CN118289067A - Train operation safety test method and system based on virtual-real combination - Google Patents
Train operation safety test method and system based on virtual-real combination Download PDFInfo
- Publication number
- CN118289067A CN118289067A CN202410384056.5A CN202410384056A CN118289067A CN 118289067 A CN118289067 A CN 118289067A CN 202410384056 A CN202410384056 A CN 202410384056A CN 118289067 A CN118289067 A CN 118289067A
- Authority
- CN
- China
- Prior art keywords
- test
- scene
- virtual
- information
- real
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000009781 safety test method Methods 0.000 title claims abstract description 13
- 238000012360 testing method Methods 0.000 claims abstract description 298
- 238000011156 evaluation Methods 0.000 claims abstract description 72
- 238000004088 simulation Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 238000011076 safety test Methods 0.000 claims abstract description 17
- 230000000007 visual effect Effects 0.000 claims abstract description 14
- 238000001212 derivatisation Methods 0.000 claims description 18
- 230000002068 genetic effect Effects 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 9
- 230000001502 supplementing effect Effects 0.000 claims description 8
- 238000013209 evaluation strategy Methods 0.000 claims description 7
- 239000000470 constituent Substances 0.000 claims description 6
- 238000003066 decision tree Methods 0.000 claims description 6
- 239000013589 supplement Substances 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 5
- 238000004220 aggregation Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 230000007614 genetic variation Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 4
- 238000007726 management method Methods 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 9
- 238000012544 monitoring process Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 7
- 230000009545 invasion Effects 0.000 description 7
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000005856 abnormality Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 2
- 238000013523 data management Methods 0.000 description 2
- 238000013100 final test Methods 0.000 description 2
- 230000009469 supplementation Effects 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a train operation safety test method and system based on virtual-real combination, and relates to the field of safety test. The method comprises the following steps: importing information to be tested into a platform access layer for demand analysis, and determining test scene demand information; matching in a scene library to obtain a virtual test scene, and building a test channel by combining a real test scene; sequentially performing simulation test in the test channel based on the test tasks received by the management layer, and performing intelligent evaluation on simulation test results; and inputting the test evaluation result into an application layer for visual display. The method solves the technical problems of poor generalization capability, long test period and high cost of the test results due to the fact that the limited test scenes can not cover high-risk and high-complexity scenes in the prior art, and achieves the technical effects of improving the accuracy of the test results by improving the number of the test scenes and reducing the test period time and the cost by a virtual-real combined simulation method.
Description
Technical Field
The application relates to the technical field of safety test, in particular to the field of train safety test, and specifically relates to a train operation safety test method and system based on virtual-real combination.
Background
Before the railway train running environment safety detection and monitoring system is put into use, a large number of tests are required to be carried out to meet the application requirements. At present, the safety detection monitoring equipment and system test of the train running environment mainly comprise actual test environments which are actually built, and have the advantages of long test period, high cost, limited test coverage scenes, difficulty in covering boundary scenes with high risk, extreme, high complexity and the like, and incapability of meeting the test requirements of different layers of algorithms, software, hardware and systems. With the application of artificial intelligence technology, the safety detection and monitoring result of the train running environment is more dependent on the complexity of the testing environment and the completeness of the testing data, the hardware-in-the-loop testing based on the real environment is difficult to meet the testing requirement, and the safety detection and monitoring equipment, the system testing technology and the method of the train running environment based on complex scene driving and virtual-real combination are required to be researched.
In summary, in the prior art, because the test scenario is limited, the high-risk and high-complexity scenario cannot be covered, so that the generalization capability of the test result is poor, the test period is long, and the cost is high.
Disclosure of Invention
Based on the above, it is necessary to provide a train operation safety test method and system based on virtual-real combination, which can increase the number of test scenes and reduce the test cycle time and cost.
In a first aspect, a method for testing train operation safety based on virtual-real combination is provided, the method comprising: importing information to be tested into a platform access layer for demand analysis, and determining test scene demand information; the test scene demand information is transferred to a base layer, matching is carried out in a scene library based on virtual test scene demands, a virtual test scene is obtained, and a test channel is built by combining a real test scene; sequentially performing simulation test in the test channel based on the test tasks received by the management layer, and performing intelligent evaluation on simulation test results; and inputting the test evaluation result into an application layer for visual display.
In a second aspect, a train operation safety test system based on virtual-real combination is provided, the system comprising: the test scene demand information determining module is used for importing information to be tested into a platform access layer for demand analysis and determining test scene demand information; the test channel construction module is used for streaming the test scene demand information to a base layer, matching the test scene demand information in a scene library based on virtual test scene demands to obtain a virtual test scene, and constructing a test channel by combining a real test scene; the intelligent evaluation module is used for sequentially carrying out simulation tests in the test channel based on the test tasks received by the management layer and carrying out intelligent evaluation on simulation test results; and the visual display module is used for inputting the test evaluation result into the application layer for visual display.
