Journal of Physics: Conference Series

PAPER • OPEN ACCESS

Application of Artificial Intelligence Technology in the Teaching of

Mechanical Education Courses in Universities
To cite this article: Honghua Liu et al 2021 J. Phys.: Conf. Ser. 1992 042065

View the article online for updates and enhancements.

This content was downloaded from IP address 216.19.206.98 on 25/08/2021 at 15:26

CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

Application of Artificial Intelligence Technology in the

Teaching of Mechanical Education Courses in Universities

Honghua Liu, Wenping Tan*, Hongmei Li, Jingzhong Gong, Xiangling Liu

College of Information and Mechatronical Engineering, Hunan International

Economics University, Changsha, 410205, Hunan, China

*Corresponding author e-mail: [email protected]

Abstract. With the continuous innovation of artificial intelligence technology, the

teaching level of mechanical education courses in colleges and universities is also
constantly improving. In the context of the era of big data, the teaching of mechanical
education courses in colleges and universities has become more intelligent and
informatized. This article mainly introduces BP neural network method and hill
climbing algorithm. This paper uses BP neural network method to analyze the
application of artificial intelligence technology in the teaching of mechanical
education courses in colleges and universities, and establishes a potential
mathematical model of BP neural network method. The model is solved by BP neural
network method, and the current situation analysis and application status of the
teaching mode of mechanical education courses in colleges and universities are
analyzed, and the model is revised using historical data to improve the application
research of artificial intelligence technology in the teaching of mechanical education
courses in colleges and universities Accuracy. The experimental results of this paper
show that the BP neural network method has increased the effect of artificial
intelligence technology in the teaching of mechanical education courses in colleges
and universities by 33%. Finally, by comparing the value analysis of the application of
artificial intelligence technology in the teaching of mechanical education courses in
colleges and universities, and the data analysis of artificial intelligence technology in
the teaching of mechanical education courses in colleges and universities, the system
shows that artificial intelligence technology is used in mechanical education courses in
universities. Application in teaching.

Keywords. Artificial Intelligence Technology, Mechanical Professional Education,

Literature Reference Research Method, Analytic Hierarchy Process

1. Introduction

1.1 Background and Significance

In recent years, with the development of a new generation of information and communication
technology, mobile Internet is everywhere [1]. The theory of artificial intelligence, the large-scale
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

computing power of the Internet cloud platform and continuous optimization and algorithm
improvements enable computer systems to completely rely on deep learning to autonomously
complete the most complex and meticulous tasks, and artificial intelligence has begun to be integrated
into all walks of life [2]. The combination of artificial intelligence and other new technologies is also
an important foundation for many developing industries [3-4]. The purpose of artificial intelligence is
to develop abilities, improve learning, improve intelligence and meet everyone's needs [5]. However,
artificial intelligence technology cannot replace teachers, but in big data, the use of teachers to change
the type of education is essential for improving the quality of university education in the future [6-7].
Today, the teaching system of Chinese higher education institutions has been steadily increasing in
terms of technology, education, scientific research and management.

1.2 Related Work

Breines MR provides a method that can evaluate related innovations with participation in a complex
related environment to solve essential problems [8]. Based on the principle of common value creation,
he proposed a research framework. This framework illustrates the application research process of
artificial intelligence technology in the teaching of mechanical education courses in colleges and
universities. In this process, the relevant parties integrate their resources and capabilities to develop
innovative BP neural network methods [9-10]. In order to evaluate this research framework, Breines
MR collected multiple data in the research. This case also represents the significance of the application
research and system realization of artificial intelligence technology in the teaching of mechanical
education courses in colleges and universities [11-12]. Due to the error in the experimental process of
this paper, the result is not very consistent with the actual prediction.

1.3 Main Content

The innovation of this paper is the use of BP neural network and hill climbing algorithm. According to
the theory of artificial intelligence, combined with the mechanical education courses of colleges and
universities, as a basis for judging the application effect of artificial intelligence technology in the
teaching of mechanical education courses in colleges and universities, the BP neural network and
mountain climbing algorithm models are established. Mechanical professional teachers provide certain
reference and guidance.

2. Application Method of Artificial Intelligence Technology in the Teaching of Mechanical

Education Courses in Colleges and Universities

2.1 BP Neural Network

BP neural network is a multi-layer forward network with three layers, such as: input layer, middle
layer, output layer, or more than three layers of unidirectional propagation. Its main feature is that the
upper and lower layers are completely connected, and there is no connection between neurons in each
layer. Its input and output relationship is a highly non-linear mapping relationship. If the number of
input nodes is m and the number of output nodes is n, the BP network is a mapping from m-
dimensional Euclidean space to n-dimensional Euclidean space. The neural network state is activated.
Initial network setup with small random number, where n is the connection weight between the input
layer and the second layer, and the threshold is m, which is the connection weight between the second
layer and the output layer. The formula is as follows:

E  , E X  X 
Q

O
G 2
2
1  3 (1)
O 1
Q Q 3

Therefore, first standardize the value of each index used to evaluate teaching quality. Then use it as
the input vector of the BP neural network model, use the evaluation result as the output vector, and use
enough samples when training the network to adapt it to expert experience knowledge and judgment.
Understand the index weight, so that the BP neural network model automatically adjusts the value of

2
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

the weight statement until it reaches the true internal knowledge model, and then adds the specific
value of each indicator that has been evaluated by quality learning to the BP neural network model,
and receives training to Accept the results of evaluating the quality of learning. Finally, the experience
and knowledge of experts are used for reference to ensure the objectivity of evaluating teaching
quality.

2.2 Hill Climbing Algorithm

The hill climbing algorithm starts from the current node and compares it with the values of
surrounding nodes. If the value of the current node is higher, return to that row. As the maximum
value, if the surrounding nodes have a larger value, please change the current node to the maximum
value node, and then change the value of the surrounding nodes, the hill climbing algorithm will end
and the best solution will be obtained.
Since the hill climbing algorithm cannot guarantee the optimal solution, it is not suitable for
finding the optimal path in mobile applications. The advantage of the hill climbing algorithm is that it
discards smaller values and then loops around nodes. After many times, the highest point can be
obtained. When using the hill-climbing algorithm, when the value of a node is greater than half a node,
the depth of the first search will be refined to use the manual search method to optimize the search to
the best local solution. This article uses the maximum-minimum method because the data processing
performed by this method is a linear transformation, which can better retain its original meaning
without generating information.
The formula is as follows:
E  Emin (2)

Emax  Emin

In the formula, E is the processed neural network input value; E is the unprocessed neural network
input value; is the minimum neural network input value.

3. Application Experiment of Artificial Intelligence Technology in the Teaching of Mechanical

Education Courses in Colleges and Universities

3.1 Application Design of Artificial Intelligence Technology in the Teaching of Mechanical Education
Courses in Colleges and Universities
At present, every classroom in colleges and universities is equipped with advanced multimedia
teaching equipment, and the campus network has been updated, and a computer-assisted information
unit has been established to meet the needs of one person and one computer according to the
requirements of mechanical information teaching. In addition, mechanical training includes various
CNC equipment, industrial production equipment, etc. Therefore, colleges and universities equipped
CNC lathes and machining centers with the industrial operating system i50S, opened API interfaces to
all users, created industrial apps, and communicated directly between smart terminals and the cloud, in
order to promote the integration of mechanical engineering and information technology. At the same
time, the structure of the software environment cannot be ignored.
Universities need to establish a distance online education system to meet the needs of online
communication between teachers and students. According to the demand for product information, it is
also equipped with various tools and software related to this profession. In teaching, teachers need to
focus on teaching problems and provide extracurricular teaching resources for students according to
their learning level and development needs. In addition, it is necessary to fully demonstrate the
strength of the teaching and research team, jointly develop teaching materials, and incorporate public
courses, case studies, classic courses and learning results, and school learning content into the
electronic resource database, electronic curriculum database, course resource database, test question
database and documents. In the library, resources are aggregated using information technology. In the

3
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

classroom, the integration of information technology and the classroom is mainly reflected in the
integration of teaching methods, teaching materials and teaching processes.

3.2 Data Collection of Students' Various Skills

This article conducts an experimental design on 100 students and randomly divides them into two
groups of 50 people each, an experimental group and a control group. The variables of the
experimental group are the combination of traditional teaching methods and artificial intelligence
technology, while the control group only uses traditional methods. The subjects of the test are the
students’ four intelligences. These four intelligences are selected from the eight intelligences proposed
by the American psychologist Gardner. They are interpersonal intelligence, introspective intelligence,
and Spatial intelligence and language intelligence use experimental methods to obtain experimental
data.
By counting the average scores of 100 students in the four intelligent feature parts, and using the
excel software to draw the following figure. Therefore, it can be clearly seen from Table 1 that the
average scores of the scores obtained by 100 students in each intelligence are approximately equal,
indicating that each intelligence is reflected in all students, and the overall level is maintained. The
average score of the four types of intelligence in the experimental group exceeded 80 points. Among
them, the outstanding performance was verbal intelligence with an average score of 91 points,
followed by spatial intelligence and interpersonal intelligence with average scores of 90 and 89 points,
respectively. There is a big difference between the average score of four kinds of intelligence and the
average score of the four kinds of intelligence in the experimental group, and the difference of space
intelligence is 29 points. Because the subjects of the survey are mainly students born in 2002 and
2003, they are more prominent in interpersonal communication, language expression, spatial
imagination and introspective ability, which is in line with the characteristics of contemporary college
students who dare to discover and express themselves. In this respect, the results of the questionnaire
survey are roughly in line with the actual situation of the students. The results of the students’ smart
data collection are shown in Table 1:
Table 1. Smart data sheet of students
Interpersonal Introspective Linguistic
Group Spatial intelligence
intelligence intelligence intelligence
test group 89 88 90 91
Control group 71 76 61 78
Set the number range from 1 to 100, and the result of data analysis after the algorithm operation is
shown in Figure 1.
100 89 88 90 91
76 78
80 71
61
60

40

20

0
Interpersonal Introspective Spatial Linguistic
intelligence intelligence intelligence intelligence
test group Control group

Figure 1. A graph of student intelligence data

It can be seen from Figure 1 that there are almost no students with a score of 0-1, that is, for
individuals, each intelligence has its performance, but the possible performance methods are different.
Secondly, the proportion of people in the score range of 2-4 is also small, indicating that there are also

4
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

a small number of students who are more prominent in a certain intelligence, and through
understanding, it is found that these students with more outstanding performance are more or less good
in learning Something has been shown, so in the classroom teaching process, on the one hand, we
must cultivate students' multiple intelligences, and on the other hand, we must encourage students to
cultivate their own intellectual advantages. In addition, the proportion of students in the middle score
segment is more prominent in the various intelligences, indicating that this distribution law is closer to
the normal distribution in mathematics, and it also reflects the evaluation of students from the
perspective of multiple intelligences. There are good and bad points, but they can’t perform well in
learning.

4. Application Analysis of Artificial Intelligence Technology in the Teaching of Mechanical

Education Courses in Colleges and Universities

4.1 Application Value Analysis of Artificial Intelligence Technology in the Teaching of Mechanical
Education Courses in Universities
Thanks to the latest developments in artificial intelligence technology, today's labor market has
changed. The impact of technological change can be reformed in two ways. First of all, most
automated operations will replace workers, which will reduce the number of jobs in the market, the
loss of job opportunities, and the company has fewer employees engaged in low-skilled jobs.
However, these phenomena bring inefficiency and cost reductions. , And ultimately stimulate the
growth of wealth and employment opportunities. Another phenomenon is that the results of the
development of science and technology have changed the rules of the market. It is precisely because of
this indirect effect that ultimately makes the traders and enterprises in the market reorganize and
reconstruct their operating methods. Improvements in market processes and organizational changes
will result in the loss of many repetitive tasks and well-trained personnel. For the mechanical
education field of colleges and universities, they are also facing the same problem, but fortunately, like
other science and technology, artificial intelligence technology will gradually eliminate university
mechanical training teachers in the market, thereby regulating the mechanical education market of
colleges and universities.
Artificial intelligence technology will help improve teaching quality and efficiency. In the teaching
process, most teachers are "students", and they should respect the individuality of students.
Demonstrate teaching, training and curriculum plans based on the understanding of students’
psychological development. If teachers cannot understand students well, or violate the laws of
students' physical and mental development and students' personality, they will not be able to achieve
good results in the teaching process. Obviously, this is not suitable for students' own development and
psychological maturity. When a teacher encounters a new student or a new class in the classroom, the
teacher can use artificial intelligence technology to perform big data analysis, so that the teacher can
accurately understand the student's learning background, learning process, and the strengths and
weaknesses of the students in a short time. In this way, teachers can quickly assume the role of
education and organize teaching content in a reasonable way.

4.2 Pre- and Post-Test Data Analysis of Artificial Intelligence Technology in the Teaching of
Mechanical Education Courses in Colleges and Universities
In the experiment, 200 contemporary college students were randomly selected for grouping and
divided into two groups. The pre-test was conducted in the first semester and the post-test was
conducted at the end of the semester. The researchers used a set of midterm exam assignments from
the university to conduct preliminary exams on 14 mechanical majors. This set of test assignments can
check students' mechanical professional skills from different angles. In order to describe the changes
before and after the test subjects' mechanical expertise, the researchers performed descriptive statistics
and independent test T tests on the experimental group and the control group. The test data are shown
in Table 2:

