一种新型联合训练仿真评估框架及其关键技术研究

doi:10.16182/j.issn1004731x.joss.24-0488

系统仿真学报 ›› 2025, Vol. 37 ›› Issue (10): 2557-2567.doi: 10.16182/j.issn1004731x.joss.24-0488

• 论文 • 上一篇

一种新型联合训练仿真评估框架及其关键技术研究

鞠儒生, 陈栋栋, 曾云秀, 刘吉元, 邱思航, 周鹏

国防科技大学系统工程学院，湖南长沙 410073

收稿日期:2024-05-07 修回日期:2024-07-01 出版日期:2025-10-20 发布日期:2025-10-21
通讯作者: 陈栋栋
第一作者简介:鞠儒生(1976-)，男，研究员，博士，研究方向为分布式仿真测试与评估、智能控制及其应用。
基金资助:
国家自然科学基金(2206306);国家自然科学基金(62202477);国家自然科学基金(62306324)

A Novel Joint Training Simulation Evaluation Framework and Its Key Techniques

Ju Rusheng, Chen Dongdong, Zeng Yunxiu, Liu Jiyuan, Qiu Sihang, Zhou Peng

College of Systems Engineering, National University of Defense Technology, Changsha 410073, China

Received:2024-05-07 Revised:2024-07-01 Online:2025-10-20 Published:2025-10-21
Contact: Chen Dongdong

摘要/Abstract

摘要：

为解决传统评估系统内在模块相互耦合、无法重复利用、难以适应多领域评估需求的难题，设计了一个“数据+服务+应用”三层解耦的评估技术框架。提出基于关键事件对音视频数据进行切片并抽取高价值信息的策略，解决了海量非结构化评估数据加工问题；给出通用与专用相结合的评估模型模板设计 方法，有效应对评估模型的普适性应用难题；构建了专家知识驱动的定性定量相结合的综合集成研讨评估环境，方便不同领域专家从不同维度对关键评估指标进行客观评判，提升了关键指标评估的全面性与客观性。基于技术框架设计开发联合训练仿真评估原型系统，验证了方法的有效性。

关键词: 评估框架, 评估模型模板, 关键事件, 定性定量, 综合集成研讨评估

Abstract:

To address the challenges of traditional evaluation systems, such as internal module coupling, lack of reusability, and poor adaptability to multi-domain evaluation needs, a three-tier decoupled technical evaluation framework of "data + service + application" was designed. A strategy was proposed for extracting high-value information from massive audio and video data based on key events, resolving the problem of unstructured evaluation data processing. A design method combining general and dedicated evaluation model templates was proposed, improving the general applicability of the evaluation model. An expert knowledge-driven comprehensive integrated discussion and evaluation environment was constructed using qualitative and quantitative methods, which allowed experts from different domains to objectively assess key evaluation indicators from different dimensions, enhancing both the comprehensiveness and objectivity of the evaluation. A joint training simulation prototype system, developed based on the three-tier technical framework, verified the effectiveness of the proposed method.

Key words: evaluation framework, evaluation model template, key event, qualitative and quantitative method, comprehensive integrated discussion and evaluation

中图分类号:

TP391.9

鞠儒生,陈栋栋,曾云秀等 . 一种新型联合训练仿真评估框架及其关键技术研究[J]. 系统仿真学报, 2025, 37(10): 2557-2567.

Ju Rusheng,Chen Dongdong,Zeng Yunxiu,et al . A Novel Joint Training Simulation Evaluation Framework and Its Key Techniques[J]. Journal of System Simulation, 2025, 37(10): 2557-2567.

