Available via license: CC BY 4.0
Content may be subject to copyright.
Scalable Computing: Practice and Experience, ISSN 1895-1767, http://www.scpe.org
© 2024 SCPE. Volume 25, Issues 6, pp. 5131–5141, DOI 10.12694/scpe.v25i6.3245
THE SECURITY AND PROTECTION SYSTEM OF ELECTROMECHANICAL
EQUIPMENT IN SMART CAMPUS USING THE IMPROVED DATA MINING
ALGORITHM
ANYUAN HE∗
Abstract. In order to improve the maintenance and management eciency of campus electromechanical equipment and
reduce or even avoid the safety risks brought by campus electromechanical equipment, this work uses the data mining algorithm
to design the security and protection system of campus electromechanical equipment. First, this work constructs the campus
electromechanical equipment classication model using the Bayesian algorithm of data mining algorithm and designs a simulation
experiment to verify the eect of the classication model. Then, the security and protection system for the campus electromechanical
equipment is designed. It includes the system business process, system function design, system core module’s function design and
system implementation. Finally, a simulation experiment is designed to verify the system’s performance. The results show that:
(1) Bayesian algorithm is superior to the K-Nearest Neighbor (KNN) algorithm in both classication eect and running time. (2)
When the browser concurrency in the system increases, the server processor and memory usage also increases, but the value meets
the expected requirements. It shows that the system has a certain browser concurrency-bearing capacity. Moreover, as the browser
concurrency of the system increases, the response time of the test also increases, but the value meets the expected requirements.
This work aims to improve the maintenance eciency of campus electromechanical equipment and provide a reference for the safety
protection work of electromechanical equipment in other enterprises or units.
Key words: data mining, campus electromechanical safety, Bayesian algorithm, K-Nearest Neighbor algorithm, simulation
experiment
1. Introduction. Foreign developed countries have studied the informatization of equipment management
since the 1970s. At present, the equipment management system has been quite popular abroad, especially in
the United States. Many public places, such as hospitals, schools and enterprises, have used information
technology to varying degrees, and the management of instruments and equipment is quite standardized [1].
The Asset Management Solution equipment operation management system developed by Fisher-Rosemount
is relatively complete [2]. The Maximo equipment management system developed in the United States takes
preventive maintenance as the means and cost management as the core to achieve economic optimization
and benet maximization. Currently, the system has been applied worldwide, with over 5000 installations
in major enterprises, especially in safety, education, aerospace, medical, and machinery manufacturing [3].
Moreover, foreign scholars consider using the knowledge base to summarize common faults, thereby helping users
automatically nd and solve faults. Such systems are often referred to as expert systems [4]. The early expert
system is adopted to help medical sta make medical diagnoses and help managers make decisions on problems.
They usually need to establish a perfect knowledge base and inference rules to give corresponding conclusions
according to the problem phenomenon. This kind of system has been widely used in fault management systems
worldwide.
China starts relatively late in the research of equipment information management. Since the 1980s, in
response to enterprise development needs, some domestic scholars have conducted research on equipment man-
agement informatization. Over the years, enterprises and universities in some regions of China have developed
lots of excellent equipment management software, which has improved the eciency of xed asset management
and made certain technological progress. The software all have the functional modules required by general
enterprises to manage equipment and instruments, including basic equipment information management, equip-
ment bidding and purchasing, equipment operation and maintenance prompts, equipment repair forecast and
∗Information Technology Center, Nanjing University of Aeronautics and Astronautics, Nanjing, 210016, Jiangsu, China.
(AnyuanHe9@163.com)
5131
5132 Anyuan He
repair cost estimation, equipment depreciation and scrapping, which basically achieve automatic management
[5]. For example, Guangzhou Zhengtai PMISS.0 equipment integrated management system can realize the inte-
grated management of equipment information [6]. HSWZ001 construction equipment management software of
Tsinghua Sware Company computerizes the establishment of equipment archives account, account accounting,
equipment maintenance and equipment scrapping [7]. The College Teaching and Scientic Research Instrument
and Equipment Management System developed by the Beijing University of Chemical Technology, which many
colleges currently use, can complete the accounting management of equipment and the task of reporting various
data required by the Ministry of Education [8]. Besides, Zhao et al. (2021) pointed out that the equipment
manager can use the electronic information platform to access the equipment data with the highest eciency
through the establishment of electronic equipment drawings, technical manuals, and parameter report archives
for the convenience of equipment repair and maintenance. Moreover, this kind of electronic archive is easier
to preserve than paper materials [9]. Huazhong University of Science and Technology has developed an equip-
ment information management system based on Java EE technology architecture on the basis of traditional
architecture. Java EE is a crucial development platform directly built based on the Java 2 standard [10].