5
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

Table 2. Data table before and after testing

Group test The average score Standard deviation variance
Pre-test of experimental group 65 0.2 0.0400
Control group pre-test 63 0.21 0.0441
Experimental group posttest 91 0.12 0.0144
Control group post-test 78 0.22 0.0484
As shown in Table 2, under the same conditions, the experimental group and the control group had
similar pre-test average scores, standard deviations and variances. However, the post-test average
score of the experimental group was 13 points higher than the average score of the control group. At
the same time, the corresponding standard deviations of the two categories were 0.12 and 0.22
respectively. The difference was significant. This shows that the artificial intelligence technology is
applied to the mechanical specialty. It can improve the teaching effect of mechanical education
courses in colleges and universities. In order to be able to see the relationship between the two more
clearly, we analyzed the data in Table 2, and the analysis results are shown in Figure 2:
The average score Standard deviation variance
100
80
60
40
20
0 Pre-test of
Control group Experimental Control group
experimental
pre-test group posttest post-test
group
The average score 65 63 91 78
Standard deviation 0.2 0.21 0.22 0.12
variance 0.04 0.0441 0.0144 0.0484

Figure 2. Independent sample test for before and after testing

From Figure 2, we can see that. The standard score difference is also nearly one point away. The
above data all show that under the teaching mode guided by artificial intelligence theory. During the
teaching experiment, compared with the class implementing the new teaching model, the mechanical
professional level of the two classes actually at the same level has been significantly improved.

5. Conclusion
Although this article has made some achievements in the application of artificial intelligence
technology in the teaching of mechanical education courses in colleges and universities, there are still
shortcomings on the basis of BP neural network and hill climbing algorithm. There are still many
methods that can be studied in depth in the application research of artificial intelligence technology in
the teaching of mechanical education courses in colleges and universities. Due to space and personal
skills, there are many steps in the research process that have not been involved. In addition, from the
perspective of theory and simulation, the actual application effect of the improved algorithm can only
be compared with traditional models.

Acknowledgements
This work was supported by scientific research project of Hunan Provincial Department of Education,
Project number: 20A287; General education project of teaching reform in Colleges and universities of
Hunan Province, project number: HNJG-2020-1018.

6
CETCE 2021 IOP Publishing
Journal of Physics: Conference Series 1992 (2021) 042065 doi:10.1088/1742-6596/1992/4/042065

References
[1] Hu Rong. Research on the Application of Computer Virtual Technology in the Teaching of
Basic Courses. SME Management and Technology, 2018, 000(023):63-64.
[2] Hong Changchun. A Study on the Construction of College English Ecological Teaching Mode
in the Age of Artificial Intelligence. Foreign Language Audiovisual Teaching, 2018,
000(006):29-34.
[3] Zheng J, Hu X W. The Application of Big Data and Artificial Intelligence in Higher Education.
Lingdong Journal, 2019(45):35-44.
[4] Cox A M . Exploring the impact of Artificial Intelligence and robots on higher education
through literature-based design fictions. International Journal of Educational Technology in
Higher Education, 2021, 18(1):1-19.
[5] Song Ye. Research on the Application of Case Teaching Method in the Course of Computer
Application Technology% "Case-based" Teaching Method in the Course of Computer
Application Technology. Science & Technology Information, 2016, 014(036): 194-195.
[6] Deng Wei, Yao Guwen. Research on the Application of Pro/E Software in the Teaching of
Mechanical Drawing. Journal of Guiyang University (Natural Science Edition), 2017,
012(003):29-31.
[7] Xia Y . Big data based research on the management system framework of ideological and
political education in colleges and universities. Journal of Intelligent and Fuzzy Systems,
2020(6):1-10.
[8] Breines M R, Gallagher M. A return to Teacherbot: rethinking the development of educational
technology at the University of Edinburgh. Teaching in Higher Education, 2020(3):1-15.
[9] Mazurov M E , Mikryukov A A , Titov V A , et al. Neuroeducational Environment for
Acquisition of Competencies in the Field of End-To-End Digital Technologies
(Neurotechnology) in the Conditions of Digital Transformation. Open Education, 2020,
24(6):31-40.
[10] Bucher A M , Kleesiek J . Künstliche Intelligenz in der onkologischen Radiologie. Der
Radiologe, 2021, 61(1):52-59.
[11] Bunyatova F , Karimova Y , Najafov R . Transformation of education system supported by the
government Integrity (whole) and fuzzy education model. Azerbaijan Journal of Educational
Studies, 2019, 229(4):47-64.
[12] Zhang Chunran, Ma Qian, Peng Hui. The Reform and Thinking of Artificial Intelligence
Technology on Teaching. Office Automation (Office Equipment and Consumables), 2019,
024(010):35-37.