A Mobile Application Platform Based on BP Artificial Neural

Network Model Algorithm

Juanping Shen and Guozhong Wang

Wenshan Power Supply Bureau of Yunnan Power Grid Co., Ltd, Wenshan Yunnan, China

Keywords: BP Artificial Neural Network Model Algorithm, Infrastructure Site Control, Mobile Application Platform.

Abstract: In order not to consider the application requirements of on-site control in infrastructure, which leads to the

unreasonable distribution of mobile application platforms, in order to meet the application needs of on-site

control of mobile application platforms, this paper studies the reasonable distribution of on-site control

applications in infrastructure. First of all, the number of mobile application platforms for infrastructure site

control in the research area is calculated on the basis of the redivision of land types, and the application

demand coefficient of mobile application platforms is calculated. Then the BP artificial neural network model

algorithm is proposed to enrich the diversity of the population, and the non-linear convergence factor is used

to prevent the algorithm from falling into the local optimal, and complete the reasonable distribution research

of the application of the field control in infrastructure. Simulation experiments prove that the proposed method

in this paper can effectively meet the application needs of infrastructure site control mobile application

platform and the convenience needs of infrastructure site control mobile application platform.

1 INTRODUCTION

In recent years, the consumption of some fossil fuels,

such as oil and natural gas, has led to an increasingly

serious energy crisis and environmental problems

(Cheng, and Hu, et al. 2022). The traditional

infrastructure site uses fossil fuels, while the mobile

application platform is different from the traditional

infrastructure site, which has unique development

advantages, and can convert chemical energy into

electric energy stored in the rechargeable control pool

group (Gulcu, 2022). In order to respond reasonably,

the goal of carbon neutrality proposed by China is to

establish a low-carbon and efficient energy system

with electric energy as the main core, and to

vigorously promote the infrastructure on-site control

of mobile application platform is a very important

link (Han, 2023). Different from the traditional

infrastructure site, the endurance of the mobile

application platform for infrastructure site control is

relatively poor. Therefore, it is necessary to set up the

mobile application platform for infrastructure site

control in each city (Li, and Ding, et al. 2022).

However, due to the limitations of buildings in

different cities and the differences in different

demands, the reasonable distribution of mobile

application platforms has the corresponding

randomness and non-uniformity (Liu, and Xue, et al.

2022). For example, if the power supply line between

the infrastructure field mobile application platform

and the power dispatching center is not properly

planned, it will increase the construction cost of the

mobile application platform and the loss generated in

the power transmission process (Wu, and Zeng, et al.

2023). Therefore, it is of great research significance

to rationalize the distribution of infrastructure to the

on-site control of the mobile application platform,

improve the application efficiency of the mobile

application platform and reduce the construction cost

(Zhang, and Wang, et al. 2022).

2 RELATED WORKS

The rationalization of the distribution of mobile

application platform for on-site control needs to

consider many factors such as power grid,

transportation and economy, which is a non-linear

optimization problem. At present, many experts and

scholars have conducted in-depth research on this

problem (Zhang, 2022). The reasonable distribution

of the mobile application platform based on the

200

Shen, J. and Wang, G.

A Mobile Application Platform Based on BP Artiﬁcial Neural Network Model Algorithm.

DOI: 10.5220/0013538300004664

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Futuristic Technology (INCOFT 2025) - Volume 1, pages 200-205

ISBN: 978-989-758-763-4

simulated annealing algorithm is proposed. First of

all, the application of on-site control is in line with the

impact on the distribution network and the reasonable

distribution of the mobile application platform on the

infrastructure site, and analyzes the penetration rate at

different moments and different loads (Zhang, and

Wang, 2023). The application load of on-site control

will have different effects on the offsets of the

distribution network voltage. According to the

queuing theory, the waiting time of the infrastructure

field control mobile application platform and the

utilization rate of the mobile application platform are

taken as the satisfaction function, and the optimal

number of infrastructure field mobile application

platforms is obtained (Zheng, 2022). The simulated

annealing algorithm is used to obtain the optimal

distribution of the mobile application platform.