In summary, the current equipment management informatization in China has just started. Moreover,
equipment management in China is deeply aected by the traditional management mode. The equipment au-
tomation level of many schools and units is still decient, and they do not pay enough attention to equipment
information processing, so there is still a big gap compared with foreign countries. Based on this, this work
studies the safety protection of campus electromechanical equipment based on a data mining algorithm. The
innovation is to use the improved statistical classication algorithm of data mining algorithm - Bayesian algo-
rithm to classify and recognize campus electromechanical equipment. This algorithm has a good classication
eect. This work aims to improve the management level of the electromechanical equipment fault prediction
and maintenance plan generation system in colleges in the maintenance process to reduce or even avoid the
security risks brought by campus electromechanical equipment.
2. Theoretical basis and model design.
2.1. Research on classication model of electromechanical equipment based on Bayesian.
(1) Bayesian classication. Bayesian classication is a classication method based on the classical Bayesian
probability theory using the knowledge of probability statistics [11]. Its main idea is to predict the possibility
that an item to be classied belongs to each category through Thomas Bayes, and take the category with the
greatest possibility as the category of the item to be classied. Thomas Bayes is the conditional probability
of random events h and D:
P(h|D) = [P(D|h)P(h)]/P (D)(2.1)
P(h|D)refers to the conditional probability of event D when event h occurs. It indicates the probability
of event h under the premise that event D has occurred [12].
The Bayesian classication covers many types of algorithm models, of which naive Bayesian classication
algorithm and Bayesian network are two commonly used. Naive Bayes classier (NB algorithm), which is a
supervised learning method, is a simple and eective classier. In some application elds, its performance is
comparable to that of articial neural networks and decision trees. The algorithm assumes that attributes are
independent based on Bayesian probability. This assumption greatly reduces the construction complexity of
the Bayesian application model and makes it suitable for classication tasks in data mining. Meanwhile, this
assumption also limits the application scope of the algorithm to some extent [13, 14].
(2) Bayesian-based equipment classication model. The specic goal of the classication management of
campus electromechanical equipment is to maximize the accuracy of equipment classication for existing or
new campus equipment and minimize the time spent on equipment classication.
Thereby, the NB algorithm (naive Bayesian classication algorithm) is used as the classier of campus
equipment to classify and manage thousands of school equipment [15]. Figure 2.1 displays each module of the
classier and the specic classication process.
Figure 2.1 suggests that the overall classication process of the equipment management system can be
divided into three modules.
The Security and Protection System of Electromechanical Equipment in Smart Campus 5133
Fig. 2.1: Each module of the classier and the specic classication process (a) Text classier module diagram;
(b) Classication process of the equipment management system
The rst module is the pretreatment module of classier construction. The main goal of this module is to
select the characteristics that can best represent the equipment category from the numerous electromechanical
equipment lists of the school. First, each item of data on the equipment list is input as a set of text data
to be classied. Then, according to the classication problem’s specic situation, the set’s feature attributes
(these attributes are the attributes that can describe the equipment features) are selected. The classied feature
attributes are appropriately taken as the feature attribute set of the classier to be constructed in the next
step. Moreover, it is also essential to randomly select a part of the data to be classied from all the unclassied
data object sets to be classied as the sample set, and manually classify the sample set, so that the training
sample set is obtained after classication.
The above description reveals that this stage requires manual classication of the training sample set, which
is the only part of the NB algorithm that requires manual implementation. This result will greatly impact the
construction of the subsequent classication model. It can be said that the quality of the relevant equipment
feature attribute set output by this module and the training sample set obtained through manual classication
determines the classication quality of the constructed NB algorithm. This module is the basis of the NB
algorithm.
The second module is the classier training module. This module aims to generate the classier of instances,
so this module is also the core of the entire NB algorithm. At this time, the classier calculates the occurrence
frequency of each category in the training samples manually classied previously. The prior probabilities of
each feature attribute corresponding to each category are also calculated together, and then these probabilities
are recorded for the next classication stage. In this stage, the input is the feature attribute set and the training
sample set, and the output is the classier.
