CN116956642A - Threat assessment algorithm performance dynamic simulation verification method - Google Patents
Threat assessment algorithm performance dynamic simulation verification method Download PDFInfo
- Publication number
- CN116956642A CN116956642A CN202311209126.5A CN202311209126A CN116956642A CN 116956642 A CN116956642 A CN 116956642A CN 202311209126 A CN202311209126 A CN 202311209126A CN 116956642 A CN116956642 A CN 116956642A
- Authority
- CN
- China
- Prior art keywords
- threat assessment
- value
- assessment algorithm
- algorithm
- threat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 113
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000012795 verification Methods 0.000 title claims abstract description 19
- 238000005094 computer simulation Methods 0.000 title claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 36
- 230000008859 change Effects 0.000 claims abstract description 30
- 238000004088 simulation Methods 0.000 claims abstract description 23
- 238000011156 evaluation Methods 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 21
- 238000005070 sampling Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims description 34
- 238000012935 Averaging Methods 0.000 claims description 9
- 238000013461 design Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006872 improvement Effects 0.000 abstract description 5
- 238000005457 optimization Methods 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/042—Backward inferencing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a threat assessment algorithm performance dynamic simulation verification method, which comprises simulation and data acquisition, sampling error analysis, change error analysis and comprehensive evaluation and analysis, and realizes the threat assessment algorithm performance dynamic simulation verification based on real total element whole process complex disk deduction of an opposing game scene through sampling error analysis, change error analysis and comprehensive evaluation and analysis, thereby comprehensively measuring the accuracy and stability of the threat assessment algorithm. The method provides support for overall process data tracking and analysis, threat assessment optimization and improvement direction searching and threat assessment algorithm performance improvement searching, and further lays a technical foundation for accurate and stable assessment of high-value target threat degree in a future high-strength game environment.
Description
Technical Field
The application relates to the technical field of system simulation, in particular to a dynamic simulation verification method for threat assessment algorithm performance.
Background
Threat assessment in the wide-area sense is a means for assessing the threat level of a target according to current game scene information, and further defining the intention of an enemy. Threat assessment concepts are derived from data fusion models proposed by the united states military, are in the third level in the models, and are far more difficult to model than the first and second levels. The threat assessment algorithm is applied to game scenes, and has irreplaceable positions for supporting commanders to accurately judge danger, accurately decide and deploy global resources. The information space of the current game scene expands rapidly, the quantization mode of the information elements is increasingly complex, and the updating iteration of the threat assessment algorithm is accelerated.
The traditional threat assessment method comprises a multi-attribute decision method, fuzzy reasoning, gray correlation, bayesian reasoning and the like. Most of the methods construct threat assessment index systems in scene space, and quantitative and qualitative methods are adopted to quantify threat. The method generally requires a large amount of expert experience and experimental statistical data to form reasonable threat factor membership functions and weight coefficients, but the establishment of the threat factor membership functions still has difficulty in covering all scene elements, and the setting of the weight coefficients also has difficulty in verifying the data consistency. With the development of technology, researchers begin to apply new technologies such as neural networks and deep learning to threat assessment, hope to train to obtain reasonable network system weights through methods such as sample learning and the like, and aim to solve uncertain influences caused by human factor participation.
It is not difficult to find that both types of algorithms belong to the data-driven multi-attribute decision problem in the core problem, and in the process of algorithm verification, verification of data consistency and algorithm interpretability is considered. The traditional threat assessment algorithm checking method only uses a few samples to verify the correctness of the threat assessment algorithm in a static scene, but lacks a dynamic threat assessment process in a real game scene.
Disclosure of Invention
The application aims to at least solve one of the technical problems that only a few samples are used for verifying the correctness of a threat assessment algorithm in a static scene in the prior art, but the dynamic threat assessment process in a real game scene is lacked.
Therefore, the application provides a threat assessment algorithm performance dynamic simulation verification method aiming at the problems of poor interpretability and reliability of the current threat assessment algorithm, and realizes the threat assessment algorithm performance dynamic simulation verification based on the whole-process complex deduction of the real countermeasure game scene whole elements through sampling error analysis, change error analysis and comprehensive evaluation and analysis, thereby providing support for whole-process data tracking and analysis, threat assessment optimization and improvement direction searching and threat assessment algorithm performance promotion searching, and further laying a technical foundation for accurate and stable assessment of high-value target threat degree in a future high-strength game environment.
