A seminar-style simulation review system and method

Abstract

This application relates to a discussion-based simulation review system and method. The system includes: a simulation object state saving module, a simulation running control module, and a simulation playback control module. The simulation object state saving module is used to save the states of all simulation objects in the data-driven mode to obtain a state list of each simulation object. The simulation running module is used to update its own state according to the corresponding mechanism model in the model-driven mode of the system during the simulation running process. The simulation playback control module is used to restore the attribute values of each simulation object at the current simulation clock according to the state list of each simulation object, evaluate the attribute values, and modify the attribute values of the simulation objects in the model-driven mode according to the evaluation results. Using this system can achieve discussion-based review, save a large amount of running time, and improve the review efficiency.

Description

Technical Field

[0001] This application relates to the field of simulation technology, and particularly to a discussion-based simulation review system and method. Background Art

[0002] Both simulation and training have a review process, the purpose of which is to reproduce the simulation and training processes to discover problems or highlights in the simulation and training. It can be said that review is a necessary link to realize the value of simulation and training. The review is generally carried out after the event, that is, key state data is saved in chronological order during the simulation and training process, and these state data are reproduced in the same chronological order after the simulation and training are over. From this process, it can be seen that the review is to reproduce the simulation and training processes, which is also the basic requirement of the review, namely the so-called "reproduction-based review".

[0003] However, in practical applications, the reproduction-based review is not sufficient for discovering problems and values in the simulation and training processes, especially in analysis and demonstration simulations and trainings. For example, at a certain time point in a certain simulation review (referred to as the discussion intervention point), it is found that there are 5 ammunition stocks on position A and they are not enough to eliminate the enemy target B, and at the same time, it is found that if there are 10 ammunition stocks on this position, the target can be completed. At this time, the most direct idea is to change the stock on this position to 10, that is, data in the simulation or training process can still be intervened during the review process. This is the process of the discussion-based review. It can be seen that the discussion-based review can shorten the time to demonstrate the optimal solution by iteratively correcting the simulation and training processes during the post-event review. The discussion-based review is different from modifying the simulation scenario and then running the simulation again. This involves the problem of simulation reproduction. Especially after a large number of random numbers are used in the simulation, repeated running may obtain a simulation state different from that before the discussion intervention point. Then, even if the initial scenario is modified, the effect of the discussion cannot be verified. In addition, compared with running the simulation again after modifying the scenario, the discussion-based review can save a large amount of running time and improve the review efficiency. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a discussion-based simulation review system and method that can implement the discussion-based review, save a large amount of running time, and improve the review efficiency.

[0005] A discussion-based simulation review system, the system includes a simulation object state saving module, a simulation running control module, and a simulation playback control module:

[0006] The simulation object state saving module is used to save the states of all simulation objects in the data-driven mode to obtain a state list of each simulation object;

[0007] The simulation running module is used to update the state of the simulation object according to the corresponding mechanism model in the model-driven mode of the system during the simulation running process;

[0008] The simulation playback control module is used to restore the attribute values of each simulation object at the current simulation clock according to the state list of each simulation object, evaluate the attribute values, and modify the attribute values of the simulation object in the model-driven mode according to the evaluation results.

[0009] In one embodiment, the model-driven mode means that the simulation and training run forward according to the logic contained in the model.

[0010] In one embodiment, the data-driven mode means reading the saved state data according to the timestamp sequence and restoring the state of the simulation object according to the read data; the state includes the simulation clock and important attribute values of the simulation object.

[0011] In one embodiment, the simulation object state saving module is also used to update the state of the simulation object when processing events and advance the simulation clock.

[0012] In one embodiment, the simulation playback control module is also used to start reading the state list of all simulation objects from the data-driven mode, determine each simulation clock for playback, and restore the attribute values of each simulation object by using the state of the simulation object saved at the simulation clock, thus obtaining a snapshot of the simulation system at the simulation clock.

[0013] In one embodiment, the simulation playback control module includes a visual situation system and an evaluation system; the visual situation system displays the state of the simulation object in a graphical interface; the evaluation system gives the contribution degree of the simulation object to the evaluation indexes of the simulation experiment.

[0014] In one embodiment, the simulation playback control module is also used to evaluate the attribute values of the simulation object, evaluate whether the states of all simulation objects at the current simulation moment are normal and whether modification is needed, and when it is found that the state of the simulation object needs to be corrected, a discussion-based review intervention is carried out.

[0015] In one embodiment, when the discussion-based review intervention is carried out, the simulation playback control module is also used to continue running the playback after the evaluator changes the attributes of the simulation object, while the attribute changes of the simulation object during the simulation running are restored to the model-driven mode, and the model determines the attribute values of the simulation object.

[0016] In one embodiment, a discussion-based review method, the method includes:

[0017] Read the list of simulation object instances in the scenario file and generate all simulation objects. If the current state is the simulation playback state, sequentially read the first timestamp in the saved state list of each simulation object, select the smallest one as the current playback time t_cur, read the states whose timestamps are not greater than the sum of the current playback time and the preset minimum playback time interval, and push these states to each simulation object. After receiving the saved state, the simulation object restores its state and pushes it to the external situation system to determine whether there is playback intervention. If not, continue the above playback process; if so, pause reading the state list, set the playback flag display to 0, create a new branch S12 after the current state list, and call the existing state list S11 as the current branch. Record the branch time as t_branch, whose value is the current timestamp of the simulation object.

[0018] If the current state is the simulation running state, take the pending events from the event queue. The simulation object reads the local simulation time, serializes the key attributes, and adds this state to the end of the current state list; determine whether there is playback intervention. If not, process the events; if so, set the playback flag display to 1 to enter the simulation playback state, delete the states in other branches of the state list whose timestamps are between [t_branch, t_cur], connect the current state list to the historical state list, update the historical state list by connecting the head after t_branch and the tail before t_cur.

[0019] The simulation analyst makes a seminar-style intervention in a timely manner according to the changes in the simulation situation, changes the state of the simulation object online, and resumes the simulation running process.

[0020] In the above-mentioned discussion-based simulation review system and method, through the combination of data-driven simulation object state preservation, mechanism model-driven state update, and flexible simulation playback control in this application, the discussion-based review system can quickly evaluate and optimize simulation results without re-executing the entire simulation. It can achieve dynamic intervention in the simulation process, reduce time and resource waste in traditional simulations, improve the review efficiency, and thus accelerate the formulation of optimization decisions. At the same time, control elements such as playback flags are set in the discussion-based simulation review method to achieve seamless switching between simulation operation and playback. When problems are found during simulation operation, the playback can be paused and switched to the simulation mode to modify the state, and then the playback can continue to ensure timely intervention and analysis during review. The state list of the simulation object is set to record and save the key attributes of the simulation object in detail during simulation operation, providing rich data for review. During playback, the state can be efficiently read and restored, and the playback rhythm can be controlled according to specific rules and parameters. After the discussion intervenes to modify the state, the state list is properly managed, branches are created or deleted to ensure state consistency and traceability, and the state of the simulation object is pushed to the external situation system in real time, providing an intuitive visualization basis for the discussion personnel and improving the efficiency and quality of simulation analysis and training review. Brief Description of the Drawings

[0021] Figure 1 It is the internal structure diagram of a discussion-based simulation review system in an embodiment;

[0022] Figure 2 It is the flowchart of a discussion-based simulation review method in an embodiment. Detailed Embodiment

[0023] In order to make the purpose, technical solutions and advantages of this application clearer, the following further details this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0024] In an embodiment, as Figure 1 shown, a discussion-based simulation review system is provided. The system includes a simulation object state preservation module 102, a simulation operation control module 104, and a simulation playback control module 106:

[0025] The simulation object state preservation module 102 is used to save the states of all simulation objects in the data-driven mode to obtain the state list of each simulation object;

[0026] The simulation operation module 104 is used to update the state of the simulation object according to the corresponding mechanism model in the model-driven mode of the system during the simulation operation process;

[0027] The simulation playback control module 106 is used to restore the attribute values of each simulation object at the current simulation clock according to the state list of each simulation object, evaluate the attribute values, and modify the attribute values of the simulation objects in the model-driven mode according to the evaluation results.

[0028] In the above-mentioned discussion-based simulation review system, the simulation object state saving module 102 is set to record the states of all simulation objects in chronological order during the simulation process. The state list of each simulation object includes data such as its attributes, behaviors, and interactions. The states of the simulation objects are saved in the system in a data-driven mode, which means that it does not solely rely on the physical model but is recorded according to the actual data during the simulation process. Timestamps or clocks can be used to mark each state recording point so that the states of the simulation objects at specific time points can be accurately restored during subsequent playback. The state information of all simulation objects (such as ammunition, targets, weapon systems, etc.) is serialized and stored in a unified database for subsequent retrieval and use. By recording the states of each simulation object at each time step, the simulation object state saving module can avoid restarting the entire simulation from scratch every time a review is conducted. During the review, only the existing state data needs to be loaded, the states of the simulation objects are restored and analyzed, without having to re-execute the entire simulation process. Through the archiving of the simulation states, it is possible to trace back to a specific time point during the simulation process at any time as needed and start analysis and adjustment based on that point. As long as the states of each simulation object are recorded, they can be quickly played back during the review and adjusted as needed, thus avoiding the time cost of repeated runs in traditional simulation methods.

[0029] The simulation operation control module 104 is set to be responsible for updating the state of each simulation object according to the mode driven by the simulation model. Each step in the simulation operation process is updated according to the corresponding mechanism model. The mechanism model may include dynamic equations, control strategies, interaction rules, etc., which are used to drive the behavioral changes of the simulation objects. The simulation operation control module updates the state by parsing the behavioral model of the simulation object, based on the current time step and the changes in the simulation environment. The state update may include information such as the physical position, speed, state value, health status, etc. of the simulation object. In the seminar-style review, the simulation operation control module can be enabled to re-evaluate the state of the simulation object. The role of the simulation operation control module is to automatically adjust the behavior of the simulation object according to the environmental changes. Through the model-driven simulation process, different input changes can be quickly responded to, and the data can be modified immediately during playback, avoiding repeated simulation processes. During the review, the behavior of the simulation object can be modified and evaluated immediately based on the existing data. This kind of adjustment and optimization does not require re-running the simulation, but quickly tests different solutions through parameter adjustment and immediate feedback, thus significantly shortening the decision-making time. At the same time, by gradually adjusting the behavior or state of individual simulation objects, the optimization process can be carried out in real time without having to wait until after the simulation ends to make modifications.

[0030] The simulation playback control module 106 is set to perform simulation playback based on the recorded list of simulation object states. During playback, the system restores the attribute values of the simulation objects and gradually resumes the simulation state according to the current simulation clock. During the playback process, the state of each simulation object is evaluated, and based on the analysis results, it is allowed to modify the attribute values in the simulation. The modified attribute values are updated according to the model-driven mechanism and further simulation is performed. The playback process requires gradually restoring the state of each simulation object, which can be completed by accessing the state record list in the database. During playback, the system synchronously updates the state of the simulation objects according to the simulation clock. Through playback, data at specific time points can be observed and decisions can be made for intervention at this moment. For example, when it is found that the ammunition quantity is insufficient at a certain time point, the inventory quantity can be modified and the simulation can continue. The core advantage of the seminar-style review lies in the ability to intervene in the simulation data in real time during the playback process. During the review, if it is found that the result of a certain link does not meet the expectation, the analyst can immediately modify the simulation parameters (such as ammunition quantity, target distribution, etc.), and then continue the simulation without having to run the entire simulation again. This method significantly improves the efficiency and saves the time of repeated simulations. The playback control module allows specific moments during the simulation process to be reproduced after the simulation ends without having to run the entire simulation again. By restoring the state data and making immediate modifications, "post-event" rapid intervention can be achieved, saving the time of restarting the simulation and calculation. By dynamically adjusting the state during the playback process, the behavior of the simulation objects can be immediately modified according to the evaluation results (such as adjusting the ammunition inventory, etc.), avoiding starting the simulation from scratch and enhancing the flexibility and efficiency of the optimization process.

