Application method of emergency experience in educational safety based on virtual reality

Abstract

The invention discloses an application method of emergency experience in education safety based on virtual reality, and relates to the field of virtual safety education. Corresponding virtual scenes are constructed for a safety accident type, an on-site hidden danger checking type and a safety skill simulation teaching type; the employees can be helped to perform simulation drilling before high-risk operation; different degrees of setting are provided for virtual scenes of each safety accident type and field hidden danger checking type, so that employees can select simulation training of different degrees and difficulties according to actual conditions of the employees, and the best drilling effect can be achieved; wherein the clustering operation is matched to generate a plurality of different types of virtual scenes corresponding to different labels, so that different virtual simulation requirements of different employees can be met, each employee can be matched with virtual scene simulation suitable for own conditions, and the employees can be trained in a targeted manner.

Description

Technical Field

[0001] The present invention belongs to the field of virtual safety education, and in particular, relates to an application method of emergency experience based on virtual reality in education safety. Background Art

[0002] Safety accidents, such as falling from heights, falling during fan maintenance, being hit by objects, etc., on-site hidden danger inspections, such as electrical inspection hidden danger inspections, temporary power use hidden danger inspections, and safety skills simulation teaching. These are the necessary skills and possible accidents for high-risk operations. If employees are not given targeted simulation training for these situations, the quality of the work may not be high, and may even cause accidents to employees, resulting in loss of life. Summary of the invention

[0003] In view of the problems in the related technology, the present invention proposes an application method of emergency experience based on virtual reality in educational safety to overcome the above-mentioned technical problems existing in the existing related technology.

[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0005] The present invention is an application method of emergency experience based on virtual reality in educational safety, comprising the following steps:

[0006] S1. Collecting characteristic data of multiple groups of different types of safety accident scenes, hidden danger inspection scenes and simulation teaching scenes in history, and obtaining a historical safety accident characteristic data matrix set, a historical hidden danger inspection characteristic data matrix set and a simulation teaching characteristic data matrix set;

[0007] S2, clustering the historical safety accident feature data matrix set and the historical hidden danger investigation feature data matrix set and calculating the central data of each classification to obtain a historical safety accident feature classification central data matrix set and a historical hidden danger investigation feature classification central data matrix set;

[0008] S3, setting the category labels corresponding to the historical safety accident feature classification center data matrix set and the historical hidden danger investigation feature classification center data matrix set, and obtaining the safety accident severity level data matrix and the hidden danger investigation standard level data matrix;

[0009] S4, generating a plurality of virtual scenes corresponding to different types and different labels according to the historical safety accident feature data matrix set, the historical hidden danger inspection feature data matrix set, the simulation teaching feature data matrix set, the safety accident severity level data matrix and the hidden danger inspection specification level data matrix, and obtaining the final safety accident virtual scene matrix, the final hidden danger inspection virtual scene matrix and the final simulation teaching scene set;

[0010] S5. Use the final safety accident virtual scenario matrix, the final hidden danger investigation virtual scenario matrix and the final simulation teaching scenario set to simulate the operation of employees;

[0011] By collecting characteristic data of multiple groups of different types of historical safety accident scenes, hidden danger inspection scenes and simulation teaching scenes, data support is provided for subsequent characteristic data clustering operations; by clustering the historical safety accident characteristic data matrix set and the historical hidden danger inspection characteristic data matrix set and calculating the central data of each category, a data basis is provided for the subsequent generation of labels for the characteristic data of each category, and thus data support is provided for the subsequent generation of virtual scenes of different levels; by generating virtual scenes of multiple different types and corresponding to different labels, in order to meet the different virtual simulation requirements of different employees, each employee can be matched with a virtual scene simulation suitable for his or her own situation, so that all employees can receive targeted training.

[0012] Preferably, the S1 comprises the following steps:

[0013] S11, setting multiple safety accident types, on-site hidden danger investigation types, and safety skill simulation teaching types to obtain a safety accident type set, a hidden danger investigation type set, and a simulation teaching type set;

[0014] Then set the characteristic data types of each safety accident type, hidden danger investigation type and simulation teaching type in the safety accident type set, hidden danger investigation type set and simulation teaching type set to obtain the safety accident characteristic data type matrix a1, the hidden danger investigation characteristic data type matrix a2 and the simulation teaching characteristic data type matrix a3; respectively as follows,

[0015]

[0016]

[0017] Among them, a 1ij 、a 2ij 、a 3ij a1′ represents the jth characteristic data type of the i-th safety accident type, the on-site hidden danger investigation type, and the safety skill simulation teaching type respectively; i , a′ 2i , a3′ i Respectively represent the total number of characteristic data types of the i-th safety accident type, on-site hidden danger inspection type, and safety skill simulation teaching type; b represents the total number of set safety accident types, on-site hidden danger investigation types, and safety skills simulation teaching types;

[0018] S12, according to the safety accident characteristic data type matrix a1, the hidden danger inspection characteristic data type matrix a2 and the simulation teaching characteristic data type matrix a3, the characteristic data of multiple groups of historical safety accident scenes of different types, hidden danger inspection scenes and simulation teaching scenes are collected to obtain the historical safety accident characteristic data matrix set Historical hidden danger investigation feature data matrix set And the simulation teaching characteristic data matrix set b3′={b3′1,...,b3′ i ,...,b3′ b}; b1′ i , b2′ i , b3′ i They represent the characteristic data matrices of the set i-th safety accident type, on-site hidden danger investigation type, and safety skills simulation teaching type respectively; they are as follows,

[0019]

[0020]

[0021] Among them, b1′ ijk b2′ represents the kth type of feature data of the jth group of safety accident scenes collected for the ith safety accident type; ijk b3′ represents the kth type of characteristic data collected at the jth group of hidden danger inspection sites for the ith hidden danger inspection type; ijk represents the kth type of feature data of the jth group of simulation teaching scenes collected for the ith simulation teaching type; Indicates the total number of groups of characteristic data collected from different types of historical safety accident scenes, hidden danger inspection scenes, and simulation teaching scenes;

[0022] The types of safety accidents mainly include high-altitude falling accidents, wind turbine maintenance falling accidents, object impact accidents, etc.; the types of hidden danger investigation mainly include electrical inspection hidden danger investigation, temporary power hidden danger investigation, etc.; the types of simulation teaching mainly include temporary power operation simulation teaching, high-altitude operation simulation teaching, etc.;

[0023] The safety accident feature data type matrix includes falling height, falling radius, and falling type, such as falling from scaffolding, falling from suspended high-altitude operations, falling during climbing, and falling from ladders; casualties, accident time, accident location, use of personal protective equipment, safety facility conditions, operator characteristics, such as age, health status, psychological state, and fatigue level, which will also affect the occurrence of accidents; accident causes, etc.