The application provides a threat assessment algorithm performance dynamic simulation verification method, which comprises the following steps:
simulation and data acquisition: designing a simulation entity and a wanted scene based on real game countermeasure scene data, and summarizing threat assessment factors of a threat assessment algorithm in the wanted scene; carrying out complex disk deduction by using a simulation entity and a designed scene, and collecting deduction whole-process data in a fixed period to obtain a deduction whole-process data set, wherein the deduction whole-process data set comprises input data, a threat assessment element item calculation value, a threat assessment result output value and a threat assessment element item calculation value;
sampling error analysis: randomly extracting a plurality of sample data from the deduction whole process data set, comparing the calculated value of the threat assessment element in each sample data with a reference value, obtaining a reference comparison error by combining the weight value of the threat assessment element, and averaging the reference comparison errors of the plurality of sample data to obtain a reference comparison average error value, thereby measuring the accuracy of the threat assessment algorithm;
and (3) change error analysis: calculating the variation of the threat assessment algorithm result output values at two adjacent moments in the whole process data set to obtain a threat assessment algorithm result variation value, comparing and calculating the threat assessment algorithm result variation value with expert scoring results, and averaging the multiple comparison calculation results to obtain an average deviation value so as to measure the credibility of the threat assessment algorithm;
comprehensive evaluation and analysis: and comprehensively calculating the calculation results of the sampling error analysis and the variation error analysis to obtain the comprehensive evaluation value of the threat assessment algorithm, thereby comprehensively measuring the accuracy and the stability of the threat assessment algorithm.
According to the technical scheme, the threat assessment algorithm performance dynamic simulation verification method can also have the following additional technical characteristics:
in the above technical solution, the simulating and data collecting includes:
designing simulation entities and thinking scenes: designing a simulation entity and a wanted scene based on real game countermeasure scene data, and respectively storing design results into a target model library and an wanted scene library;
configuring a threat assessment index system: based on real game countermeasure scene data and designed simulation entities and a wanted scene, carrying out multi-level and multi-angle summarization on threat assessment essential items of a threat assessment algorithm under the wanted scene by using expert knowledge to form a layered threat assessment index system, and storing the result into a threat assessment essential item library;
and carrying out the multiple-disk deduction by using the simulation entity and the wanted scene, acquiring the deduction whole-process data according to the set data acquisition period to obtain a deduction whole-process data set, and storing the deduction whole-process data set into a database.
In the above technical solution, the simulating and data collecting further includes:
expert scoring is carried out on each element item of threat assessment according to real game countermeasure data, and a normal distribution method is adopted to determine the first step of expert scoring resultsLayer->Confidence intervals for the benchmark values of the threat assessment elements.
In the above technical solution, the sampling error analysis includes:
random extraction from a derived whole process datasetThe individual sample data form a sample data set +.>Calculating a deviation value of the threat assessment element calculation value from the confidence interval of the threat assessment element benchmark value, the deviation value +.>The calculation method comprises the following steps:
wherein ,indicate->Layer->Infinitesimal of confidence interval of individual threat assessment element benchmark value,/->Indicate->Layer->Upper bound of confidence interval of reference value of each evaluation element,/->Indicate->Corresponding +.>Layer->A calculated value of the threat assessment element;
calculating the first according to the weight value of the threat assessment element and the deviation value of the threat assessment element calculation valueThe reference comparison error of the sample data is calculated as follows:
wherein ,represent the firstkReference contrast error of individual sample data, +.>Layer number representing threat assessment element set, +.>Indicate->The number of threat assessment element items within a layer; />Indicate->Layer->Weights of the threat assessment elements;
the reference contrast errors of the plurality of sample data are averaged to obtain a reference contrast average error value, and the calculation method is as follows:
wherein ,represents the reference contrast mean error value, +.>Representing the number of sample data extracted.
In the above technical solution, the reference contrast average error valueThe smaller the threat assessment algorithm, the higher the accuracy.
