Intelligent simulation engine management method, device, equipment and storage medium

Abstract

The present invention relates to the field of intelligent simulation technology, and discloses an intelligent simulation engine management method, device, equipment and storage medium. When a simulation task is received, the received simulation task is analyzed, the priority of the simulation task is determined, and the resource demand of the simulation task is evaluated; a genetic algorithm is used to schedule the thread of the intelligent simulation engine, and the simulation task is assigned to a corresponding computing node for processing; the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technology, and the resource usage of each computing node is monitored in real time. When the usage is abnormal, the execution strategy of the simulation task is adjusted; when the simulation task is completed, the simulation result is analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized through the key performance indicators; the present invention can share resources among multiple simulation domains through the support of multiple driving modes, and significantly improve the simulation efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the field of intelligent simulation technology, and in particular to an intelligent simulation engine management method, device, equipment and storage medium. Background Art

[0002] With the development of simulation technology, the scale and demand of complex simulation tasks are growing, especially in the fields of military, aerospace, industrial design and urban planning. Traditional simulation engines usually rely on a single drive mode and have difficulty handling large-scale, high-complexity multi-threaded simulation tasks. In addition, the existing simulation engines have poor scene adaptability, cannot support the parallel execution of multiple drive modes, and cannot efficiently support multiple simulation domains sharing one engine, resulting in a waste of computing resources and low simulation efficiency. Summary of the invention

[0003] The purpose of the present invention is to solve the above problems and to design an intelligent simulation engine management method, device, equipment and storage medium.

[0004] A first aspect of the present invention provides an intelligent simulation engine management method, the intelligent simulation engine management method comprising:

[0005] When receiving a simulation task, analyzing the received simulation task, determining the priority of the simulation task, and evaluating the resource requirements of the simulation task;

[0006] According to the priority and resource requirements of the simulation task, the genetic algorithm is used to schedule the threads of the intelligent simulation engine, and the simulation task is assigned to the corresponding computing node for processing, where the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven;

[0007] During the simulation process, the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technologies to monitor the resource usage of each computing node in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted;

[0008] When the simulation task is completed, the simulation results are analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized based on the key performance indicators.

[0009] Optionally, in a first implementation of the first aspect of the present invention, when a simulation task is received, analyzing the received simulation task, determining the priority of the simulation task, and evaluating the resource requirements of the simulation task include:

[0010] When receiving a simulation task, analyze the received simulation task to determine the field, application scenario, type and task time requirement of the simulation task, and evaluate the complexity of the simulation task and the task quality requirements;

[0011] Determine the priority of the simulation task according to a pre-set rule, wherein the pre-set rule is priority = type weight × task type + complexity weight × task complexity + time weight × task time requirement + quality weight × task quality requirement;

[0012] Determine the required number of CPU cores, memory capacity requirements, and network bandwidth requirements based on the complexity of the simulation task, and adjust the configuration of the intelligent simulation engine based on the number of CPU cores, memory capacity requirements, and network bandwidth.

[0013] Optionally, in a second implementation of the first aspect of the present invention, the method of using a genetic algorithm to schedule threads of the intelligent simulation engine according to the priority and resource requirements of the simulation task and allocating the simulation task to a corresponding computing node for processing includes:

[0014] Sort the simulation tasks according to their priorities, and for simulation tasks with the same priority, allocate them according to their resource requirements;

[0015] Initialize the population, and the individuals in the population are used to represent different simulation task allocation schemes. The fitness function is:

[0016] Fitness=w p ×Priority+w r ×ResourceDemand / NodeCapacity

[0017] In the formula, w p and w r They are the weight coefficients of priority and resource demand, respectively. Priority is the priority of the task, ResourceDemand is the resource demand of the task, and NodeCapacity is the processing capacity of the computing node;

[0018] The roulette wheel selection method is used to select individuals with higher fitness from the current population as parents, and the selected parent individuals are crossovered. A computing node is randomly selected as the crossover point in the task allocation scheme, and the task allocations of the two parent individuals on the node are exchanged to generate new offspring individuals.