In a third aspect, there is provided a computer device comprising a memory storing a computer program and a processor implementing the steps of the first aspect when the computer program is executed.
In a fourth aspect, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the first aspect.
The train operation safety test method and the train operation safety test system based on virtual-real combination solve the technical problems that in the prior art, high-risk and high-complexity scenes cannot be covered due to limited test scenes, so that the generalization capability of test results is poor, the test period is long and the cost is high, and the technical effects of improving the accuracy of the test results and reducing the test period time and the cost by a virtual-real combination simulation method are achieved by improving the number of the test scenes.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a flow chart of a train operation safety test method based on virtual-real combination in one embodiment;
fig. 2 is a schematic flow diagram of an application platform of a train operation safety test method based on virtual-real combination in one embodiment;
FIG. 3 is a block diagram of a train operation safety test system based on virtual-real combination in one embodiment;
Fig. 4 is an internal structural diagram of a computer device in one embodiment.
Reference numerals illustrate: the system comprises a test scene demand information determining module 11, a test channel constructing module 12, an intelligent evaluation module 13 and a visual display module 14.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
As shown in fig. 1, the application provides a train operation safety test method based on virtual-real combination, which comprises the following steps:
importing information to be tested into a platform access layer for demand analysis, and determining test scene demand information;
The virtual-real combination is a comprehensive method or technology, combines the characteristics and advantages of the virtual environment and the real environment to provide a more comprehensive, efficient and accurate solution. The virtual and real combination is applied to the train operation safety test, and aims to combine the advantages of the virtual test and the real test so as to provide a more comprehensive, more efficient and more accurate test method. The method is beneficial to improving the safety of train operation, reducing the accident risk and promoting the sustainable development of the railway transportation industry.
As shown in fig. 2, a virtual-real combined test platform is developed to meet the requirements of safety guarantee detection, monitoring algorithm, hardware, software and system test of a train running environment, wherein the information to be tested comprises a model algorithm, software, hardware and system to be tested, which are target information to be tested, the platform consists of an access layer, a base layer, a management layer and an application layer, wherein the platform access layer is used for receiving the information to be tested, and when the information to be tested meets the conditions of the model algorithm, software, hardware and system test access, the information to be tested is accessed to the platform through middleware, and the requirement analysis is performed based on the purpose and the requirement of the test of the information to be tested; the foundation layer comprises a real test scene and a virtual test scene, wherein the real test scene comprises environments of different working conditions such as railway roadbed, tunnel, side slope and the like, and detection equipment such as a video camera and radar is installed. The foundation layer comprises simulation conditions of different invasion types such as strong wind, strong rain, strong fog and the like, debris flow, foreign matter invasion limit, perimeter invasion and the like, and a scene library under dangerous working conditions is generated through simulation and historical data; the management layer comprises test task management, test management, data management, test evaluation, platform management and the like. The task management is to manage the task to be tested, and the test management is to manage the process of a specific test task. The data management is recording of test data, and the like. The test evaluation gives the test result. Platform management includes user management, platform maintenance, etc.; the application layer provides visual test result display according to the user requirements, and meets the requirements of model algorithm, software, hardware and system test. The test scene refers to a virtual-real combined real test scene and a virtual test scene, the test scene of the information to be tested is determined based on the demand analysis, for example, different weather conditions, track types, selected models, load conditions and the like, the determined test scene is converted into specific test scene demand information, for example, specific description of the test scene, required data sets, test steps and the like, and by determining the test scene demand information of the information to be tested, specific directions and data are provided for subsequent virtual-real combined tests, so that the validity and pertinence of the tests are ensured.
Performing test attribute identification on the information to be tested, and determining the attribute of the object to be tested;
based on the attribute of the object to be tested, extracting the characteristics of the information to be tested according to the attribute preset dimension to obtain the test characteristics of the object to be tested;
Setting a clustering center, wherein the clustering center corresponds to a preset test scene demand characteristic, and clustering the test characteristic of the object to be tested based on the clustering center to obtain the test scene demand information.