In the above technical solution, the simulating and data collecting further includes:
expert scoring is carried out on threat assessment algorithm results at all moments according to real game countermeasure data, the variable quantity of threat assessment algorithm results at two adjacent moments is determined based on expert scoring, and the confidence interval of the threat assessment algorithm result variable value is determined through a normal distribution method.
In the above technical solution, the change error analysis includes:
the variable quantity of threat assessment result output values of two adjacent moments in the whole deduction process data set is calculated, and the calculation method is as follows:
wherein ,representation->Time and->Threat assessment algorithm result change value between moments, < +.>Representation->Output value of threat assessment algorithm result at moment, +.>Representation->Outputting a result output value of a threat assessment algorithm at the moment;
calculating the deviation value of the threat assessment algorithm result change value according to the confidence interval of the threat assessment result change valueThe calculation method comprises the following steps:
wherein ,representation->Time and->The infinitesimal bounds of the confidence interval of the time threat assessment result variation value,representation->Time and->Time threat assessment result change value confidence interval upper bound +.>Representation ofTime and->A deviation value of threat assessment result change values between moments;
and averaging the deviation values of the threat assessment algorithm result change values obtained by multiple comparison to obtain an average deviation value:
wherein ,representing the average deviation value>For counting the moments>Time indication->Time; />Representing the total number of data acquisition moments.
In the above technical solution, the average deviation valueThe smaller the threat assessment algorithm, the higher the stability.
In the above technical solution, the threat assessment algorithm synthesizes an evaluation valueThe calculation method of (2) is as follows:
wherein ,representing a reference contrast average error value; />Representing the average deviation value; />And representing the evaluation coefficient for balancing the weight of the accuracy and the stability of the algorithm.
In the above technical solution, the threat assessment algorithm synthesizes an evaluation valueTakes on a value between 0 and 1,/for>The closer to 0 the value of (c) indicates the better performance of the threat assessment algorithm>The closer to 1 the value of (c) indicates the poorer the performance of the threat assessment algorithm.
In summary, due to the adoption of the technical characteristics, the application has the beneficial effects that:
based on real game countermeasure data, by using complex disk deduction, a more abundant and large amount of process data can be obtained through periodic acquisition, and data support is provided for threat assessment algorithm performance verification;
through sampling error analysis, change error analysis, comprehensive evaluation and analysis, the accuracy and stability of the threat assessment algorithm under the real game countermeasure scene can be evaluated, and references and directions are provided for optimizing and improving the algorithm and improving the performance of the algorithm for users.
Additional aspects and advantages of the application will be set forth in part in the description which follows, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the application will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of a method for dynamically simulating and verifying the performance of a threat assessment algorithm in accordance with one embodiment of the application.
Detailed Description
In order that the above-recited objects, features and advantages of the present application will be more clearly understood, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present application and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, however, the present application may be practiced otherwise than as described herein, and therefore the scope of the present application is not limited to the specific embodiments disclosed below.
A method and system for dynamically simulating verification of threat assessment algorithm performance provided in accordance with some embodiments of the application is described below with reference to fig. 1.
Some embodiments of the application provide a threat assessment algorithm performance dynamic simulation verification method.
As shown in fig. 1, an embodiment of the present application provides a method for dynamically simulating and verifying performance of a threat assessment algorithm, including:
simulation and data acquisition: designing a simulation entity and a wanted scene based on real game countermeasure scene data, and summarizing threat assessment factors of a threat assessment algorithm in the wanted scene; and carrying out complex disk deduction by using the simulation entity and the wanted scene, and collecting deduction whole-process data in a fixed period to obtain a deduction whole-process data set, wherein the deduction whole-process data set comprises input data, a threat assessment element item calculation value, a threat assessment result output value and a threat assessment element item calculation value.
In some embodiments, the simulating and data collecting includes the steps of:
designing simulation entities and thinking scenes: and designing the simulation entity and the wanted scene based on the real game countermeasure scene data, and respectively storing the design results into a target model library and an wanted scene library.
Configuring a threat assessment index system: based on real game countermeasure scene data and designed simulation entities and a wanted scene, carrying out multi-level and multi-angle summarization on threat assessment essential items of a threat assessment algorithm under the wanted scene by using expert knowledge to form a layered threat assessment index system, and storing the result into a threat assessment essential item library; it will be understood that, in this embodiment, summarizing by using expert knowledge, scoring by expert, etc. refer to objective evaluation of an application object according to industry standards, industry well-known, conventional experience summary, etc., which has relatively clear scoring criteria, even if performed by different experts, the same or similar results can be obtained, and such result errors do not affect the accuracy of the verification method of this embodiment.
And carrying out the multiple-disk deduction by using the simulation entity and the wanted scene, acquiring the deduction whole-process data according to the set data acquisition period to obtain a deduction whole-process data set, and storing the deduction whole-process data set into a database.
In some embodiments, the simulating and data collecting further comprises:
expert scoring is carried out on each element item of threat assessment according to real game countermeasure data, and a normal distribution method is adopted to determine the first step of expert scoring resultsLayer->Confidence interval of benchmark values of individual threat assessment elements +.>, wherein ,indicate->Layer->Infinitesimal of confidence interval of reference value of individual threat assessment element,/->Indicate->Layer->Upper bounds of confidence intervals for individual threat assessment element benchmark values.
Expert scoring is carried out on threat assessment algorithm results at all moments according to real game countermeasure data, the variable quantity of threat assessment algorithm results at two adjacent moments is determined based on expert scoring, and confidence intervals of threat assessment algorithm result variable values are determined through a normal distribution method, wherein ,/>Representation->Time and->Infinitesimal bounds of confidence interval of time threat assessment algorithm result change value,/-for>Representation->Time and->The moment threat assessment algorithm results change the upper bound of the confidence interval.
And storing the obtained threat assessment element benchmark value confidence interval and threat assessment algorithm result change value confidence interval and the deduction whole process data set into a database.
Sampling error analysis: randomly extracting a plurality of sample data from the deduction whole process data set, comparing the calculated value of the threat assessment element in each sample data with a reference value, obtaining a reference comparison error by combining the weight value of the threat assessment element, and averaging the reference comparison errors of the plurality of sample data to obtain a reference comparison average error value, thereby measuring the accuracy of the threat assessment algorithm;
in some embodiments, the sampling error analysis includes:
random extraction from a derived whole process datasetThe individual sample data form a sample data set +.>Calculating a deviation value of the threat assessment element calculation value from the confidence interval of the threat assessment element benchmark value, the deviation value +.>The calculation method comprises the following steps:
wherein ,indicate->Layer->Infinitesimal of confidence interval of individual threat assessment element benchmark value,/->Indicate->Layer->Upper bound of confidence interval of reference value of each evaluation element,/->Indicate->Corresponding +.>Layer->A calculated value of the threat assessment element;
calculating the first according to the weight value of the threat assessment element and the deviation value of the threat assessment element calculation valueThe reference comparison error of the sample data is calculated as follows:
wherein ,represent the firstkReference contrast error of individual sample data, +.>Layer number representing threat assessment element set, +.>Indicate->The number of threat assessment element items within a layer; />Indicate->Layer->Weights of the threat assessment elements;
the reference contrast errors of the plurality of sample data are averaged to obtain a reference contrast average error value, and the calculation method is as follows:
wherein ,representing the reference versus average error value. Specifically, the reference contrast mean error value +.>The smaller the threat assessment algorithm, the higher the accuracy.
And (3) change error analysis: calculating the variation of the threat assessment algorithm result output values at two adjacent moments in the whole process data set to obtain a threat assessment algorithm result variation value, comparing and calculating the threat assessment algorithm result variation value with expert scoring results, and averaging the multiple comparison calculation results to obtain an average deviation value so as to measure the credibility of the threat assessment algorithm;
in some embodiments, the variation error analysis includes:
the variable quantity of threat assessment result output values of two adjacent moments in the whole deduction process data set is calculated, and the calculation method is as follows:
wherein ,representation->Time and->Threat assessment algorithm result change value between moments, < +.>Representation->Output value of threat assessment algorithm result at moment, +.>Representation->Outputting a result output value of a threat assessment algorithm at the moment;
calculating the deviation value of the threat assessment algorithm result change value according to the confidence interval of the threat assessment result change valueThe calculation method comprises the following steps:
wherein ,representation->Time and->The infinitesimal bounds of the confidence interval of the time threat assessment result variation value,representation->Time and->Time threat assessment result change value confidence interval upper bound +.>Representation ofTime and->A deviation value of threat assessment result change values between moments;
and averaging the deviation values of the threat assessment algorithm result change values obtained by multiple comparison to obtain an average deviation value:
wherein ,representing the average deviation value>For counting the moments>Time indication->Time; />The total number of data acquisition time is represented, and the total number of data in the whole process data set is correspondingly deduced. Specifically, the average deviation value +.>The smaller the threat assessment algorithm, the higher the stability.
Comprehensive evaluation and analysis: and comprehensively calculating the calculation results of the sampling error analysis and the variation error analysis to obtain the comprehensive evaluation value of the threat assessment algorithm, thereby comprehensively measuring the accuracy and the stability of the threat assessment algorithm. In some embodiments, the threat assessment algorithm comprehensive evaluation value can also be used for algorithm analysis, and the threat assessment comprehensive capability in the game countermeasure process is further improved through improvement and optimization.
Specifically, the threat assessment algorithm synthesizes an assessment valueThe calculation method of (2) is as follows:
wherein ,representing a reference contrast average error value; />Representing the average deviation value; />Representing the evaluation coefficient, weight for balancing the accuracy and stability of the algorithm, value +.>。
In a specific embodiment, the threat assessment algorithm synthesizes an assessment valueTakes on a value between 0 and 1,/for>The closer to 0 the value of (c) indicates the better performance of the threat assessment algorithm>The closer to 1 the value of (c) indicates the poorer the performance of the threat assessment algorithm.
In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (10)
1. The threat assessment algorithm performance dynamic simulation verification method is characterized by comprising the following steps:
simulation and data acquisition: designing a simulation entity and a wanted scene based on real game countermeasure scene data, and summarizing threat assessment factors of a threat assessment algorithm in the wanted scene; carrying out complex disk deduction by using a simulation entity and a designed scene, and collecting deduction whole-process data in a fixed period to obtain a deduction whole-process data set, wherein the deduction whole-process data set comprises input data, a threat assessment element item calculation value, a threat assessment result output value and a threat assessment element item calculation value;
sampling error analysis: randomly extracting a plurality of sample data from the deduction whole process data set, comparing the calculated value of the threat assessment element in each sample data with a reference value, obtaining a reference comparison error by combining the weight value of the threat assessment element, and averaging the reference comparison errors of the plurality of sample data to obtain a reference comparison average error value, thereby measuring the accuracy of the threat assessment algorithm;
and (3) change error analysis: calculating the variation of the threat assessment algorithm result output values at two adjacent moments in the whole process data set to obtain a threat assessment algorithm result variation value, comparing and calculating the threat assessment algorithm result variation value with expert scoring results, and averaging the multiple comparison calculation results to obtain an average deviation value so as to measure the credibility of the threat assessment algorithm;
comprehensive evaluation and analysis: and comprehensively calculating the calculation results of the sampling error analysis and the variation error analysis to obtain the comprehensive evaluation value of the threat assessment algorithm, thereby comprehensively measuring the accuracy and the stability of the threat assessment algorithm.
2. The method of claim 1, wherein the simulating and data collecting comprises:
designing simulation entities and thinking scenes: designing a simulation entity and a wanted scene based on real game countermeasure scene data, and respectively storing design results into a target model library and an wanted scene library;
configuring a threat assessment index system: based on real game countermeasure scene data and designed simulation entities and a wanted scene, carrying out multi-level and multi-angle summarization on threat assessment essential items of a threat assessment algorithm under the wanted scene by using expert knowledge to form a layered threat assessment index system, and storing the result into a threat assessment essential item library;
and carrying out the multiple-disk deduction by using the simulation entity and the wanted scene, acquiring the deduction whole-process data according to the set data acquisition period to obtain a deduction whole-process data set, and storing the deduction whole-process data set into a database.
3. The method of claim 2, wherein the simulating and data collecting further comprises:
expert scoring is carried out on each element item of threat assessment according to real game countermeasure data, and a normal distribution method is adopted to determine the first step of expert scoring resultsLayer->Confidence intervals for the benchmark values of the threat assessment elements.