[0019] Perform mutation operations on the newly generated offspring individuals to move a task from one computing node to another;

[0020] When the genetic algorithm reaches the set maximum number of iterations, the algorithm stops and the individual with the highest fitness value is selected from the last generation of the population as the optimal simulation task allocation scheme. Based on the optimal simulation task allocation scheme, the simulation tasks are allocated to the corresponding computing nodes for processing.

[0021] Optionally, in a third implementation of the first aspect of the present invention, during the simulation process, an intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technology, and resource usage of each computing node is monitored in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted, including:

[0022] Using server virtualization technology, the server of the intelligent simulation engine is divided into multiple virtual servers, and the intelligent simulation engine is packaged into a container image. It is deployed and run on multiple computing nodes through containerization technology, where each virtual server independently runs a simulation domain.

[0023] Real-time monitoring of computing node resource usage, where the computing node resource usage includes at least CPU usage, memory usage, storage usage, network bandwidth usage, and task execution progress;

[0024] Set the usage threshold of each resource indicator. When the resource usage exceeds the set threshold, calculate the node load index:

[0025]

[0026] Where CurrentResourceUsage represents the resource usage of the current computing node, and TotalResoourceCapacity represents the total resource capacity of the computing node;

[0027] The execution strategy of simulation tasks is adjusted using the dwarf mongoose optimization algorithm according to load differences and resource requirements;

[0028] Among them, the CPU usage threshold is 70%, the memory usage threshold is set to 60%, the storage usage threshold is set to 80%, and the network bandwidth usage threshold is set to 70%. The threshold of the task execution progress is determined according to the urgency of the task.

[0029] Optionally, in a fourth implementation of the first aspect of the present invention, the step of adjusting the execution strategy of the simulation task by using the dwarf mongoose optimization algorithm according to load differences and resource requirements includes:

[0030] Initialize the population of the simulation task and set the initial parameters, including at least the resource requirements and load differences of the simulation task. The dwarf mongoose population is divided into three groups: the Alpha group, the scout group, and the nanny group.

[0031] The individuals in the Alpha group perform foraging operations under the leadership of the female leader to search for food. The probability of individual j becoming the female leader is:

[0032]

[0033] Where L is the number of individuals in the Alpha group, fit j is the fitness value of the jth individual, is the sum of all individual fitness values ​​in the Alpha group;

[0034] Individuals in the Alpha group, after searching for food, will update the solution if the newly generated solution is better than the original one, otherwise it will remain unchanged. When the foraging is over, the average value of the sleeping mound is updated, and the scouting group is responsible for finding the next sleeping mound;

[0035] The worst individual in the Alpha group will be exchanged for the best individual in the nanny group, adjusting the roles of the members in the population. The Alpha group will become a scouting group, searching for new locations.

[0036] When the number of iterations reaches the maximum value, the current optimal solution is output, and the optimal simulation task execution strategy is obtained based on the current optimal solution.

[0037] Optionally, in a fifth implementation of the first aspect of the present invention, after the simulation task is completed, analyzing the simulation results, extracting key performance indicators, and optimizing the parameters of the intelligent simulation engine according to the key performance indicators, including:

[0038] When the simulation task is completed, all data generated during the execution of the simulation task are collected, and the collected data are sorted and classified to obtain the execution process data;

[0039] Calculate based on the execution process data to obtain simulation accuracy index, execution efficiency index and stability index, and integrate to obtain key performance indicators;

[0040] Regression analysis is used to establish a relationship model between the key performance indicators and the intelligent simulation engine parameters. Assuming that the intelligent simulation engine parameters are P1, P2 and P3, the key performance indicator is Y, and the linear regression model is:

[0041] Y=aP1+bP2+cP3+d

[0042] In the formula, a, b, and c are coefficients, which represent the impact on the key performance indicator Y, and d is the intercept term, which represents the key performance indicator value when all parameters are 0;

[0043] The simulated annealing algorithm is used to find the optimal parameter combination, and the optimal parameter combination obtained by the simulated annealing algorithm is the optimized intelligent simulation engine parameters.