[0031] Traditional simulation methods usually require running the entire simulation from scratch. Especially when dealing with complex systems, the simulation time is extremely long. The discussion-based review system avoids re-running the simulation by saving and replaying state data and directly modifying and optimizing based on historical states. Discussion-based review allows for dynamically adjusting the attribute values of simulation objects during replay and continuing the simulation based on the modified states. This method enables simulation personnel to test different hypotheses and scenarios in real time without having to rewrite the simulation settings or start the entire simulation process. By saving state data and replaying, analysts can more flexibly analyze various results during the simulation process, adjust different parameter combinations, and quickly evaluate the effects of various optimization schemes. This flexibility enables the review not only to help identify problems but also to explore new optimization paths, thereby improving decision-making efficiency. Traditional simulation requires running different experiments or scenarios multiple times, while discussion-based review avoids repeated calculations by directly adjusting the simulation state, thus saving a large amount of computing resources and time. Through combining data-driven simulation object state saving, mechanism model-driven state updating, and flexible simulation replay control, the discussion-based review system can quickly evaluate and optimize simulation results without re-executing the full simulation. It can achieve dynamic intervention in the simulation process, reduce time and resource waste in traditional simulation, improve the review efficiency, and thus accelerate the formulation of optimized decisions.

[0032] Meanwhile, this application designs a combination of model-driven and data-driven. By combining model-driven (state updating based on physical or mechanism models) and data-driven (replay and analysis based on historical state data), it can provide a more flexible control mechanism for the simulation. Model-driven can ensure the physical rationality of the simulation process, while data-driven provides efficient review and intervention tools. During the review process, the combined data-driven and model-driven approach can provide greater flexibility for analysts, not only being able to reproduce the simulation process but also adjusting parameters according to actual situations to meet specific requirements. Such a mechanism makes the review process more efficient and can promptly identify and adjust potential problems.

[0033] In one of the embodiments, the model-driven mode means that the simulation and training run forward according to the logic contained in the model.

[0034] In a specific embodiment, for example, an aircraft iteratively updates its position according to the set flight path and speed based on the time stamp (simulation time always moves forward). The model-driven is the main mode for the simulation and training to run. To support the review, it is necessary to save the state of the simulation object in the model-driven mode. Also, for example, an aircraft determines its spatial position at each moment according to its direction of operation, engine power, and hydrodynamic characteristics. So at this time, the control right of the attributes of the simulation object lies with the model.

[0035] In one embodiment, the data-driven mode means reading the saved state data in timestamp order and restoring the state of the simulation object according to the read data; the state includes the simulation clock of the simulation object and important attribute values (such as the "position" attribute of the aircraft simulation object in this example).

[0036] In one embodiment, the simulation object state saving module is also used to update the self-state of the simulation object when processing events and advance the simulation clock.

[0037] In a specific embodiment, each simulation object has its own simulation clock. During the simulation run, the simulation object updates its self-state and advances the simulation clock when processing events.

[0038] Assume that the position of the aircraft is represented by a floating-point triple (x, y, z), and the simulation clock of the aircraft simulation object is represented by a floating-point number (starting from 0). After the simulation starts, the aircraft simulation object flies along a set trajectory: updating the clock and position. Then the simulation system needs to save the clock and position at each moment simultaneously to form a state list of the aircraft simulation object, as shown in Table 1.

[0039] Table 1

[0040]

[0041] The simulation system saves the states of all simulation objects to obtain a state list for each simulation object. The state lists of all simulation objects need to be persistently stored, such as stored in a database or a disk file.

[0042] In one embodiment, the simulation playback control module is also used to start reading the state lists of all simulation objects from the data-driven mode, determine each simulation clock for playback, and use the state of the simulation object saved at the simulation clock to restore the attribute values of each simulation object, thus obtaining a snapshot of the simulation system at the simulation clock.

[0043] In a specific embodiment, during the playback process, the control of the simulation object attributes lies in the data, that is, the state saved in the model state list. For example, if the state list of the position of the aircraft simulation object is saved during the simulation run, the saved state can be read out sequentially from the list according to the simulation clock at this time.

[0044] In one embodiment, the simulation playback control module includes a visual situation system and an evaluation system; the visual situation system displays the state of the simulation object in a graphical interface; the evaluation system gives the contribution degree of the simulation object to the simulation experiment evaluation index.

[0045] In a specific embodiment, during the simulation playback process, in coordination with the visualization situation system and the evaluation system, where the situation system displays the status of the simulation objects in a graphical interface, and the evaluation system gives the contribution degree of the simulation objects to the simulation experiment evaluation indicators. Combining these two systems, during the playback process, the evaluator can view and evaluate whether the status of all simulation objects at the current simulation time is normal and whether modification is required. For example, during the playback process, it is found that the position of the aircraft at a certain moment is not conducive to bombing, which results in the bombing missing the target.