图/表 8

图1

图2

图3

图4

表1

表2

图5

表3

参考文献 25

[1]	胡晓峰, 司光亚, 罗批, 等. 战争复杂系统与战争模拟研究[J]. 系统仿真学报, 2005, 17(11): 2769-2774.
	Hu Xiaofeng, Si Guangya, Luo Pi, et al. Study on War Complex System and War Gaming & Simulation[J]. Journal of System Simulation, 2005, 17(11): 2769-2774.
[2]	邹静, 高智慧, 杨巍. 大数据处理技术在现代战争中的应用[C]//第十届中国指挥控制大会论文集(上册)会议论文集. 北京: 中国指挥与控制学会, 2022: 871-873.
[3]	Birney E. The Making of ENCODE: Lessons for Big-data Projects[J]. Nature, 2012, 489(7414): 49-51.
[4]	张道海, 袁雪梅, 李丹丹, 等. 大数据处理技术及案例应用[M]. 北京: 机械工业出版社, 2022.
	Zhang Daohai, Yuan Xuemei, Li Dandan, et al. Big Data Processing Technology and Case Application[M]. Beijing: China Machine Press, 2022.
[5]	刘伟, 朱雅璇. 数据挖掘在军事研究中的应用[C]//中国国防科学技术信息学会第十三届学术年会. 北京: 国防工业出版社, 2014: 27.
[6]	李德伟, 黄高明. 多源异构海量训练数据实时处理技术研究[J]. 计算机与网络, 2019, 45(11): 58-61.
	Li Dewei, Huang Gaoming. Research on Real-time Processing Technology of Multi-source Heterogeneous Massive Training Data[J]. Computer and Network, 2019, 45(11): 58-61.
[7]	王佩骐, 鞠儒生, 张淼, 等. 基于非支配遗传算法的HLA仿真系统数据采集策略[J]. 系统工程与电子技术, 2024, 46(9): 3103-3111.
	Wang Peiqi, Ju Rusheng, Zhang Miao, et al. Data Collection Strategy of HLA Simulation System Based on Non-dominated Genetic Algorithm[J]. Systems Engineering and Electronics, 2024, 46(9): 3103-3111.
[8]	Hou Benzhong, Zhang Yongqiang, Shang Ying, et al. Research on Unstructured Data Processing Technology in Executing Audit Based on Big Data Budget[J]. Journal of Physics: Conference Series, 2020, 1650(3): 032100.
[9]	王磊, 陈磊, 张明儒, 等. 面向数据库查询的非结构化数据融合存储系统[J]. 电子设计工程, 2022, 30(24): 148-152.
	Wang Lei, Chen Lei, Zhang Mingru, et al. Unstructured Data Fusion Storage System for Database Query[J]. Electronic Design Engineering, 2022, 30(24): 148-152.
[10]	吴杰, 孙加森, 储军飞, 等. 数据包络分析交叉效率评价理论、方法及应用[M]. 北京: 科学出版社, 2023.
[11]	王继武. 新版失效模式及影响分析(FMEA)实施指南[M]. 北京: 中国铁道出版社有限公司, 2022.
[12]	张权, 毕于慧, 刘茂林. 多层次灰色评价法在军事数据质量评价中的应用[J]. 计算机系统应用, 2017, 26(4): 280-283.
	Zhang Quan, Bi Yuhui, Liu Maolin. Application of Multilevel Grey Evaluation Method in the Military Data Quality Evaluation[J]. Computer Systems & Applications, 2017, 26(4): 280-283.
[13]	张超, 马存宝, 胡云兰, 等. 基于多指标决策的武器装备作战效能综合评估[J]. 兵工学报, 2006, 27(6): 1081-1085.
	Zhang Chao, Ma Cunbao, Hu Yunlan, et al. Composite Evaluation of Operational Effectiveness for Weapon Equipment Based on Multiple Criterion Decision Making[J]. Acta Armamentarii, 2006, 27(6): 1081-1085.
[14]	贾跃, 赵学涛, 林贤杰, 等. 基于BP神经网络的鱼雷作战效能模糊综合评估模型及其仿真[J]. 兵工学报, 2009, 30(9): 1232-1235.
	Jia Yue, Zhao Xuetao, Lin Xianjie, et al. Fuzzy Multifactorial Evaluation Model of Torpedo Operational Effectiveness Based on BP Neural Network and Its Simulation[J]. Acta Armamentarii, 2009, 30(9): 1232-1235.
[15]	苏丽娜, 李广. 层次分析与模糊综合评判法在军事情报评估的应用[J]. 21世纪(理论实践探索), 2010(3): 84-85.
[16]	易晓宇, 易绵竹. 基于SWOT分析的非结构化数据文本分类算法[J]. 科技创新与应用, 2022, 12(29): 25-28, 33.
	Yi Xiaoyu, Yi Mianzhu. Text Classification Algorithm for Unstructured Data Based on SWOT Analysis[J]. Technology Innovation and Application, 2022, 12(29): 25-28, 33.
[17]	姜滟稳, 江自昊. 基于人工智能算法对非结构化数据使用率的研究与应用[J]. 绵阳师范学院学报, 2023, 42(2): 80-85.
	Jiang Yanwen, Jiang Zihao. Research and Application of Unstructured Data Utilization Based on Artificial Intelligence Algorithm[J]. Journal of Mianyang Teachers' College, 2023, 42(2): 80-85.
[18]	刘祥凯, 张云, 张欢, 等. 视频大数据研究综述[J]. 集成技术, 2016, 5(2): 41-56.
	Liu Xiangkai, Zhang Yun, Zhang Huan, et al. A Review on Video Big Data[J]. Journal of Integration Technology, 2016, 5(2): 41-56.
[19]	钱学明. 基于语义的视频浏览系统中的关键技术研究[D]. 西安: 西安交通大学, 2007.
	Qian Xueming. Research on the Key Technologies of Semantic Based Video Browsing System[D]. Xi'an: Xi'an Jiaotong University, 2007.
[20]	邓正良. 基于FFmpeg和SDL的视频流播放存储研究综述[J]. 现代计算机, 2019(22): 47-50.
	Deng Zhengliang. Summary of Research on Video Streaming Playback and Storage Based on FFmpeg and SDL[J]. Modern Computer, 2019(22): 47-50.
[21]	Xu Yungeng, Cao Sanxing. Real-time Video Acquisition and Frame Compression Processing Technology Based on FFmpeg[J]. Applied Mechanics and Materials, 2014, 631/632: 494-497.
[22]	孟利, 沈郑燕, 张泰雯. 基于FFmpeg提取目标人物语音的应用研究[J]. 信息系统工程, 2023(3): 74-76.
[23]	唐宇枫. 基于FPGA的高速数据采集系统设计[D]. 上海: 华东师范大学, 2022.
	Tang Yufeng. Design of High-speed Data Acquisition System Based on FPGA[D]. Shanghai: East China Normal University, 2022.
[24]	郭晶晶, 刘光尧, 汪磊, 等. FFmpeg在视频图像处理中的应用[J]. 刑事技术, 2020, 45(3): 234-237.
	Guo Jingjing, Liu Guangyao, Wang Lei, et al. FFmpeg Software and Its Applying into Video Image Processing[J]. Forensic Science and Technology, 2020, 45(3): 234-237.
[25]	钱学森, 于景元, 戴汝为. 一个科学新领域—开放的复杂巨系统及其方法论[J]. 自然杂志, 1990, 13(1): 3-10, 64.

姓名	类型	命中率/%
受训者A	射击手	80
受训者B	射击手	70
受训者C	射击手	65
受训者D	射击手	80
受训者E	射击手	60
受训者F	射击手	80
受训者G	射击手	95

姓名	类型	命中率/%	完成任务平均时长/s
受训者A	射击手	80	15
受训者D	射击手	80	18
受训者F	射击手	80	10

专家	专家类别	专家评价意见	改进意见
李××	训练专家	1.整体表现良好，动作熟练规范... 2.存在问题: ...	1.×××××× 2.××××××
汪××	军事专家	××××××	××××××
刘××	装备专家	××××××	××××××
何××	仿真专家	××××××	××××××

一种新型联合训练仿真评估框架及其关键技术研究

A Novel Joint Training Simulation Evaluation Framework and Its Key Techniques

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 25

相关文章 1

编辑推荐

Metrics

本文评价