[0044] Optionally, in a sixth implementation of the first aspect of the present invention, the simulation accuracy index is obtained by calculating the mean absolute error and the root mean square error:

[0045]

[0046] In the formula, y i is the actual value, is the simulation value, n is the number of data points;

[0047] The stability index is obtained by calculating the standard deviation:

[0048]

[0049] In the formula, x i is a single data point and μ is the mean of the data.

[0050] A second aspect of the present invention provides an intelligent simulation engine management device, the intelligent simulation engine management device comprising:

[0051] The simulation task analysis model is used to analyze the received simulation task, determine the priority of the simulation task, and evaluate the resource requirements of the simulation task when the simulation task is received;

[0052] The thread scheduling model is used to schedule the threads of the intelligent simulation engine using a genetic algorithm according to the priority and resource requirements of the simulation task, and to assign the simulation task to the corresponding computing node for processing, wherein the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven;

[0053] The resource monitoring model is used to share the resource usage of each computing node in real time through virtualization and containerization technologies using an intelligent simulation engine among multiple simulation domains during the simulation process. When abnormal usage is found, the execution strategy of the simulation task is adjusted.

[0054] The result analysis model is used to analyze the simulation results after the simulation task is completed, extract key performance indicators, and optimize the parameters of the intelligent simulation engine through the key performance indicators.

[0055] The third aspect of the present invention provides an intelligent simulation engine management device, which includes a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the intelligent simulation engine management device executes each step of the intelligent simulation engine management method as described in any one of the above items.

[0056] A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored thereon, and when the instructions are executed by a processor, the various steps of the intelligent simulation engine management method as described in any of the above items are implemented.

[0057] In the technical solution provided by the present invention, when a simulation task is received, the received simulation task is analyzed to determine the priority of the simulation task and evaluate the resource requirements of the simulation task; according to the priority and resource requirements of the simulation task, a genetic algorithm is used to schedule the threads of the intelligent simulation engine, and the simulation task is assigned to the corresponding computing node for processing, wherein the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven; during the simulation process, the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technology, and the resource usage of each computing node is monitored in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted; when the simulation task is completed, the simulation results are analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized through the key performance indicators; the present invention can share resources among multiple simulation domains and flexibly adjust the priority, time span and execution mode of the simulation task through the support of multiple driving modes, thereby significantly improving the simulation efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the following detailed description of the preferred embodiment.The drawings are only for the purpose of illustrating the preferred embodiments and are not to be construed as limiting the invention.

[0059] Figure 1 A schematic diagram of a first embodiment of the intelligent simulation engine management method provided by an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of a second embodiment of the intelligent simulation engine management method provided by an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of a third embodiment of the intelligent simulation engine management method provided by an embodiment of the present invention;

[0062] Figure 4 A schematic diagram of the structure of an intelligent simulation engine management device provided by an embodiment of the present invention;

[0063] Figure 5 A schematic diagram of the structure of an intelligent simulation engine management device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0065] The present invention provides an intelligent simulation engine, which can perform intelligent simulation management in a large-scale, multi-threaded environment. Through the support of multiple driving modes (such as time-driven, event-driven, scenario-driven and knowledge-driven), it can share resources between multiple simulation domains and flexibly adjust the priority, time span and execution mode of simulation tasks, thereby significantly improving simulation efficiency and accuracy.

[0066] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 A first embodiment schematic diagram of an intelligent simulation engine management method provided by an embodiment of the present invention, the method specifically comprises the following steps:

[0067] Step 101: When a simulation task is received, the received simulation task is analyzed, the priority of the simulation task is determined, and the resource requirement of the simulation task is evaluated;

[0068] In this embodiment, when a simulation task is received, the received simulation task is analyzed to determine the field, application scenario, type and task time requirement of the simulation task, and evaluate the complexity of the simulation task and the task quality requirement; the priority of the simulation task is determined according to pre-set rules, wherein the pre-set rules are priority = type weight × task type + complexity weight × task complexity + time weight × task time requirement + quality weight × task quality requirement; the required number of CPU cores, memory capacity requirements and network bandwidth requirements are determined according to the complexity of the simulation task, and the configuration of the intelligent simulation engine is adjusted according to the number of CPU cores, memory capacity requirements and network bandwidth.