The third module is the classication module. The main goal of this module is to use the classier produced
in the second stage to classify other object data to be classied except for the training sample set. The working
objects of this module are the constructed equipment classier and other data objects to be classied. The
data object to be classied has a functional mapping relationship with its category, which is the nal output.
5134 Anyuan He
(3) Bayesian-based mathematical model for equipment classication. The classication of campus electrome-
chanical equipment is briey described as follows. A large number of teaching equipment purchased by the
school can actually be classied into dierent equipment categories. For example, according to the category
of electromechanical equipment, they can be divided into electronic products and instruments. They can be
divided into special equipment and general equipment according to the department using the equipment. They
can be divided into consumables and low-value consumables according to equipment cost. Therefore, the above
equipment classication problem can be described by the following formalized mathematical model:
1. x={a1, a2, a3, , am}is set as a campus device to be classied. Each a is a feature attribute of x used
to describe the characteristics of this device, such as projection, computer, and optical ber.
2. There is equipment category set C={y1, y2, y3, , yn}, such as electronic products and instruments,
special equipment and general equipment.
3. The conditional probability P(y1|x), P (y2|x), , P (yn|x)of the equipment to be classied for each equip-
ment classication is calculated.
4. If P(yk|x) = max{P(y1|x), P (y2|x), , P (yn|x)},x∈yk.
The above problem description shows that the key to solving the problem now is how to calculate the
conditional probability of the device object to be classied for each existing device classication in step 3.
The process of step 3 can be divided into the following sub-processes to solve this problem:
1) From all the equipment sets to be classied, it is necessary to randomly nd a subset of the equipment objects
to be classied, and manually classify them into each equipment category. The set thus obtained is
called the training sample set.
2) The conditional probabilities of the characteristic attributes a1, a2, a3, , amof each device corresponding to
each device category y1, y2, y3, , ynare calculated, counted and estimated.
3) If the feature attributes of each device are conditionally independent, the following deduction can be made
according to Thomas Bayes:
P(yi|x) = P(x|yi)P(yi)
P(x)(2.2)
For all equipment categories, the denominator is equivalent to a constant, and the numerator can be
maximized. It is assumed that each feature attribute is conditionally independent, so the following
equation is obtained:
P(x|yi)P(yi) = P(yi)
m
∏
j=1
P(aj|yi)(2.3)
Hence, the classication of campus equipment can be solved through the above mathematical calcula-
tion logic. Figure 2.2 is the architecture of the electromechanical equipment classication mathematical
model based on this.
(4) Design of simulation experiment.
Naive Bayesian and K-Nearest Neighbor (KNN) classication algorithms are the most commonly used in
text classication. They are both relatively simple and eective classication algorithms. The two algorithms
have the same advantages and disadvantages in most text classication experiments. In this experiment, the
two algorithms are applied to the actual equipment classication, and the common equipment data in the
school equipment inventory are adopted to compare the classication accuracy and time eciency. According
to this, this experiment selects a college as the research object and conducts a simulation experiment design
for 200 pieces of electromechanical equipment owned by it. In the experiment, 100 pieces of electromechanical
equipment are randomly selected as the training sample set, and the equipment names in the training sample set
are manually sorted and classied. The training sample set is divided into six categories: consumables, low-value
consumables, instruments and meters, electronic products, special equipment, and general equipment.
2.2. Design and research of campus electromechanical equipment management system. The
last section is to use the Bayesian algorithm in the data mining algorithm to design and research the campus
electromechanical equipment classication model. This section will further design the security and protection
The Security and Protection System of Electromechanical Equipment in Smart Campus 5135
Fig. 2.2: The mathematical model
system of campus electromechanical equipment. It aims to realize fault prediction and intelligent maintenance
of campus electromechanical equipment through the system.
2.2.1. System business process analysis.
1) Business process of equipment routine check management. Figure 2.3 presents the designed business pro-
cess of routine equipment checks according to college’s current management characteristics of electromechanical
equipment.
Figure 2.3 suggests that the participants in the process of routine equipment checks include the equipment
supervisor and the equipment administrator. Equipment routine check refers to the periodic inspection and
maintenance of various equipment parameters in equipment management [16]. The equipment supervisor assigns
routine check business to specic equipment administrators in routine equipment checks. The equipment
administrator needs to set the routine check object, add specic routine check parameters, set specic check
cycles, process the routine check data eectively, and nally view the corresponding check report.
2) Process analysis of fault declaration and maintenance management business. Figure 2.4 shows the
designed fault declaration and maintenance business process.