The test attribute identification means that the information to be tested is analyzed, the specific condition of the information to be tested is obtained, for example, the information to be tested is obtained, the test attribute identification is carried out on the information to be tested, the test attribute comprises an algorithm, software, hardware and a system, and the attribute of an object to be tested such as the algorithm is determined; the attribute preset dimension is data which are required to be used when the information to be tested is tested, and refers to data which are set by a worker, such as a scene, a target, safety requirements and the like of the application of the information to be tested, corresponding characteristics of the application scene, the running environment, the safety and the accuracy of the information to be tested are determined according to the information to be tested, and the corresponding characteristics of the information to be tested are extracted based on to obtain test characteristics of an object to be tested; the clustering center is preset by staff according to the requirement characteristics of the test scene, and represents different test scenes and conditions, such as different environments for train operation, different operation conditions, specific safety requirements and the like; after the clustering center is set, the test features of the objects to be tested are required to be clustered by using a clustering algorithm, the distance or similarity between the test features of the objects to be tested and the clustering center is analyzed to cluster, information corresponding to the preset test scene demand features is obtained through cluster analysis, and the test scene demand information is determined according to the information. By setting a clustering center and carrying out clustering analysis based on the clustering center, the test characteristics of the object to be tested can be matched with the preset test scene demand characteristics, so that information corresponding to the test scene demand is determined, and powerful support is provided for subsequent test work.
Collecting a record data set of train operation, classifying the record data set through a decision tree model, and constructing a multi-level historical data set, wherein the multi-level historical data set comprises a weather category, an environment scene category and an operation safety category;
performing multi-category scene aggregation based on the multi-level historical data set to construct a multi-category scene data set;
Extracting scene elements from the multi-category scene data set to construct an element extraction set;
Traversing the multi-category scene data set and the element extraction set respectively based on preset scene elements to determine missing data description information;
and supplementing the scene and the elements according to the missing data description information, and constructing the element library by utilizing the scene-element library data after supplementing and the mapping relation thereof.
Collecting a record data set of a train in the running process, wherein the record data set comprises relevant information such as weather conditions, environment scenes, running safety events and the like, the environment scenes such as railway roadbed, tunnels, side slopes and other environments with different working conditions can be obtained through a control system of the train, monitoring equipment such as a video camera and the like, the record data set is classified through a decision tree model, the decision tree is a prediction model and represents a mapping relation between object attributes and object values, a root node in the decision tree represents the record data set, three leaf nodes respectively represent weather categories, environment scene categories, running safety categories, namely a multi-level historical data set, and classification subsets corresponding to the leaf nodes of the weather categories such as sunny days, rainy days, snowy days and the like; the classification subsets corresponding to the environmental scene categories are city, country, mountain area and the like; the operation safety category corresponds to a plurality of levels such as normal operation, fault early warning, emergency braking and the like, and each level comprises a corresponding data subset for subsequent scene aggregation and element extraction; based on a multi-level historical dataset, multi-category scene aggregation is carried out, namely weather, environment scenes and operation safety category data are aggregated together to form a multi-category scene dataset, the multi-category scene dataset comprises train operation data under various different scenes, specifically, simulation conditions of different invasion categories such as strong wind, strong rain, strong fog and the like, debris flow, foreign matter invasion and perimeter invasion and the like and the operation state of a train under the conditions, scene element extraction is carried out on the multi-category scene dataset, namely key elements related to train operation safety such as weather conditions, environment scene characteristics, train operation state and the like are extracted from the dataset, and an element extraction set is constructed; the scene elements are set by the staff, based on the preset scene elements, a multi-category scene data set and an element extraction set are traversed, missing data which possibly exist in the data set can be found, the missing data description information refers to data which are not actually collected, supplementary simulation is needed, for example, a target scene does not have a debris flow natural disaster, the monitoring device does not monitor related information, but does not need to simulate the environment scene where the debris flow occurs, corresponding missing data description information is generated, data supplementary work is carried out according to the missing data description information, a final element library is constructed by using the supplementary scene-element library data and the mapping relation of the supplementary scene-element library, and the element library contains train operation safety related elements under various scenes and provides powerful support for subsequent train operation safety tests. By constructing a complete train operation safety test element library, powerful guarantee is provided for the safety operation of the train.
According to the missing data description information, data acquisition is carried out, and the number evaluation is carried out on acquisition results;
When the data quantity evaluation result does not meet the supplement requirement, core description information of the missing data description information is obtained;
Performing characteristic genetic derivatization based on the core description information to obtain a genetic derivatization data set, wherein the genetic derivatization data set is obtained by performing characteristic genetic variation on the core description information according to a preset derivatization step length and a derivatization direction;
Based on the genetic derivative data set, carrying out cooperative change on the missing data description information except the core description information to construct a missing data derivative set;
And acquiring a historical record data set based on the core description information, constructing a simulation module, performing simulation through the simulation module based on the missing data derivative set, and screening derivative data meeting a scene simulation result to supplement.
Acquiring corresponding data acquisition schemes according to the specific description information of the missing data, including types, ranges, influences and the like of the missing data, acquiring corresponding data, including acquisition methods, acquisition tools, acquisition periods and the like, namely acquiring the acquisition results, performing quantity evaluation on the acquisition results, wherein the quantity evaluation comprises checking whether the quantity of the data is enough, whether the distribution is uniform or not and the like, and acquiring core description information of the missing data description information if the quantity evaluation result does not meet the supplementary requirement, wherein the core description information refers to description information with the largest influence on the missing data and the most critical description information, and is a main reason or key factor of data missing; after the core description information is obtained, new data are generated by using a characteristic genetic derivatization method, characteristic genetic change is carried out on the core description information according to a preset derivatization step length and a derivatization direction, new data which are similar to original data but are not identical are generated, the new data are used for expanding a data set, for example, derivatization is carried out by using existing collected data according to the requirement of the data, the data which meet the scene requirement are obtained for supplementation, besides the core description information, description information comprising other missing data is obtained, the description information is carried out on the description information based on the genetic derivatization data set, collaborative change is carried out on the description information, the missing data derivatization set is constructed, data adjustment or generation is carried out simultaneously when the interrelation among a plurality of variables or factors is considered, after the genetic derivatization data set and the missing data derivatization set are obtained, a simulation module is constructed based on the data, different scene requirements should have corresponding description characteristics, if the simulation results meet the description characteristics, the derivatized data are indicated to be correct and usable if the description of the data of the scene is not met, and the requirement of the simulation results is not met; the derivative data meeting the scene simulation result is screened and supplemented into the original data set, so that the data set is perfected and the accuracy and reliability of the test are improved. The method has the advantages that targeted data acquisition, evaluation and supplementation are carried out according to the description information of the missing data, a more complete and accurate data set can be constructed, and the effectiveness and reliability of train operation safety test can be improved.
The test scene demand information is transferred to a base layer, matching is carried out in a scene library based on virtual test scene demands, a virtual test scene is obtained, and a test channel is built by combining a real test scene;
the method comprises the steps of transmitting test scene demand information obtained through cluster analysis to a base layer, wherein a pre-built scene library is arranged in the base layer and comprises various virtual test scenes, the virtual scenes are created through simulation and emulation technology and are used for simulating actual operation environments, matching is carried out in a field Jing Ku according to the test scene demand information, a virtual test scene matched with the demand is found, the virtual test scene is combined with the actual test scene, a test channel is built, the actual test scene is the actual operation environment, such as a specific railway line, a station, a roadbed and the like, and test data and results in the virtual scene can be interacted with and verified with data in the actual scene through building the test channel. By combining the virtual test scene and the real test scene, the test channel is built, so that the safety test can be effectively performed, and the test efficiency and accuracy are improved.
Based on the test scene requirement information, performing modularized decoupling of scene requirements, and extracting scene constituent elements of a test scene;
Traversing and matching in the scene library based on the scene constituent elements to obtain a matched scene element set, analyzing internal relations among scene elements of the matched scene element set, establishing a modularized test scene library, and establishing a test channel.
Based on the test scene demand information, carrying out detailed analysis and modularized decoupling, wherein the modularized decoupling refers to dividing a complex system into mutually independent modules so as to reduce the complexity of the system and improve the maintainability of codes; the extracted scene components are subjected to traversal matching with the virtual test scenes in the scene library. The scene library is a pre-constructed database containing various virtual test scenes, and each virtual scene in the scene library is traversed to find and extract the virtual test scene with matched scene components. And forming a matched virtual scene into a matched scene element set for subsequent testing work. The matching scene element set is obtained, and the internal relation among the scene elements is analyzed, so that the overall structure and characteristics of the test scene can be more comprehensively understood, and the operation safety performance can be more accurately simulated and tested; the method comprises the steps of establishing a modularized test scene library, wherein the modularized test scene library comprises a plurality of independent and manageable modules or sub-scenes, each module or sub-scene represents a specific test requirement or test scene component, and constructing a test channel, wherein the test channel is used for connecting a virtual test scene with an actual test scene and is used for transmitting test data, monitoring a test process and analyzing a test result. Based on the test scene demand information, the modularized decoupling extraction scene constituent elements are performed, traversal matching and internal connection analysis are performed in the field Jing Ku, and finally, a modularized test scene library and a test channel are established, so that the operation safety performance of a train can be simulated and tested more accurately, and the test efficiency and accuracy are improved.
Based on virtual test scene requirements, extracting a collaborative real scene, wherein the collaborative real scene is a real scene with highest similarity with the virtual test scene requirements;
Carrying out cooperativity and difference analysis on the cooperativity real scene according to the modularized test scene library;
And obtaining the components of the collaborative real scene based on the synergy, combining the differences to perform difference component fusion to obtain virtual and real scene components, and building the test channel.
Specific requirements of virtual test scenes, such as weather conditions, track types, train running states and the like, are clarified, and scenes with highest similarity with the requirements of the virtual test scenes, namely collaborative real scenes, are extracted from the existing real scenes; the method comprises the steps of selecting a real scene which meets the requirements best as a cooperative scene by comparing the characteristics and parameters of each scene in a virtual test scene and a real scene library; the collaborative analysis aims at finding out common characteristics and parameters of the virtual scene and the collaborative real scene, ensuring that the virtual scene and the collaborative real scene have similar behavior in the test, and the difference analysis focuses on differences between the virtual scene and the collaborative real scene, wherein the differences comprise weather conditions, track characteristics, train performances and the like; extracting components of a collaborative real scene through collaborative analysis, wherein the components are basic units for forming the real scene, combining with a result of the differential analysis, fusing the differential components in the virtual scene and the collaborative real scene, and obtaining a virtual-real scene containing virtual and real scene components after the differential components are fused, wherein the virtual-real scene combines the controllability of a virtual test scene and the authenticity of the real scene, provides a more practical environment for safety test, and builds a test channel based on the virtual-real scene. By constructing the test channel combining the virtual scene and the real scene, the accuracy and the effectiveness of the safety test are improved.
Sequentially performing simulation test in the test channel based on the test tasks received by the management layer, and performing intelligent evaluation on simulation test results;
The test tasks comprise different test scenes, test targets and test requirements, namely the test requirements of the information to be tested, after the test tasks are received, the test channels are required to be configured according to the specific requirements of the test tasks, the test channels are ensured to simulate the environment conforming to the requirements of the test tasks, in the simulation test process, data such as test management, performance indexes, safety performance and the like are required to be recorded so as to be convenient for subsequent analysis and evaluation, after the simulation test is completed, the test results are analyzed, namely the test results are objectively and comprehensively evaluated according to the specific requirements of the test tasks and the factors such as the performance indexes, the safety performance and the like comprehensively considered. By acquiring the intelligent evaluation of the simulation test result, the accuracy and the effectiveness of the safety test are improved.
Extracting discrete test results and continuous test results according to the simulation test results;
building a discrete evaluation module, and evaluating the discrete test result to obtain a discrete evaluation result;
Constructing a continuous evaluation module, and evaluating the continuous test result to obtain a continuous evaluation result;
setting a multi-level evaluation strategy, and generating a test evaluation result according to the strategy corresponding relation of the discrete evaluation result and the continuous evaluation result.
After the simulation test is completed, a large amount of test data, namely a simulation test result, is obtained, wherein the simulation test result comprises a discrete test result and a continuous test result, the discrete test result generally refers to different test data, different test directions or indexes, the discrete test result is the test result of one index or time point in the middle, whether abnormality exists or not is checked, for example, the target index is normal between 20 and 40, the risk level is between 10 and 20 and 40 and 50, 60 and 80 belong to a risk level two-level, the risk level is greater than 80, the risk level three-level is determined by comparing the test result of a single parameter with the data of each level, and if the abnormality exists, the abnormal risk level is determined; the continuous test result is a test result of a train running period, whether the continuous test result is good or not is judged, if a certain discrete parameter is abnormal, the corresponding evaluation result of the simulation test result is a first-level, and if the whole period or the whole continuous parameter is abnormal, the risk of the corresponding evaluation result is judged to be higher. The discrete evaluation module is used for judging whether a test result is input into a preset value range, for example, whether the evaluation synthesis of different monitoring data is tested in a normal range, and determining the test result of the discrete test parameter according to the input range probability or the ratio; the continuous evaluation module can evaluate the test result of the period by constructing a period time sequence chain, such as a Markov chain model or a neural network prediction model. And setting a multi-level evaluation strategy, wherein the multi-level evaluation strategy is set by a worker according to experience, and the final test evaluation result is generated by synthesis. The multi-level evaluation strategy may include different evaluation hierarchies and weight assignments to reflect the importance of different test scenarios and performance indicators. And carrying out deep analysis and evaluation on the simulation test result to obtain a discrete evaluation result and a continuous evaluation result, and generating a final test evaluation result according to a multi-level evaluation strategy. The method provides powerful data support and analysis basis for train operation safety, and improves the overall working efficiency and quality.
And inputting the test evaluation result into an application layer for visual display.
It is necessary to select a suitable visualization tool or platform, and input the test evaluation result into a visualization interface, for example, using a chart, a graph, a color code, etc. to display data, so that a user can quickly identify a performance trend, an abnormal value, or a potential problem. The test evaluation result is effectively input into the application layer for visual display, so that an intuitive and easy-to-understand mode is provided for a decision maker to acquire the test evaluation result. The method solves the technical problems of poor generalization capability, long test period and high cost of the test results due to the fact that the limited test scenes can not cover high-risk and high-complexity scenes in the prior art, and achieves the technical effects of improving the accuracy of the test results by improving the number of the test scenes and reducing the test period time and the cost by a virtual-real combined simulation method.
In summary, the invention has at least the following advantages:
1. Through a virtual test scene library generation technology, high-risk scenes such as strong wind, heavy rain, falling rocks, personnel invasion and the like which are possibly encountered by train operation and historical accident data are simulated based on a real test environment, a test scene library is generated, virtual tests are carried out based on the scene library, and the coverage rate of the test scene is improved.
2. The virtual-real fusion test scene is subjected to modularized decoupling, the test scene constituent elements are extracted, the internal relations among different virtual-real test scene elements under the complex test requirements are analyzed, a modularized test scene library is established, and based on the virtual-real environment, the reconstruction technology of different scene modules is provided according to the test requirements so as to cover the safe typical and boundary scene of the train running environment.
3. The dynamic and static elements of the test scenes such as weather, perimeter environment, personnel/animal activities and the like are comprehensively considered, the uncertainty of an artificial intelligent algorithm is combined, the safety guarantee system test technology of the train operation environment is carried out based on the expected functional safety analysis, an unknown risk scene prediction and safety evaluation method is established, and the test evaluation of the railway operation environment safety detection monitoring system equipment is realized.
As shown in fig. 3, an embodiment of the present application includes a train operation safety test system based on virtual-real combination, the system including:
The test scene demand information determining module 11 is used for importing information to be tested into a platform access layer for demand analysis, and determining test scene demand information;
The test channel construction module 12 is configured to stream the test scene requirement information to a base layer, match the test scene requirement information in a scene library based on a virtual test scene requirement to obtain a virtual test scene, and construct a test channel in combination with a real test scene;
the intelligent evaluation module 13 is used for sequentially performing simulation tests in the test channel based on the test tasks received by the management layer, and performing intelligent evaluation on simulation test results;
The visual display module 14 is used for inputting the test evaluation result into the application layer for visual display by the visual display module 14.
Further, the embodiment of the application further comprises:
the attribute determining module of the object to be tested is used for identifying the test attribute of the information to be tested and determining the attribute of the object to be tested;
The test feature acquisition module is used for extracting features of the information to be tested according to the attribute preset dimension based on the attribute of the object to be tested to obtain test features of the object to be tested;
The test scene demand information acquisition module is used for setting a clustering center, the clustering center corresponds to preset test scene demand characteristics, and the test characteristics of the object to be tested are clustered based on the clustering center to acquire the test scene demand information.
Further, the embodiment of the application further comprises:
The multi-level historical data set construction module is used for collecting a record data set of train operation, classifying the record data set through a decision tree model and constructing a multi-level historical data set, wherein the multi-level historical data set comprises a weather category, an environment scene category and an operation safety category;
The multi-category scene data set construction module is used for conducting multi-category scene aggregation based on the multi-level historical data set to construct a multi-category scene data set;
the element extraction set construction module is used for extracting scene elements from the multi-category scene data set and constructing an element extraction set;
The missing data description information determining module is used for traversing the multi-category scene data set and the element extraction set respectively based on preset scene elements to determine missing data description information;
The element library construction module is used for supplementing scenes and elements according to the missing data description information, and constructing the element library by utilizing the scene-element library data after supplementing and the mapping relation of the scene-element library data.
Further, the embodiment of the application further comprises:
the quantity evaluation module is used for collecting data according to the missing data description information and evaluating the quantity of the collection result;
The core descriptive information acquisition module is used for acquiring the core descriptive information of the missing data descriptive information when the data quantity evaluation result does not meet the supplement requirement;
The genetic derivative data set acquisition module is used for carrying out characteristic genetic derivative on the basis of the core description information to obtain a genetic derivative data set, wherein the genetic derivative data set is obtained by carrying out characteristic genetic variation on the core description information according to a preset derivative step length and a derivative direction;
The missing data derivative set construction module is used for carrying out cooperative change on missing data description information except the core description information based on the genetic derivative data set to construct a missing data derivative set;
And the derived data supplementing module is used for acquiring a historical record data set based on the core description information, constructing a simulation module, simulating and simulating the missing data derived set through the simulation module based on the missing data derived set, and screening derived data meeting the scene simulation result to supplement.
Further, the embodiment of the application further comprises:
the scene component extraction module is used for carrying out modularized decoupling of scene requirements based on the test scene requirement information and extracting scene components of the test scene;
The test channel construction module is used for performing traversal matching in the scene library based on the scene components to obtain a matched scene component set, performing internal relation analysis among scene components on the matched scene component set, establishing a modularized test scene library and constructing a test channel.