[0069] Step 102: according to the priority and resource requirements of the simulation task, a genetic algorithm is used to schedule the threads of the intelligent simulation engine, and the simulation task is assigned to the corresponding computing node for processing, wherein the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven;

[0070] In this embodiment, time drive includes time span, multi-scale and nonlinear time drive. Time span: this engine can support the execution of simulation models in different time dimensions, including the ability to simulate time span, that is, the simulation task can evolve in a leap-forward manner on the time axis; multi-scale: supports simulation at different time scales, and can handle tasks of various time scales from micro to macro, short to long term; nonlinear time drive: the simulation engine can support nonlinear time flow, so that the occurrence of simulation events is independent of the passage of time, and can simulate more complex time nonlinear relationships.

[0071] Event-driven includes multiple roles, multiple events and multiple behaviors. Multiple roles: the engine supports the interaction of multiple simulation roles in the same environment, and can drive the progress of simulation tasks according to the behaviors and events of different roles; Multiple events: the simulation engine can determine the execution process of simulation tasks according to the occurrence of multiple independent or dependent events, including concurrent events, priority events, etc.; Multiple behaviors: supports simulation tasks with multiple behavior modes, including decision-making behavior, motion behavior, perception behavior, etc., and can flexibly respond to different types of behavior drives.

[0072] Scenario-driven: This engine supports simulation task management based on different scenarios. For example, in different geographical environments, physical conditions, weather changes and other scenarios, the behavior and state of the simulation model will have different responses, ensuring that the simulation results can adapt to different actual environments.

[0073] Knowledge-driven includes knowledge representation and reasoning and knowledge self-learning. Knowledge representation and reasoning: the simulation engine can reason and make decisions based on existing knowledge and intelligently drive simulation tasks by integrating technologies such as knowledge graphs and expert systems; Knowledge self-learning: supports automatic learning and self-updating of knowledge based on generative AI to enhance the adaptability and autonomy of the simulation engine; the knowledge base is the core part of knowledge-driven simulation, which contains the domain knowledge, rules, historical data, expert experience and best practices required for simulation tasks. This knowledge includes not only static technical documents and standard operating procedures (SOPs), but also dynamically updated real-time data, experience summaries and other information.

[0074] Step 103: During the simulation process, the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technologies to monitor the resource usage of each computing node in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted;

[0075] Step 104: After the simulation task is completed, the simulation results are analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized based on the key performance indicators.

[0076] In this embodiment, when the simulation task is completed, all data generated during the execution of the simulation task are collected, and the collected data are sorted and classified to obtain the execution process data; based on the execution process data, calculations are performed to obtain simulation accuracy indicators, execution efficiency indicators and stability indicators, and key performance indicators are integrated to obtain key performance indicators; regression analysis is used to establish a relationship model between key performance indicators and intelligent simulation engine parameters, assuming that the intelligent simulation engine parameters are P1, P2 and P3, the key performance indicator is Y, and the linear regression model is:

[0077] Y=aP1+bP2+cP3+d

[0078] In the formula, a, b, and c are coefficients, representing the impact on the key performance indicator Y, and d is the intercept term, representing the key performance indicator value when all parameters are 0. The simulated annealing algorithm is used to find the optimal parameter combination, and the optimal parameter combination obtained by the simulated annealing algorithm is the optimized intelligent simulation engine parameters.

[0079] In this embodiment, the simulation accuracy index is obtained by calculating the mean absolute error and the root mean square error:

[0080]

[0081] In the formula, y i is the actual value, is the simulation value, n is the number of data points;

[0082] The stability index is obtained by calculating the standard deviation:

[0083]

[0084] In the formula, x i is a single data point and μ is the mean of the data.

[0085] In this embodiment, the simulated annealing algorithm is a random search algorithm based on the physical annealing process. It avoids falling into the local optimal solution by accepting inferior solutions with a certain probability during the search process, thereby gradually approaching the global optimal solution. The objective function can be set to the inverse of the performance index, and a set of initial intelligent simulation engine parameters are randomly generated as the initial solution. At the current temperature, the current solution is perturbed to generate a new solution, and the objective function value of the new solution and the objective function value of the current solution are calculated. If the objective function value of the new solution is greater than the objective function value of the current solution, the new solution is accepted as the current solution; otherwise, the new solution is accepted with a certain probability. When the temperature drops to the termination temperature, the algorithm stops, and the solution at this time is the optimized intelligent simulation engine parameters.