Figure 2.4 suggests that in the fault declaration and maintenance management business process, the equip-
ment user department rst applies for the equipment fault, then lls in the corresponding application form and
submits it to the equipment management department for declaration and approval. The approved information
shall be fed back to the equipment user department to formulate a maintenance plan for the failure. The failure
repair needs to be approved and submitted to the equipment repair department. Then, it is submitted to the
repair and material preparation department. After the repair is completed, it needs to be submitted to the
equipment user department for acceptance and settlement.
3) Consigned processing business management process. Consigned processing refers to directly delegating
the corresponding equipment maintenance work to the relevant third-party units for specic equipment mainte-
nance when the corresponding technical diculties or the maintenance department’s maintenance workload is
too large to complete the corresponding maintenance tasks on time [17]. In the process of consigned processing,
the user department of the equipment rst makes a repair plan for the equipment, and then the equipment
management department approves the plan. After that, the equipment maintenance department proposes
corresponding consigned processing suggestions based on the repair task.
5136 Anyuan He
Fig. 2.3: The business process of routine check management
Fig. 2.4: The process of fault declaration and maintenance management business
2.2.2. System functional requirements analysis.
1) Analysis of routine check management requirements. Equipment routine check management is a signi-
cant module in the management system of college equipment fault prediction and maintenance plan generation.
It mainly assists in the management of basic equipment routine check information during the college equipment
failure prediction and the generation of maintenance plans, including the management of routine check position
information, equipment routine check object information, equipment routine check data entry management,
The Security and Protection System of Electromechanical Equipment in Smart Campus 5137
Fig. 2.5: Flow chart of fault prediction and early warning management
and routine check data maintenance.
2) Demand analysis of failure maintenance management. Fault maintenance management is an important
module in the management system of college equipment fault prediction and maintenance plan generation. It
is mainly used to assist in the management of basic fault repair information during the college equipment
fault prediction and maintenance plan generation, including the management of account information in the
repair process, the management of consigned processing settlement information, the management of consigned
processing order information, and the management of acceptance processing information.
3) Demand analysis of fault prediction and early warning management. The risk early warning management
of failures is a crucial module in the management system of college equipment fault prediction and maintenance
plan generation. This module is mainly adopted to predict, warn, analyze and manage equipment failures in
the process of equipment failure prediction and maintenance plan generation in colleges. The failure risk in the
college equipment failure prediction and maintenance plan generation can be avoided in advance by using the
model and algorithms.
2.2.3. Function design of main modules of the system.
1) Fault maintenance management design. The fault repair management module rst needs to process the
newly added fault repair application form, and then click the information to save it. If it fails to save, it needs
to submit it again for saving. For those saved successfully, it is essential to go directly to the next step to
approve the application form, and then apply for conrmation of fault repair. Only those approved can apply
for fault repair. Otherwise, the application needs to be submitted again.
2) Design of fault prediction and early warning management. In order to complete the prediction of equip-
ment failures and the generation of maintenance plans in colleges, it is necessary to establish a perfect risk
assessment system and information-based evaluation method. Figure 2.5 is the designed ow chart for fault
prediction and early warning management process.
2.2.4. System function realization. The platform of this system includes a hardware platform and a
software platform. Table 3.1 shows the specic composition of the platform.
5138 Anyuan He
Table 2.1: Composition of system hardware platform and software platform
Composition of system hardware platform
Platform classication Specic parameters/models
Processor Intel(R) Core(TM)2 Auad CPU Q9500 @2.83GHz
Memory 4GB
Composition of system software platform
Platform classication Specic parameters/models
Operating system Windows 7
Database SQL Server 2005
Development environment JDK6.0
Development tool Myeclipse10
Development language Java
Web server Tomcat 6.0
Table 2.2: Hardware and software test environment conguration of the system
System test hardware conguration
Name Parameter
Server IBM System X3550 M4
disk space: 600.0G; RAM: 8G;
Client Dell OptiPleax 3020
disk space: 400.0G; RAM: 2G;
System test software environment
Name Parameter
Client operating system Windows 10
Database management system SQL Server 2012
Application server Tomcat
Browser software Chrome
2.2.5. Simulation test of system performance. Table 2.2 shows the hardware and software test envi-
ronment of the system.