[0024] The matrix of characteristic data types for hidden danger investigation includes data on the operating status of electrical equipment, such as the voltage, current, power, temperature, residual current, insulation resistance to ground and other parameters of the equipment; line status data, such as the voltage, current, insulation resistance, grounding resistance and other parameters of the line; safety measures data, such as the operating status and parameter settings of leakage protection devices, overload protection devices, short-circuit protection devices and other devices; environmental data, such as the temperature, humidity, dust, corrosive gases and other parameters of the operating environment of electrical equipment; maintenance record data, such as the maintenance time, maintenance content, replaced parts and other information of the equipment; fault record data, such as the type of fault, fault time, fault cause and other information that has occurred in the equipment or line;

[0025] The matrix of characteristic data types for simulation teaching includes scene data, such as the layout of factories and construction sites, equipment placement, and safety sign locations; equipment data, such as the parameters, operating specifications, and safety requirements of various electrical equipment; personnel data, such as the division of responsibilities, operating procedures, and safety precautions of temporary power workers; safety data, such as safety standards for temporary power operations, risk assessment data, and accident cases; and interactive data, such as user operation records and feedback data in virtual scenes.

[0026] Preferably, S2 comprises the following steps:

[0027] S21. Set the number of clusters According to the number of clusters, each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set are clustered to obtain a historical safety accident feature classification center data matrix set. And the historical hidden danger investigation feature classification center data matrix set c 1i 、c 2i They represent the characteristic classification center data matrix corresponding to the set i-th safety accident type and hidden danger investigation type respectively; they are as follows,

[0028]

[0029] Among them, c 1ijk 、c 2ijk Respectively represent the k-th type of feature data in the j-th feature classification center data set of the feature data matrix of the set i-th safety accident type and the on-site hidden danger inspection type;

[0030] By setting the number of clusters, a clustering basis is provided for subsequent clustering operations; by performing clustering operations on each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger inspection feature data matrix in the historical hidden danger inspection feature data matrix set, similar data therein are combined into one category, and the virtual scenes subsequently constructed based on these data of the same category are virtual scenes of the same type.

[0031] Preferably, in S21, clustering each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set according to the number of clusters includes the following steps:

[0032] S211, Constructing high-risk operation characteristic data to cluster locust populations represents the i-th locust in the locust population clustered by the high-risk operation characteristic data, represents the size of the high-risk operation feature data clustering locust population; sets the maximum iteration number of the high-risk operation feature data clustering locust population to d1 and the current iteration number to d2, which are recorded as the maximum iteration number of feature data clustering and the current iteration number of feature data clustering respectively; the search space dimension of the high-risk operation feature data clustering locust population is

[0033] S212, randomly selecting the same number of rows of data from each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set as the initial position matrix of each locust in the high-risk operation feature data clustering locust population, to obtain a first initial position matrix set d k ″ represents the initial position matrix of the k′th locust in the high-risk operation characteristic data clustering locust population; as follows,

[0034]

[0035] Among them, d k ″ 1i d k ″ 2i Respectively represent d k ″The position component matrix on the dimension of the feature classification center data matrix corresponding to the i-th safety accident type and d k ″The position component matrix on the dimension of the feature classification center data matrix corresponding to the i-th on-site hidden danger inspection type; d k ″ 1ijk d k ″2ijk Indicates d k ″ 1i The position component on the feature data dimension of the kth type in the jth feature classification center data set of the feature data matrix of the i-th safety accident type and the on-site hidden danger inspection type;

[0036] S213, calculating the Euclidean distance between each row of data in each historical safety accident feature data matrix in the historical safety accident feature data matrix set and the initial safety accident feature classification center data set corresponding to the initial position matrix of the k'th locust in the high-risk operation feature data cluster locust population, and then calculating the Euclidean distance between each row of data in each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set and the initial hidden danger investigation feature classification center data set corresponding to the initial position matrix of the k'th locust in the high-risk operation feature data cluster locust population, to obtain a first Euclidean distance matrix as follows,

[0037]

[0038] in, The Euclidean distance matrix represents the initial safety accident feature classification center data set corresponding to each row of data in the i-th historical safety accident feature data matrix in the historical safety accident feature data matrix set and the initial position matrix of the k′th locust in the high-risk operation feature data clustering locust population; Represents the Euclidean distance matrix between each row of data in the i-th historical hidden danger inspection feature data matrix set and the initial hidden danger inspection feature classification center data set corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population; It represents the Euclidean distance between the j-th row data in the ith historical safety accident feature data matrix in the historical safety accident feature data matrix set and the k-th row of initial safety accident feature classification center data corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population; It represents the Euclidean distance between the j-th row data in the ith historical hidden danger inspection feature data matrix in the historical hidden danger inspection feature data matrix set and the k-th row of initial hidden danger inspection feature classification center data corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population;

[0039] According to the first Euclidean distance matrix Constructing the fitness function of the k′th locust in the locust population clustered with high-risk operation characteristic data as follows,

[0040]