[0086] In the embodiment of the present invention, multiple simulation driving modes such as time-driven, event-driven, scenario-driven, and knowledge-driven are innovatively integrated, so that the simulation engine can flexibly respond to different types of simulation tasks; through multi-threaded management and intelligent scheduling algorithms, multiple simulation tasks in large scenarios can be efficiently managed to ensure the coordination and optimization of task execution; it is proposed to support the sharing of multiple simulation domains through the same simulation engine, which significantly improves the resource utilization efficiency and reduces the system construction cost; the intelligent simulation engine can realize automatic learning and reasoning through knowledge-driven, and can dynamically adapt to different simulation needs; in terms of military simulation, the intelligent simulation engine can help the military simulation system to intelligently select strategies, optimize tactics, and respond to the complex and rapidly changing environment on the battlefield; in the field of intelligent manufacturing, the intelligent simulation engine can adjust and optimize each link in the production process in real time to improve production efficiency and product quality; in the construction of smart cities, the intelligent simulation engine can provide data-based decision support to help optimize and predict systems such as transportation and energy; medical simulation can optimize clinical decision-making, improve diagnosis and treatment efficiency, and improve patient health management and treatment effects through the intelligent simulation engine.

[0087] See also Figure 2 , a schematic diagram of a second embodiment of the intelligent simulation engine management method provided by an embodiment of the present invention, the method comprising:

[0088] Step 201: sort the simulation tasks according to their priorities, and allocate the simulation tasks with the same priority according to their resource requirements;

[0089] Step 202, initializing the population, where individuals in the population are used to represent different simulation task allocation schemes;

[0090] In this embodiment, the fitness function is:

[0091] Fitness=w p ×Priority+w r ×ResourceDemand / NodeeCapacity

[0092] In the formula, w p and w r They are the weight coefficients of priority and resource demand, respectively. Priority is the priority of the task, ResourceDemand is the resource demand of the task, and NodeCapacity is the processing capacity of the computing node;

[0093] Step 203: Use the roulette wheel selection method to select individuals with higher fitness from the current population as parents, perform a crossover operation on the selected parent individuals, randomly select a computing node as a crossover point in the task allocation scheme, and exchange the task allocations of the two parent individuals on the node, thereby generating new child individuals;

[0094] Step 204: Perform mutation operation on the newly generated offspring individuals to move a task from one computing node to another computing node;

[0095] Step 205: When the genetic algorithm reaches the set maximum number of iterations, the algorithm is stopped, and the individual with the highest fitness value is selected from the last generation population as the optimal simulation task allocation scheme, and the simulation tasks are allocated to the corresponding computing nodes for processing based on the optimal simulation task allocation scheme.

[0096] See also Figure 3 , a third embodiment schematic diagram of the intelligent simulation engine management method provided by an embodiment of the present invention, the method comprising:

[0097] Step 301: using server virtualization technology, divide the server of the intelligent simulation engine into multiple virtual servers, package the intelligent simulation engine into a container image, deploy and run it on multiple computing nodes through containerization technology, where each virtual server independently runs a simulation domain;

[0098] Step 302: Real-time monitoring of the resource usage of the computing nodes, wherein the resource usage of the computing nodes includes at least CPU usage, memory usage, storage usage, network bandwidth usage, and task execution progress;

[0099] In this embodiment, the CPU usage threshold is 70%, the memory usage threshold is set to 60%, the storage usage threshold is set to 80%, and the network bandwidth usage threshold is set to 70%. The threshold of the task execution progress is determined according to the urgency of the task. For urgent simulation tasks with strict time limits, the task execution progress threshold can be set more strictly. For example, within the first hour after the start of the task, at least 30% of the task progress is required to be completed. If the total task duration is 4 hours, then when the task is halfway through (2 hours), at least 70% of the progress should be completed. When the task is executed for 3 hours, 90% of the progress should be completed. This ensures that the task can be successfully completed before the deadline. For important tasks that do not have strict real-time requirements, the progress threshold can be relatively loose. For example, within the first 20% of the time after the task starts, at least 10% of the task progress is required to be completed, when the task is halfway executed, at least 40% of the progress should be completed, and when the task is 80% of the time, 80% of the progress should be completed. For ordinary simulation tasks, the progress threshold can be more relaxed. For example, within the first 30% of the time after the task starts, at least 5% of the task progress is required to be completed, when the task is halfway executed, at least 30% of the progress should be completed, when the task is 70% of the time, 60% of the progress should be completed, and when the task is 90% of the time, 80% of the progress should be completed.

[0100] Step 303: Set the usage rate threshold of each resource indicator. When the resource usage exceeds the set threshold, calculate the node load index:

[0101]

[0102] In the formula, CurrentResourceUsage represents the resource usage of the current computing node, and TotalResourceCapacity represents the total resource capacity of the computing node;

[0103] Step 304: The execution strategy of the simulation task is adjusted using the dwarf mongoose optimization algorithm according to the load difference and resource requirements.

[0104] In this embodiment, the population of the simulation task is initialized, and the initial parameters are set, including at least the resource requirements and load differences of the simulation task. The dwarf mongoose group is divided into three groups: the Alpha group, the scout group, and the nanny group. The individuals in the Alpha group perform foraging operations under the leadership of the female leader to search for food. The probability of individual j becoming the female leader is:

[0105]

[0106] Where L is the number of individuals in the Alpha group, fit jis the fitness value of the jth individual, It is the sum of the fitness values ​​of all individuals in the Alpha group. After the individuals in the Alpha group search for food, if the newly generated solution is better than the original solution, the solution will be updated, otherwise it will remain unchanged. After the foraging is completed, the average value of the sleeping mound is updated, and the reconnaissance group is responsible for finding the next sleeping mound. The worst individual in the Alpha group will be exchanged with the best individual in the nanny group, and the roles of the members in the population will be adjusted. The Alpha group will become the reconnaissance group and search for a new position. When the number of iterations reaches the maximum value, the current optimal solution is output, and the optimal simulation task execution strategy is obtained based on the current optimal solution.

[0107] See also Figure 4 , a structural diagram of an intelligent simulation engine management device provided by an embodiment of the present invention, the device comprises:

[0108] The simulation task analysis model 401 is used to analyze the received simulation task, determine the priority of the simulation task, and evaluate the resource requirements of the simulation task when the simulation task is received;

[0109] The thread scheduling model 402 is used to schedule the threads of the intelligent simulation engine using a genetic algorithm according to the priority and resource requirements of the simulation task, and to assign the simulation task to the corresponding computing node for processing, wherein the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven;

[0110] The resource monitoring model 403 is used to share the resource usage of each computing node in real time by using the intelligent simulation engine through virtualization and containerization technology among multiple simulation domains during the simulation process, and to adjust the execution strategy of the simulation task when abnormal usage is found;

[0111] The result analysis model 404 is used to analyze the simulation results after the simulation task is completed, extract key performance indicators, and optimize the parameters of the intelligent simulation engine through the key performance indicators.

[0112] above Figure 4 The intelligent simulation engine management device in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the intelligent simulation engine management device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0113] Figure 5It is a structural diagram of an intelligent simulation engine management device provided by an embodiment of the present invention. The intelligent simulation engine management device 600 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 610 (for example, one or more processors) and a memory 620, and one or more storage media 630 (for example, one or more mass storage devices) storing application programs 633 or data 632. Among them, the memory 620 and the storage medium 630 can be short-term storage or permanent storage. The program stored in the storage medium 630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the intelligent simulation engine management device 600. Furthermore, the processor 610 can be configured to communicate with the storage medium 630, and execute a series of instruction operations in the storage medium 630 on the intelligent simulation engine management device 600 to implement the method provided by the above implementation.

