Simulation scene processing method and device, computing equipment and computer program product

Abstract

The embodiment of the invention provides a simulation scene processing method, and the method comprises the steps: carrying out the anomaly detection in a simulation process of a target simulation scene, and carrying out the similarity comparison of a scene parameter obtained through detection and an anomaly record stored in a knowledge base, so as to obtain a target record; obtaining a reproduction scene based on the target record; and similarity comparison is carried out on the reproduction parameter and a scene parameter corresponding to the target record to call an optimization operation corresponding to the target record, so that the target simulation scene is processed based on the optimization operation. Through the process of scene reproduction and automatic comparison optimization based on the knowledge base when the exception occurs, rapid and accurate exception detection and reproduction capability can be provided on the premise of not increasing a large amount of computing resources, and the exception handling accuracy in the simulation scene is improved.

Description

Technical Field

[0001] The present specification relates to the field of computer application technology, specifically, to the processing technology of simulation scenes under the field of computer application technology, and more specifically, to the processing method, device, computing equipment and computer program product of simulation scenes. Background Art

[0002] Multi-body simulation software is widely used in aerospace, automotive, mechanical engineering and other fields to simulate the interactions and dynamic behaviors of complex multi-objects. However, due to the complexity of simulation models and environments, the reliability and stability of simulation software become key issues during development, testing and optimization. Especially when simulation results deviate from expectations or produce unrealistic situations, how to quickly troubleshoot problems, correct models and retest is a challenge.

[0003] Generally, manual reproduction can be used, that is, R&D personnel manually set parameters similar to the abnormal scenario for abnormal results in the simulation, in order to reproduce the problem and analyze the cause of the abnormality.

[0004] However, although the manual reproduction process helps to locate the problem, it is inefficient and it is difficult to ensure that all reproduction conditions are accurate. Summary of the invention

[0005] The embodiments of this specification provide a method, apparatus, computing device and computer program product for processing a simulation scenario, which improve the accuracy of exception processing in the simulation scenario.

[0006] To achieve the above technical objectives, the embodiments of this specification provide the following technical solutions:

[0007] In a first aspect, an embodiment of the present specification provides a method for processing a simulation scene, including:

[0008] During the simulation of the target simulation scene, key physical quantities in the target simulation scene are detected to obtain detection information;

[0009] If the detection information indicates that the key physical quantity is abnormal, then the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record;

[0010] Adjusting parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene;

[0011] Determine a reproduction parameter through the reproduction scene, and perform a similarity comparison between the reproduction parameter and the scene parameter corresponding to the target record to determine scene comparison information;

[0012] The optimization operation corresponding to the target record is called according to the scenario comparison information, so as to process the target simulation scenario based on the optimization operation.

[0013] Optionally, in some possible implementations, during the simulation of the target simulation scene, detecting key physical quantities in the target simulation scene to obtain detection information includes:

[0014] Obtaining detection rules corresponding to the target simulation scenario;

[0015] Determine threshold configuration information corresponding to each of the key physical quantities according to the detection rule;

[0016] During the simulation of the target simulation scenario, the key physical quantity is compared with the threshold configuration information to obtain the detection information.

[0017] Optionally, in some possible implementations, during the simulation of the target simulation scenario, performing parameter comparison between the key physical quantity and the threshold configuration information to obtain the detection information includes:

[0018] During the simulation of the target simulation scenario based on the parameter information, determining the change information of the key physical quantity within a preset time period;

[0019] determining a reference trend indicated by the threshold configuration information;

[0020] The trend parameter indicated by the change information is compared with the reference trend to obtain the detection information.

[0021] Optionally, in some possible implementations, if the detection information indicates that the key physical quantity is abnormal, a scene parameter corresponding to the target simulation scene is compared with an abnormal record stored in a knowledge base for similarity to obtain a target record, including:

[0022] If the detection information indicates that the key physical quantity is abnormal, determining a data index corresponding to the abnormal record stored in the knowledge base;

[0023] Determining key features in scenario parameters corresponding to the target simulation scenario;

[0024] Calculating the data similarity between the key feature and the data index;

[0025] If the data similarity reaches a similarity threshold, the corresponding abnormal record is determined through the data index to obtain the target record;

[0026] Optionally, in some possible implementations, the calculating the data similarity between the key feature and the data index includes:

[0027] Calculating the parameter similarity between each of the key features and the corresponding component in the data index respectively;

[0028] Obtaining first weight information corresponding to the characteristic parameter;

[0029] The parameter similarity is weightedly calculated based on the first weight information to obtain the data similarity.

[0030] Optionally, in some possible implementations, adjusting parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene includes:

[0031] Obtaining the fault tolerance standard corresponding to the target record;

[0032] Performing parameter iteration on the target simulation scene so that the scene parameters of the target simulation scene are close to the scene parameters corresponding to the target record;

[0033] A parameter iteration process is controlled based on the fault tolerance standard to obtain the recurrence scenario.

[0034] Optionally, in some possible implementations, determining a reproduction parameter by using the reproduction scene, and performing a similarity comparison between the reproduction parameter and a scene parameter corresponding to the target record to determine scene comparison information includes:

[0035] Determine a reproduction parameter through the reproduction scene, and compare the reproduction parameter with the scene parameter corresponding to the target record item by item to obtain parameter comparison information;

[0036] Obtaining the error range for each of the reproduction parameter configurations;

[0037] Screening the parameter comparison information based on the error range to obtain confidence comparison information;

[0038] The various similarities indicated in the confidence comparison information are integrated to obtain the scene comparison information.

[0039] Optionally, in some possible implementations, integrating the various similarities indicated in the confidence comparison information to obtain the scene comparison information includes:

[0040] Obtaining importance information for each of the recurrence parameter configurations;

[0041] Determining second weight information corresponding to the importance information;

[0042] The various similarities indicated in the confidence comparison information are weighted based on the second importance information to obtain the scene comparison information.

[0043] Optionally, in some possible implementations, the method further includes:

[0044] If the scenario comparison information indicates that the target simulation scenario does not match the abnormal record in the knowledge base, the abnormal information corresponding to the target simulation scenario is recorded to obtain a current record;

[0045] A data index is generated based on the current record and stored in the knowledge base.

[0046] In a second aspect, an embodiment of the present specification provides a processing device for a simulation scene, including:

[0047] A detection unit, used to detect key physical quantities in the target simulation scene during the simulation of the target simulation scene to obtain detection information;

[0048] A processing unit, configured to compare the scene parameters corresponding to the target simulation scene with the abnormal records stored in the knowledge base for similarity to obtain a target record if the detection information indicates that the key physical quantity is abnormal;

[0049] The processing unit is further used to adjust the parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene;

[0050] A reproduction unit, configured to determine a reproduction parameter through the reproduction scene, and perform a similarity comparison between the reproduction parameter and a scene parameter corresponding to the target record to determine scene comparison information;

[0051] The processing unit is further used to call the optimization operation corresponding to the target record according to the scenario comparison information, so as to process the target simulation scenario based on the optimization operation.

[0052] Optionally, in some possible implementations, the detection unit is used to detect key physical quantities in the target simulation scene during simulation of the target simulation scene, and to obtain detection rules corresponding to the target simulation scene when obtaining detection information; determine threshold configuration information corresponding to each of the key physical quantities according to the detection rules; and perform parameter comparison between the key physical quantities and the threshold configuration information during simulation of the target simulation scene to obtain the detection information.

[0053] Optionally, in some possible implementations, the detection unit is used to perform parameter comparison between the key physical quantity and the threshold configuration information to obtain the detection information during the simulation of the target simulation scenario. During the simulation of the target simulation scenario based on the parameter information, the detection unit determines the change information of the key physical quantity within a preset time period; determines the reference trend indicated by the threshold configuration information; and compares the trend parameters indicated by the change information with the reference trend to obtain the detection information.

[0054] Optionally, in some possible implementations, the processing unit is used to, when the detection information indicates that the key physical quantity is abnormal, perform a similarity comparison between the scene parameters corresponding to the target simulation scene and the abnormal records stored in the knowledge base to obtain the target record. If the detection information indicates that the key physical quantity is abnormal, determine the data index corresponding to the abnormal record stored in the knowledge base; determine the key features in the scene parameters corresponding to the target simulation scene; calculate the data similarity between the key features and the data index; if the data similarity reaches a similarity threshold, determine the corresponding abnormal record through the data index to obtain the target record;

[0055] Optionally, in some possible implementations, the processing unit is used to calculate the parameter similarity of each key feature and the corresponding component in the data index when calculating the data similarity between the key feature and the data index; obtain first weight information corresponding to the feature parameter; and perform weighted calculation on the parameter similarity based on the first weight information to obtain the data similarity.

[0056] Optionally, in some possible implementations, the processing unit is used to obtain a fault-tolerance standard corresponding to the target record when adjusting parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene; perform parameter iteration on the target simulation scene to make the scene parameters of the target simulation scene close to the scene parameters corresponding to the target record; and control the parameter iteration process based on the fault-tolerance standard to obtain the reproduced scene.

[0057] Optionally, in some possible implementations, the reproduction unit is used to determine the reproduction parameters through the reproduction scene, and perform a similarity comparison between the reproduction parameters and the scene parameters corresponding to the target record item by item to obtain parameter comparison information; obtain an error range for each of the reproduction parameter configurations; filter the parameter comparison information based on the error range to obtain confidence comparison information; and integrate the similarities indicated in the confidence comparison information to obtain the scene comparison information.

[0058] Optionally, in some possible implementations, the processing unit is used to obtain importance information for each of the reproduction parameter configurations when integrating the various similarities indicated in the confidence comparison information to obtain the scene comparison information; determine second weight information corresponding to the importance information; and weight the various similarities indicated in the confidence comparison information based on the second importance information to obtain the scene comparison information.

[0059] Optionally, in some possible implementations, the processing unit is used to record the exception information corresponding to the target simulation scenario to obtain the current record if the scenario comparison information indicates that the target simulation scenario does not match the exception record in the knowledge base; generate a data index based on the current record and store it in the knowledge base.

[0060] In a third aspect, an embodiment of the present specification further provides a computing device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-described method for processing the simulation scenario when executing the computer program.

[0061] In a fourth aspect, an embodiment of the present specification further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the processing method of the simulation scenario as described above is implemented.

[0062] In a fifth aspect, an embodiment of the present specification provides a computer program product or a computer program, wherein the computer program product includes a computer program, and the computer program can be stored in a computer-readable storage medium or in the cloud; the processor of the computer device reads the computer program, and when the processor executes the computer program, the steps of the above-mentioned simulation scenario processing method are implemented.

[0063] It can be seen from the above technical scheme that the processing method of the simulation scene provided by the embodiment of this specification detects the key physical quantity in the target simulation scene during the simulation of the target simulation scene to obtain the detection information; if the detection information indicates that the key physical quantity is abnormal, the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record; then the target simulation scene is adjusted based on the scene parameters corresponding to the target record to obtain the reproduction scene; and the reproduction parameters are determined through the reproduction scene, and the reproduction parameters are compared with the scene parameters corresponding to the target record for similarity to determine the scene comparison information; and then the optimization operation corresponding to the target record is called according to the scene comparison information to process the target simulation scene based on the optimization operation. Thereby, an automated simulation exception processing process is realized, and through the process of scene reproduction and automatic comparison optimization based on the knowledge base when an exception occurs, it is possible to provide fast and accurate exception detection and reproduction capabilities without increasing a large amount of computing resources, thereby improving the accuracy of exception processing in the simulation scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0065] Figure 1 A schematic diagram of an application environment of a method for processing a simulation scenario provided in one embodiment of this specification;

[0066] Figure 2 A flowchart of a method for processing a simulation scene provided for one embodiment of this specification;

[0067] Figure 3 A system structure diagram of a method for processing a simulation scene provided for one embodiment of this specification;

[0068] Figure 4 A scenario flow chart of a method for processing a simulation scenario provided for one embodiment of this specification;

[0069] Figure 5 A schematic diagram of functional modules of a simulation scene processing device provided in one embodiment of this specification;

[0070] Figure 6 A schematic diagram of the structure of a computing device provided for one embodiment of the present specification. DETAILED DESCRIPTION

[0071] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of this specification shall have the common meanings understood by persons with ordinary skills in the field to which this specification belongs. The words "first", "second" and similar words used in the embodiments of this specification do not indicate any order, quantity or importance, but are only used to avoid confusion of constituent elements.

[0072] Unless the context requires otherwise, throughout the specification, "plurality" means "at least two", and "including" is interpreted as an open, inclusive meaning, that is, "including, but not limited to". In the description of the specification, the terms "one embodiment", "some embodiments", "exemplary embodiments", "examples", "specific examples" or "some examples" are intended to indicate that a particular feature, structure, material or characteristic associated with the embodiment or example is included in at least one embodiment or example of the specification. The schematic representation of the above terms does not necessarily refer to the same embodiment or example.

[0073] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0074] Multi-body simulation software is widely used in aerospace, automotive, mechanical engineering and other fields to simulate the interactions and dynamic behaviors of complex multi-objects. However, due to the complexity of simulation models and environments, the reliability and stability of simulation software become key issues during development, testing and optimization. Especially when simulation results deviate from expectations or produce unrealistic situations, how to quickly troubleshoot problems, correct models and retest is a challenge.

[0075] Generally, manual reproduction can be used, that is, R&D personnel manually set parameters similar to the abnormal scenario for abnormal results in the simulation, in order to reproduce the problem and analyze the cause of the abnormality.

[0076] However, although the manual reproduction process helps to locate the problem, it is inefficient and it is difficult to ensure that all reproduction conditions are accurate.

[0077] To solve the above problems, the embodiments of this specification provide a processing system for simulation scenarios, and the processing method for simulation scenarios provided in the embodiments of this specification is applied to the processing system for simulation scenarios. The processing system performs scene reproduction and automatic comparison optimization through a knowledge base, and can provide fast and accurate anomaly detection and reproduction capabilities without increasing a large amount of computing resources, thereby improving the accuracy of anomaly processing in simulation scenarios.

[0078] Specifically, the vehicle information processing system may include: Figure 1 The operating environment formed by the server 110 and the vehicle 120 in the embodiment. The vehicle 120 communicates with the server 110 through a network connection. The functions implemented in the vehicle 120 can be local processing by the vehicle 120, or the vehicle 120 receives information and uploads it to the server 110 in the cloud, and the server 110 in the cloud processes and sends it to the vehicle 120. Among them, the server 110 can be an electronic device with certain computing and processing capabilities. It can have a network communication module, a processor, and a memory, etc. Of course, the server 110 can also refer to the software running in the electronic device. The server 110 can also be a distributed server, which can be a system with multiple processors, memories, network communication modules, etc. operating in collaboration. Alternatively, the server 110 can also be a cluster formed by several servers 110. Alternatively, with the development of science and technology, the server 110 can also be a new technical means that can realize the corresponding functions of the implementation method of the specification. For example, it can be a new form of "server" based on quantum computing.

[0079] Specifically, when the above-mentioned processing system is performing simulation, the key physical quantity in the target simulation scene is detected during the simulation of the target simulation scene to obtain detection information; if the detection information indicates that the key physical quantity is abnormal, the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record; the parameters of the target simulation scene are adjusted based on the scene parameters corresponding to the target record to obtain the reproduction scene; the reproduction parameters are determined through the reproduction scene, and the reproduction parameters are compared with the scene parameters corresponding to the target record for similarity to determine the scene comparison information; the optimization operation corresponding to the target record is called according to the scene comparison information to process the target simulation scene based on the optimization operation. Thereby, an automated simulation exception handling process is realized, and through the process of scene reproduction and automatic comparison optimization based on the knowledge base when an exception occurs, it is possible to provide fast and accurate exception detection and reproduction capabilities without increasing a large amount of computing resources, thereby improving the accuracy of exception handling in the simulation scene.

[0080] Based on the above concept, an embodiment of the present specification provides a method for processing a simulation scene. The following will exemplarily describe the method for processing a simulation scene provided in the embodiment of the present specification in conjunction with the accompanying drawings.

[0081] To be applied to Figure 1 Taking the server in the example, some implementation methods of this specification exemplarily illustrate the processing method of the simulation scenario, such as Figure 2 As shown, Figure 2 A flowchart of a method for processing a simulation scene provided for one embodiment of this specification; the method for processing a simulation scene includes:

[0082] 201. During the simulation of the target simulation scene, key physical quantities in the target simulation scene are detected to obtain detection information.

[0083] In this embodiment, the target simulation scenario is a multi-body simulation process. Multi-body simulation is a technology used to analyze and simulate the dynamic behavior of a complex mechanical system composed of multiple interacting objects. This technology has a wide range of applications in engineering design, scientific research, and product development, especially in fields that require accurate prediction of system dynamic characteristics, such as vehicles, aerospace, and robotics. This embodiment is described using a vehicle as an example.

[0084] It can be understood that since the target simulation scenario is obtained based on multiple scenario parameter configurations, the key physical quantities therein can be detected when performing anomaly detection. The key physical quantities are the parameters in the target simulation scenario that are directly related to the simulation results. For example, in a vehicle scenario, the key physical quantities are position, velocity, acceleration, torque, etc.

[0085] Specifically, the abnormality detection process of key physical quantities can be carried out through threshold comparison, that is, first obtaining the detection rules corresponding to the target simulation scenario; then determining the threshold configuration information corresponding to each key physical quantity according to the detection rules; and in the process of simulating the target simulation scenario, performing parameter comparison between the key physical quantities and the threshold configuration information to obtain detection information.

[0086] Optionally, for the configuration of detection rules, that is, the process of rule benchmark detection, predefined detection rules can be loaded when the simulation starts, for example: a. Position deviation: If the distance of the object position deviating from the expected position exceeds the set threshold, it is judged to be abnormal. The threshold can be adjusted according to the actual application needs; b. Speed ​​and acceleration abnormalities: The speed or acceleration exceeds the reasonable range, or suddenly changes sharply. c. Force and torque range detection: The force or torque value exceeds the physical range (such as too high collision torque).

[0087] Furthermore, for the threshold detection process, different threshold standards can be set for each key parameter for different simulation model parameters to improve the flexibility of detection. For example, the detection range of torque or acceleration can be dynamically adjusted based on the quality and material properties of the model.

[0088] Considering that the parameters in the simulation scenario are changing, the process of anomaly detection can also be a trend detection process, that is, first, in the process of simulating the target simulation scenario based on the parameter information, determine the change information of the key physical quantity within the preset time period; then determine the reference trend indicated by the threshold configuration information; and compare the trend parameters indicated by the change information with the reference trend to obtain detection information. That is, use the algorithm to analyze the parameter trend over a period of time to identify whether the continuous upward or downward trend conforms to the normal physical law, such as the abnormal situation of sudden stop or acceleration to the limit value.

[0089] Through the above anomaly detection process, when an anomaly is detected, the module automatically queries the knowledge base to determine whether there is a historical record similar to the current anomaly. For example, in a multi-body simulation, if the speed of a component is detected to increase to 50 meters per second, exceeding the upper limit of the reasonable speed of the model (such as 30 meters per second), it is determined to be abnormal and trigger a knowledge base query. Among them, the knowledge base query mainly determines whether the current anomaly is similar to the historical record through similarity calculation, thereby determining the subsequent steps.

[0090] Specifically, the configuration of the knowledge base is the basis for the automatic anomaly detection and processing of this embodiment, and is configured in the following system architecture, such as Figure 3 As shown, Figure 3A system structure diagram of a method for processing a simulation scenario provided for an embodiment of the present specification; the figure shows a knowledge base module, anomaly detection module, scenario reproduction module, data comparison and analysis module, and feedback and optimization module. The knowledge base module is used to store various simulation scenarios, parameters, test results, and abnormal modes. The knowledge base classifies and indexes the scenario data to make reproduction and comparison more efficient. For the anomaly detection module, the module is responsible for real-time monitoring of data deviations in the simulation process, mainly real-time monitoring of the output of key physical quantities during the simulation operation, and judging whether anomalies occur through a series of preset thresholds and rules. For the scenario reproduction module, the module calls similar scenario data and parameters in the knowledge base to rebuild the reproduction scenario of the current abnormal situation. The reproduction module can adjust subtle parameters in multiple iterations to accurately match the original abnormal environment. For the data comparison and analysis module, it is used to compare the current simulation results with the historical abnormal records stored in the knowledge base, and calculate the data deviation and similarity. By comparing multiple feature parameters, the nature and type of the anomaly are judged according to the similarity and deviation. When the similarity exceeds the preset threshold, the system determines it as a repeated anomaly and automatically prompts the user. For the feedback and optimization module, the system automatically feeds back the detected abnormal patterns and recurrence data to the knowledge base, gradually optimizing the accuracy of scene recurrence and anomaly detection. The core is to record the recurrence of each anomaly and the corresponding parameter fine-tuning, so as to reproduce or warn more quickly and accurately when similar anomalies occur in the future.

[0091] It can be seen that this embodiment stores abnormal patterns based on the knowledge base, establishes a knowledge base system, and records various scenario data and abnormal patterns in the multi-body simulation process. During the test process, the system will automatically compare the current abnormal scenario with the known abnormalities in the knowledge base and quickly detect similarities. This method avoids the process of manually identifying wrong patterns and has a high degree of automation. Thereby improving the accuracy of abnormality detection, by comparing and reproducing scenarios through the knowledge base, it is possible to quickly identify possible abnormal patterns when deviations occur, ensuring the accuracy of abnormality detection.

[0092] Next, in conjunction with the above system architecture, the process of performing anomaly detection and automatic comparison processing based on the knowledge base is explained.

[0093] 202. If the detection information indicates that the key physical quantity is abnormal, the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record.

[0094] In this embodiment, when the key physical quantity is abnormal, it means that the target simulation scene is abnormal. At this time, record matching is performed through the knowledge base to quickly obtain a scene similar to the abnormal event of the target simulation scene, that is, to obtain the target record.

[0095] Specifically, the composition of the knowledge base includes an abnormal event database, which is used to store each detected abnormal event and its parameters (such as speed, position, torque, etc.) and the specific environment in which the abnormality occurs (such as simulation conditions, model parameters, etc.). It also includes a similarity index library, which constructs a feature vector index for each abnormal event to facilitate quick search for similar anomalies. The index library can be generated by hash values ​​or clustering algorithms of feature parameters (such as speed, acceleration, torque, etc.). It also includes an optimized parameter library, which records the optimal parameter adjustment information for each abnormal event so that it can be directly referenced the next time a similar anomaly recurs. The parameter library not only records parameter values, but also includes adjustment strategies and processes (such as gradually fine-tuned parameter change values).

[0096] In one possible scenario, since the simulation process of the target simulation scenario is continuous, for the exception handling process, after a new exception event is detected, its complete parameters and related information can be first stored in the exception event database for subsequent reference. Then after the exception is detected, the knowledge base module compares the similarity between the feature vector of the current exception and all historical records in the knowledge base, usually using algorithms such as Euclidean distance and cosine similarity for calculation. If the similarity is higher than the set threshold (for example, 90%), it is considered that a similar historical record has been found. Moreover, during the anomaly reproduction and analysis process, if the system finds a new set of optimization parameters (i.e., more accurate reproduction parameters), the optimization parameter library in the knowledge base will be automatically updated.

[0097] It is understandable that the data (abnormal records) in the knowledge base are dynamically updated, that is, the update process of the knowledge base module is carried out through the following steps. After a new abnormality is detected, it is saved to the abnormal event database. And in the process of reproducing the same type of abnormality, if a new reproduction scheme or parameter configuration is obtained, the old parameter record is replaced. Furthermore, the old data is cleaned or archived regularly to avoid the query efficiency being reduced due to the excessive capacity of the knowledge base.

[0098] Through the configuration process of the above knowledge base, the knowledge base has the ability to match multiple abnormal records. Therefore, for the similarity comparison process of determining the target record, when the detection information indicates that the key physical quantity is abnormal, the data index corresponding to the abnormal record stored in the knowledge base can be determined; then the key features in the scene parameters corresponding to the target simulation scene are determined; and the data similarity between the key features and the data index is calculated; if the data similarity reaches the similarity threshold, the corresponding abnormal record is determined through the data index to obtain the target record.

[0099] It can be seen that the query process of the knowledge base is carried out through data indexing, that is, the system establishes a data index for each abnormal record in the knowledge base, and can use feature parameters (position, speed, torque) and abnormal conditions to facilitate rapid retrieval. Among them, for the similarity calculation process, the Euclidean distance, cosine similarity or other similarity algorithms are used to calculate the similarity between the current abnormality and each historical record, so as to compare the similarity of key features.

[0100] It can be understood that the key features are the parameters and conditions in the target simulation scene that can directly affect the simulation results. The similarity judgment standard for the key features can be set by threshold setting, that is, a similarity threshold (such as 90%) is set. If the similarity exceeds this value, it is judged as a similar anomaly and enters the scene reproduction module.

[0101] Furthermore, considering that different parameters can have different weights (such as torque and speed are more important than position error) to meet different anomaly detection requirements, weighted feature matching can be performed for the calculation process of weighted feature matching, that is, firstly, the parameter similarity between each key feature and the corresponding component in the data index is calculated respectively; then the first weight information corresponding to the feature parameter is obtained; and then the parameter similarity is weightedly calculated based on the first weight information to obtain the data similarity, thereby improving the accuracy of similarity calculation in different simulation scenarios.

[0102] 203. Adjust parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene.

[0103] In this embodiment, after determining a target record similar to the target simulation scene when an abnormality occurs, the target simulation scene can be controlled through the target record to reproduce the scene.

[0104] Specifically, considering the instability caused by a large number of parameters in the simulation process, a certain fault tolerance standard can be set, that is, error control can be performed. Therefore, for the construction process of the reproduction scene, the fault tolerance standard corresponding to the target record can be obtained; then the parameters of the target simulation scene are iterated to make the scene parameters of the target simulation scene close to the scene parameters corresponding to the target record; and then the parameter iteration process is controlled based on the fault tolerance standard to obtain the reproduction scene. Among them, for the error control process, the allowable error range can be set, such as position error ±5%, speed error ±3%, etc., to determine whether to reproduce. In addition, during the reproduction process, the similarity between the current parameters and the historical parameters is kept within a certain range (such as not less than 90%), and the error is controlled within the allowable range.

[0105] Furthermore, for the parameter iteration process, i.e., for adjusting subtle parameters, an iterative algorithm (such as binary search or gradient descent) can be used to fine-tune the parameters so that the reproduced scene gradually approaches the historical record. After each adjustment, the similarity of the simulation output is compared in real time until the set standard is met. In addition, real-time feedback adjustment can also be used, such as control parameters such as speed and torque, and the input parameters can be reversely adjusted according to the deviation after each simulation output.

[0106] In one possible scenario, if the historical record shows an abnormal speed of 40 m / s for an object, and the current simulated speed is 45 m / s, the system can gradually reduce the input torque until the speed approaches 40 m / s.

[0107] It can be seen that this embodiment has automatically reproduced the scene, and can automatically reconstruct the abnormal scene according to the knowledge base data, and automatically call the corresponding reproduction parameters when the abnormality is found, accurately restoring the abnormal situation. At the same time, the system can also adjust the scene parameters in real time to ensure the accuracy of the reproduction of the abnormal scene, reducing the time for manual adjustment. And significantly improve the efficiency of retesting, that is, realize automatic reproduction and real-time comparison, the user does not need to perform multiple manual operations, and greatly shortens the test time.

[0108] 204. Determine a reproduction parameter by reproducing the scene, and perform a similarity comparison between the reproduction parameter and the scene parameter corresponding to the target record to determine scene comparison information.

[0109] In this embodiment, after reproducing the scene, it is determined whether the anomaly of the current scene is similar to the anomaly of the target record, that is, the dimension of abnormal parameter detection is advanced to the dimension of abnormal scene, and the fault is compared from the perspective of the entire scene to ensure the reliability of the similarity comparison.

[0110] Specifically, the scene comparison information process can be Figure 3 The data comparison and analysis module shown in the figure is executed, and the module includes a feature parameter comparison unit, a similarity calculation module, and an abnormality type discrimination unit. Among them, the feature parameter comparison unit is mainly used to compare the parameters in the current abnormal event and the recurrence history record one by one, such as position, speed, acceleration, torque, etc. Each parameter will have a corresponding threshold to determine whether it meets the similarity requirements. The similarity calculation module is used to input the compared feature parameter values ​​into the similarity calculation module and comprehensively calculate the similarity value.

[0111] Specifically, similarity calculation can adopt a variety of algorithms, such as Euclidean distance: calculating the geometric distance of abnormal event feature vectors; cosine similarity: used to measure the angular similarity between feature vectors; Mahalanobis distance: measuring the similarity of abnormal events while considering data covariance.

[0112] In addition, the abnormality type determination unit is used to determine whether the current abnormality belongs to a known abnormality type or a completely new abnormality pattern based on the similarity calculation result.

[0113] Through the configuration of the above-mentioned module units, for the process of determining the scene comparison information, parameter comparison can be performed item by item, that is, first determine the reproduction parameters through the reproduction scene, and compare the reproduction parameters with the scene parameters corresponding to the target record item by item to obtain parameter comparison information; then obtain the error range for each reproduction parameter configuration; and filter the parameter comparison information based on the error range to obtain confidence comparison information; and then integrate the various similarities indicated in the confidence comparison information to obtain the scene comparison information. Through the judgment of multi-level conditions, each parameter is compared step by step. For example, the comparison of position can tolerate an error of ±5%, while the comparison of speed may only have an error of ±2%.

[0114] In addition, since the importance of different parameters in the simulation scene is different, similarity calculation with dynamic weight adjustment can be performed, that is, first obtain the importance information for each recurrence parameter configuration; then determine the second weight information corresponding to the importance information; and then weight the various similarities indicated in the confidence comparison information based on the second importance information to obtain the scene comparison information. That is, when calculating the comprehensive similarity, the weights of different parameters can be adjusted dynamically. For example, in some scenes, the importance of torque may be higher than acceleration, so in the similarity calculation, torque will be given a higher weight.

[0115] In one possible scenario, for the above feature comparison process, the traditional knowledge base comparison method can also be replaced by a deep learning abnormal pattern recognition algorithm. In the alternative, the system can use a neural network training model to convert historical abnormal data into feature vectors for real-time judgment of the abnormal probability of current simulation data. However, although this alternative can improve detection accuracy, it consumes higher computing resources and is difficult to apply to small-scale data.

[0116] Therefore, for the above data comparison and analysis to determine the process of scene comparison information, first receive the abnormal data transmitted by the anomaly detection module and the historical data returned by the knowledge base module. Then compare the current abnormal data with the historical data one by one to check whether each parameter is within the allowable error range. Perform similarity calculation and discrimination, then calculate the similarity of each feature parameter, and perform comprehensive calculation. If the comprehensive similarity value is higher than the threshold (for example, 90%), the current anomaly is considered to be a known anomaly and enters the optimization module. If it is lower than the threshold, it is marked as a new abnormal pattern and stored in the knowledge base.

[0117] It can be seen that in the scenario where the scenario comparison information indicates that the target simulation scenario does not match the abnormal record in the knowledge base, the abnormal information corresponding to the target simulation scenario is recorded to obtain the current record; and a data index is generated based on the current record and stored in the knowledge base. This realizes the automatic update of the knowledge base, that is, the knowledge base has intelligent learning capabilities, and the knowledge base can gradually accumulate abnormal patterns and optimize test scenarios, and has self-learning capabilities, thereby continuously improving the test effect.

[0118] It can be seen from the above embodiments that during the exception handling process, the system will not only reproduce the abnormal scenario, but also compare the current simulation with the historical abnormal results in real time, and determine whether a repetitive abnormality has occurred and issue a real-time warning through data similarity matching and error calculation. This method effectively improves the efficiency and accuracy of multiple retests. And by reducing human intervention, the knowledge base and automatic reproduction mechanism reduce the possibility of human errors and improve the stability and consistency of simulation tests.

[0119] 205. Call the optimization operation corresponding to the target record according to the scene comparison information, so as to process the target simulation scene based on the optimization operation.

[0120] In this embodiment, for the execution process of the optimization operation, the parameter tuning record can be called, that is, the tuning parameter range and adjustment sequence (such as torque adjustment range, speed fine-tuning ratio, etc.) are recorded in the knowledge base for each abnormal scenario. Then, the abnormal pattern is updated, that is, the existing abnormal pattern library is expanded according to the new abnormal data to improve the accuracy of judgment on similar abnormalities.

[0121] Specifically, for the fine-tuning process in the optimization operation, the optimal parameter set can be updated. If a better parameter setting is found during the reproduction process to improve the reproduction accuracy, the parameter set in the knowledge base is updated so that it can be directly called during subsequent reproduction. In addition, the average value of multiple reproduction records can be calculated. If an anomaly occurs multiple times, the system can record the average parameters of multiple reproductions and use this as a standard to further optimize the accuracy of reproduction.

[0122] Therefore, combined with the above embodiments, it can be obtained as follows Figure 4 The scenario flow chart shown, Figure 4A scenario flow chart of a method for processing a simulation scenario provided as an embodiment of the present specification; first, data input and simulation are performed, the user inputs the simulation model and initial parameters, and the system starts the simulation operation. Then, anomaly detection and knowledge base query are performed. During the simulation process, the anomaly detection module monitors the simulation results in real time. When abnormal data is detected, similar scenarios in the knowledge base are automatically queried. Then, scene reproduction and automatic comparison are performed, that is, the scene reproduction module automatically reproduces similar abnormal scenarios to ensure that the reproduction is consistent with the original anomaly. The comparison module compares the current simulation and the reproduction scenario in real time and calculates the similarity. Feedback and knowledge base update are further performed, that is, the system feeds back the current anomaly and analysis results to the knowledge base to optimize the parameter setting for the next reproduction.

[0123] In summary, this embodiment detects the key physical quantities in the target simulation scene during the simulation of the target simulation scene to obtain detection information; if the detection information indicates that the key physical quantity is abnormal, the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record; then the target simulation scene is adjusted based on the scene parameters corresponding to the target record to obtain the reproduction scene; and the reproduction parameters are determined through the reproduction scene, and the reproduction parameters are compared with the scene parameters corresponding to the target record for similarity to determine the scene comparison information; and then the optimization operation corresponding to the target record is called according to the scene comparison information to process the target simulation scene based on the optimization operation. Thereby, an automated simulation exception handling process is realized, and through the process of scene reproduction and automatic comparison optimization based on the knowledge base when an exception occurs, it is possible to provide fast and accurate exception detection and reproduction capabilities without increasing a large amount of computing resources, thereby improving the accuracy of exception handling in the simulation scene.

[0124] It should be noted that the multiple embodiments in this specification emphasize the parts that are different from other embodiments, and the various embodiments can be explained in comparison with each other. Any combination of the multiple embodiments in this specification by technicians in the relevant field based on general technical knowledge is included in the disclosure scope of this specification.

[0125] In an exemplary embodiment of the present specification, a simulation scene processing device 500 is also provided. Figure 5 As shown, Figure 5 A functional module diagram of a processing device for a simulation scene provided in one embodiment of this specification, wherein the interaction device 500 includes:

[0126] The detection unit 501 is used to detect key physical quantities in the target simulation scene during the simulation of the target simulation scene to obtain detection information;

[0127] The processing unit 502 is configured to perform a similarity comparison between the scene parameters corresponding to the target simulation scene and the abnormal records stored in the knowledge base to obtain the target record if the detection information indicates that the key physical quantity is abnormal;

[0128] The processing unit 502 is further configured to adjust parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene;

[0129] A reproduction unit 503, configured to determine a reproduction parameter through the reproduction scene, and perform a similarity comparison between the reproduction parameter and a scene parameter corresponding to the target record to determine scene comparison information;

[0130] The processing unit 502 is further configured to call the optimization operation corresponding to the target record according to the scenario comparison information, so as to process the target simulation scenario based on the optimization operation.

[0131] Optionally, in some possible implementations, the detection unit 501 is used to detect key physical quantities in the target simulation scene during the simulation of the target simulation scene, and to obtain detection rules corresponding to the target simulation scene when obtaining detection information; determine the threshold configuration information corresponding to each of the key physical quantities according to the detection rules; and perform parameter comparison between the key physical quantities and the threshold configuration information during the simulation of the target simulation scene to obtain the detection information.

[0132] Optionally, in some possible implementations, the detection unit 501 is used to perform parameter comparison between the key physical quantity and the threshold configuration information to obtain the detection information during the simulation of the target simulation scenario. During the simulation of the target simulation scenario based on the parameter information, the detection unit 501 determines the change information of the key physical quantity within a preset time period; determines the reference trend indicated by the threshold configuration information; and compares the trend parameters indicated by the change information with the reference trend to obtain the detection information.

[0133] Optionally, in some possible implementations, the processing unit 502 is used to perform a similarity comparison between the scene parameters corresponding to the target simulation scene and the abnormal records stored in the knowledge base to obtain the target record when the detection information indicates that the key physical quantity is abnormal. If the detection information indicates that the key physical quantity is abnormal, then determine the data index corresponding to the abnormal record stored in the knowledge base; determine the key features in the scene parameters corresponding to the target simulation scene; calculate the data similarity between the key features and the data index; if the data similarity reaches a similarity threshold, determine the corresponding abnormal record through the data index to obtain the target record.

[0134] Optionally, in some possible implementations, the processing unit 502 is used to calculate the parameter similarity between each key feature and the corresponding component in the data index when calculating the data similarity between the key feature and the data index; obtain first weight information corresponding to the feature parameter; and perform weighted calculation on the parameter similarity based on the first weight information to obtain the data similarity.

[0135] Optionally, in some possible implementations, the processing unit 502 is used to obtain a fault tolerance standard corresponding to the target record when adjusting parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene; perform parameter iteration on the target simulation scene to make the scene parameters of the target simulation scene close to the scene parameters corresponding to the target record; and control the parameter iteration process based on the fault tolerance standard to obtain the reproduced scene.

[0136] Optionally, in some possible implementations, the reproduction unit 503 is used to determine the reproduction parameters through the reproduction scene, and perform a similarity comparison between the reproduction parameters and the scene parameters corresponding to the target record item by item to obtain parameter comparison information; obtain an error range for each of the reproduction parameter configurations; filter the parameter comparison information based on the error range to obtain confidence comparison information; and integrate the similarities indicated in the confidence comparison information to obtain the scene comparison information.

[0137] Optionally, in some possible implementations, the processing unit 502 is used to obtain importance information for each of the reproduction parameter configurations when integrating the various similarities indicated in the confidence comparison information to obtain the scene comparison information; determine second weight information corresponding to the importance information; and weight the various similarities indicated in the confidence comparison information based on the second importance information to obtain the scene comparison information.

[0138] Optionally, in some possible implementations, the detection unit is used to record the exception information corresponding to the target simulation scenario to obtain the current record if the scenario comparison information indicates that the target simulation scenario does not match the exception record in the knowledge base; generate a data index based on the current record and store it in the knowledge base.

[0139] Specifically, the detection unit, processing unit, and reproduction unit in this embodiment may correspond to physical components. For example, the processing unit may be a processing module such as a CPU, a GPU, or an FPGA. The specific physical components may be any components or combinations of components having the above functions. The specific method depends on the actual scenario and is not limited here.

[0140] The above-mentioned processing device detects the key physical quantity in the target simulation scene during the simulation of the target simulation scene to obtain detection information; if the detection information indicates that the key physical quantity is abnormal, the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record; then the target simulation scene is adjusted based on the scene parameters corresponding to the target record to obtain the reproduction scene; and the reproduction parameters are determined through the reproduction scene, and the reproduction parameters are compared with the scene parameters corresponding to the target record for similarity to determine the scene comparison information; and then the optimization operation corresponding to the target record is called according to the scene comparison information to process the target simulation scene based on the optimization operation. Thereby, an automated simulation exception handling process is realized, and through the process of scene reproduction and automatic comparison optimization based on the knowledge base when an exception occurs, it can provide fast and accurate exception detection and reproduction capabilities without increasing a large amount of computing resources, thereby improving the accuracy of exception handling in the simulation scene.

[0141] For the specific definition of the processing device of the simulation scene, please refer to the definition of the processing method of the simulation scene above, which will not be repeated here. The various unit modules in the processing device of the above simulation scene can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0142] Another embodiment of the present application further provides a computing device, see Figure 6 As shown, an exemplary embodiment of the present specification also provides a computing device, including: a memory and a processor, the memory storing a computer program, and the processor executing the computer program when executing the computer program performs the steps of the processing method of the simulation scene according to various embodiments of the present specification described in the above embodiments of the present specification.

[0143] The internal structure of the computing device can be as follows Figure 6As shown, the computing device includes a processor, a memory, a network interface and an input device connected through a system bus. Among them, the processor of the computing device is used to provide computing and control capabilities. The memory of the computing device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computing device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in the processing method of the simulation scenario according to various embodiments of the present specification described in the above embodiments of the present specification are performed.

[0144] The processor may include a main processor and may also include a baseband chip, a modem, etc.

[0145] The memory stores a program for executing the technical solution of the present invention, and may also store an operating system and other key services. Specifically, the program may include a program code, and the program code includes computer operation instructions. More specifically, the memory may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk storage, a flash, and the like.

[0146] The processor may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the scheme of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.

[0147] The input device may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0148] Output devices may include means that allow information to be output to a user, such as display screens, printers, speakers, etc.

[0149] The communication interface may include using any transceiver or the like to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0150] The processor executes the program stored in the memory and calls other devices, which can be used to implement the various steps of the processing method of any simulation scenario provided in the above embodiments of the present application.

[0151] The computing device may also include a display component and a voice component. The display component may be a liquid crystal display or an electronic ink display. The input device of the computing device may be a touch layer covered on the display component, or a button, trackball or touchpad provided on the housing of the computing device, or an external keyboard, touchpad or mouse.

[0152] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of this specification, and does not constitute a limitation on the computing device to which the scheme of this specification is applied. The specific computing device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0153] In addition to the above-mentioned methods and devices, the method for processing simulation scenarios provided in the embodiments of this specification may also be a computer program product, which includes a computer program, which, when executed by a processor, enables the processor to execute the steps of the method for processing simulation scenarios according to various embodiments of this specification described in the above-mentioned "Exemplary Method" section of this specification.

[0154] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the embodiments of the present specification, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0155] In addition, an embodiment of the present specification also provides a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to execute the steps of the processing method of the simulation scenario according to various embodiments of the present specification described in the above "Exemplary Method" section of the present specification.

[0156] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0157] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] The above-mentioned embodiments only express several implementation methods of this specification, and the descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the solutions provided by the embodiments of this specification. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of this specification, which all belong to the protection scope of this specification. Therefore, the protection scope of the patent of this specification shall be based on the attached claims.

Claims

1. A method for processing a simulation scene, characterized in that: include: During the simulation of the target simulation scene, key physical quantities in the target simulation scene are detected to obtain detection information; If the detection information indicates that the key physical quantity is abnormal, then the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record; Adjusting parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene; Determine a reproduction parameter through the reproduction scene, and perform a similarity comparison between the reproduction parameter and the scene parameter corresponding to the target record to determine scene comparison information; The optimization operation corresponding to the target record is called according to the scenario comparison information, so as to process the target simulation scenario based on the optimization operation.

2. The method according to claim 1, characterized in that During the simulation of the target simulation scene, detecting key physical quantities in the target simulation scene to obtain detection information includes: Obtaining detection rules corresponding to the target simulation scenario; Determine threshold configuration information corresponding to each of the key physical quantities according to the detection rule; During the simulation of the target simulation scenario, the key physical quantity is compared with the threshold configuration information to obtain the detection information.

3. The method according to claim 2, characterized in that In the process of simulating the target simulation scenario, performing parameter comparison between the key physical quantity and the threshold configuration information to obtain the detection information includes: During the simulation of the target simulation scenario based on the parameter information, determining the change information of the key physical quantity within a preset time period; Determining a reference trend indicated by the threshold configuration information; The trend parameter indicated by the change information is compared with the reference trend to obtain the detection information.

4. The method according to claim 1, characterized in that: If the detection information indicates that the key physical quantity is abnormal, then the scene parameters corresponding to the target simulation scene are compared with the abnormal records stored in the knowledge base for similarity to obtain the target record, including: If the detection information indicates that the key physical quantity is abnormal, determining a data index corresponding to the abnormal record stored in the knowledge base; Determining key features in scenario parameters corresponding to the target simulation scenario; Calculating the data similarity between the key feature and the data index; If the data similarity reaches a similarity threshold, the corresponding abnormal record is determined through the data index to obtain the target record.

5. The method according to claim 4, characterized in that The calculating the data similarity between the key feature and the data index includes: Calculating the parameter similarity between each of the key features and the corresponding component in the data index respectively; Obtaining first weight information corresponding to the characteristic parameter; The parameter similarity is weightedly calculated based on the first weight information to obtain the data similarity.

6. The method according to claim 1, characterized in that The step of adjusting the parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene includes: Obtaining the fault tolerance standard corresponding to the target record; Performing parameter iteration on the target simulation scene so that the scene parameters of the target simulation scene are close to the scene parameters corresponding to the target record; A parameter iteration process is controlled based on the fault tolerance standard to obtain the recurrence scenario.

7. The method according to claim 1, characterized in that The determining of the reproduction parameters by the reproduction scene, and performing a similarity comparison between the reproduction parameters and the scene parameters corresponding to the target record to determine the scene comparison information, includes: Determine a reproduction parameter through the reproduction scene, and compare the reproduction parameter with the scene parameter corresponding to the target record item by item to obtain parameter comparison information; Obtaining the error range for each of the reproduction parameter configurations; Screening the parameter comparison information based on the error range to obtain confidence comparison information; The various similarities indicated in the confidence comparison information are integrated to obtain the scene comparison information.

8. The method according to claim 7, characterized in that The integrating the various similarities indicated in the confidence comparison information to obtain the scene comparison information includes: Obtaining importance information for each of the recurrence parameter configurations; Determining second weight information corresponding to the importance information; The various similarities indicated in the confidence comparison information are weighted based on the second importance information to obtain the scene comparison information.

9. The method according to any one of claims 1 to 8, characterized in that: The method further comprises: If the scenario comparison information indicates that the target simulation scenario does not match the abnormal record in the knowledge base, the abnormal information corresponding to the target simulation scenario is recorded to obtain a current record; A data index is generated based on the current record and stored in the knowledge base.

10. A simulation scene processing device, characterized in that: include: A detection unit, used to detect key physical quantities in the target simulation scene during the simulation of the target simulation scene to obtain detection information; A processing unit, configured to compare the scene parameters corresponding to the target simulation scene with the abnormal records stored in the knowledge base for similarity to obtain a target record if the detection information indicates that the key physical quantity is abnormal; The processing unit is further used to adjust the parameters of the target simulation scene based on the scene parameters corresponding to the target record to obtain a reproduced scene; A reproduction unit, configured to determine a reproduction parameter through the reproduction scene, and perform a similarity comparison between the reproduction parameter and a scene parameter corresponding to the target record to determine scene comparison information; The processing unit is further used to call the optimization operation corresponding to the target record according to the scenario comparison information, so as to process the target simulation scenario based on the optimization operation.

11. A computing device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for processing a simulation scene according to any one of claims 1 to 9 when executing the computer program.

12. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the method for processing a simulation scenario according to any one of claims 1 to 9 is implemented.

Images