Electric locomotive electrical control training based on virtual reality technology

The recent rapid development of the computer‐aided design technologies has provided concrete support for the education and research in the field of railway transportation. Virtual reality (VR) technology is becoming particularly important in the field of experimental teaching due to its low cost, low restriction and immersion. In this work, we have designed a virtual simulation untried system of the electric joint control in electric locomotives based on VR technology. Based on the user's understanding of the three circuit principles in the experiment, together with the recognition and operation ability to control appliances and equipment, the system assessment system has been scored and evaluated accordingly. The tests of the virtual simulation system have shown the experimental teaching can reduce the cost while effectively improving the learning effect.


INTRODUCTION
Railway is a key infrastructure construction object in the national economy. Countries like China have a vast territory, complex and diverse terrain, and diverse distribution of various resources, all of which make the advantages of railway transportation more obvious. With the strong staunch of the country and the rapid development of the decades, the continuous advancement of the high-speed railway electrification, railway and power network have been inseparable. The failure of the power system will directly lead to the breakdown of the railway system. The quality of the system will directly affect the running quality and service performance of the whole railway system. Electrical joint control virtual simulation experiment is an important experimental link for the students of the track electrical major. The traditional experimental method is inefficient, and the content is complex and difficult to understand. The goal of this system is to improve the operation training efficiency of students through a scientific way. Therefore, this article designs a virtual simulation test system for electric locomotive electrical joint control based on virtual reality technology. It is a meaningful teaching method for students of electrical major to understand the circuit composition of electric locomotive, understand the joint control relationship between the three major circuits of locomotive, and analyze the interlocking state between electrical appliances.
With the development of human-computer interaction, simulation, image processing, and programming technologies, VR technology has been extensively used in various fields. VR technology has created a virtual world composed of computer simulation technology and achieved a highly realistic and immersion effect. Nowadays, VR technology has been broadly applied into medical, engineering, aerospace and the other fields. With the development of science and technology, virtual teaching mode has become an important mode in current education methods. It can impart systematic experimental principles to students in a better expression way and strengthen students' understanding of theoretical knowledge. At the same time, the operation, function demonstration and the fault diagnosis in the virtual environment can enhance students' independent interest and improve learning efficiency. For those complicated and difficult experimental research projects under objective conditions, a virtual experimental platform can be built to achieve the desired experimental teaching effect with the advantages of low cost and high quality.
For the joint electric control system of electric locomotives, the key technologies are only mastered by the railway group and related track universities. There exist many problems such as high risk, poor visibility, multiple limitation, high cost, poor repeatability in the electric locomotive electric joint control system used in the field. These problems can be solved quickly and effectively through the electric joint control virtual simulation experiment of electric locomotive, and the basic concept of electric traction transmission system, speed regulation principle of electrical multiple unit (EMU), and other related theories can be mastered at the same time. The experiment can specifically achieve the following objects: (1) Through the study of traction transmission system, students can explain the composition of electric traction transmission system, the working principles of traction motor and traction converter, and can compare and analysis different transmission systems. (2) Through the study of the electrical circuit system, students can correctly analyze the electrical circuit schematic diagram of typical EMU traction control system, and can judge, analyze and simply deal with the electrical fault phenomenon. (3) Through the use of Multisim 7 and PLC (programmable logic controller) software, students are equipped with the ability to explore and improve the design, manufacture, application and maintenance of related equipment of EMU (electric multiple units) traction control system.

Recent developments of the VR technology based simulation
VR has been widely used in education and training. Alfadil explored the influence of the virtual reality (VR) game house of languages on the vocabulary acquisition when English as a Foreign Language (EFL) for intermediate school students. The results show that the use of the new VR technology as an effective vocabulary acquisition method in the learning process could be engaged to improve vocabulary acquisition and outperforming previous methods to enhance the achievement in classroom. 1 In the field of VR training effect and mode, Çankaya review with content analysis on the articles about the use of VR headsets in education from the Scopus database between 2014 and 2019, the results show that the number of studies about the use of virtual reality headsets in education tends to increase by years. In addition, they found that quantitative methods are as major research methods. 2 Chander believes that the degree of realism of virtual height in virtual reality matches the degree of the experimenter's response to SSQ and PQ. 3 Wang believes that in the development of VR technology, the trend is to shift from a simple UI desktop style to a style that enhances immersion and interaction. 4 RS Calabrò demonstrated that through hardware intervention the Rivermead mobility index and Tinetti performance oriented mobility assessment were improved for all patients in the RAGT + VR group compared with the RAGT-VR group. 5 David believes that VR simulations and experiments should be designed based on evidence. The potential effectiveness of training in virtual scenarios is huge, but this aspect is susceptible to interference from clumsy applications and lack of evidence testing. 6 Gallagher believes that if a training program meets the requirements, it needs to strictly follow the defined indicators, even weight training. Regular interval training is more suitable for virtual reality technology to increase the success rate of training. 7 Chen proposed a distributed algorithm based on echo state network machine learning framework to solve the quality of service of virtual reality users in SCN. 8 Wang designed an application framework of VR rides to integrate different hardware into the same immersive system with the convenience of modular programming. 9 Daria Vlah believes that compared with desktop modeling tools, virtual reality modeling is more intuitive and fast, but lacks precision and accuracy. 10 Lee conducted a study based on the case of Korea University of Technical Education, and the results show that virtual reality courses had significantly improved the development of students' abilities with its advantages of sustainable development. 11 Juan surveys consumers on VR technology through four evaluation indicators. According to statistical analysis, virtual display technology can reduce spatial perception differences and increase the interaction with users to obtain product information. 12 In the field of engineering education, Fu using quantitative theory I and KE regression analysis model for virtual reality system of human-computer interaction interface design elements were analyzed, and thus optimize the VR visual interface design. 13 Soliman believes that VR technology has a positive effect on cognition and teaching, and it can also be a good substitute for physics laboratory to reduce the institution's investment in cost. From the perspective of maximizing user experience. 14 Guo proposed a distributed learning method composed of offline training stage and online operation stage, which maintained scalability and adaptive ability. 15 According to Li, immersive VR applications need extremely high data rates and low latency to run smoothly, and enhancing the multi-user wireless VR experience is one of the main solutions. 16 Fussell believes that the TAM and TPB validated models can be applied to dynamic learning environments and assess users' perception of the environment, while PEXP and PENJ are considered to be important factors in assessing attitudes. 17 Wei combined optimizes the duration of the observation window of the prediction block and the duration of calculating and transmitting the prediction block, aiming to balance the performance of the three tasks to maximize the QoE of any given predictor and allocated resources. 18 Based on deep learning related algorithms, Chen developed four modules of interactive classroom, including learning objectives, video introduction, review and summary, and interactive test, and designed corresponding models. 19 Taking intelligent virtual rowing machine as an example, Bu developed a virtual reality technology system of intelligent value-added game machine, which is user-centered, and proved the feasibility and effectiveness of the method. 20 Wang introduced a BIM (building information modeling) based virtual environment supported by VR and a serious of game engines to address several key issues in emergency management. 21 Xiao proposed a general classification method composed of VR/AR technical characteristics, application fields, safety scenarios and evaluation methods, according to the safety characteristics and types under different construction conditions, hazard identification, safety training and inspection. 22

VR based train simulations
Bodnar aimed to evaluate the influence of locomotives' operation and reliability indicators on the system of locomotives repair organization in depot, using the methods of queue theory, and graphical simulation environment Simulink as software simulation tool. 23 Gironimo illustrated the use and management of heterogeneous product information (2D drawings, technical documentation, photos), virtual reality tools and digital human models for the re-designing of a locomotive, using a collaborative approach with a total absence of the reference digital models. 24 Luo proposed a simulation for power quality in V/V traction system of the 350-km/h high-speed railway, a kind of railway static power conditioner (RPC) is discussed, which is used to carry out the comprehensive compensation of negative sequence and harmonic currents in the traction substation. 25 Many applications of VR in train simulation have reflected the advantages of VR. Bezin described the work performed in Work Package 6 of the European project DynoTRAIN. Its task was to investigate the effects that uncertainties present within the track and running conditions have on the simulated behavior of a railway vehicle. Methodologies and frameworks for using virtual simulation and statistical tools, in order to reduce both the cost and time required for the certification of new or modified railway vehicles. 26 Xu developed a training simulator for the recovery of a railway accident through designing a set of task-specific hand gestures, the simulator helps the rescuers learn and practice rescue skills in a life-like environment and gain the firsthand experience. 27 Xu built a crane training system for railway locomotives, using the PHusX engine to build a visual framework that can reconstruct real accident scenes. 28 Yang studied the application of virtual reality panorama technology in rail transit engineering simulation system, and discussed the performance of virtual reality technology, rail transit system risk analysis and simulation system participating in virtual reality panorama technology. 29

Design ideas
This virtual experiment platform breaks through the traditional teaching mode, takes the actual electric locomotive model, machine room layout, and teaching content as the blueprint, displays the electric locomotive completely in the virtual environment through 3D modeling, and flexibly arranges multidisciplinary laboratories in the virtual scene, which can be reached through roaming, or accessed through menu selection. The virtual roaming developed by this technology gives people an immersive and realistic effect.

System technology platform architecture
The virtual environment is modeled by 3D Max, and the design system is Unity 3D. Unity 3D system can perform real-time processing and dynamic dyeing on a large number of models and lighting. Unity also supports C #, JS, BO and other development languages. The exhibition platform is divided into five layers: data model, database, Unity 3D service, development interface and presentation layer. Unity 3D: system operation logic. Including VR camera control, collision detection, action trigger, Bluetooth connection, multiplatform control and so forth.

3.3
System main functions 1. Scene roaming function: provide an excellent display platform for the machine room and its cab. Visitors can realize automatic roaming and manual roaming in electric locomotives by wearing virtual devices, and turn their heads to look around. Visitors can also move and jump through the Bluetooth handle. 2. Virtual laboratory. This system has designed two versions: PC version and mobile version. The PC version is controlled by the PC mouse and keyboard, and the mobile version enables users to interact with the virtual scene through virtual glasses, Bluetooth handles and other virtual reality hardware devices.

OVERVIEW
In the process of the experimental teaching, the teaching concept of taking students as the main body and the teaching method of task-driven is adopted to fully mobilize students' learning enthusiasm. Before the experiment, we allow the students to determine the experimental teaching objectives, to ensure the students a clear object in mind. Based on the simulation software, students can complete the locomotive bow raising, main break, auxiliary engine, starting locomotive and other operations, fully stimulate students' interest, and cultivate students' practical ability in the interest. During the experiment, students can see their own operation in the virtual scene and the linkage effect on the circuit, so that students can understand the interlocking relationship between the three circuits, exercise students' autonomous learning ability. This can inspire students to think positively, and improve their ability to analyze and solve problems.
By simulating the control process and action of a real locomotive in the virtual environment, students can intuitively learn the locomotive, understand the joint control relationship between the three major circuits of the locomotive, and analyze the interlocking state between electrical appliances. The system working pipeline is shown in Figure 1.
This system is mainly built through the 3D design software by rendering virtual models. The substance painter is applicable to draw the map and the UI (user interface), integrate in the Unity 3D, build three electric locomotive electrical spreading into virtual simulation experiment module, placed in a virtual server. Through the Internet, we can let the students visit anytime and anywhere.
The Internet module is mainly composed of the web part, the Unity part and the back end part. This module interacts with the data through the unified external data interface. The data resources of Unity come from the interface of Web API. Through the interface call, the download speed of resources is accelerated and the experience effect of users is improved. The platform is deployed and published on the network in the framework of the Web application, integrates various virtual experimental projects, and provides a unified remote access interface. This enables students to access the platform anytime and anywhere through a browser and use the system online.

Reconstruction of the training scenes in driver's control room
In the electric joint control virtual simulation training system, there are a large number of equipment models and complex surface structures. In order to ensure the accuracy of the essential equipment and the fidelity of the scene, the locomotive model and the practical training scene presented in the system are jointly constructed by the three dimensional modeling software Croe, Maya and the texture making software substance painter. The mechanical modeling software Croe is used for parametric modeling of important parts such as the locomotives, to ensure the accuracy and proportion of the model. The basic model is then imported into Maya for surface reduction and optimization. Meanwhile, UV expansion of the model is carried out in Maya for later production of maps in substance painter. The optimized model from Maya is imported into Unity 3D engine together with the texture of resource synthesis and baking rendering. In order to reduce high realistic model overhead in the rendering process, the map presents the details of the locomotive model, the UV equipment mode. A three dimensional devices through a complex surface texture map can be displayed. Meanwhile, a bump map is used to show the details of the device surface, without adding the number of model surfaces to show the details, which can achieve the effect of low mode and high-precision, reduce the occupation of computer resources, and improve the smoothness of the system operation.
In order to create more realistic environment and the surface of the device with light and shadow effect, lighting components are added for rendering, as shown in Figure 2.

Trigger monitoring technology
In Unity 3D, there are two ways to detect the occurrence of a collision, collider and trigger. There is also a built-in PhysX physics engine, which has nested algorithms related to the collision, barrier, friction and other physical phenomena occurring in the contact between rigid bodies. Trigger detectors by having ontology and objects clearly defined in scripts. In the operation mode, collision detection will be carried out for each frame. Once the collision condition is reached, the collider will be triggered to carry out internal logic operation and start the physical operation trajectory of the next frame. When the collision condition is not reached, the object will continue to move along the original state in the consecutive frame.

SYSTEM FUNCTION SIMULATION
The main contents of the experimental system include: the cognition of space equipment, the cognition of three major circuits, combined control operation of circuit and three modules of gas circuit, students study in turn. By establishing a highly realistic 3D model and high precision assembly relationship, the experimental content is highly aligned with the field. This will be convenient for students to simulate the field control of the high voltage circuit the virtual scene, familiar with the equipment in the control room and the corresponding circuit flow direction. Through the unique operation of the equipment, changing the logic connection mode of the circuit, completing the operation of the locomotive lifting bow, closing main break, starting locomotive. Through the human-machine interface, the experimental parameters are set and the experimental steps are controlled. The students' autonomous learning ability is trained. The students can be inspired to consider positively, and the ability to analyze and solve problems can be improved. System built-in information log will record each step of the operation, and give the corresponding score after all operations to give the individual and class as a whole operation process report. From the micro point of view, students will fully realize whether there are doubts in a certain knowledge point, and make up for the deficiency point by point in time. From a macro point of view, this is all right with teachers to understand the overall situation of the class and guide students.

Spatial cognitive experiment
The purpose of the machine room cognitive experiment is to help students form a comprehensive understanding of the structure of the machine room and driver's room of electric locomotives, understand the physical appearance and spatial arrangement of instruments, buttons, LCDM, CIR and other equipment, and the inspection sequence and precautions. Using multimedia technology to embed high precision text, picture, video and other information into the scene, not only can complete the introduction of the structure of each module between the machines, but also can strengthen the feelings of different aspects, enhance the students' immersion experience. Students roam the space according to the prompt flow of the system, flow 360 • comprehensive observation surrounding model. All models are assembled from separate level, 1:1 highly restored site environment. This allows students to further study the working principles of local devices. For complex and difficult to observe parts, we can switch to free mode after the end of learning mode, and understand the parts in multiple dimensions by perspective or analysis alone. The effect diagram is shown in Figure 3.

Cognitive experiments on three major circuits
The three major circuits include the principal circuit, the auxiliary circuit and the control circuit. In the experiment, students can directly observe the dynamic flow effect of the line, as shown in Figure 4. Different from the traditional electrical circuit diagram, the logical relationship between parts has been processed through 3D Max in the basic modeling link. After the model being imported into the Unity, the dynamic circuit simulation diagram is presented according to the different degrees of connection with different devices. This is very effective for students to understand the obscure internal electrical structure.
VR technology provides a safe, quiet and highly recognizable environment to simulate the electronic flow in different conditions of electric locomotives. It not only completes the visualization of the electron flow direction, but also resets to the original environment numerous times for the unknown logic node and learns repeatedly. It helps to develop students' ability to analyze and solve problems independently in tricky situations. Three major circuits, as classic electric locomotive circuit experiment, include most of the original electrical circuit. The design process of the experiment not only requires students to master the basic circuit principles, but also has the part of innovative thinking.

Control linkage experiment
Manipulation experiment is the process of combining theory with operation. The test process will simulate the switch button, the response of the joint control circuit, the closure of the main circuit breaker. Then users check the network voltage meter, read the data of the network voltage meter, and judge whether the network voltage is normal to check the joint control circuit. The main circuit, control circuit should get electricity. The charging contactor and working contactor of the main transformer will display their electric states as shown in Figure 5.

SYSTEM EVALUATION
In order to evaluate whether the virtual simulation experiment system of EMU electric joint control can effectively improve the effect of experimental learning, we choose two classes of track electrical specialty with close levels, and test all students before and after the experiment. In order to exclude the influence of irrelevant variables, there was no significant difference between the two classes' scores and the number of students. In order to avoid the influence of other subjects, we choose the weekend as the experimental stage. Before the beginning of the experiment, each student received a questionnaire to get the basic information of the participants. After the experiment, the teachers and students were asked to participate in the system evaluation. At the end of the experiment, teachers and students were examined again. Students in the test group are required to participate in the system evaluation.
Before the experiment, we are required to investigate the basic information of these two classes. As shown in Figure 6, we can clearly see each person's basic level and the overall distribution of results used for statistical analysis of the test results of students in two classes.
First, the Skewness and Kurtosis of the two arrays are obtained, and the prerequisites for macroscopic analysis of normal distribution are shown in Table 1. Then the Kolmogorov-Smirnov test and the Shapiro-Wilk method is used to detect whether the data has significant characteristics of normal distribution, as shown in Table 2. Table 2 is the evaluation results of the generated KS test (D test) and the SW test (W test). Here we concentrate on the significance of the Sig. value, especially the Sig, P value. When P ≥ 0.05, the data can be regarded as normally distributed. The results show that the two groups of data meet the standard of normal distribution.
We also used the independent sample T test to analyze the difference between the two classes before and after testing. The results are shown in Table 3. The control class is 59.71 and the experimental class is 60. 19. Test results showed no statistical difference between the two classes. Meanwhile, the control class score is 61.18, the experimental class score is F I G U R E 6 The distribution of student achievement as a scatter diagram. (The statistics and fitting of the two groups of students before and after the test were carried out, and the macroscopic results showed that the use of VR for experimental teaching could effectively improve the learning effect.) 69.31, and the test result is P < 0.05. The results show that there are significant differences between the two, the results of the control class fluctuate at lower level. The results of the experimental class are obviously improved, which indicates that the virtual simulation experiment can effectively improve the experimental teaching effect. Based on constructive teaching theory, this experiment applies the advantages of high efficiency, high precision and low cost to complex traditional experimental courses. The virtual simulation laboratory has built a new type of scientific and technological educational laboratory with the price significantly lower than that of the traditional laboratory. The way of active operation and system prompt enable students change from passive teaching to active learning, multi-person cooperative online mode enables students change from thinking alone to group cooperation and communication, and the multi-integrated test and assessment system ensure the completion of all teaching tasks. Meanwhile, the high interesting of the experiment greatly attract and enhance the enthusiasm of students. The concept of education and fun is truly integrated ⋅experimental project can be published on many platforms, such as PC, Vive, Zspace and mobile. This allows to realize the learning process in many manners, and thus breaking through the limitation of time and space. The mode of access to the Internet can not only be aimed at teachers and students of this major, but also can be easily accessed and used to other colleges and universities in need to realize the social value of the virtual simulation laboratory.

CONCLUSION
The virtual simulation system of the electric control in EMU is mainly used in the auxiliary teaching system of railway transportation majors in colleges and universities. Combined with field experiment teaching content, virtual simulation technology is utilized to build the network platform for students. The platform is an extension and supplement of the experimental project, and it is also a rich and practical learning resource for students. The platform will realize the teaching function of electronic control virtual simulation. This experimental project makes full use of modern science and technology to realize the operation. Invisible visual areas of internal structure and complex circuits and the disadvantages of dynamic logic which cannot be reasonably presented will be operated through virtual reality technology and simulation. It faithfully simulates the intricate field conditions. While obtaining the same teaching effect and experience, the system alleviates many problems such as high cost, multiple limitations, low visibility, very low repeatability and unique test data. Through task-driven learning, students can conduct independent preparation before the experiment online, accomplish the experiment operation independently, fill in the experiment report, learn and communicate, and conduct offline practice to consolidate and strengthen knowledge. In this project, students can control the virtual locomotives' motion state in real time, cooperate with the interaction of circuit joint control and core electric movements and so forth to realize a comprehensive cognition of the locomotive circuit.

DISCUSSION
Design of the experimental scheme: The project adopts a student-oriented and problem-oriented experimental teaching concept, which makes the electric circuit and its working principle of electric locomotive invisible. The virtual simulation experiment of electric joint control of electric locomotive is an open teaching platform. We aim at the observation learning phenomenon which is common to the students in the practical training of locomotive electric experiment, making full use of modern science and technology to realize the visualization of invisible internal circuits of locomotives. Inaccessible driver consoles, machines. In order to meet the needs of professional personnel training and cultivate students' ability to solve complex engineering problems, the experimental operation is carried out in order to solve the problems existing in the actual vehicle experiment by simulating the complex working conditions in the field. Innovative teaching methods: realistic 3D effect, rich teaching content, fully inspire students' interest in learning. The project overcomes the shortcomings of passive observation and teaching in traditional experimental teaching, such as less interaction, poor intuition, one-sided content and boring and tedious content. Through realistic three-dimensional model and rich human-computer interaction experience, students can use various forms of interaction to achieve immersion experience effect in the process of teaching experiment learning, to realize the rapid understanding of electric circuit of electric locomotive and can think independently, stimulate students' initiative.
Evaluation system of this project is scientific and reasonable, which meets the requirements of professional teaching and training and improves students' independent learning ability and learning interest. The link in the system according to the students operating conditions, comprehensive user usage time, qualification examination content, data operation and data analysis results form an experiment report; report generated based on the experimental teachers grading rules to review report anonymously. Finally, based on the evaluation of the cognitive modules, operations, a report based on the scores of three grades is generated as the final report.
Extension and expansion of traditional teaching: the project adopts a hybrid teaching method combining production, teaching and research based on Internet technology to improve students' ability of independent learning and active exploration of systematic research. Combined with the characteristics of production practice teaching and industrial simulation software, the performance experiment results are close to the real trains.