[0114] The intelligent simulation engine management device 600 may also include one or more power supplies 640, one or more wired or wireless network interfaces 650, one or more input and output interfaces 660, and / or one or more operating devices 631, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 5 The structure of the intelligent simulation engine management device shown does not constitute a limitation on the computer device provided by the present invention, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0115] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the various steps of the intelligent simulation engine management method provided in the above-mentioned embodiments.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described equipment, devices, and units can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0117] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0118] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for managing an intelligent simulation engine, characterized in that: The intelligent simulation engine management method comprises: When receiving a simulation task, analyzing the received simulation task, determining the priority of the simulation task, and evaluating the resource requirements of the simulation task; According to the priority and resource requirements of the simulation task, the genetic algorithm is used to schedule the threads of the intelligent simulation engine, and the simulation task is assigned to the corresponding computing node for processing, where the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven; During the simulation process, the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technologies to monitor the resource usage of each computing node in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted. When the simulation task is completed, the simulation results are analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized based on the key performance indicators.

2. The intelligent simulation engine management method according to claim 1, characterized in that: When a simulation task is received, analyzing the received simulation task, determining the priority of the simulation task, and evaluating the resource requirements of the simulation task include: When receiving a simulation task, analyze the received simulation task to determine the field, application scenario, type and task time requirement of the simulation task, and evaluate the complexity of the simulation task and the task quality requirements; Determine the priority of the simulation task according to a pre-set rule, wherein the pre-set rule is priority = type weight × task type + complexity weight × task complexity + time weight × task time requirement + quality weight × task quality requirement; Determine the required number of CPU cores, memory capacity requirements, and network bandwidth requirements based on the complexity of the simulation task, and adjust the configuration of the intelligent simulation engine based on the number of CPU cores, memory capacity requirements, and network bandwidth.

3. The intelligent simulation engine management method according to claim 1, characterized in that: The method uses a genetic algorithm to schedule threads of the intelligent simulation engine according to the priority and resource requirements of the simulation task, and distributes the simulation task to the corresponding computing node for processing, including: Sort the simulation tasks according to their priorities, and for simulation tasks with the same priority, allocate them according to their resource requirements; Initialize the population, and the individuals in the population are used to represent different simulation task allocation schemes. The fitness function is: Fitness=w p ×Priority+w r ×ResourceDemand / NodeCapacity In the formula, w p and w r They are the weight coefficients of priority and resource demand, respectively. Priority is the priority of the task, ResourceDemand is the resource demand of the task, and NodeCapacity is the processing capacity of the computing node; The roulette wheel selection method is used to select individuals with higher fitness from the current population as parents, and the selected parent individuals are crossovered. A computing node is randomly selected as the crossover point in the task allocation scheme, and the task allocations of the two parent individuals on the node are exchanged to generate new offspring individuals. Perform mutation operations on the newly generated offspring individuals to move a task from one computing node to another; When the genetic algorithm reaches the set maximum number of iterations, the algorithm stops and the individual with the highest fitness value is selected from the last generation of the population as the optimal simulation task allocation scheme. Based on the optimal simulation task allocation scheme, the simulation tasks are allocated to the corresponding computing nodes for processing.

4. The intelligent simulation engine management method according to claim 1, characterized in that: In the simulation process, the intelligent simulation engine is shared among multiple simulation domains through virtualization and containerization technology, and the resource usage of each computing node is monitored in real time. When abnormal usage is found, the execution strategy of the simulation task is adjusted, including: Using server virtualization technology, the server of the intelligent simulation engine is divided into multiple virtual servers, and the intelligent simulation engine is packaged into a container image. It is deployed and run on multiple computing nodes through containerization technology, where each virtual server independently runs a simulation domain. Real-time monitoring of computing node resource usage, where the computing node resource usage includes at least CPU usage, memory usage, storage usage, network bandwidth usage, and task execution progress; Set the usage threshold of each resource indicator. When the resource usage exceeds the set threshold, calculate the node load index: In the formula, CurrentResourceUsage represents the resource usage of the current computing node, and TotalResourceCapacity represents the total resource capacity of the computing node; The execution strategy of simulation tasks is adjusted using the dwarf mongoose optimization algorithm according to load differences and resource requirements; Among them, the CPU usage threshold is 70%, the memory usage threshold is set to 60%, the storage usage threshold is set to 80%, and the network bandwidth usage threshold is set to 70%. The threshold of the task execution progress is determined according to the urgency of the task.

5. The intelligent simulation engine management method according to claim 4, characterized in that: The dwarf mongoose optimization algorithm is used to adjust the execution strategy of the simulation task according to the load difference and resource requirements, including: Initialize the population of the simulation task and set the initial parameters, including at least the resource requirements and load differences of the simulation task. The dwarf mongoose population is divided into three groups: the Alpha group, the scout group, and the nanny group. The individuals in the Alpha group perform foraging operations under the leadership of the female leader to search for food. The probability of individual j becoming the female leader is: Where L is the number of individuals in the Alpha group, fit j is the fitness value of the jth individual, is the sum of all individual fitness values ​​in the Alpha group; Individuals in the Alpha group, after searching for food, will update the solution if the newly generated solution is better than the original one, otherwise it will remain unchanged. When the foraging is over, the average value of the sleeping mound is updated, and the scouting group is responsible for finding the next sleeping mound; The worst individual in the Alpha group will be exchanged for the best individual in the nanny group, adjusting the roles of the members in the population. The Alpha group will become a scouting group, searching for new locations. When the number of iterations reaches the maximum value, the current optimal solution is output, and the optimal simulation task execution strategy is obtained based on the current optimal solution.

6. The intelligent simulation engine management method according to claim 1, characterized in that: After the simulation task is completed, the simulation results are analyzed, key performance indicators are extracted, and the parameters of the intelligent simulation engine are optimized according to the key performance indicators, including: When the simulation task is completed, all data generated during the execution of the simulation task are collected, and the collected data are sorted and classified to obtain the execution process data; Calculate based on the execution process data to obtain simulation accuracy index, execution efficiency index and stability index, and integrate to obtain key performance indicators; Regression analysis is used to establish a relationship model between the key performance indicators and the intelligent simulation engine parameters. Assuming that the intelligent simulation engine parameters are P1, P2 and P3, the key performance indicator is Y, and the linear regression model is: Y=aP1+bP2+cP3+d In the formula, a, b, and c are coefficients, which represent the impact on the key performance indicator Y, and d is the intercept term, which represents the key performance indicator value when all parameters are 0; The simulated annealing algorithm is used to find the optimal parameter combination, and the optimal parameter combination obtained by the simulated annealing algorithm is the optimized intelligent simulation engine parameters.

7. An intelligent simulation engine management method as claimed in claim 6, characterized in that: The simulation accuracy index is obtained by calculating the mean absolute error and the root mean square error: In the formula, y i is the actual value, is the simulation value, n is the number of data points; The stability index is obtained by calculating the standard deviation: In the formula, x i is a single data point and μ is the mean of the data.

8. An intelligent simulation engine management device, characterized in that: The intelligent simulation engine management device comprises: The simulation task analysis model is used to analyze the received simulation task, determine the priority of the simulation task, and evaluate the resource requirements of the simulation task when the simulation task is received; The thread scheduling model is used to schedule the threads of the intelligent simulation engine using a genetic algorithm according to the priority and resource requirements of the simulation task, and to assign the simulation task to the corresponding computing node for processing, wherein the intelligent simulation engine includes at least time-driven, event-driven, scenario-driven, and knowledge-driven; The resource monitoring model is used to share the resource usage of each computing node in real time through virtualization and containerization technologies using an intelligent simulation engine among multiple simulation domains during the simulation process. When abnormal usage is found, the execution strategy of the simulation task is adjusted. The result analysis model is used to analyze the simulation results after the simulation task is completed, extract key performance indicators, and optimize the parameters of the intelligent simulation engine through the key performance indicators.

9. An intelligent simulation engine management device, characterized in that: The intelligent simulation engine management device includes a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the intelligent simulation engine management device executes each step of the intelligent simulation engine management method as described in any one of claims 1-7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the various steps of the intelligent simulation engine management method as described in any one of claims 1-7 are implemented.

Images