4. A threat assessment algorithm performance dynamic simulation verification method according to claim 3 wherein said sampling error analysis comprises:
random extraction from a derived whole process datasetThe individual sample data form a sample data set +.>Calculating a deviation value of the threat assessment element calculation value from the confidence interval of the threat assessment element benchmark value, the deviation value +.>The calculation method comprises the following steps:
wherein ,indicate->Layer->Infinitesimal of confidence interval of individual threat assessment element benchmark value,/->Indicate->Layer->Upper bound of confidence interval of reference value of each evaluation element,/->Indicate->Corresponding +.>Layer->A calculated value of the threat assessment element;
calculating the first according to the weight value of the threat assessment element and the deviation value of the threat assessment element calculation valueSample dataThe calculation method is as follows:
wherein ,represent the firstkReference contrast error of individual sample data, +.>Layer number representing threat assessment element set, +.>Indicate->The number of threat assessment element items within a layer; />Indicate->Layer->Weights of the threat assessment elements;
the reference contrast errors of the plurality of sample data are averaged to obtain a reference contrast average error value, and the calculation method is as follows:
wherein ,representing the reference versus average error value.
5. Root of Chinese characterThe method for dynamic simulation verification of threat assessment algorithm performance of claim 4, wherein said reference-to-average error valueThe smaller the threat assessment algorithm, the higher the accuracy.
6. The method of claim 2, wherein the simulating and data collecting further comprises:
expert scoring is carried out on threat assessment algorithm results at all moments according to real game countermeasure data, the variable quantity of threat assessment algorithm results at two adjacent moments is determined based on expert scoring, and the confidence interval of the threat assessment algorithm result variable value is determined through a normal distribution method.
7. The threat assessment algorithm performance dynamic simulation verification method of claim 6, wherein the variation error analysis comprises:
the variable quantity of threat assessment result output values of two adjacent moments in the whole deduction process data set is calculated, and the calculation method is as follows:
wherein ,representation->Time and->Threat assessment algorithm result change value between moments, < +.>Representation ofOutput value of threat assessment algorithm result at moment, +.>Representation->Outputting a result output value of a threat assessment algorithm at the moment;
calculating the deviation value of the threat assessment algorithm result change value according to the confidence interval of the threat assessment result change valueThe calculation method comprises the following steps:
wherein ,representation->Time and->Infinitesimal of time threat assessment result change value confidence interval,/->Representation->Time and->Time threat assessment result change value confidence interval upper bound +.>Representation->Time and->A deviation value of threat assessment result change values between moments;
and averaging the deviation values of the threat assessment algorithm result change values obtained by multiple comparison to obtain an average deviation value:
wherein ,representing the average deviation value>For counting the moments>Time indication->Time; />Representing the total number of data acquisition moments.
8. The threat assessment algorithm performance dynamic simulation verification method of claim 7, wherein the average deviation valueThe smaller the threat assessment algorithm, the higher the stability.
9. The method for dynamically simulating and verifying the performance of a threat assessment algorithm according to claim 1, wherein the threat assessment algorithm integrates the evaluation valuesThe calculation method of (2) is as follows:
wherein ,representing a reference contrast average error value; />Representing the average deviation value; />And representing the evaluation coefficient for balancing the weight of the accuracy and the stability of the algorithm.