Further, the embodiment of the application further comprises:
The collaborative real scene extraction module is used for extracting a collaborative real scene based on virtual test scene requirements, wherein the collaborative real scene is a real scene with highest similarity with the virtual test scene requirements;
The collaborative difference analysis module is used for carrying out collaborative and difference analysis on the collaborative real scene according to the modularized test scene library;
And the difference component element fusion module is used for obtaining the components of the collaborative real scene based on the synergies, carrying out difference component element fusion by combining the differences to obtain virtual and real scene components, and building the test channel.
Further, the embodiment of the application further comprises:
the test result extraction module is used for extracting discrete test results and continuous test results according to the simulation test results;
The discrete evaluation result obtaining module is used for building a discrete evaluation module and evaluating the discrete test result to obtain a discrete evaluation result;
The continuous evaluation result acquisition module is used for constructing a continuous evaluation module, evaluating the continuous test result and acquiring a continuous evaluation result;
the test evaluation result generation module is used for setting a multi-level evaluation strategy and generating a test evaluation result according to the strategy corresponding relation of the discrete evaluation result and the continuous evaluation result.
For specific embodiments of the train operation safety test system based on the virtual-real combination, reference may be made to the above embodiments of the train operation safety test method based on the virtual-real combination, which are not described herein. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 4. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing news data, time attenuation factors and other data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by the processor to implement a virtual-real combination based train operation safety test method.
It will be appreciated by persons skilled in the art that the architecture shown in fig. 4 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In one embodiment, a computer device is provided that includes a memory having a computer program stored therein and a processor that when executing the computer program performs the steps of a method for testing train operation safety based on a combination of virtual and real.
In one embodiment, a computer readable storage medium having a computer program stored thereon is provided, which when executed by a processor implements the steps of a virtual-real combination based train operation safety test method.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
Claims (10)
1. The train operation safety test method based on virtual-real combination is characterized by comprising the following steps:
importing information to be tested into a platform access layer for demand analysis, and determining test scene demand information;
The test scene demand information is transferred to a base layer, matching is carried out in a scene library based on virtual test scene demands, a virtual test scene is obtained, and a test channel is built by combining a real test scene;
sequentially performing simulation test in the test channel based on the test tasks received by the management layer, and performing intelligent evaluation on simulation test results;
And inputting the test evaluation result into an application layer for visual display.
2. The method of claim 1, wherein importing information to be tested into a platform access layer for demand analysis, determining test scenario demand information, comprises:
Performing test attribute identification on the information to be tested, and determining the attribute of the object to be tested;
based on the attribute of the object to be tested, extracting the characteristics of the information to be tested according to the attribute preset dimension to obtain the test characteristics of the object to be tested;
Setting a clustering center, wherein the clustering center corresponds to a preset test scene demand characteristic, and clustering the test characteristic of the object to be tested based on the clustering center to obtain the test scene demand information.
3. The method of claim 1, comprising, prior to matching in the element library based on virtual test scenario requirements:
Collecting a record data set of train operation, classifying the record data set through a decision tree model, and constructing a multi-level historical data set, wherein the multi-level historical data set comprises a weather category, an environment scene category and an operation safety category;
performing multi-category scene aggregation based on the multi-level historical data set to construct a multi-category scene data set;
Extracting scene elements from the multi-category scene data set to construct an element extraction set;
Traversing the multi-category scene data set and the element extraction set respectively based on preset scene elements to determine missing data description information;
and supplementing the scene and the elements according to the missing data description information, and constructing the element library by utilizing the scene-element library data after supplementing and the mapping relation thereof.
4. The method of claim 3, wherein scene, element supplementing based on the missing data description information comprises:
according to the missing data description information, data acquisition is carried out, and the number evaluation is carried out on acquisition results;
When the data quantity evaluation result does not meet the supplement requirement, core description information of the missing data description information is obtained;
Performing characteristic genetic derivatization based on the core description information to obtain a genetic derivatization data set, wherein the genetic derivatization data set is obtained by performing characteristic genetic variation on the core description information according to a preset derivatization step length and a derivatization direction;
Based on the genetic derivative data set, carrying out cooperative change on the missing data description information except the core description information to construct a missing data derivative set;
And acquiring a historical record data set based on the core description information, constructing a simulation module, performing simulation through the simulation module based on the missing data derivative set, and screening derivative data meeting a scene simulation result to supplement.
5. The method of claim 1, wherein streaming the test scenario requirement information to a base layer, matching in a scenario library based on virtual test scenario requirements to obtain a virtual test scenario, and building a test channel in combination with a real test scenario, comprises:
based on the test scene requirement information, performing modularized decoupling of scene requirements, and extracting scene constituent elements of a test scene;
Traversing and matching in the scene library based on the scene constituent elements to obtain a matched scene element set, analyzing internal relations among scene elements of the matched scene element set, establishing a modularized test scene library, and establishing a test channel.
6. The method of claim 5, wherein the creating a modular test scenario library, building a test channel, comprises:
based on virtual test scene requirements, extracting a collaborative real scene, wherein the collaborative real scene is a real scene with highest similarity with the virtual test scene requirements;
Carrying out cooperativity and difference analysis on the cooperativity real scene according to the modularized test scene library;
And obtaining the components of the collaborative real scene based on the synergy, combining the differences to perform difference component fusion to obtain virtual and real scene components, and building the test channel.
7. The method of claim 1, wherein the intelligently evaluating the simulated test results comprises:
Extracting discrete test results and continuous test results according to the simulation test results;
building a discrete evaluation module, and evaluating the discrete test result to obtain a discrete evaluation result;
Constructing a continuous evaluation module, and evaluating the continuous test result to obtain a continuous evaluation result;
setting a multi-level evaluation strategy, and generating a test evaluation result according to the strategy corresponding relation of the discrete evaluation result and the continuous evaluation result.
8. Train operation safety test system based on virtual and real combination, which is characterized in that the system comprises:
the test scene demand information determining module is used for importing information to be tested into a platform access layer for demand analysis and determining test scene demand information;
The test channel construction module is used for streaming the test scene demand information to a base layer, matching the test scene demand information in a scene library based on virtual test scene demands to obtain a virtual test scene, and constructing a test channel by combining a real test scene;
The intelligent evaluation module is used for sequentially carrying out simulation tests in the test channel based on the test tasks received by the management layer and carrying out intelligent evaluation on simulation test results;
And the visual display module is used for inputting the test evaluation result into the application layer for visual display.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410384056.5A CN118289067A (en) | 2024-04-01 | 2024-04-01 | Train operation safety test method and system based on virtual-real combination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410384056.5A CN118289067A (en) | 2024-04-01 | 2024-04-01 | Train operation safety test method and system based on virtual-real combination |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118289067A true CN118289067A (en) | 2024-07-05 |
Family
ID=91677080
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410384056.5A Pending CN118289067A (en) | 2024-04-01 | 2024-04-01 | Train operation safety test method and system based on virtual-real combination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118289067A (en) |
-
2024
- 2024-04-01 CN CN202410384056.5A patent/CN118289067A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111259947A (en) | Power system fault early warning method and system based on multi-mode learning | |
| CN110703057A (en) | Power equipment partial discharge diagnosis method based on data enhancement and neural network | |
| CN115423009A (en) | Cloud edge coordination-oriented power equipment fault identification method and system | |
| CN111967712B (en) | Traffic risk prediction method based on complex network theory | |
| CN116862081B (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN117934998A (en) | Tunnel fire digital twin model credibility assessment method and system | |
| CN116308883A (en) | Regional power grid data overall management system based on big data | |
| CN117522652B (en) | Human living environment vulnerability evaluation method, system, intelligent terminal and storage medium | |
| CN117200449B (en) | Multi-dimensional algorithm analysis-based power grid monitoring management method and system | |
| CN113505980A (en) | Reliability evaluation method, device and system for intelligent traffic management system | |
| CN118289067A (en) | Train operation safety test method and system based on virtual-real combination | |
| CN116187932A (en) | Information system engineering supervision project risk self-adaptive assessment method | |
| CN113723478B (en) | Track circuit fault diagnosis method based on priori knowledge | |
| CN117172138B (en) | Urban traffic carbon emission prediction method and device based on deep learning | |
| CN117726053B (en) | Carbon emission monitoring method and system applied to digital platform system | |
| KR102473115B1 (en) | System and method for analysing report data | |
| CN118447459B (en) | Landslide accumulation monitoring method and system based on deep learning | |
| CN117993894B (en) | Rail transit operation and maintenance state data processing method and system based on artificial intelligence | |
| CN117952321B (en) | Soil erosion intelligent monitoring and early warning method and system based on land engineering | |
| CN116701912B (en) | Bearing fault diagnosis method and system based on one-dimensional convolutional neural network | |
| CN118313495A (en) | Al data modeling method based on distributed new energy power prediction | |
| Huang et al. | Graph embedding and its application in defect detection system | |
| CN118283663A (en) | Base station fault diagnosis method and device | |
| CN118800044A (en) | Automatic monitoring method for radioactive tailing pond based on multi-source heterogeneous data fusion | |
| CN118627407A (en) | Waterlogging model creation method and system applied to urban water management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |