Research on the Operation Modes of Electric Vehicles in Association with a 5G Real-Time System of Electric Vehicle and Traffic

Abstract

With the popularity of 5G technology and electric vehicles, many countries around the world have adopted 5G technology to build sustainable smart city systems, and intelligent transportation is an important part of smart cities. From the perspective of 5G technology innovation bringing changes to traditional industries, in this paper, we analyze the mechanism by which 5G technology drives the transformation and upgrading of the electric vehicle industry. Based on the changes brought by 5G technology to the three industries of agriculture, industry and services, we analyzed the transformation of business models brought about by 5G with respect to electric vehicle operation. Furthermore, we analyzed the data of a 5G real-time system of electric vehicle and traffic operating in Nanjing, China, for a month in 2021, with a total of 10,610 electric vehicles and 1,048,575 cases to model the modes of electric vehicle operation associated with the platform. Based on the frequency density method, we identified three typical operating modes of urban electric vehicles: private electric vehicle use instead of walking accounts for 24.8%, passenger vehicles (Uber/Didi and taxi) account for 64.4% and logistic distribution electric vehicles account for 10.8%. We developed a method to automatically identify the operating mode of electric vehicles using data from a 5G real-time electric vehicle traffic platform, which provide a reference for the operation of electric vehicles associated with the platform. This work also provides data that can be used to support the establishment of models for the commercial operation of charging points.

1. Introduction

Motivation: With the maturity of fifth-generation mobile communication technology(5G) its applications have been promoted around the world. The number of smart phone users will reach 5.9 billion in 2025 [1], and the number of 5G devices on the Internet will exceed 10 billion [2]. Global deployment status of the 5G network by June 2020 shows most developing countries and all developed countries have operational 5G networks [3,4], with varying degrees of impact on various fields. Tang et al. (2021) conducted in-depth research on the differences in technological development between 2G and 5G [4]. Castells (1997) studied the driving factors and network symbiosis of communication technology from the perspective of social organization development [5], and Fonzone et al. (2018) built 5GRTS-ECR intelligent mobile network solutions [6].

Sun et al. (2020) studied vehicle-to-everything (V2X) technology in 5G technical scenarios [7]. Figure 1 summarizes the number of studies on smart cities with 5G in recent years, with a sharp increase each year since 2016 [8]. The main focus of contemporary literature is the feasibility of 5G technology and the economic development of various industries. Little research has been conducted on business model changes associated with 5G, with even less research on 5G and the electric vehicle (EV) industry. In this paper, we analyze the impact of 5G on three major industries and explore the changes in business models with respect to EV operation.

Looking back at the first four generations of communication technology, the users of the first and second generations were limited to professional and technical knowledge groups; the content was limited to document delivery and personal publication, with the scope of us limited to the government and public interconnection. As for the third and fourth generations of communication technology, the number of users gradually increased worldwide, with the content upgraded from search engines to mobile intelligent Internet and the scope of use extended to include global interconnection and intelligent Internet of Things. Sureephong et al. (2017) proposed an integrated detection system based on the Internet of Things [9]. As the interval between technology iterations has gradually shortened and the integration of information technology in various industries has intensified, the demand for more efficient communication technology has increased. Tang et al. (2021) introduced a “theory of telecommunications needs“ analogous to Maslow’s Hierarchy of Needs [4]. In this theory, Maslow’s five-tiered needs with respect to physiology, safety, socialization, respect and self-worth are reduced to the three categories of survival, belonging and growth [10], as shown in

Table 1. Theory of telecommunications needs analogous to Maslow’s Hierarchy of Needs.

Growth	Self-Realization	Self-Liberation	Intelligence
Ascription	Respect (confidence, achievement)	Sensory extension (human–computer interconnection, etc.)	People–things, things–things
	Social needs (sense of belonging, emotion, etc.)	Information consumption (mobile Internet, etc.)	People–information
Survival	Security requirements (safety of persons and property, etc.)	Universal communication (SMS, etc.)	People–people
	Physiological needs (breathing, water, food)	Necessary communication (112/119 etc.)

.

Some scholars have asserted that telecommunication needs can be divided into five categories: necessary communication, universal communication, information consumption, sensory extension and self-liberation, which correspond to the human–human level, the human–information level, the human–thing level, the thing–thing level and the intelligence level. Gangadhar and Chandra Sekhar (2022) elaborated the evolution of 5G technology, as well as its merits, drawbacks and architecture [3].

The innovative benefits of 5G for the whole society are not only reflected in by meeting existing needs but also promoting new industrial scenarios. The standards of fifth-generation communication technology proposed in 2019 enhanced mobile networks, large-scale machine communication and ultra-reliable, low-latency communication. The above standards not forced various industries update their operations but also promoted the emergence of subversive technological innovations and industries. On the one hand, the connection of wireless devices through network equipment and optical fiber connections considerably improves transmission efficiency by integration with existing technical hardware. On the other hand, some new products, such as 8K real-time transmission, immersive game experiences, remote conferences and other tertiary industry products; smart grids, smart cities, autonomous driving and other secondary industry products; and unmanned driving [11], smart agriculture and other primary industry products, are supported by 5G standards and network services. 5G technology will also have a positive impact on the further maturation of the Chinese socialist governance system, such as organizational, institutional, operational, evaluation and quality control systems. As a result, countries around the world are aggressively developing 5G applications, with the technology becoming popular globally.

The changes brought about by 5G have not only met the increased demand for communication technology in various industries but also induced profound changes in the upper, middle and lower reaches of industry. Upstream changes are mainly reflected in the construction of communication base stations, midstream changes are reflected in network construction and downstream changes are reflected in the application of products and terminals. The fifth generation of communication technology, with its unique technical advantages, affects both technological (processes and products) and non-technological (organization and marketing) innovation in industry. Tragos et al. (2008) conducted an in-depth study on the limitations of 2G, 3G and 4G and found that the efficiency of 5G communication is four times higher than that of the original four generations of technology [12]. The innovative contribution of such efficient communication technology depends on the technical advantages of 5G; the lack of speed and storage capacity of previous technology is alleviated through fast and efficient transmission rates. With respect to products and processes, new technologies reduce network costs and improve the performance of the Internet of Things which was previously lacking. The application of artificial intelligence in all areas can reduce the cost of information flow, with more rapid information dissemination speeds prompting producers to pay more attention to the needs of consumers and provide more personalized and intelligent products.

As for organization, the communication capacity of old technology is limited, and existing architecture will inevitably face the risk of deconstruction, with the novel organizational structure of 5G technology bound to replace existing insufficient architecture. On one hand, some scholars believe that the innovation of industrial organizations based on (Khan et al., 2020) 5G requires the support of a large number of high-level technical personnel [13] and that industrial hierarchies will tend to be centralized. Other scholars point out that with reliance on the high-efficiency transmission rate of 5G [3], technical personnel can be dispersed around the world so that innovative activities will be distributed. With respect to market innovation, 5G technology will be reflected in the integration of “5G+” industrial activities. The resources of different industries and platforms will produce commercial interaction, and the industrial chain formed by big data will be more in line with market demand. On the other hand, the high speed and high efficiency of 5G will further digitize resources in order to form cloud resources, and low transmission costs and big data linkage may promote a free business model.

5G technology was developed on the basis of 4G. In this era of interconnection of everything and the information economy, individual requirements for communication quality and speed are increasing. The popularity of 5G is unstoppable because the demand of individuals for communication has increased to Maslow’s fifth layer: intelligence. Existing technical and economic strength, in combination with the capabilities of software and hardware, can support the development of the EV industry and 5G technology; the development of related upstream and downstream industries also requires the support of EVs and 5G.

Related work: In this paper, we mainly focus on the feasibility of 5G technology and the economic development of various industries. Little research has been conducted on business model changes associated with 5G, with even less research on 5G and the electric vehicle (EV) industry. From the perspective of 5G technology innovation bringing changes to traditional industries, in this paper, we analyze the mechanism by which 5G technology drives the transformation and upgrading of the electric vehicle industry. Based on the changes brought by 5G technology to the three industries of agriculture, industry and services, we analyzed the transformation of business models brought about by 5G with respect to EV operation. Furthermore, we analyzed the data of a 5G real-time system of electric vehicle and traffic (5gRTS-ET) operating in Nanjing, China, for a month in 2021, with a total of 10,610 electric vehicles and 1,048,575 cases to model the modes of electric vehicle operation associated with the platform. Based on the frequency density method, we identified three typical operating modes of urban electric vehicles.

Paper contribution: With the promotion of 5gRTS-ET in an increasing number of countries and cities, the popularity of ride-hailing platforms and the strengthening of people’s understanding of environmental protection, the use of private EVs instead of walking will gradually decrease, whereas the operation of passenger EVs (Uber/Didi) will increase so that people can work as part-time drivers during weekends or during their commute [14]. Rapid and automatic identification of EV operation modes based on data is the key to adaptation of charging points to dynamic modes of EV operation associated with the 5gRTS-ET platform, also representing an effective way to improve the utilization rate of charging points. The employed research methods and conclusions presented in this paper provide a reference for the operation of electric vehicles associated with platform under investigation, in addition to providing data support for the establishment of models for the commercial operation of charging points.

In Section 2 of this paper, we introduce data collection and research methods and analyze the impact of 5G technology on the three major industries, as well as the changes in the EV operating model brought about by the transition from traditional EV management platforms to the 5gRTS-ET real-time interactive system. Furthermore, we summarize the results of statistical analysis. Using a large amount of real-world 5gRTS-ET data, we established a mathematical model and verified the three typical operating modes of EVs associated with the platform. Finally, in Section 4, we present conclusions and suggestions for future research directions.

2. Materials and Methods

Data were collected from the 5gRTS-ET platform, a system operator in Nanjing, China, that collects operational data from EVs, with the ability to monitor EV status and analyze traffic data. Real-time monitoring of EVs, path recording, abnormal warnings and charging data not only guarantee the scientific, intelligent and safe operation of EVs but also provide a theoretical and practical basis for future commercial applications of EVs. The platform 5gRTS-ET has accumulated a massive database including battery data, longitude and latitude data, EV data, path data, etc.

As a data acquisition device, EV information acquisition and transmission systems are integrated in EV vehicle information systems, collecting real-time data through CAN-BUS modules, which carry out local storage and operation. Then, the useful results obtained by this system are encrypted and transmitted through 5G networks to the remote 5gRTS-ET platform with a sampling cycle of 1s. The 5gRTS-ET platform parses the data sent by the EV terminal according to the communication protocol and stores them in a local SQL server database. The data are stored in two databases: the local storage unit of the EV vehicle terminal and the 5gRTS-ET platform SQL server database. The data stored locally by the EV terminal constitutes raw data, whereas that stored by the 5gRTS-ET platform is converted to usable data following transmission from the EV terminal. Furthermore, the result is stored in a straightforward, easy-to-access and legible format according to the specifications of the database. The data used in the present study were retrieved from the SQL server database of the 5gRTS-ET platform, a system operator in Nanjing, China, for a month in 2021, with a total of 10,610 EVs and 1,048,575 cases.

3. Results

3.1. The Impact Mechanism of 5G Technology on the Innovative Activities of Primary Industry

Agricultural informalization has achieved rapid development; the concept of smart agriculture is well-established, but its development has been considerably hindered due to the remote terrain of vast rural areas, low network coverage, insufficient network capacity and frequent problems such as communication failures between equipment, Even under 4G network architecture, a series of problems related the development of agricultural informalization persists, such as chaotic data with a low integration rate, difficulties in horizontal communication with low compatibility, poor timeliness of information transmission with insufficient sharing, etc. Problems that cannot be solved within the 4G network are compensated by other technical means [15]. The development of 5G can provide strong technical support for smart agriculture and, being much smarter and more efficient than 4G, is bound to constitute the dominant development trend in agricultural informalization. Data availability for the purpose of automated learning requires improvements in hardware technology [13].

Smart agriculture based on the Internet and 5G technology is a new development model that relies on the Internet of Things, uses sensors and software to comprehensively control the agricultural production process through mobile or computer platforms and achieves the goal of developing a more scientific and reliable agricultural development strategy with information and data. Based on the Internet of Things, cloud computing services in 5G networks provide flexible and efficient solutions for smart agriculture [4].

5G technology contributes to agriculture mainly by promoting the construction of an integrated agricultural service cloud platform. To build a smart agricultural system, technology is necessary as a support to form resources, facilities, applications and service modules, which have enabled agricultural information resources to be generated, analyzed and applied. 5G presents fewer obstacles between agricultural information networks and the construction of resources. Agricultural resource information can be transmitted more efficiently and smoothly through 5G networks, achieving a high degree of integration, which makes comprehensive utilization of agricultural information more efficient. 5G can also facilitate agricultural production management so that massive amounts of agricultural data can be analyzed and utilized more accurately, providing more effective data support services for agricultural producers and operators [4].

5G is also helpful in improving agricultural information service systems and boosting industrial structures in agriculture. 5G cloud services and integrated information technology services have been added to agricultural information systems based on 4G, promoting the timely, efficient and accurate transmission of information so that agricultural information service systems have a higher degree of sharing and more complete functions. Informatization construction is not limited only to production and management but is also integrated into the construction and development of rural areas. Due to the promotion of network systems, public services such as insurance, finance and medical care are continuously integrated into smart agriculture so that market resources continue to flow into agriculture, making rural construction and development more convenient, economic and effective. In addition, the integration of 5G networks and agricultural business entities has effectively promoted the optimization and innovation of the agricultural industry. On one hand, farmers, specialized farmer cooperatives, agricultural enterprises, etc., are gradually building a modern agricultural production chain integrating production, circulation, processing, storage, transportation, sales and service; on the other hand, the interaction and integration of secondary and tertiary industries with agriculture have promoted rural tourism. The development of emerging industries, such as rural leisure tourism, has optimized the traditional agricultural structure and promoted the transformation and upgrading of agriculture.

5G can also promote the development of smart agriculture and support the development of a standard smart agriculture system. On 11 October 2019, the first 5G Smart Agriculture Development Forum was held in Hohhot, China. During this forum, the development status of smart agriculture in the 5G era was elaborated, and the prospect of smart agriculture was discussed. Tang et al. (2021) studied China’s agricultural model of complex landforms, as both sensing technology and AI require 5G wireless support [4]. At present, the problem that must be faced in the process of global agricultural development is the establishment of standards for relevant industries. Relevant departments can establish basic, common and special standards associated with six main aspects, i.e., resources, networks, applications, technologies, talents and regulations, according to the practical experience of smart agriculture so as to develop a framework for smart agriculture and to promote standardization with the support of technology.

5G technology can also improve the efficiency of big data applications to provide new development space for the collection, analysis and utilization of agricultural information data. For example, biological surveys can already be carried out by computer operations. Massive streams of genetic information can be created and analyzed in the cloud, including hypothesis verification, test planning, definition and development. Furthermore, decisions pertaining to drought resistance, flood resistance, disease and pest resistance can be made so as to reduce production costs and effectively control the environmental risks associated with crop planting.

5G is can also be used to build an e-commerce system for agricultural products and improve the marketization level of agriculture. Under the influence of 5G technology, the agricultural industry has access to the Internet and the Internet of Things, further expanding the development space of agriculture. The development of an e-commerce system has shortened the distance from production to sales, such as by continuously improving rural logistics distribution. The collection system of online delivery channel data of agricultural products has also shortened the distance between consumers and farmers, as well as the distance in the time and space traveled by agricultural products. 5G enables accurate monitoring of the delivery process. The development of e-commerce for agricultural products will further open up the sales market and improve the level of agricultural marketization.

3.2. The Impact Mechanism of 5G Technology on the Innovation Activities of Secondary Industry

The development of the Internet and information technology has played a vital role in the digitization, networking and intelligence of the manufacturing industry. Compared with traditional industries that rely on capital as a driving force, the development of new industries relies more on technological innovation. Countries around the world have proposed the concepts of the “Industrial Internet” and “Industry 4.0”, and China subsequently proposed the “Made in China 2025” strategy in 2014 to promote the deep integration of industry with new Internet and communications technology. In digital industry, based on the breakthrough of 5G and Internet + technology, the smart factory model with 5G + industrial Internet is gradually emerging. In addition, the “new infrastructure” industry based on 5G commercial and artificial intelligence, industrial Internet and Internet of Things has also emerged as a strong short-term measure to cope with the economic downturn and stabilize economic growth. As a general-purpose technology (GPT), 5G contributes to secondary industry, mainly owing to its characteristics of extensiveness, continuous improvement and continuous innovation, producing positive externalities associated with innovation activities. On the supply side, secondary industry has a very low marginal cost, exhibiting the characteristics of public goods in relation to the innovation of products and services [16].

At the level of technological innovation, the industrial Internet includes a framework layer of device awareness, an IaaS layer, PaaS layer, SaaS layer and cloud-side collaboration layer [17]. New technologies, such as the Internet of Things, cloud computing, big data and artificial intelligence, have brought out the new function of “perception” in originally non-intelligent production line control systems to facilitate product life-cycle management. Furthermore, through the perception data of artificial intelligence supplemented by big data technology, the reliability and stable operation of production will be greatly improved, reducing the engagement of workers. Thus, 5G can be used to improve product quality, with artificial intelligence visual recognition technology and big data traceability replacing manual quality inspection, which can improve the accuracy of quality inspection and considerably reduce labor costs. For example, in current electronic product production lines, visual recognition technology is used in the early stages of circuit board manufacturing to identify potential technical problems. On the other hand, with respect to large-scale production of machines, enterprises will inevitably increase investment in fixed assets. From a long-term point of view, the aging and wear of equipment will inevitably result in losses for enterprises. With intelligent big data, equipment management, such as remote diagnosis, as well as operation and maintenance of equipment, becomes more efficient and intelligent. For example, the German company Bosch collects the operation data of its workers’ tools through 5G. By such means, the Internet can record the work processes of workers in order to ensure product quality and verify new innovative activities through big data experiments. Finally, driverless transportation, plant safety monitoring and remote-control systems can be realized with 5G to achieve high reliability and low latency, considerably improving production efficiency.

At the level of non-technological innovation, the “Industrial Internet” requires enterprises to achieve intelligent linking of materials and equipment, as well as timely adjustment of production plans through real-time data sharing so as to achieve flexible production. From the perspective of the industrial chain, the integration of industrial manufacturing into the Industrial Internet will help to form a complete industrial chain with upstream equipment, material supply enterprises and downstream consumer enterprises. From the perspective of production organization, the upgrading of the Industrial Internet of Things will help to realize interconnection between basic production procedures and production materials. First of all, working data can be transmitted to the system for the first time and passed to all levels of production to achieve instant data sharing; secondly, integration of upstream and downstream industrial chain systems is necessary to make the internal personnel and material coordination organization of enterprises more orderly, thereby improving production efficiency. The Industrial Internet of Things can also integrate market planning and supply chains. Through MRP, ERP, MES and other management systems, enterprises can control internal and external resources more comprehensively, dynamically and instantly. Finally, blockchain and 5G technology can be applied to make the internal process of enterprises tighter in order to meet the allocation requirements of different departments, such as research and development, planning, procurement, production, transportation, after sales and others. Explicit demand and distribution can facilitate relatively distributed management between enterprise departments.

3.3. The Impact Mechanism of 5G Technology on the Innovation Activities of Tertiary Industry

The effect of 5G on e-commerce activities directly and effectively promotes the development of tertiary industry, producing a considerable spillover effect. According to a study by Zhen et al. (2021), for every percentage point of economic growth in e-commerce, fixed benefits have a 0.13% spillover effect on overall GDP growth [18]. China’s classification of tertiary industry is defined as industries other than primary and secondary industries. In view of its universal characteristic, the innovative contributions of 5G to cultural media, publishing, consumption, cloud education and other fields are particularly prominent. First, the acceleration of network transmission speed can provide producers in tertiary industry with richer and more personalized products and services. Currently, producers of products include enterprises or even individuals, for example, the self-media industry allows anyone become a producer of cultural products. Secondly, efficient information circulation promotes new products in tertiary industry with improved relevance with respect to market demand in the early design stages, prompting the supply side to deepen transformation and upgrading in response to market demand, which in turn promotes the introduction of products and services of higher quality. Third, the digital characteristics of culture and education industries are becoming increasingly obvious, which was highlighted following the outbreak of the COVID-19 pandemic, resulting in large-scale factory shutdowns and school closures. The promotion of cloud education platforms, such as cloud classrooms and MOOCs, reduced the impact of the pandemic on the education industry relative to that on industrial factory production. Fourth, 5G technology combined with artificial intelligence and virtual reality promotes expanded application scenarios and technologies; for example, museums in various countries around the world, as well as other cultural industries, have successfully provided online VR experiences, such as AI restoration of cultural relics, “letting national treasures speak”, and other new experiences; film and television, online shopping, games and other industries have successively launched intelligent immersive experiences, which have considerably enriched people’s cultural and entertainment needs. The tourism industry has also produced a large number of VR travel programs. Fifth, with the introduction of a new generation of communication technology, the original immature and imperfect technology of the 4G era has been considerably improved, e.g., 4K and 8K live broadcast mode with higher-definition video transmission. CCTV implemented the first 5G+4K live broadcast during the military parade for the 70th anniversary of the founding of the People’s Republic of China. During the 2022 Winter Olympic Games held in Beijing, all stadiums achieved full coverage with a 5G network and adopted 5G+8K technology for global live broadcasting. Sixth, the cost of using 5G networks is cheaper than previous generations. On the one hand, 5G promotes further development of mobile terminal networks; on the other hand, it can reduce the circulation costs associated with culture-, science- and education-related industries, facilitating more efficient dissemination of information. Seventh, 5G and artificial intelligence in combination with the Internet of Things can promote the development of ecological industry chains. New technologies can be developed and adopted in the service industry, such as the application of various intelligent service robots. The rapid development of smart cloud homes, smart cities, smart scenic locations, etc., also benefits from the development of artificial intelligence and communication technology. Ninth, in the intelligent era, the innovation activities of industrial markets and organizations develop rapidly. Instant information exchange will certainly promote the transformation and upgrading of production and sales models, promoting the intelligence of industrial chains, thereby increasing the volume and structure of the industry. On the other hand, a multiplatform linkage model can better meet the cultural, consumption, and service needs of consumers. The protection of intellectual property rights by big data can encourage individuals to to provide intellectual products and accelerate industrial development. Immersive consumer experiences can considerably enrich consumers’ sense of experience by precisely matching consumer groups. Finally, 5G technology promotes free business models, such as video live broadcasting. 5G’s high dissemination speed has considerably lowered the transmission and production costs of high-definition video and live broadcasting, which may in turn gradually reduce or even eliminate value-added service fees. Instead, merchants are likely to adopt other profit models, such as advertising.

3.4. The Relationship between the EV Industry and the Three Major Industries

The EV industry constitutes a standard secondary industry (i.e., manufacturing); however, with the penetration of 5G technology into the EV sector, technological breakthroughs have been made with respect to autopilot. [7] Together, 5G and autopilot are driving change in the EV business model, introducing characteristics of the service industry, such as with intelligent network operation of service providers [19]. In [20], a total 58 family interviews were conducted; the results showed that with the goal of contributing to environmental sustainability, sharing of electric vehicles was promoted in Oslo, Sweden, whereas in Rotterdam, the Netherlands, focus was placed on the improvement of charging infrastructure with respect to access to different traffic modes. According to statistics on car-sharing projects in several countries in the Americas [21], from 1998 to 2017, 94 car-sharing projects were developed in North America, with nearly 50% reaching the operation stage. As of 1 January 2018, 18 car-sharing projects were active in Canada, 21 in the U.S. and 1 in Mexico. Yang et al. studied the EV user sharing problem through guidance and travel path modeling methods [22].

With respect to primary industry, the global large rural market requires a large number of agricultural machines. However, rural land is vast, so parking is not a problem as it is in cities; every family in the countryside has space to install a home charging station, and farmers can use a slow charging station at home, charging vehicles during periods of low peak electricity, which would not only solve the problem of uneven loads during high peak electricity but could also reduce charging costs. In summary, with the popularity of 5G, the EV industry has the characteristics of three major industries: agriculture, industry and service.

3.5. Research on EV Operating Model

Hypothesis formation and model establishment:

According to the research of Lee et al. (2020) [23] exploring the latest EV developments, the following assumptions were made with respect to EV operation mode:

Hypothesis: The relationship between EV operation mode and weekly charging frequency (f_w) in the 5gRTS-ET platform is:

H1: Private EV mode:

$$0\le f_w < 5$$

(1)

H2: Operating EV mode:

$$f_w\ge 5$$

(2)

The number of days a year (365)/the number of days a week (7)/the number of months per year (12) = 4.35.

Therefore, the monthly charging frequency, f_m, during the year is:

$$f_m = 4.35f_w$$

(3)

In combination with (3), Formula (1) becomes:

$$0< f_m < 22$$

(4)

In combination with (3), Formula (2) becomes:

$$f_m\ge 22$$

(5)

The frequency density and group correlation of 1,048,575 cases were statistically analyzed according to Equations (4) and (5).

The charging frequency of each EV in a month was generated by grouping statistics of all charging order data. A monthly charging frequency density distribution diagram of 17,610 EVs was calculated, and normal distribution analysis was performed to analyze the monthly charge frequency of each EV ID. Figure 2 shows the ID number of 17,610 EVs in terms of charging frequency density for one month. The most frequent charging distribution is in the range of 50–150, accounting for 59.68% of the total. Figure 3 shows a charging frequency density distribution map for typical EV operating modes in 2021; the most common charging frequency is 1, which occurs 3197 times in a month, accounting for 18.2% of the total distribution.

The reasons for this situation are as follows: (1) passing vehicles in other areas, (2) vehicles with family charging points. Because they account for a very small proportion of charging volume and have no impact on the business model of charging points, we did not conduct further research on such cases. Cases conforms to the normal distribution are emphasized here. Most EV charging frequencies are captured in the normal distribution. This is very important for studying EV operation modes.

The 17,610 EV charging frequencies were grouped according to Equations (4) and (5). We identified 4371 EVs with a monthly charging frequency of less than 22, accounting for 24.8%, defined as private cars. We identified a total of 13,239 with a charging frequency greater than or equal to 22, accounting for 75.2%, defined as operating vehicles. Figure 4 shows a comparison chart of EV operating models.

Figure 5 shows an EV charging frequency area chart for different operating models (class 21–26 is operating EVs); the charging frequency of operating vehicles includes a wide class, so it was necessary to further subdivide the data, starting from the positive skew point, 150, of normal distribution to 22 of Formula (5), with 22 as the interval. The complete mathematical model from 0 to150 is as follows:

Identified:

Class1 (private EVs):

$$0<f_m<22$$

(6)

Class2 (operating EVs):

$$f_m\ge 22$$

(7)

Take Class2 for an in-depth analysis model:

Class21 (unknown):

$$22\le f_m<44$$

(8)

Class22 (unknown):

$$44\le f_m<66$$

(9)

Class23 (unknown):

$$66\le f_m<88$$

(10)

Class24 (unknown):

$$88\le f_m<110$$

(11)

Class25 (unknown):

$$110\le f_m<132$$

(12)

Class26 (unknown):

$$f_m\ge 132$$

(13)

According to the mathematical model of Equations (4)–(10), the results of the in-depth analysis are as follows:

3.6. Research Results

As shown in Figure 6, except for private EVs and class 26, the frequency density is close to normally distributed, with class 21, class 22, class 23, class 24 and class 25 accounting for 72% of all frequencies.

As shown in Figure 5, operating EV classes 21 to class 26 account for most of the total charging frequency area.

The above operation modes are summarized in

Table 2. Summary of data analysis.

Class	fm	Frequency Amount	Frequency Percent %	Hypothesis Operating Model
1	<22	4371	24.8	Private EV
21	22–43	1861	10.6	Unknown
22	44–65	2856	16.2	Unknown
23	66–87	3704	21.0	Unknown
24	88–109	2918	16.6	Unknown
25	110–131	1330	7.6	Unknown
26	>132	570	3.2	Unknown
Total		17,610	100

.

3.7. EV Operating Model Data Validation

In this paper, 5gRTS-ET data were compared to verify their existence, and new EV operation modes were tested to obtain more accurate investigation results. Finally, EV operation modes were verified, as presented.

Table 3. Summary of verified EV operating modes.

Class	fm	Frequency Amount	Frequency Percent %	>60% EV Operating Model
1	<22	4371	24.8	Private EV
21	22–43	1861	10.6	Uber/Didi
22	44–65	2856	16.2	Uber/Didi
23	66–87	3704	21.0	Taxi
24	88–109	2918	16.6	Taxi
25	110–131	1330	7.6	Logistic distribution
26	>132	570	3.2	Logistic distribution
Total		17,610	100

shows a distribution map of frequency density for typical EV operating modes in 2021.

To further verify this conclusion, the Kolmogorov–Smirnov test (K-S test) was performed on monthly charging frequency distribution [24]. The K–S test is a method to test whether a single sample is subject to a certain presupposed distribution and is suitable for large data samples. The original assumption was that charging frequency follows a log-normal distribution. If H = 0, the null hypothesis cannot be rejected. If H = 1, the null hypothesis can be rejected. The log-normal distribution probability density function can be expressed as:

$$p\left(x\right)=\frac{1}{x\sigma \sqrt{2\pi }}{e}^{\frac{{\left(In-\mu \right)}^2}{2{\sigma }^2}}$$

(14)

$$\mu =In\left(E\left(X\right)\right)-\frac{1}{2}In\left[1+\frac{var\left(X\right)}{E{\left(X\right)}^2}\right]$$

(15)

$$\sigma =\sqrt{In\left[1+\frac{var\left(X\right)}{E{\left(X\right)}^2}\right]}$$

(16)

μ—mean of the lognormal distribution.

σ—standard deviation of the log-normal distribution.

E(X)—sample data mean; the expression is shown in Equation (15).

var(X)—standard deviation of the sample data; the expression is shown in Equation (16).

$$E\left(X\right)=\frac{{{\displaystyle \sum }}_{i=1}^n{x}_i}{n}$$

(17)

$$var\left(X\right)=\sqrt{\frac{{{\displaystyle \sum }}_{i=0}^n{\left(x_i-E\left(X\right)\right)}^2}{n}}$$

(18)

The K–S test was performed for a large number of sample data points, and the result shows H = 1, indicating that the null hypothesis that charging frequency follows a log-normal distribution can be rejected, thus proving that charging frequency do not follow a log-normal distribution (Figure 7).

As shown in Figure 7, EV charging frequency above 22 (Figure 3 shows classes 21–26) presents close to a normal distribution, so it is assumed that the charging frequency of the class obeys a normal distribution. If H = 0, the null hypothesis cannot be rejected. If H = 1, the null hypothesis can be rejected. The normal distribution probability density function can be expressed as:

$$f\left(x\right)=\frac{1}{\sigma \sqrt{2\pi }}{e}^{\frac{{\left(x-\mu \right)}^2}{2{\sigma }^2}}, x>0$$

(19)

$$\mu =E\left(X\right)$$

(20)

$$\sigma =var\left(X\right)$$

(21)

μ—mean of the normal distribution.

σ—standard deviation of the normal distribution.

E(X)—sample data mean; the expression is shown in Equation (17).

var(X)—standard deviation of the sample data; the expression is shown in Equation (18).

A large number of sample data can be substituted into Equations (20), (21), (17) and (18). The results of the calculations are μ = 76.65, σ = 26.338. The corresponding normal distribution probability density function is shown in the Formula (19).

$$f\left(X\right)=\frac{1}{\sigma \sqrt{2\pi }}{e}^{\frac{{\left(x-76.65\right)}^2}{2\ast {26.338}^2}},x>0$$

(22)

A K–S test was performed for a large number of sample data points of classes 21–26; the result shows H = 0, indicating that the null hypothesis that charging frequency follows a normal distribution cannot be rejected, proving that charging frequency follows a normal distribution. Figure 8 shows a comparison of the density function of normal distribution and the charging frequency of classes 21–26.

4. Conclusions

In the present study, we analyzed the innovation impact mechanism in primary, secondary and tertiary industries brought about by the development of 5G technology, as well as technological changes in the EV market resulting from the introduction of 5G. The results show that the introduction of 5G technology introduced features of the service industry to the electric vehicle market. 5G changed the original single manufacturing sales process with the introduction of the EV leasing model to acquire the right of use by paying a leasing fee on the basis of acquiring the ownership of an EV from a single purchase. 5G also promotes the replacement of electric energy for mobility vehicles and agricultural machinery in rural markets, introducing characteristics of the agriculture, manufacturing and service industries in the EV market.

A total of 1,048,575 cases and 17,610 EV IDs were obtained in the form of real-world operational data of the 5gRTS-ET platform. Based on these data, in the present study, we used the frequency density and normal distribution method to build an EV operation model for the 5gRTS-ET platform; furthermore, we applied the Kolmogorov–Smirnov inspection (K-S inspection) to test the validity of the sample.

Finally, three typical EV operation modes—private EVs, passenger EVs and logistics distribution EVs—were identified:

• Private EVs instead of walking account for 24.8% of the total EV use;
• Passenger EVs (Uber/Didi and taxi) account for 64.4%; and
• Logistics distribution EVs account for 10.8%.

We developed a research method to automatically identify the operating mode of electric vehicles using data from a 5G real-time electric vehicle and traffic platform. This result provides a reference for the operation of EVs in association with the 5gRTS-ET platform, as well as support data for the establishment models for the commercial operation of charging points.