According to the simulation experiment results, the

mobile application platform can effectively reduce

the impact of the site control application load on the

voltage of the distribution network system, but this

method does not improve the utilization rate of the

mobile application site control platform. Put forward

a new analysis method, first of the construction site

control mobile application platform construction

factors comprehensive analysis, the construction site

mobile application platform operators of the early

stage of the investment economic cost and late

economic cost of operation and maintenance, and

construction site control mobile application platform

application waiting time cost minimum as a target

function to build construction site control mobile

application platform distribution model. Analyze the

quantity configuration of mobile application platform

under the known demand mode of infrastructure site

control application. Taking a commercial district in a

city as an example, the analysis results show that the

comprehensive cost of the operators of the mobile

application platform installed in the commercial

district is low, and the utilization rate of the mobile

application platform in the infrastructure site is also

high, but the method has the overall process. The

reasonable distribution method of mobile application

platform based on multi-stage collaborative planning

is proposed. Considering the timing characteristics,

the infrastructure site control distribution and load

power consumption planning model are constructed.

Taking the satisfaction and carbon emission of mobile

application platform as the target function, build an

infrastructure field mobile application platform

planning model for data envelope analysis. Taking a

system composed of a transportation network and

distribution network as an example, the free search

algorithm is used to solve the nonlinear optimization

problem, and get the optimal configuration scheme of

each stage. The results of the simulation experiment

show that the proposed method can achieve the

reasonable distribution goal of the mobile application

platform, but there is the problem of equipment

redundancy in the early stage.

3 DISTRIBUTION ANALYSIS OF

MOBILE APPLICATION

PLATFORM FOR ON-SITE

CONTROL OF

INFRASTRUCTURE

3.1 Demand Analysis of Mobile

Application Platform

Based on the analysis of the application

characteristics of on-site control, the application time

of on-site control in large application power and small

application power within one day is found, which is

expressed as:（1）



)

In the formula, it means that the power obtained

by the infrastructure site control in the corresponding

application mode, and Pi represents the power of the

infrastructure site control application.

Through the daily average service time of high-

power mobile application platform and low-power

mobile application platform, the number of mobile

application platforms required for infrastructure on-

site control can be calculated. Because in real daily

life, the application behavior of infrastructure on-site

control will be concentrated in a certain period of

time, resulting in a large number of infrastructure on-

site control with application needs in the same period

of time. Therefore, in order to meet the timely

application needs of infrastructure site control, when

clarifying the number of mobile application platforms

for infrastructure site control, it is necessary to

comprehensively consider the simultaneous

application of a large number of infrastructure site

control, and test and correct the number of mobile

application platforms.

Each city will be divided into many

administrative areas, at the same time the

administrative area economic development situation

also different, economic indicators to reflect the

A Mobile Application Platform Based on BP Artiﬁcial Neural Network Model Algorithm

201

number of the infrastructure site control, therefore,

according to the administrative area GDP in the city

in the proportion of GDP calculation for the

administrative area infrastructure site control the

number of mobile application platform. The number

of mobile application platforms for infrastructure site

control required by each administrative region is

expressed as:（2）

GDP

(2)

In the formula, N i represents the number of

mobile application platforms in the i administrative

region, and N indicates the total number of mobile

application platforms in the infrastructure control in

the city.

3.2 Division of urban areas

The location accuracy of the mobile application

platform mainly depends on the rationality of the

urban area division. The division of urban land types

is applicable to urban construction and planning, but

not applicable to power grid planning. According to

the characteristics of electric power and the load

density, the urban land types are divided into seven

types. They are housing, government, culture,

medical and education, industrial and commercial

land, public facilities and municipal transportation,

and green waters. These seven types of urban land use

not only consider the diversity of land use types, but

also conform to the planning of the power grid.

3.3 On-site control application demand

After the division of urban areas, the demand

coefficient of infrastructure site control application in

each region is calculated according to the

characteristic index of normalized attributes and the

traffic congestion index.

There is a direct relationship between the attribute

characteristic index of urban divided areas and the

region type, while different regional types need to

correspond to different application needs. The larger

the attribute characteristic index, the higher the

application needs. The attribute characteristic index

of the social public application network is shown in

Table 1.

The size of the traffic congestion index can

indicate the size of the traffic flow. The larger the

traffic congestion index, the greater the traffic flow,

and the higher the application demand. The traffic

congestion index of different roads should be

averaged over the same time period.

Table 1: The attribute feature index of the divided

regions

Area T

e Characteristic Index

House 0.1

Cultural And Medical

Education

0.4

Industrial Business

District

0.7

Government 0.1

Municipal Traffic 1

Communal Facilities 0.4

Green Wate

0.3

The demand coefficient of infrastructure site

control application in each divided area is expressed

as:（3）

pwpwp

2211

)

In the formula, i is the number of the divided area,

indicating the attribute characteristic index of the area

where the mobile application platform is located, and

indicating the traffic congestion coefficient of the

area where the mobile application platform is located.

The w 1, w 2 indicates the index weights.

4 RATIONALIZED

DISTRIBUTION OF

INFRASTRUCTURE FIELD

CONTROL AND MOBILE

APPLICATION PLATFORM

BASED ON BP ARTIFICIAL

NEURAL NETWORK MODEL

ALGORITHM

Combined with the above comprehensive analysis

results of the reasonable distribution of the

infrastructure site control mobile application

platform, the BP artificial neural network model

algorithm is used to plan the reasonable distribution

of the infrastructure field control mobile application

platform.

Suppose that the search space of the mobile

application platform on the infrastructure site is D

dimension and N represents the Wolf pack composed

of gray wolves. In the process of tracking and

surrounding the prey, the mathematical model of

controlling the distribution of mobile application

INCOFT 2025 - International Conference on Futuristic Technology

202

platforms on the infrastructure site is expressed as

follows:

In the above formula, t represents the number of

iterations, and the vector of the representation

coefficient, represents the position vector of the target

prey, and represents the position vector of the gray

Wolf. The coefficient vector can be defined as:

In the above formula, representing the factor of

convergence, the number of accompanying iterations

decreases from two to zero.

After the pack locks in the position of the target

prey, the pack surrounds the prey. Generally, the

leader of a Wolf pack will have a better understanding

of the potential location of the target prey, so he can

judge the location of the prey according to their own

location, and update their position at the same time,

and gradually approach the target prey. The

mathematical model of the hunting behavior of

wolves is expressed as:（4）

DCXX

αα

=•−

   

)

The initial population is introduced into the BP

artificial neural network model algorithm. The

chaotic map is a piecewise linear map, and the

function can be expressed as:（5）

,0 u

≤<

(5)

Chaos maps have a uniformly distributed

sequence and can have the same distribution density

for different parameters, thus referencing u=

The 1 / 2 chaos mapping formula, expressed as:（

6）

x≤<

(6)

The nonlinear decreasing method is adopted to

improve the convergence factor. The decline speed in

the early stage is slow, the large convergence factor

can effectively enhance the global search ability,

prevent falling into the local optimal solution, the

convergence speed in the later stage is slow, and the

small convergence factor can enhance the local search

ability, so as to accelerate the convergence speed of

the algorithm. This nonlinear convergence

mechanism can reasonably coordinate the local and

global search ability of BP artificial neural network

model algorithm, and the convergence factor A value

is adaptive. The improved convergence factor is

expressed as:（7）

*22

max

−

−=

)

In the formula, k is the adjustment coefficient, the

specific value determines the speed of the

convergence factor with increasing the number of

iterations, and t is the current number of iterations, t

ｍ

ａｘ

Represents the maximum number of iterations.

Different weights are used to refer to the location

information of the Wolf pack leaders. The higher the

level, the greater the reference weight. Random

disturbance is added to avoid falling into the local

optimal solution, which is expressed as:（8）

10632

321

randnXXX

tX +++=+（

)

The above process will complete the reasonable

distribution of the infrastructure field control mobile

application platform based on BP artificial neural

network model algorithm. Figure 1 shows the whole

process of BP artificial neural network model

algorithm.

Figure 1: Network structure of the BP neural network

5 EXPERIMENTAL RESULTS

In order to verify the validity of the reasonable

distribution of the mobile mobile application platform

based on the BP artificial neural network model

algorithm proposed in this paper, the simulation

experiment is carried out in the Matlab simulation

A Mobile Application Platform Based on BP Artiﬁcial Neural Network Model Algorithm

203

environment. Table 2 shows the parameters of the

field control of the infrastructure.

Table 2: Parameters of infrastructure site control

Paramete

Numeric Value

Infrastructure Qualit

1610kg

The Rest Of The

Application

13.5kwh

Full Power A

lication 60kwh

Internal Temperature In

Infrastructure

24℃

Using BP artificial neural network model

algorithm will shortest total travel time, mobile

application platform cost and total travel time and

total cost of mobile application platform

rationalization distribution target, infrastructure site

control mobile application platform rationalization

distribution results expressed by Figure 2.

Figure 2: Flowchart of attribute reduction based on rough

set theory

Table 3: Reasonable distribution results information of

mobile application platforms

Optimization

Objective

The Total

Travel Time

Is The

Shortest

The Lowest Cost Of

The Total Cost For

Mobile Application

Platforms

Distance From

Drivin

5.86km 5.70km

The Time Of

The Total Tri

0.31h 0.43h

Initial

lication

10.39kwh 8.59kwh

Application

lication

13.59kwh 15.29kwh

Total Cost

Cost

Thirty-Eight

Point Four

Three Yuan

Twenty-Nine Point

Four Eight Yuan

The details during the rationalization of the

distribution of mobile application platforms under

different optimization objectives are presented in

Table 3.

Analysis table 3 can see, the infrastructure site

control total travel time and mobile application

platform overall optimal as a reasonable distribution

target, than the construction site control total travel

time as a reasonable distribution target shortest

construction site control total cost reduce 10.12 yuan,

than the construction site control mobile application

platform total cost as a reasonable distribution target

reduce 1 yuan, and the total travel time also reduced.

It can be seen that the method proposed in this paper

can effectively rationalize the distribution of the

mobile application platform for the field control of

infrastructure.

In order to compare the impact of infrastructure

on-site control on the power grid after access to the

mobile application platform under different

optimization objectives, the analysis results are

shown in Figure 3 .

Figure 3: The impact of on-site control of different

optimization objectives

Analyze of figure 3 can see that the construction

site control the shortest total travel time and

construction site control the minimum cost of mobile

application platform as infrastructure site control

mobile application platform rationalization

distribution of optimization target, relative than the

construction site control total travel time as a shortest

rationalization distribution target in the morning and

evening peak, the overall power will decrease, thus,

the construction site control the shortest total travel

time and infrastructure site control the total cost of the

lowest as a reasonable distribution optimization target

is more conducive to the stable operation of the power

grid. Figure 4 compares the optimal distribution

results of the proposed method and the traditional

method.

INCOFT 2025 - International Conference on Futuristic Technology

204

Figure 4: The optimal distribution results of the proposed

method

From the analysis of Figure 4, it can be seen that

the curve convergence of the reasonable distribution

length of the mobile application platform of the

proposed method is 39.1m, while the curve

convergence of the reasonable distribution length of

the traditional method is 41.1m, indicating that the

convergence rate of the proposed method is fast. This

is because the traditional method is easy to fall into

the local optimal solution, and the method proposed

in this paper is to use the chaotic mapping to produce

the initial solution of the population, so as to enrich

the diversity of the population, by introducing the

non-linear convergence factor and adding the

disturbance position formula to prevent the algorithm

from falling into the local optimal, and increase the

convergence speed of the later algorithm. This shows

that the overall stability of the method proposed in

this paper is good, and can be better applied to the

reasonable distribution of mobile application

platform on infrastructure control.