According to the system test environment given in Table 3.2, the nal performance test of the system is
required during the delivery of the nal system. After the corresponding tests, it can be conrmed whether
the system can be synchronized with the user’s requirements in the actual operation process. The main tool
used is Load Runner when testing the performance of the designed management system for college equipment
failure prediction and maintenance plan generation. The main indicator information in the performance aspect
mainly includes the capacity of the relevant business system in terms of load, some capabilities in terms of
the corresponding capacity, and the response time information of the specic business system. By running the
Load Runner tool, the system is tested in dierent time periods, and dierent access frequency levels (10/min,
100/min, 1000/min and 5000/min) are tested, dierent operation types (specic operations such as adding,
deleting, modifying and querying data information) are tested, and nally the indicator information of the
business system in the actual operation process is recorded.
3. Model and system performance test results.
3.1. Classication results of college electromechanical equipment based on the Bayesian algo-
rithm.
3.1.1. Comparison of equipment classication eects under dierent algorithms. Figure 3.1(a)
displays the specic distribution of training sample equipment data under each equipment category through
simulation experiments. x-axis is the equipment category and y-axis is the number of training samples. Figure
The Security and Protection System of Electromechanical Equipment in Smart Campus 5139
Fig. 3.1: Equipment distribution results of training samples and comparison of equipment classication eects
under dierent algorithms (a) Equipment distribution diagram of training sample; (b) Comparison results of
equipment classication eects under dierent algorithms
3.1(b) shows the classication eect under the two algorithms.
Figure 3.1 shows that under the same sample set, the classication eects of the two classication algorithms
are signicantly dierent, and the accuracy of the Bayesian algorithm is higher. It suggests that the Bayesian
classication (NB algorithm) is obviously superior to the KNN algorithm in terms of equipment classication.
3.1.2. Comparison of running time required for equipment classication under dierent al-
gorithms. In order to further compare the classication eciency of the two algorithms, the running time of
the two algorithms for classication is compared based on the same equipment sample set. In the experiment
process, the method of gradually expanding the size of the equipment sample set is adopted, so that the running
time of the two classication algorithms under equipment sample sets with dierent sizes is obtained. Figure
3.2 presents the specic experimental comparison results.
In Figure 3.2, the x-axis represents the extent of the sample set increase, and the y-axis represents the
time spent running the algorithm. Figure 3.2(a) represents the KNN algorithm, and Figure 3.2(b) is the NB
classication algorithm. When the x-axis sample set value is smaller than 100, the running time of the two
algorithms is similar. Because of the simplicity of the algorithm, there is little dierence. With the increasing
number of sample sets, when the value reaches 2500, the running time of the KNN algorithm is signicantly
longer than that of NB. It shows that the NB classication algorithm is better than the KNN algorithm.
The above experimental results show that the Bayesian-based equipment classication model studied has
a shorter running time, higher classication accuracy and better performance than the KNN algorithm.
3.2. Performance test results of security and protection system for campus electromechanical
equipment. The system is tested according to the given hardware and software environment. Figure 3.3
displays the system performance.
Figure 3.3 reveals that when the browser concurrency in the system is 10, 100, 500, 1000, and 5000, the
server processor utilization reaches 23%, 33%, 49%, 54%, and 67%, respectively, while the memory utilization
reaches 23%, 43%, 56%, 64%, and 75%, respectively. It shows that the system has a certain browser concurrency-
bearing capacity. Moreover, as the browser concurrency in the system increases, the response time of the test
also increases, but the values are in the standard state. To sum up, the system has achieved the expected goal
through the experimental test, the degree of perfection of the whole system has passed the test, and the whole
system is in a stable operation state.
5140 Anyuan He
Fig. 3.2: Comparison of the algorithm running time (a) Running time of KNN algorithm; (b) Running time of
NB algorithm
Fig. 3.3: System performance test results (a) System processor and memory usage; (b) System response time
4. Conclusion. In order to achieve the ecient management of campus electromechanical equipment
security, this work uses the Bayesian algorithm in data mining technology to eectively classify the electrome-
chanical equipment in colleges. Then, the safety management system of electromechanical equipment in colleges
is designed. Finally, the simulation experiment is set to test the classication model and the system’s perfor-
mance.