[0046] In one of the embodiments, the simulation playback control module is further configured to evaluate the attribute values of the simulation objects, evaluate whether the status of all simulation objects at the current simulation time is normal and whether modification is required, and when it is found that the status of the simulation objects needs to be corrected, initiate a discussion-based review intervention.

[0047] In one of the embodiments, the simulation playback control module is further configured to, when a discussion-based review intervention is initiated, continue the playback when the evaluator changes the attributes of the simulation objects, and during the simulation operation, the attribute changes of the simulation objects return to the model-driven mode, and the model determines the attribute values of the simulation objects.

[0048] In a specific embodiment, at the intervention point, if the position of the aircraft simulation object is manually modified, its subsequent position will be determined by the flight model of the aircraft. The evaluator can also switch to the data-driven mode as needed during the discussion-based review.

[0049] In one embodiment, a discussion-based review method, the method includes:

[0050] Read the list of simulation object instances in the scenario file and generate all simulation objects. If the current state is the simulation playback state, sequentially read the first timestamp in the saved state list of each simulation object, select the smallest as the current playback time t_cur, read the states with timestamps not greater than the sum of the current playback time and a pre-set minimum playback time interval, and push these states to each simulation object. After receiving the saved state, the simulation object restores its state and pushes it to the external situation system to determine whether there is a playback intervention. If not, continue the above playback process; if so, pause reading the state list, set the playback flag display to 0, create a new branch S12 after the current state list, and the existing state list is called S11 and used as the current branch. Record the branch time as t_branch, and its value is the current timestamp of the simulation object.

[0051] If the current state is the simulation running state, take the event to be processed from the event queue. The simulation object reads the local simulation time, serializes the key attributes, and adds this state to the end of the current state list. Determine whether there is replay intervention. If not, process the event. If there is, set the replay flag display to 1 to enter the simulation replay state, delete the states in other branches of the state list whose timestamps are between [t_branch, t_cur], connect the current state list to the historical state list, update the historical state list by connecting the head after t_branch and the tail before t_cur.

[0052] The simulation analyst makes a discussion-based intervention in a timely manner according to the changes in the simulation situation, online changes the state of the simulation object, and re-enters the simulation running process.

[0053] In a specific embodiment, as Figure 2 shown, first, the first run can only be the simulation running process (because there is no replay data at this time). During the simulation running process, all simulation objects will save their states (for simulation replay) before processing events. In this process, the proposed file will be read to generate instances, and then the event to be processed will be taken from the event queue. The simulation object reads the local simulation time, serializes the key attributes and adds them to the end of the current state list, that is, in this way, the state of the simulation object is saved to prepare for subsequent replay.

[0054] When the replay flag display is 1, read the state list saved by each simulation object instance in turn, select the minimum timestamp as the current replay time t_cur, read the corresponding state and push it to the simulation object instance for restoration, and at the same time push it to the external situation system. The basic process of replay is to read the states saved by all simulation objects in turn and restore the state of each simulation object. At this time, it can be seen through the situation system that all simulation objects change over time, which is an embodiment of the data-driven mode. The replay process is driven by reading the saved data. By setting control elements such as the replay flag (such as display), it is possible to flexibly switch according to needs during the simulation running and replay processes. This enables, during the review process, when a point that requires discussion and intervention is found, the replay can be immediately paused, switched to the simulation mode to modify and adjust the state, and then returned to the replay mode to continue observing the subsequent effects. This flexible process control is the basis for realizing the discussion-based review, ensuring that intervention and analysis can be carried out at the appropriate time without being restricted by the process.