6 CONCLUSIONS

Driven by the low-carbon target, because

infrastructure site control is better than traditional

infrastructure sites in terms of greenhouse gas

emissions and energy consumption, infrastructure site

control is considered to be a very promising

sustainable transportation mode in the future.

However, with the rapid development of on-site

control, there are still some problems from the

perspective of mobile application platform and

mobile application platform. Therefore, this paper

carries out in-depth research on the reasonable

distribution of on-site control of mobile application

platform. First of all, through economic indicators to

research area can calculate the infrastructure site

control mobile application platform, on the land type

again on the basis of urban area, and the construction

site control mobile application platform application

demand coefficient calculation, and then use BP

artificial neural network model algorithm of

infrastructure site control mobile application platform

mobile application platform rationalization

distribution analysis. And the simulation experiments

prove that the proposed method is practical.

REFERENCES

Cheng, X. Y., Hu, X. P., Li, Z. Z., Geng, C. H., Liu, J. X.,

Liu, M., Zhu, B. K., Li, Q., and Chen, Q. G., Using

Genetic Algorithm and Particle Swarm Optimization

BP Neural Network Algorithm to Improve Marine Oil

Spill Prediction: Water Air and Soil Pollution, vol. 233,

no. 8,2022.

Gulcu, S., Training of the feed forward artificial neural

networks using dragonfly algorithm: Applied Soft

Computing, vol. 124,2022.

Han, L. M., Gao, H., And Zhai, R. J., Optimization Of

Machine Translation Model Based On Dboa-Bp Neural

Network: University Politehnica Of Bucharest

Scientific Bulletin Series C-Electrical Engineering and

Computer Science, vol. 85, no. 2, pp. 63-78,2023.

Li, Y., Ding, F., and Tian, W. J., Optimization of 3D

Printing Parameters on Deformation by BP Neural

Network Algorithm: Metals, vol. 12, no. 10,2022.

Liu, L., Xue, M. Y., Guo, N., Wang, Z. L., Wang, Y. W.,

and Tang, Q. X., Investigating the Path Tracking

Algorithm Based on BP Neural Network: Sensors, vol.

23, no. 9,2023.

Wu, S. X., Zeng, X. Y., Li, C. M., Cang, H. Z., Tan, Q. C.,

and Xu, D. W., CO<sub>2</sub> emission forecasting

based on nonlinear grey Bernoulli and BP neural

network combined model: Soft Computing, vol. 27, no.

21, pp. 15509-15521,2023.

Zhang, G. B., Wang, Y. H., Duan, Q., Huang, Y. M., Ma,

C. Y., Xu, R. B., and Chai, L., Evaluation Model of

Urban Smart Energy System Based on Improved

Genetic Algorithm-Bp Neural Network: International

Journal of Pattern Recognition and Artificial

Intelligence, vol. 36, no. 09,2022.

Zhang, G. N., Research on safety simulation model and

algorithm of dynamic system based on artificial neural

network: Soft Computing, vol. 26, no. 15, pp. 7377-

7386,2022.

Zhang, H. Y., and Wang, J., A Smart Knowledge Discover

System for Teaching Quality Evaluation via Genetic

Algorithm-Based BP Neural Network: Ieee Access,

vol. 11, pp. 53615-53623,2023.

Zheng, L., Research On Application Of Improved Genetic

Algorithm And Bp Neural Network In Air Quality

Evaluation: Fresenius Environmental Bulletin, Vol. 31,

No. 6a, Pp. 6043-6052,2022.

A Mobile Application Platform Based on BP Artiﬁcial Neural Network Model Algorithm

205