[0041] S214, start iteration, before iteration, set the current iteration number of the characteristic data clustering to 1; in the first round of iteration, use the high-risk operation characteristic data clustering locust population k'th locust fitness function Calculate the fitness value of the initial position matrix of each locust in the first initial position matrix set to obtain a first fitness value set; use the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the locust as the first global optimal fitness and the first global optimal position respectively; update the initial position matrix of each locust in the first initial position matrix set according to the first global optimal fitness and the first global optimal position; after the update is completed, add 1 to the current iteration number of the feature data clustering and perform the next round of iteration;

[0042] In each other round of iteration, the fitness function of the k′th locust in the locust population is clustered using the high-risk operation feature data 1. Calculate the fitness value of the position matrix of each locust in the high-risk operation feature data clustering locust population updated in the previous round of iteration to obtain a second fitness value set; use the maximum fitness value in the second fitness value set and the corresponding initial position matrix of the locust as the second global optimal fitness and the second global optimal position respectively; update the position matrix of each locust in the high-risk operation feature data clustering locust population updated in the previous round of iteration according to the second global optimal fitness and the second global optimal position; after the update is completed, add 1 to the current iteration number of the feature data clustering and perform the next round of iteration;

[0043] S215. When d2≥d1, stop iteration and output the first final global optimal position; otherwise, continue iteration until d2≥d1; use the position component of the historical safety accident feature classification center data corresponding to the first final global optimal position as the historical safety accident feature classification center data matrix set, and use the position component of the historical hidden danger investigation feature classification center data corresponding to the first final global optimal position as the historical hidden danger investigation feature classification center data matrix set;

[0044] The locust optimization algorithm can effectively explore different areas of the solution space, which helps to find the global optimal solution or the approximate optimal solution; it can usually find a better solution within a smaller number of iterations; it has fewer parameters, which reduces the complexity of parameter adjustment and is easier to use; it can effectively search the solution space and avoid falling into the local optimal solution; based on the above advantages, the locust optimization algorithm is used in this scheme to iterate and adjust multiple clustering center data of each type of safety accident feature data and hidden danger inspection feature data in the historical safety accident feature data matrix set and the historical hidden danger inspection feature data matrix set at the same time, and the overall discrete degree of the historical safety accident feature data matrix set and the historical hidden danger inspection feature data matrix set is used as the fitness function; therefore, as the iteration proceeds, the overall discrete degree of the historical safety accident feature data matrix set and the historical hidden danger inspection feature data matrix set becomes smaller and smaller, indicating that the classification effect of the multiple classification center data obtained by iteration is better.

[0045] Preferably, S3 comprises the following steps:

[0046] S31, setting a corresponding safety accident severity level for each row of historical safety accident feature classification center data in each historical safety accident feature classification center data matrix in the historical safety accident feature classification center data matrix set, and obtaining a safety accident severity level data matrix c1′; then setting a corresponding hidden danger investigation standard level for each row of historical hidden danger investigation feature classification center data in each historical hidden danger investigation feature classification center data matrix in the historical hidden danger investigation feature classification center data matrix set, and obtaining a hidden danger investigation standard level data matrix c′2; c1′ and c′2 are respectively as follows,

[0047]

[0048] Among them, c1′ ij , c′ 2ij They respectively represent the severity level data corresponding to the j-th feature classification center data set of the set i-th safety accident type and the standard level data corresponding to the j-th feature classification center data set of the feature data matrix of the on-site hidden danger inspection type;

[0049] By setting the severity level of safety accidents and the standard level of hidden danger inspection, a variety of options are provided for subsequent employees to simulate virtual scenes, allowing employees to choose appropriate virtual scenes for simulation based on their actual situation.

[0050] Preferably, S4 comprises the following steps:

[0051] S41, constructing corresponding safety accident virtual scenes respectively according to the historical safety accident feature data in the historical safety accident feature data matrix set corresponding to each row of safety accident feature classification center data in the historical safety accident feature classification center data matrix set, to obtain a final safety accident virtual scene matrix;

[0052] Then, according to the historical hidden danger inspection feature data in the historical hidden danger inspection feature data matrix set corresponding to each row of hidden danger inspection feature classification center data in the historical hidden danger inspection feature classification center data matrix set, corresponding hidden danger inspection virtual scenes are respectively constructed to obtain a final hidden danger inspection virtual scene matrix;

[0053] S42, constructing a simulation teaching scene according to the simulation teaching feature data matrix set to obtain a final simulation teaching scene set;

[0054] Preferably, the S41 comprises the following steps:

[0055] S411, constructing a first initial virtual scene generation model and a second initial virtual scene generation model;

[0056] S412, training and testing the first initial virtual scene generation model according to the historical safety accident feature data in the historical safety accident feature data matrix set corresponding to each row of safety accident feature classification center data in the historical safety accident feature classification center data matrix set and in conjunction with the historical safety accident feature classification center data matrix set and the safety accident severity level data matrix, to obtain a first final virtual scene generation model;

[0057] Then, according to the historical hidden danger inspection feature data in the historical hidden danger inspection feature data matrix set corresponding to each row of hidden danger inspection feature classification center data in the historical hidden danger inspection feature classification center data matrix set, and in conjunction with the historical hidden danger inspection feature classification center data matrix set and the hidden danger inspection specification level data matrix, the second initial virtual scene generation model is trained and tested to obtain a second final virtual scene generation model;

[0058] S413, generating a plurality of virtual safety accident scenes of different severity levels according to the safety accident severity level data matrix and using the first final virtual scene generation model to obtain a final safety accident virtual scene matrix; generating a plurality of virtual hidden danger investigation scenes of different specification levels according to the hidden danger investigation specification level data matrix and using the second final virtual scene generation model to obtain a final hidden danger investigation virtual scene matrix;

[0059] The first initial virtual scene generation model and the second initial virtual scene generation model are preferably generative adversarial networks;

[0060] By training and testing the first initial virtual scene generation model and the second initial virtual scene generation model, the virtual scenes generated by the first final virtual scene generation model and the second final virtual scene generation model are more accurate, thereby better facilitating accurate simulation operations for employees who subsequently select different virtual scenes.

[0061] Preferably, the S412 comprises the following steps:

[0062] S4121, classifying the historical safety accident feature data matrix set according to the historical safety accident feature classification center data matrix set to obtain a historical safety accident feature data classification matrix; and then classifying the historical hidden danger investigation feature data matrix set according to the historical hidden danger investigation feature classification center data matrix set to obtain a historical hidden danger investigation feature data classification matrix;

[0063] S4122, setting a first training data ratio and a second training data ratio; performing data division on the historical safety accident feature data classification matrix, the safety accident severity level data matrix, the historical hidden danger investigation feature data classification matrix, and the hidden danger investigation specification level data matrix according to the first training data ratio and the second training data ratio, respectively, to obtain the historical safety accident feature training data classification matrix, the safety accident severity level training data matrix, the historical safety accident feature test data classification matrix, the safety accident severity level test data matrix, the historical hidden danger investigation feature training data classification matrix, the hidden danger investigation specification level training data matrix, the historical hidden danger investigation feature test data classification matrix, and the hidden danger investigation specification level test data matrix;

[0064] S4123, setting a first training error threshold and a second training error threshold;

[0065] Input the historical safety accident feature training data classification matrix and the safety accident severity level training data matrix as training data and training labels respectively into the first initial virtual scene generation model for training; during the training process, when the training error is less than the first training error threshold, stop the training and obtain the first trained virtual scene generation model;

[0066] Input the historical hidden danger inspection feature training data classification matrix and the hidden danger inspection specification level training data matrix as training data and training labels respectively into the second initial virtual scene generation model for training; during the training process, when the training error is less than the second training error threshold, stop the training and obtain the second trained virtual scene generation model;

[0067] S4124, setting a first accuracy threshold and a second accuracy threshold;

[0068] Input the historical safety accident feature test data classification matrix and the safety accident severity level test data matrix as test data and test labels respectively into the first trained virtual scene generation model for testing; after the test is completed, obtain a first test accuracy rate; when the first test accuracy rate is greater than or equal to a first accuracy rate threshold, use the first trained virtual scene generation model as the first final virtual scene generation model;

[0069] The historical hidden danger inspection feature test data classification matrix and the hidden danger inspection specification level test data matrix are respectively input as test data and test labels into the second trained virtual scene generation model for testing; after the test is completed, a second test accuracy rate is obtained; when the second test accuracy rate is greater than or equal to a second accuracy rate threshold, the second trained virtual scene generation model is used as the second final virtual scene generation model;

[0070] Preferably, the S5 comprises the following steps:

[0071] S51, according to the employee's historical operational performance in high-risk operations, selecting a corresponding safety accident virtual scene and a hidden danger investigation virtual scene from the final safety accident virtual scene matrix and the final hidden danger investigation virtual scene matrix to simulate the employee;

[0072] S52, using the final simulation teaching scene set to simulate skill teaching for employees;

[0073] An application system of emergency experience in educational safety based on virtual reality includes a virtual project type setting module, a project feature data type setting module, a historical feature data acquisition module, a clustering calculation module, a category label setting module, a virtual scene generation module and an employee simulation module.

[0074] The present invention has the following beneficial effects:

[0075] 1. In the present invention, corresponding virtual scenes are constructed for three types of safety accident types, on-site hidden danger inspection types, and safety skill simulation teaching types, so as to help employees conduct simulation drills before performing high-risk operations; there are different levels of settings for virtual scenes of each type of safety accident and on-site hidden danger inspection type, so that employees can choose simulation training of different degrees and difficulties according to their actual conditions, so as to achieve the best drill effect; wherein, in conjunction with clustering operations, a variety of virtual scenes of different types and corresponding to different labels are generated, so as to meet the different virtual simulation requirements of different employees, so that each employee can be matched with a virtual scene simulation suitable for his or her own situation, so that all employees can receive targeted training.

[0076] 2. In the present invention, through training and testing, the virtual scenes generated by the first final virtual scene generation model and the second final virtual scene generation model are made more accurate, thereby better facilitating accurate simulation operations for employees who subsequently select different virtual scenes.

[0077] 3. In the present invention, the locust optimization algorithm is used to simultaneously iterate multiple cluster center data of each type of safety accident feature data and hidden danger inspection feature data, and the overall discrete degree is used as the fitness function; therefore, as the iteration proceeds, the overall discrete degree of the historical safety accident feature data matrix set and the historical hidden danger inspection feature data matrix set becomes smaller and smaller, thereby indicating that the classification effect of the multiple classification center data obtained by the iteration is better.

[0078] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0080] Figure 1 The present invention is a flowchart of an application method of emergency experience based on virtual reality in educational safety. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the invention to clearly and completely describe the technical solutions in the embodiments of the invention. Obviously, the described embodiments are only part of the embodiments of the invention, not all of the embodiments. Based on the embodiments in the invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the invention.

[0082] Embodiment 1

[0083] See also Figure 1 This embodiment is an application method of emergency experience based on virtual reality in educational safety, comprising the following steps:

[0084] S1. Collecting characteristic data of multiple groups of different types of safety accident scenes, hidden danger inspection scenes and simulation teaching scenes in history, and obtaining a historical safety accident characteristic data matrix set, a historical hidden danger inspection characteristic data matrix set and a simulation teaching characteristic data matrix set;

[0085] The S1 comprises the following steps:

[0086] S11, setting multiple safety accident types, on-site hidden danger investigation types, and safety skill simulation teaching types to obtain a safety accident type set, a hidden danger investigation type set, and a simulation teaching type set;

[0087] Then set the characteristic data types of each safety accident type, hidden danger investigation type and simulation teaching type in the safety accident type set, hidden danger investigation type set and simulation teaching type set to obtain the safety accident characteristic data type matrix a1, the hidden danger investigation characteristic data type matrix a2 and the simulation teaching characteristic data type matrix a3; respectively as follows,

[0088]

[0089]

[0090] Among them, a 1ij 、a 2ij 、a 3ij a1′ represents the jth characteristic data type of the i-th safety accident type, the on-site hidden danger investigation type, and the safety skill simulation teaching type respectively; i , a′ 2i , a3′ i Respectively represent the total number of characteristic data types of the i-th safety accident type, on-site hidden danger inspection type, and safety skill simulation teaching type; b represents the total number of set safety accident types, on-site hidden danger investigation types, and safety skills simulation teaching types;

[0091] S12, according to the safety accident characteristic data type matrix a1, the hidden danger inspection characteristic data type matrix a2 and the simulation teaching characteristic data type matrix a3, the characteristic data of multiple groups of historical safety accident scenes of different types, hidden danger inspection scenes and simulation teaching scenes are collected to obtain the historical safety accident characteristic data matrix set Historical hidden danger investigation feature data matrix set And the simulation teaching characteristic data matrix set b3′={b3′1,...,b3′ i ,...,b3′ b}; b1′ i , b2′ i , b3′ i They represent the characteristic data matrices of the set i-th safety accident type, on-site hidden danger investigation type, and safety skills simulation teaching type respectively; they are as follows,

[0092]

[0093]

[0094] Among them, b1′ ijk b2′ represents the kth type of feature data of the jth group of safety accident scenes collected for the ith safety accident type; ijk b3′ represents the kth type of characteristic data collected at the jth group of hidden danger inspection sites for the ith hidden danger inspection type; ijk represents the kth type of feature data of the jth group of simulation teaching scenes collected for the ith simulation teaching type; Indicates the total number of groups of characteristic data collected from different types of historical safety accident scenes, hidden danger inspection scenes, and simulation teaching scenes;

[0095] S2, clustering the historical safety accident feature data matrix set and the historical hidden danger investigation feature data matrix set and calculating the central data of each classification to obtain a historical safety accident feature classification central data matrix set and a historical hidden danger investigation feature classification central data matrix set;

[0096] The S2 comprises the following steps:

[0097] S21. Set the number of clusters According to the number of clusters, each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set are clustered to obtain a historical safety accident feature classification center data matrix set. And the historical hidden danger investigation feature classification center data matrix set c 1i 、c 2i They represent the characteristic classification center data matrix corresponding to the set i-th safety accident type and hidden danger investigation type respectively; they are as follows,

[0098]

[0099] Among them, c 1ijk 、c 2ijk Respectively represent the k-th type of feature data in the j-th feature classification center data set of the feature data matrix of the set i-th safety accident type and the on-site hidden danger inspection type;

[0100] In S21, clustering operations are performed on each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set according to the number of clusters, including the following steps:

[0101] S211, Constructing high-risk operation characteristic data to cluster locust populations represents the i-th locust in the locust population clustered by the high-risk operation characteristic data, represents the size of the high-risk operation feature data clustering locust population; sets the maximum iteration number of the high-risk operation feature data clustering locust population to d1 and the current iteration number to d2, which are recorded as the maximum iteration number of feature data clustering and the current iteration number of feature data clustering respectively; the search space dimension of the high-risk operation feature data clustering locust population is

[0102] S212, randomly selecting the same number of rows of data from each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set as the initial position matrix of each locust in the high-risk operation feature data clustering locust population, to obtain a first initial position matrix set d k ″ represents the initial position matrix of the k′th locust in the high-risk operation characteristic data clustering locust population; as follows,

[0103]

[0104] Among them, d k ″ 1i d k ″ 2i Respectively represent d k ″The position component matrix on the dimension of the feature classification center data matrix corresponding to the i-th safety accident type and d k ″The position component matrix on the dimension of the feature classification center data matrix corresponding to the i-th on-site hidden danger inspection type; d k ″ 1ijk d k ″ 2ijk Indicates d k ″ 1i The position component on the feature data dimension of the kth type in the jth feature classification center data set of the feature data matrix of the i-th safety accident type and the on-site hidden danger inspection type;

[0105] S213, calculating the Euclidean distance between each row of data in each historical safety accident feature data matrix in the historical safety accident feature data matrix set and the initial safety accident feature classification center data set corresponding to the initial position matrix of the k'th locust in the high-risk operation feature data cluster locust population, and then calculating the Euclidean distance between each row of data in each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set and the initial hidden danger investigation feature classification center data set corresponding to the initial position matrix of the k'th locust in the high-risk operation feature data cluster locust population, to obtain a first Euclidean distance matrix as follows,

[0106]

[0107] in, The Euclidean distance matrix represents the initial safety accident feature classification center data set corresponding to each row of data in the i-th historical safety accident feature data matrix in the historical safety accident feature data matrix set and the initial position matrix of the k′th locust in the high-risk operation feature data clustering locust population; Represents the Euclidean distance matrix between each row of data in the i-th historical hidden danger inspection feature data matrix set and the initial hidden danger inspection feature classification center data set corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population; It represents the Euclidean distance between the j-th row data in the ith historical safety accident feature data matrix in the historical safety accident feature data matrix set and the k-th row of initial safety accident feature classification center data corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population; It represents the Euclidean distance between the j-th row data in the ith historical hidden danger inspection feature data matrix in the historical hidden danger inspection feature data matrix set and the k-th row of initial hidden danger inspection feature classification center data corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data cluster locust population;

[0108] According to the first Euclidean distance matrix Constructing the fitness function of the k′th locust in the locust population clustered with high-risk operation characteristic data as follows,

[0109]

[0110] S214, start iteration, before iteration, set the current iteration number of the characteristic data clustering to 1; in the first round of iteration, use the high-risk operation characteristic data clustering locust population k'th locust fitness function Calculate the fitness value of the initial position matrix of each locust in the first initial position matrix set to obtain a first fitness value set; use the maximum fitness value in the first fitness value set and the corresponding initial position matrix of the locust as the first global optimal fitness and the first global optimal position respectively; update the initial position matrix of each locust in the first initial position matrix set according to the first global optimal fitness and the first global optimal position; after the update is completed, add 1 to the current iteration number of the feature data clustering and perform the next round of iteration;

[0111] In each other round of iteration, the fitness function of the k′th locust in the locust population is clustered using the high-risk operation feature data Calculate the fitness value of the position matrix of each locust in the high-risk operation feature data clustering locust population updated in the previous round of iterations to obtain a second fitness value set; use the maximum fitness value in the second fitness value set and the corresponding initial position matrix of the locust as the second global optimal fitness and the second global optimal position respectively; update the position matrix of each locust in the high-risk operation feature data clustering locust population updated in the previous round of iterations according to the second global optimal fitness and the second global optimal position; after the update is completed, add 1 to the current iteration number of the feature data clustering and perform the next round of iterations;

[0112] S215. When d2≥d1, stop iteration and output the first final global optimal position; otherwise, continue iteration until d2≥d1; use the position component of the historical safety accident feature classification center data corresponding to the first final global optimal position as the historical safety accident feature classification center data matrix set, and use the position component of the historical hidden danger investigation feature classification center data corresponding to the first final global optimal position as the historical hidden danger investigation feature classification center data matrix set;

[0113] S3, setting the category labels corresponding to the historical safety accident feature classification center data matrix set and the historical hidden danger investigation feature classification center data matrix set, and obtaining the safety accident severity level data matrix and the hidden danger investigation standard level data matrix;

[0114] The S3 comprises the following steps:

[0115] S31, setting a corresponding safety accident severity level for each row of historical safety accident feature classification center data in each historical safety accident feature classification center data matrix in the historical safety accident feature classification center data matrix set, and obtaining a safety accident severity level data matrix c1′; then setting a corresponding hidden danger investigation standard level for each row of historical hidden danger investigation feature classification center data in each historical hidden danger investigation feature classification center data matrix in the historical hidden danger investigation feature classification center data matrix set, and obtaining a hidden danger investigation standard level data matrix c′2; c1′ and c′2 are respectively as follows,

[0116]

[0117] Among them, c1′ ij , c′ 2ij They respectively represent the severity level data corresponding to the j-th feature classification center data set of the set i-th safety accident type and the standard level data corresponding to the j-th feature classification center data set of the feature data matrix of the on-site hidden danger inspection type;

[0118] S4, generating a plurality of virtual scenes corresponding to different types and different labels according to the historical safety accident feature data matrix set, the historical hidden danger inspection feature data matrix set, the simulation teaching feature data matrix set, the safety accident severity level data matrix and the hidden danger inspection specification level data matrix, and obtaining the final safety accident virtual scene matrix, the final hidden danger inspection virtual scene matrix and the final simulation teaching scene set;

[0119] The S4 comprises the following steps:

[0120] S41, constructing corresponding safety accident virtual scenes respectively according to the historical safety accident feature data in the historical safety accident feature data matrix set corresponding to each row of safety accident feature classification center data in the historical safety accident feature classification center data matrix set, to obtain a final safety accident virtual scene matrix;

[0121] Then, according to the historical hidden danger inspection feature data in the historical hidden danger inspection feature data matrix set corresponding to each row of hidden danger inspection feature classification center data in the historical hidden danger inspection feature classification center data matrix set, corresponding hidden danger inspection virtual scenes are respectively constructed to obtain a final hidden danger inspection virtual scene matrix;

[0122] The S41 comprises the following steps:

[0123] S411, constructing a first initial virtual scene generation model and a second initial virtual scene generation model;

[0124] S412, training and testing the first initial virtual scene generation model according to the historical safety accident feature data in the historical safety accident feature data matrix set corresponding to each row of safety accident feature classification center data in the historical safety accident feature classification center data matrix set and in conjunction with the historical safety accident feature classification center data matrix set and the safety accident severity level data matrix, to obtain a first final virtual scene generation model;

[0125] Then, according to the historical hidden danger inspection feature data in the historical hidden danger inspection feature data matrix set corresponding to each row of hidden danger inspection feature classification center data in the historical hidden danger inspection feature classification center data matrix set, and in conjunction with the historical hidden danger inspection feature classification center data matrix set and the hidden danger inspection specification level data matrix, the second initial virtual scene generation model is trained and tested to obtain a second final virtual scene generation model;

[0126] The S412 comprises the following steps:

[0127] S4121, classifying the historical safety accident feature data matrix set according to the historical safety accident feature classification center data matrix set to obtain a historical safety accident feature data classification matrix; and then classifying the historical hidden danger investigation feature data matrix set according to the historical hidden danger investigation feature classification center data matrix set to obtain a historical hidden danger investigation feature data classification matrix;

[0128] S4122, setting a first training data ratio and a second training data ratio; performing data division on the historical safety accident feature data classification matrix, the safety accident severity level data matrix, the historical hidden danger investigation feature data classification matrix, and the hidden danger investigation specification level data matrix according to the first training data ratio and the second training data ratio, respectively, to obtain the historical safety accident feature training data classification matrix, the safety accident severity level training data matrix, the historical safety accident feature test data classification matrix, the safety accident severity level test data matrix, the historical hidden danger investigation feature training data classification matrix, the hidden danger investigation specification level training data matrix, the historical hidden danger investigation feature test data classification matrix, and the hidden danger investigation specification level test data matrix;

[0129] S4123, setting a first training error threshold and a second training error threshold;

[0130] Input the historical safety accident feature training data classification matrix and the safety accident severity level training data matrix as training data and training labels respectively into the first initial virtual scene generation model for training; during the training process, when the training error is less than the first training error threshold, stop the training and obtain the first trained virtual scene generation model;

[0131] Input the historical hidden danger inspection feature training data classification matrix and the hidden danger inspection specification level training data matrix as training data and training labels respectively into the second initial virtual scene generation model for training; during the training process, when the training error is less than the second training error threshold, stop the training and obtain the second trained virtual scene generation model;

[0132] S4124, setting a first accuracy threshold and a second accuracy threshold;

[0133] Input the historical safety accident feature test data classification matrix and the safety accident severity level test data matrix as test data and test labels respectively into the first trained virtual scene generation model for testing; after the test is completed, obtain a first test accuracy rate; when the first test accuracy rate is greater than or equal to a first accuracy rate threshold, use the first trained virtual scene generation model as the first final virtual scene generation model;

[0134] The historical hidden danger inspection feature test data classification matrix and the hidden danger inspection specification level test data matrix are respectively input as test data and test labels into the second trained virtual scene generation model for testing; after the test is completed, a second test accuracy rate is obtained; when the second test accuracy rate is greater than or equal to a second accuracy rate threshold, the second trained virtual scene generation model is used as the second final virtual scene generation model;

[0135] S413, generating a plurality of virtual safety accident scenes of different severity levels according to the safety accident severity level data matrix and using the first final virtual scene generation model to obtain a final safety accident virtual scene matrix; generating a plurality of virtual hidden danger investigation scenes of different specification levels according to the hidden danger investigation specification level data matrix and using the second final virtual scene generation model to obtain a final hidden danger investigation virtual scene matrix;

[0136] S42, constructing a simulation teaching scene according to the simulation teaching feature data matrix set to obtain a final simulation teaching scene set;

[0137] S5. Use the final safety accident virtual scenario matrix, the final hidden danger investigation virtual scenario matrix and the final simulation teaching scenario set to simulate the operation of employees;

[0138] The S5 comprises the following steps:

[0139] S51, according to the employee's historical operational performance in high-risk operations, selecting a corresponding safety accident virtual scene and a hidden danger investigation virtual scene from the final safety accident virtual scene matrix and the final hidden danger investigation virtual scene matrix to simulate the employee;

[0140] S52: Use the final simulation teaching scene set to simulate skill teaching for employees.

[0141] Embodiment 2

[0142] This embodiment discloses an application system of emergency experience based on virtual reality in educational safety, the system can implement the method of the above embodiment, including a virtual project type setting module, a project feature data type setting module, a historical feature data collection module, a clustering calculation module, a category label setting module, a virtual scene generation module and an employee simulation module;

[0143] The virtual project type setting module is used to set a variety of safety accident types, on-site hidden danger inspection types and safety skills simulation teaching types, and obtain a safety accident type set, a hidden danger inspection type set and a simulation teaching type set;

[0144] The project feature data type setting module is used to set corresponding feature data types according to the safety accident type set, the hidden danger inspection type set and the simulation teaching type set, and obtain the safety accident feature data type matrix, the hidden danger inspection feature data type matrix and the simulation teaching feature data type matrix;

[0145] The historical feature data acquisition module is used to collect feature data of multiple groups of different types of historical safety accident scenes, hidden danger inspection scenes and simulation teaching scenes according to the safety accident feature data type matrix, the hidden danger inspection feature data type matrix and the simulation teaching feature data type matrix, to obtain a historical safety accident feature data matrix set, a historical hidden danger inspection feature data matrix set and a simulation teaching feature data matrix set;

[0146] The clustering calculation module is used to cluster the historical safety accident feature data matrix set and the historical hidden danger investigation feature data matrix set and calculate the central data of each classification to obtain the historical safety accident feature classification central data matrix set and the historical hidden danger investigation feature classification central data matrix set;

[0147] The category label setting module is used to set the category labels corresponding to the historical safety accident feature classification center data matrix set and the historical hidden danger investigation feature classification center data matrix set, and obtain the safety accident severity level data matrix and the hidden danger investigation standard level data matrix;

[0148] The virtual scene generation module is used to generate virtual scenes corresponding to multiple different types and different labels according to the historical safety accident feature classification center data matrix set, the historical hidden danger investigation feature classification center data matrix set, the historical safety accident feature data matrix set, the historical hidden danger investigation feature data matrix set, the simulation teaching feature data matrix set, the safety accident severity level data matrix and the hidden danger investigation specification level data matrix, to obtain the final safety accident virtual scene matrix, the final hidden danger investigation virtual scene matrix and the final simulation teaching scene set;

[0149] The employee simulation module is used to simulate the operation of employees using the final safety accident virtual scene matrix, the final hidden danger inspection virtual scene matrix and the final simulation teaching scene set.

[0150] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0151] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. A method for applying emergency experience based on virtual reality in educational safety, characterized in that: The following steps are involved: S1. Collecting characteristic data of multiple groups of different types of safety accident scenes, hidden danger inspection scenes and simulation teaching scenes in history, and obtaining a historical safety accident characteristic data matrix set, a historical hidden danger inspection characteristic data matrix set and a simulation teaching characteristic data matrix set; S2, clustering the historical safety accident feature data matrix set and the historical hidden danger investigation feature data matrix set and calculating the central data of each classification to obtain a historical safety accident feature classification central data matrix set and a historical hidden danger investigation feature classification central data matrix set; S3, setting the category labels corresponding to the historical safety accident feature classification center data matrix set and the historical hidden danger investigation feature classification center data matrix set, and obtaining the safety accident severity level data matrix and the hidden danger investigation standard level data matrix; S4, generating a plurality of virtual scenes corresponding to different types and different labels according to the historical safety accident feature data matrix set, the historical hidden danger inspection feature data matrix set, the simulation teaching feature data matrix set, the safety accident severity level data matrix and the hidden danger inspection specification level data matrix, and obtaining the final safety accident virtual scene matrix, the final hidden danger inspection virtual scene matrix and the final simulation teaching scene set; S5. Use the final safety accident virtual scenario matrix, the final hidden danger inspection virtual scenario matrix and the final simulation teaching scenario set to simulate operations on employees.

2. The method for applying emergency experience based on virtual reality in educational safety according to claim 1, characterized in that: The S1 comprises the following steps: S11, setting characteristic data types corresponding to multiple safety accident types, on-site hidden danger inspection types, and safety skills simulation teaching types, to obtain a safety accident characteristic data type matrix, a hidden danger inspection characteristic data type matrix, and a simulation teaching characteristic data type matrix; S12. Collect characteristic data of multiple groups of different types of historical safety accident scenes, hidden danger inspection scenes and simulation teaching scenes according to the safety accident characteristic data type matrix, hidden danger inspection characteristic data type matrix and simulation teaching characteristic data type matrix to obtain a historical safety accident characteristic data matrix set, a historical hidden danger inspection characteristic data matrix set and a simulation teaching characteristic data matrix set.