[0055] During the simulation run and playback processes, there is a judgment on "playback intervention". When there is playback intervention, corresponding operations will be carried out, such as setting the playback flag display to 1 or 0, deleting some states in the state list, creating new branches, etc. The simulation analysts will conduct discussion-based intervention in a timely manner according to the changes in the simulation situation and change the states of the simulation objects online. At this time, the simulation will resume the simulation run process. The simulation playback process can switch between the playback and run processes, and at the same time replace the existing simulation object state list (a single discussion-based review can output a simulation playback process that meets the requirements, and at the same time record the modifications to the simulation objects during the review intervention and synchronously modify the initial scenario). Through these operations, flexible switching between the simulation and playback processes is achieved, and the state list can be managed and updated according to the situation of discussion-based intervention to meet the requirements of discussion-based playback and improve the efficiency of simulation analysis. At the same time, a simulation object state list is set up to serialize and save the key attributes of the simulation objects into the state list, providing a rich data basis for subsequent playback and discussion. These state information cover various state parameters of the simulation objects at different time points, such as position, speed, energy, state variables, etc., ensuring that there are sufficient details for analysis during the review.

[0056] The playback flag display is a key flag that controls the switching between simulation and playback in the whole process. When display is 0, the process is in the simulation run mode, and event processing and state saving are carried out according to the simulation logic; when display is 1, the process enters the playback mode, and states are read and restored according to the playback logic. The flexible switching between the two processes fits well. By this flag, the direction of the process and the operations to be executed are determined, which is one of the core control elements for realizing discussion-based playback. The state list is used to save the states of the simulation objects during the simulation run process, providing a data basis for playback; in the playback process, it is used to read and restore states so that the simulation objects can be displayed according to the previously saved states. During the playback process, the states of the simulation objects can be efficiently read and restored according to certain rules (such as selecting the minimum timestamp, etc.), and the rhythm and continuity of playback can be controlled according to parameters such as delta (the minimum playback time interval). At the same time, when there is discussion intervention and the states are modified, the state list can be properly managed, such as creating new branches, deleting unnecessary states, etc., to ensure the consistency and traceability of the states.

[0057] Figure 2The process in [description] realizes the seamless switching between simulation and playback by judging the playback flag display and corresponding operations. Whether switching to playback during the simulation process as needed or re-entering the simulation due to seminar intervention during the playback process, the process can accurately manage the state and control the process, providing great flexibility for analysts to operate in different modes according to real-time analysis needs, so as to better discover problems and make decisions. At the same time, through the management and switching method of the state list, as well as the recording and updating of the state during seminar intervention, it helps to improve the efficiency of simulation analysis. By outputting a simulation playback process that meets the requirements in a single seminar review, and recording the modifications to the simulation object and synchronously modifying the initial scenario during the review intervention, it avoids repeated simulation runs and cumbersome data processing, enabling analysts to focus more on problem analysis and solution rather than wasting time on data preparation and process switching, which can greatly improve the efficiency and effectiveness of the entire simulation analysis work.

[0058] Finally, the seminar review can flexibly switch between simulation and playback as needed during the playback process (i.e., post-event review). When problems are found or new ideas emerge during the review process, one can immediately intervene and correct the state of the simulation object, and then continue with the subsequent review process. In a combat simulation review, it is found that the deployment of troops at a certain moment is unreasonable. Through the seminar review, the positions, troop allocations, etc. of the troops can be directly modified at this moment, and then the subsequent combat development can be observed to quickly verify the effect after adjustment. This real-time iterative correction does not require resetting the scenario and running the entire simulation from scratch as in the traditional method, greatly reducing the time required to try different solutions, so that the optimal solution can be found faster. For the traditional method of modifying the scenario and re-running the simulation, it is necessary to re-experience the entire process of simulation initialization, operation, etc. Even if only a small part of the scenario is modified, it may take a lot of time to recalculate and simulate the operation of the entire system, resulting in a long cycle for demonstrating the optimal solution. And in many simulation scenarios, a large number of random numbers are often used to simulate uncertain factors in reality, such as weather changes, randomness of personnel behavior, probability of equipment failure, etc. When re-running the simulation, even if the initial scenario is modified, due to the effect of random numbers, the simulation state before the seminar intervention point is very likely to be different from before. The seminar review proposed in this application is based on the data and states saved from the previously completed simulation runs. As Figure 2As shown, during the simulation run, the status lists of various simulation objects have been saved. During the review, only these statuses need to be read and local adjustments and analyses can be performed as needed, without having to re-run the large amount of data calculations and model operations during the entire simulation process. Compared with traditional methods, when re-running the simulation after modifying the scenario, all calculation and simulation steps need to be re-performed, including initializing the environment, generating random numbers, calculating the status changes of each object, etc. For complex simulation systems, this is very time-consuming. Re-running may take several hours or even days. However, through discussion-based review, based on the existing simulation results, specific links or time periods can be analyzed and adjusted quickly, and an effective review and optimization can be completed in just dozens of minutes or a few hours, significantly improving the review efficiency. Those of ordinary skill in the art can understand that all or part of the processes in the above-described embodiment systems can be completed by instructing relevant hardware through a computer program. 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 above-described system embodiments. Among them, any reference to memory, storage, database, or other media used in the various embodiments provided in this application 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. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0059] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, 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, it should be considered as falling within the scope described in this specification.

[0060] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A discussion-style review system, characterized in that: The discussion-style review system includes a simulation object state saving module, a simulation operation control module and a simulation playback control module: the simulation object state saving module is used to save the state of all simulation objects in the data-driven mode to obtain a state list of each simulation object; the simulation operation control module is used to update the state of the simulation object according to the corresponding mechanism model in the model-driven mode of the discussion-style review system during the simulation operation; the simulation playback control module is used to restore the attribute value of each simulation object under the current simulation clock according to the state list of each simulation object, evaluate the attribute value, and modify the attribute value of the simulation object in the model-driven mode according to the evaluation result; Read the list of simulation object instances in the assumption file and generate all simulation objects. If the current state is the simulation playback state, read the first timestamp in the state list saved by each simulation object in turn, select the smallest one as the current playback time t_cur, read the state whose timestamp is not greater than the sum of the current playback time and the preset time interval, and push these states to each simulation object. After receiving the saved state, the simulation object restores its state and pushes it to the external situation system to determine whether there is playback intervention. If not, continue the playback process; If there is, pause reading the state list, set the playback flag display to 0, create a new branch S12 after the current state list, the existing state list is called S11, and it is used as the current branch, and the branch time is recorded as t_branch, whose value is the current timestamp of the simulation object; If the current state is the simulation running state, take out the pending event from the event queue, read the local simulation time of the simulation object, serialize the key attributes, and add this state to the end of the current state list; determine whether there is playback intervention, if not, process the event; If there is, set the playback flag display to 1 to enter the simulation playback state, delete the states with timestamps between [t_branch, t_cur] in other branches in the state list, connect the current state list to the historical state list, connect the head to after t_branch, and connect the tail to before t_cur to update the historical state list; Simulation analysts conduct discussion-style intervention in a timely manner according to changes in the simulation situation, change the status of the simulation object online, and re-enter the simulation operation process.

2. The discussion review system according to claim 1 is characterized in that: The model-driven mode means that simulation and training run forward according to the logic contained in the model.

3. The discussion review system according to claim 1 is characterized in that: The data driven mode indicates reading the saved state data according to the time stamp sequence, and restoring the state of the simulation object according to the read data; the restored state of the simulation object includes the simulation clock and important attribute values ​​of the simulation object.

4. The discussion review system according to claim 1 is characterized in that: The simulation object state saving module is also used to update the state of the simulation object itself when processing events and advance the simulation clock.

5. The discussion review system according to claim 1 is characterized in that: The simulation playback control module is also used to read the state list of all simulation objects starting from the data drive, determine each simulation clock to be played back, and use the simulation object state saved under the simulation clock to restore the attribute value of each simulation object, thereby obtaining a simulation system snapshot under the simulation clock.

6. The discussion review system according to claim 5 is characterized in that: The simulation playback control module includes a visualization situation system and an evaluation system; the visualization situation system displays the state of the simulation object in a graphical interface; the evaluation system gives the contribution of the simulation object to the simulation experiment evaluation index.

7. The discussion review system according to claim 6 is characterized in that: The simulation playback control module is also used to evaluate the attribute values ​​of the simulation objects, to evaluate whether the status of all simulation objects at the current simulation moment is normal and whether they need to be modified, and to conduct a discussion-style review intervention when it is found that the status of the simulation objects needs to be corrected.

8. The discussion review system according to claim 7 is characterized in that: The simulation playback control module is also used for the evaluation personnel to change the properties of the simulation object during the discussion-style review intervention, and the playback continues to run. At the same time, the property changes of the simulation object during the simulation operation are restored to the model-driven mode, and the property value of the simulation object is determined by the model.

Images