10. The method for dynamically simulating and verifying the performance of a threat assessment algorithm according to claim 9, wherein the threat assessment algorithm integrates the evaluation valuesTakes on a value between 0 and 1,/for>The closer to 0 the value of (c) indicates the better performance of the threat assessment algorithm>The closer to 1 the value of (c) indicates the poorer the performance of the threat assessment algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311209126.5A CN116956642B (en) | 2023-09-19 | 2023-09-19 | Threat assessment algorithm performance dynamic simulation verification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311209126.5A CN116956642B (en) | 2023-09-19 | 2023-09-19 | Threat assessment algorithm performance dynamic simulation verification method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116956642A true CN116956642A (en) | 2023-10-27 |
CN116956642B CN116956642B (en) | 2024-01-12 |
Family
ID=88447733
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311209126.5A Active CN116956642B (en) | 2023-09-19 | 2023-09-19 | Threat assessment algorithm performance dynamic simulation verification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116956642B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090192962A1 (en) * | 2008-01-30 | 2009-07-30 | Rigdon Debra A | Intelligent threat assessment module, method and system for space situational awareness system |
CN114239728A (en) * | 2021-12-17 | 2022-03-25 | 中国航空研究院 | Multi-domain battlefield situation assessment and threat ranking method |
CN114444389A (en) * | 2022-01-25 | 2022-05-06 | 中国人民解放军空军工程大学 | Air attack target dynamic threat assessment method based on combined empowerment and improved VIKOR |
CN115544714A (en) * | 2022-07-12 | 2022-12-30 | 中国航空工业集团公司沈阳飞机设计研究所 | Time sequence dynamic countermeasure threat assessment method based on aircraft formation |
-
2023
- 2023-09-19 CN CN202311209126.5A patent/CN116956642B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090192962A1 (en) * | 2008-01-30 | 2009-07-30 | Rigdon Debra A | Intelligent threat assessment module, method and system for space situational awareness system |
CN114239728A (en) * | 2021-12-17 | 2022-03-25 | 中国航空研究院 | Multi-domain battlefield situation assessment and threat ranking method |
CN114444389A (en) * | 2022-01-25 | 2022-05-06 | 中国人民解放军空军工程大学 | Air attack target dynamic threat assessment method based on combined empowerment and improved VIKOR |
CN115544714A (en) * | 2022-07-12 | 2022-12-30 | 中国航空工业集团公司沈阳飞机设计研究所 | Time sequence dynamic countermeasure threat assessment method based on aircraft formation |
Non-Patent Citations (3)
Title |
---|
HUAJIAN ZHOU等: ""Evasive behavior-based method for threat assessment in different scenarios: A novel framework for intelligent vehicle"", 《ACCIDENT ANALYSIS & PREVENTION》, vol. 148, pages 105798 * |
师维克 等: ""基于dynFWA-IFE的空中目标威胁评估方法"", 《火力与指挥控制》, vol. 47, no. 12, pages 157 - 162 * |
焦浩洋 等: ""基于PCTBTA的有人/无人协同作战交互任务建模"", 《空军工程大学学报》, vol. 28, no. 4, pages 85 - 91 * |
Also Published As
Publication number | Publication date |
---|---|
CN116956642B (en) | 2024-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
CN112686464A (en) | Short-term wind power prediction method and device | |
CN114297036B (en) | Data processing method, device, electronic equipment and readable storage medium | |
CN109670714B (en) | Ship gas turbine comprehensive state evaluation method based on membership degree analysis | |
CN110427311A (en) | Disk failure prediction technique and system based on temporal aspect processing and model optimization | |
CN108664690A (en) | Long-life electron device reliability lifetime estimation method under more stress based on depth belief network | |
CN111143981B (en) | Virtual test model verification system and method | |
CN105760649A (en) | Big-data-oriented creditability measuring method | |
CN110909488A (en) | High-efficiency slope reliability analysis method | |
CN109981749A (en) | A kind of cloud workflow task running time prediction method promoted based on limit gradient | |
CN108879732A (en) | Transient stability evaluation in power system method and device | |
CN113411821B (en) | System reconfiguration capability evaluation method and system for complex network | |
CN115374995A (en) | Distributed photovoltaic and small wind power station power prediction method | |
CN114444910A (en) | Electric power Internet of things-oriented edge network system health degree evaluation method | |
CN110889207A (en) | System combination model credibility intelligent evaluation method based on deep learning | |
CN116956642B (en) | Threat assessment algorithm performance dynamic simulation verification method | |
CN115062551B (en) | Wet physical process parameterization method based on time sequence neural network | |
CN116842459A (en) | Electric energy metering fault diagnosis method and diagnosis terminal based on small sample learning | |
CN115618743B (en) | State evaluation method and state evaluation system of sighting telescope system | |
CN116431346A (en) | Compensation method for main memory capacity of electronic equipment | |
CN115184055A (en) | Method and system for determining test set with optimized hierarchical testability | |
CN115165332A (en) | Integrated design method and system for built-in test and comprehensive test of equipment | |
CN112348700B (en) | Line capacity prediction method combining SOM clustering and IFOU equation | |
CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
CN114240069A (en) | Power grid online risk calculation method based on regression learning and feature mining |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |