 
𝑬𝟏𝟏 Output overcurrent fault 
... ... 

𝑬𝟐𝟔 Overheated charging pile 
𝑬𝟐𝟕 Charging pile module fan overheating 

 
Table 2. Underlying event system code and interpretation 

Event codes Event name 
𝒙𝟏 Ac power outage 
𝒙𝟐 Ac input is missing 
𝒙𝟑 Communication protocol issues 
𝒙𝟒 Hardware issues 
𝒙𝟓 Communication breakdown of insulation detection module 
𝒙𝟔 Ac module communication interruption 
𝒙𝟕 The meter communication is interrupted 
𝒙𝟖 The filter is broken down in the middle 
𝒙𝟗 Rectifier drive circuit failure 
𝒙𝟏𝟎 Faulty output fuse 
𝒙𝟏𝟏 Abnormal reactor 
𝒙𝟏𝟐 Abnormal secondary coil of transformer 
𝒙𝟏𝟑 Abnormal power switch 
𝒙𝟏𝟒 Abnormal monopole switch 
... ... 

𝒙𝟒𝟐 Module fan malfunction 
𝒙𝟒𝟑 Poor/damaged switch contact 
𝒙𝟒𝟒 Faulty temperature sensor 

 

 

Figure 1. Logic structure diagram of fault tree for new energy vehicle charging equipment 

 



 

50 

 

4. Fault Diagnosis System for Electric 
Vehicle Charging Equipment 

The architecture of the whole system is shown in Figure 2, 
which consists of four layers, namely the perception layer, the 
network layer, the data layer and the application layer from 
bottom to top[13]. Figure 2. Fault diagnosis system structure 
diagram of charging equipmentThe perception layer includes 
sensors and device acquisition terminals, and its main task is 
to obtain the status information of charging devices. The 
network layer relies on the wireless communication module 
and uses the Internet of Things technology (such as 4G/5G, 
NB-IoT+4G+5G) to realize data communication with the 
monitoring and security early warning platform[14]. The data 
layer uses advanced big data technology to establish a big data 
resource center for new energy vehicle charging equipment, 
which is responsible for the centralization, storage, 
processing, analysis and mining of equipment data, 
cataloging, showing and managing equipment data, and 
providing data services and showing for the application 
layer[15]. The application layer uses data services to analyze 
and apply vehicle data, constructs a safety early warning 
model, and realizes safety early warning and data display of 
new energy vehicle charging equipment [16]. 

 

	
Figure 2. Fault diagnosis system structure diagram of 

charging equipment 

5. Conclusion 
This paper deeply studies the fault diagnosis problem of 

electric vehicle charging equipment, introduces the 
establishment process of charging equipment fault model, 
uses fuzzy fault tree algorithm to subdivide various possible 
fault events, and provides a framework for fault diagnosis. In 
addition, this paper designed a fault diagnosis system for 
charging equipment based on Internet of things technology, 
aiming at realizing real-time monitoring and rapid processing 
of charging equipment faults. 

The application of fuzzy fault tree theory provides a new 
idea for fault diagnosis, and taking uncertainty factors into 
consideration has positive significance for improving the 
accuracy and practicability of diagnosis. The fault diagnosis 
system of charging equipment based on Internet of things 
technology provides technical support for the development of 
electric vehicles, which can ensure the stability of charging 
equipment and improve user experience. 

However, it is still necessary to further improve the model 
and system and incorporate more actual data and application 
scenarios to improve the accuracy and practicality of the 
system[17]. At the same time, the continuous research and 
improvement of fault diagnosis technology for new energy 

vehicle charging equipment is crucial to promote the 
sustainable development of the electric vehicle industry. 

References 
[1] Liu, J. (2023) Common Faults and Application of maintenance 

technology for pure electric vehicles. Special Purpose Vehicles, 
(1):69-71. 

[2] Gao, D., Liang, K., Hou, J., et al. (2019) Design of mobile 
monitoring and fault diagnosis system for electric vehicle 
charging pile. Journal of Qingdao University of Science and 
Technology (Natural Science Edition), 40(3):107-113. 

[3] Song, M., Kang, Y. (2022) Application and analysis of electric 
vehicle charging pile detection technology. Electrical 
Appliances and Energy Efficiency Management Technology, 
(3):77-80. 

[4] Song, H., Zhang, H., Wang, X. (2005) T-S Fuzzy Fault tree 
analysis method. Control and Decision, 20(8):854-859. 

[5] Shao, Y., Xue, H. (2007) Application of Fault tree analysis 
method in System fault diagnosis. Chinese Manufacturing 
Information Technology, 36(1):72-74,77. 

[6] Wei, X., Bian, S. (2004) Fault tree analysis and its application. 
Electronic products reliability and environmental testing, 
(3):43-45. 

[7] Zhao, Y., Gong, W. (2001) Research on Fuzzy fault tree 
analysis and its application. Chinese Journal of Safety Science, 
11(6):31-35. 

[8] Zhu, D., Yu, S. (2002) Research on fault diagnosis method 
based on minimum cut set of fault tree. Data Acquisition and 
Processing, 17(3):341-344. 

[9] Yao, H., Li, M., Yang, X. (2021) Application of Fuzzy Fault 
tree Theory in Fault diagnosis of aviation charging equipment. 
Measurement and Testing Technology, 48(6):82-84.  

[10] Wu, W. (2023) Application of data mining in electric vehicle 
charging pile charging fault detection. Automobile and New 
Power, 6(2):81-84. 

[11] Gao, D., Zheng, X., Wang, Y., et al. (2022) Electric vehicle 
charging state monitoring and multi-level security forecasting 
police law . Transactions of China Electrotechnical Society, 
37(9):2252-2262. 

[12] Xie, J, Yu, Q. (2022) Research on electric vehicle Battery Fault 
diagnosis based on fuzzy neural network. China New 
Technology and New Products, (6):20-23. 

[13] Ruan, Z., Jiao, H., Zhou, L., et al. (2023) Prospects and 
suggestions for deepening the application of Internet of Things 
in electric vehicle industry. Hubei Electric Power, 47(1):133-
138. 

[14] Wu, J. (2019) Research on New energy vehicle operation 
monitoring Cloud platform based on big data and artificial 
intelligence. Journal of Guangdong Normal University of 
Technology, 40(6):19-25. 

[15] Miao, Y. (2020) The application of Internet of Things to 
electric vehicle charging facilities. Brand Research, (36):284. 

[16] Chen, Z. (2020) Application of artificial intelligence in 
automobile industry. Shanghai Quality, (3):28-31. 

[17] Yu, Y., Xue, X., Wang, H. (2022) Design of safety early 
warning system for new energy vehicles based on Internet of 
Things. Science and Technology and Innovation, (9):10-12,16.