The results show that: (1) the improved statistical classication algorithm - Bayesian algorithm can be used
to eciently classify electromechanical equipment. (2) The college electromechanical equipment safety manage-
ment system designed has good processing performance, and can realize the intelligent safety management of
electromechanical equipment. The research disadvantage is that the business of the actual college equipment
fault prediction and maintenance plan generation system is relatively complex, while the core function modules
of the college equipment fault prediction and maintenance plan generation system designed here are relatively
simple, and some complex functions cannot be realized. Thereby continuous improvement is needed in the
The Security and Protection System of Electromechanical Equipment in Smart Campus 5141
subsequent design. This work aims to improve the eciency of the department’s actual work, provide good
services for users, and provide benecial help for the intelligent safety management of campus electromechanical
equipment.
5. Acknowledgement. This study was supported by Jiangsu Province Modern Education Technology
Research Smart Campus Special Project (2021-R-96789).
REFERENCES
[1] Belousova M., Aleshko R., Zakieva R., et al. (2021) Development of equipment management system with monitoring of
working characteristics of technological processes[J]. Istrazivanja i Projektovanja za Privredu, 19(1), 1-7.
[2] Abu-Samra S., Ahmed M., Amador L. (2020) Asset Management Framework for Integrated Municipal Infrastructure[J].
Journal of Infrastructure Systems, 26(4), 04020039.
[3] Shankar K., Shyry S. P. (2021) A Survey of image pre-processing techniques for medical images[J]. Journal of Physics:
Conference Series, 1911(1), 012003.
[4] Merchant T., Hormozian S., Smith R. S., et al. (2022) Ethical Principles in Personal Protective Equipment Inventory Manage-
ment Decisions and Partnerships Across State Lines During the COVID-19 Pandemic:[J]. Public Health Reports, 137(2),
208-212.
[5] Khodadadi V., Bakrani A., Vafaie M. H. (2021) Factors Aecting Medical Equipment Management in the COVID-19 Pandemic
Crisis: A Mixed Qualitative and Quantitative Study[J]. Hospital Practices and Research, 6(1), 23-28.
[6] Zhao M., Pang L., Liu M., et al. (2020) Control Strategy for Helicopter Thermal Management System Based on Liquid
Cooling and Vapor Compression Refrigeration[J]. Energies, 13(9), 2177.
[7] Shao Q., Jia M., Xu C., et al. (2020) A support system for civil aviation navigation equipment security management[J]. Safety
Science, 123(2), 104578.
[8] Benson N. U., Fred-Ahmadu O. H., Bassey D. E., et al. (2021) COVID-19 Pandemic and Emerging Plastic-based Personal
Protective Equipment Waste Pollution and Management in Africa[J]. Journal of Environmental Chemical Engineering,
9(2020), 105222.
[9] Zhao X., Yan T., Li J., et al, (2021) A software design of the equipment management business in Digital Workshop[J]. Journal
of Physics: Conference Series, 1894(1), 012085.
[10] He Y., He Z., Zhang F., et al. (2022) Discussion on Management Strategy of Large Precision Equipment in Colleges and
University[J]. Open Journal of Business and Management, 10(3), 6.
[11] Long Y., Xu X. (2021) Bayesian decision rules to classication problems[J]. Australian And New Zealand Journal of Statistics,
63(2), 394-415.
[12] Dziendzikowski M., Heesch M., Gorski J., et al. (2021) Application of PZT Ceramic Sensors for Composite Structure Moni-
toring Using Harmonic Excitation Signals and Bayesian Classication Approach[J]. Materials, 14(19), 5468.
[13] Lou C., Li X., Atoui M. A. (2020) Bayesian Network Based on an Adaptive Threshold Scheme for Fault Detection and
Classication[J]. Industrial And Engineering Chemistry Research, 59(34), 15155-15164.
[14] Liu J., Zhang H., Si B. (2021) Radar Signal Classication Based on Bayesian Optimized Support-vector Machine[J]. Journal
of Physics: Conference Series, 1952(3), 032032.
[15] Ke H., Chen D., Shi B., et al. (2020) Improving Brain E-Health Services via High-Performance EEG Classication With
Grouping Bayesian Optimization[J]. IEEE Transactions on Services Computing, 13(4), 696-708.
[16] Popke M. (2022) Equipment/service update: Vegetation management[J]. Progressive railroading, 22(3), 65.
[17] Jin H., He Z., Liu L., et al. (2021) Technical Analysis of the Station Equipment IOT Management System[J]. Journal of
Physics: Conference Series, 1952(3), 032024.
Edited by: Bradha Madhavan
Special issue on: High-performance Computing Algorithms for Material Sciences
Received: Jan 27, 2024
Accepted: Mar 26, 2024