3. The method for applying emergency experience based on virtual reality in educational safety according to claim 2, characterized in that: The S2 comprises the following steps: S21. Set the number of clusters; perform clustering operations on each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set according to the number of clusters to obtain a historical safety accident feature classification center data matrix set and a historical hidden danger investigation feature classification center data matrix set.

4. The method for applying emergency experience based on virtual reality in educational safety according to claim 3 is characterized in that: In S21, clustering operations are performed on each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set according to the number of clusters, including the following steps: S211, constructing a high-risk operation characteristic data clustering locust population; setting the maximum number of iterations of the high-risk operation characteristic data clustering locust population to d1 and the current number of iterations to d2, which are recorded as the maximum number of iterations of characteristic data clustering and the current number of iterations of characteristic data clustering respectively; S212, randomly selecting the same number of rows of data from each historical safety accident feature data matrix in the historical safety accident feature data matrix set and each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set as the initial position matrix of each locust in the high-risk operation feature data clustering locust population, to obtain a first initial position matrix set; S213, calculating the Euclidean distance between each row of data in each historical safety accident feature data matrix in the historical safety accident feature data matrix set and the initial safety accident feature classification center data set corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data clustering locust population, and then calculating the Euclidean distance between each row of data in each historical hidden danger investigation feature data matrix in the historical hidden danger investigation feature data matrix set and the initial hidden danger investigation feature classification center data set corresponding to the initial position matrix of the k′th locust in the high-risk operation feature data clustering locust population, to obtain a first Euclidean distance matrix; constructing the fitness function of the k′th locust in the high-risk operation feature data clustering locust population according to the first Euclidean distance matrix; S214, start iteration; in each round of iteration, use the fitness function of the k′th locust in the high-risk operation characteristic data clustering locust population to calculate the fitness value of the position matrix of each locust in the high-risk operation characteristic data clustering locust population updated in the previous round of iteration, and update the position matrix of each locust in the high-risk operation characteristic data clustering locust population updated in the previous round of iteration; S215. When d2≥d1, stop iteration and output the first final global optimal position; otherwise, continue iteration until d2≥d1; use the position components of the historical safety accident feature classification center data and the historical hidden danger investigation feature classification center data corresponding to the first final global optimal position as the historical safety accident feature classification center data matrix set and the historical hidden danger investigation feature classification center data matrix set, respectively.

5. The method for applying emergency experience based on virtual reality in educational safety according to claim 4, characterized in that: The S3 comprises the following steps: S31. Set a corresponding safety accident severity level for each row of historical safety accident feature classification center data in each historical safety accident feature classification center data matrix in the historical safety accident feature classification center data matrix set, and obtain a safety accident severity level data matrix; then set a corresponding hidden danger investigation standard level for each row of historical hidden danger investigation feature classification center data in each historical hidden danger investigation feature classification center data matrix in the historical hidden danger investigation feature classification center data matrix set, and obtain a hidden danger investigation standard level data matrix.

6. The method for applying emergency experience based on virtual reality in educational safety according to claim 5, characterized in that: The S4 comprises the following steps: S41, constructing corresponding safety accident virtual scenes respectively according to the historical safety accident feature data in the historical safety accident feature data matrix set corresponding to each row of safety accident feature classification center data in the historical safety accident feature classification center data matrix set, to obtain a final safety accident virtual scene matrix; Then, according to the historical hidden danger inspection feature data in the historical hidden danger inspection feature data matrix set corresponding to each row of hidden danger inspection feature classification center data in the historical hidden danger inspection feature classification center data matrix set, corresponding hidden danger inspection virtual scenes are respectively constructed to obtain a final hidden danger inspection virtual scene matrix; S42, constructing a simulation teaching scene according to the simulation teaching feature data matrix set to obtain a final simulation teaching scene set.

7. The method for applying emergency experience based on virtual reality in educational safety according to claim 6, characterized in that: The S41 comprises the following steps: S411, constructing a first initial virtual scene generation model and a second initial virtual scene generation model; S412, training and testing the first initial virtual scene generation model according to the historical safety accident feature classification center data matrix set and the historical safety accident feature data matrix set in conjunction with the historical safety accident feature classification center data matrix set and the safety accident severity level data matrix to obtain a first final virtual scene generation model; Then, the second initial virtual scene generation model is trained and tested according to the historical hidden danger investigation feature classification center data matrix set and the historical hidden danger investigation feature data matrix set in conjunction with the historical hidden danger investigation feature classification center data matrix set and the hidden danger investigation specification level data matrix to obtain a second final virtual scene generation model; S413. Generate a plurality of virtual safety accident scenes of different severity levels according to the safety accident severity level data matrix and using the first final virtual scene generation model to obtain a final safety accident virtual scene matrix; generate a plurality of virtual hidden danger inspection scenes of different standard levels according to the hidden danger inspection standard level data matrix and using the second final virtual scene generation model to obtain a final hidden danger inspection virtual scene matrix.

8. The method for applying emergency experience based on virtual reality in educational safety according to claim 7, characterized in that: The first initial virtual scene generation model and the second initial virtual scene generation model adopt a generative adversarial network model.

9. The method for applying emergency experience based on virtual reality in educational safety according to claim 8, characterized in that: The S5 comprises the following steps: S51, according to the employee's historical operational performance in high-risk operations, selecting a corresponding safety accident virtual scene and a hidden danger investigation virtual scene from the final safety accident virtual scene matrix and the final hidden danger investigation virtual scene matrix to simulate the employee; S52: Use the final simulation teaching scene set to simulate skill teaching for employees.

10. A system for implementing the method for applying the virtual reality-based emergency experience in educational safety as described in any one of claims 1 to 9.

Images