基于异质网络的装备体系组合发展规划问题

doi:10.12305/j.issn.1001-506X.2025.03.17

摘要/Abstract

摘要：

装备体系组合发展规划是一项复杂的系统工程, 具有重要的军事意义和研究价值。考虑到武器装备体系的关联性和复杂性, 从体系的角度出发, 提出一种装备体系组合发展规划方法, 为装备发展论证提供思路。首先，基于装备的不同属性以及关联关系，构建了装备体系异质网络模型并识别提取功能链，为后续装备发展规划建模提供基础；其次，考虑到装备发展的不确定性，建立了多阶段装备发展规划模型；然后，设计一种双重深度Q网络(double deep Q-network, Double DQN)算法来求解模型；最后，以典型装备体系发展规划为例进行演示计算，验证了所提方法的有效性和可行性。

关键词: 装备组合, 异质网络, 深度强化学习, 装备发展规划

Abstract:

Portfolio of weapon system-of-systems(WSoS) is a complex systems engineering task, bearing significant military implications and research value. This paper considers the complexity and the correlations of the WSoS, and from the perspective of the system-of-systems, proposes a WSoS portfolio development method, which provides an idea for the equipment development demonstration. Firstly, based on the interrelationships of various types of weapons, this paper constructs a heterogeneous information network model of weapons and extracts the functional chains, providing a foundation for subsequent models. Secondly, considering the uncertainties in portfolio of WSoS, a multi-stage portfolio of WSoS model is established. Then, a double deep Q-network (Double DQN) algorithm is designed to solve the model. Finally, taking an example of a WSoS development planning, a demonstration calculation is carried out to verify the effectiveness and feasibility of the proposed method.

Key words: weapon portfolio, heterogeneous information network, deep reinforcement learning, weapon development planning

中图分类号:

TP391.9

陈夏瑢, 李际超, 陈刚, 刘鹏, 姜江. 基于异质网络的装备体系组合发展规划问题[J]. 系统工程与电子技术, 2025, 47(3): 855-861.

Xiarong CHEN, Jichao LI, Gang CHEN, Peng LIU, Jiang JIANG. Portfolio of weapon system-of-systems based on heterogeneous information networks[J]. Systems Engineering and Electronics, 2025, 47(3): 855-861.

图/表 11

图1

表1

图2

表2

表3

表4

图3

图4

表5

图5

图6

参考文献 35

1	肖文健, 王振兴, 王彦斌, 等. 基于仿真推演的光电对抗装备体系贡献率评估[J]. 航空兵器, 2022, 29 (1): 84- 89.
	XIAO W J , WANG Z X , WANG Y B , et al. Contribution eva-luation of optoelectronic countermeasure equipment based on simulation[J]. Aero Weaponry, 2022, 29 (1): 84- 89.
2	KANGASPUNTA J , LIESIOJ S . A cost-efficiency analysis of weapon system portfolios[J]. European Journal of Operational Research, 2012, 223 (1): 264- 275. doi: 10.1016/j.ejor.2012.05.042
3	DOU Y J , ZHANG P L , GE B F , et al. An integrated technology pushing and requirement pulling model for weapon system portfolio selection in defenses acquisition and manufacturing[J]. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 2015, 229 (6): 1046- 1067. doi: 10.1177/0954405414534640
4	ZHOU Z X, DOU Y J, XIA B Y, et al. Weapon systems portfolio selection based on fuzzy clustering analysis[C]//Proc. of the 3rd IEEE International Conference on Control Science and Systems Engineering, 2017: 702-705.
5	LEE J , KANG S H , ROSENBERGER J . A hybrid approach of goal programming for weapon systems selection[J]. Computers & Industrial Engineering, 2010, 58 (3): 521- 527.
6	DOU Y J , HANG P L , JING J , et al. Comparisons of hybrid multi-objective programming algorithms with grey target and PCA for weapon system portfolio selection[J]. Applied Mathematics & Information Sciences, 2014, 8 (3): 1389- 1399.
7	DOU Y J , ZHAO D L , XIA B Y , et al. System portfolio selection for large-scale complex systems construction[J]. IEEE Systems Journal, 2019, 13 (4): 3627- 3638. doi: 10.1109/JSYST.2019.2912409
8	LI J C , GE B F , ZHAO D L , et al. Meta-path-based weapon-target recommendation in heterogeneous combat network[J]. IEEE Systems Journal, 2019, 13 (4): 4433- 4441. doi: 10.1109/JSYST.2018.2890090
9	LI J C , GE B F , JIANG J , et al. High-end weapon equipment portfolio selection based on a heterogeneous network model[J]. Journal of Global Optimization, 2020, 78 (4): 743- 761. doi: 10.1007/s10898-018-0687-1
10	李瑞阳, 何明, 王智学, 等. 基于改进模糊神经推理的武器装备系统组合决策[J]. 计算机应用研究, 2020, 37 (11): 3241- 3245.
	LI R Y , HE M , WANG Z X , et al. Portfolio decision of wea-pon system based on improved fuzzy neural inference[J]. Application Research of Computers, 2020, 37 (11): 3241- 3245.
11	LIU P , LI J C , XIA B Y , et al. Weapons equipment portfolios selection based on equipment system contribution rates[J]. Journal of Systems Engineering and Electronics, 2021, 32 (3): 584- 595. doi: 10.23919/JSEE.2021.000050
12	DELLER S. Towards the quantitative analysis of the connecti-vity value of networked operations[D]. Norfolk: Old Dominion University, 2009.
13	DELLER S , RABADI G , TOLK A , et al. Organizing for improved effectiveness in networked operations[J]. Military Ope-rations Research, 2012, 17 (1): 5- 16. doi: 10.5711/1082598317105
14	DEKKER A H. Network topology and military performance[C]//Proc. of the Modsim International Congress on Modelling & Simulation, 2008.
15	DEKKER A H. Analyzing C2 structures and self-synchronization with simple computational models[C]//Proc. of the 16th International Command and Control Research and Technology Symposium, 2011.
16	周琛, 宋笔锋, 尚柏林, 等. 基于作战网络可靠度的体系贡献率评估[J]. 系统工程与电子技术, 2021, 43 (7): 1875- 1883. doi: 10.12305/j.issn.1001-506X.2021.07.19
	ZHOU C , SONG B F , SHANG B L , et al. System of systems contribution rate evaluation based on operational network reliability[J]. Systems Engineering and Electronics, 2021, 43 (7): 1875- 1883. doi: 10.12305/j.issn.1001-506X.2021.07.19
17	LI J C , JIANG J , YANG K W , et al. Research on functional robustness of heterogeneous combat networks[J]. IEEE Systems Journal, 2019, 13 (2): 1487- 1495. doi: 10.1109/JSYST.2018.2828779
18	WU W L , ZHOU X S , SHEN B . Comprehensive evaluation of the intelligence levels for unmanned swarms based on the collective OODA loop and group extension cloud model[J]. Connection Science, 2022, 34 (1): 630- 651. doi: 10.1080/09540091.2022.2026293
19	LI X, LIU D S. Modeling method of combat mission based on OODA loop[C]//Proc. of the International Conference on Physics, Computing and Mathematical, 2022, 355: 2-15.
20	罗弋洋, 赵青松, 李华超, 等. 武器装备运用知识框架及建模方法[J]. 系统工程与电子技术, 2022, 44 (3): 841- 849.
	LUO Y Y , ZHAO Q S , LI H C , et al. Framework and modeling method of weaponry utilization knowledge[J]. Systems Engineering and Electronics, 2022, 44 (3): 841- 849.
21	鲁统亮, 陈文豪, 葛冰峰, 等. 信息支援下的作战体系多层网络建模[J]. 系统工程与电子技术, 2022, 44 (2): 520- 528.
	LU T L , CHEN W H , GE B F , et al. Multi-layer network modeling for combat system-of-systems under information support[J]. Systems Engineering and Electronics, 2022, 44 (2): 520- 528.
22	HU J C , XING Y H , HAN M , et al. Nonnegative matrix tri-factorization based clustering in a heterogeneous information network with star network schema[J]. Tsinghua Science and Technology, 2022, 27 (2): 386- 395. doi: 10.26599/TST.2020.9010049
23	HU Z H , ZHANG R , LI X C , et al. Conflict detection in task heterogeneous information networks[J]. Web Intelligence, 2021, 21 (22): 7756.
24	HAO P , YANG R Y , WANG Z , et al. Lime: low-cost and incremental learning for dynamic heterogeneous information networks[J]. IEEE Trans.on Computers, 2022, 71 (3): 628- 642. doi: 10.1109/TC.2021.3057082
25	LI H , WANG Y L , LYU Z Y , et al. Multi-task learning for recommendation over heterogeneous information network[J]. IEEE Trans.on Knowledge and Data Engineering, 2022, 34 (2): 789- 802. doi: 10.1109/TKDE.2020.2983409
26	WANG C , WANG J J , CHEN W L , et al. Assessment and feedback control of paving quality of earth-rock dam based on OODA loop[J]. Sensors, 2021, 21 (22): 7756. doi: 10.3390/s21227756
27	DEREN A . The OODA loop as an instrument for developing intellectual property strategies in sems[J]. Humanities and Social Sciences, 2021, 28 (2): 7- 16.
28	MA Y , CHANG T Q , FAN W H , et al. Time management algorithm of distributed combat simulation based on OODA loop[J]. Application of Electronic Technique, 2021, 47 (5): 86- 91.
29	MILLER B , HASBROUCK S , UDREA B . Development of human-machine collaborative systems through use of observe- orient-decide-act (OODA) loop[J]. Ascend, 2021, 2021, 4092.
30	SIMOHAMMED A , SMAIL R . A decision loop for situation risk assessment under uncertainty: a case study of a gas facility[J]. Petroleum, 2021, 7 (3): 343- 348. doi: 10.1016/j.petlm.2021.01.004
31	WEN S H , LV X H , LAM H K , et al. Probability double DQN active visual SLAM for autonomous navigation in indoor environment[J]. Industrial Robot, 2021, 48 (3): 359- 365. doi: 10.1108/IR-08-2020-0160
32	LIU Z C, HU J, SONG T C, et al. A methodology based on deep reinforcement learning to autonomous driving with double Q-learning[C]//Proc. of the 7th International Conference on Computer and Communications, 2021.
33	HASSELT H V. Double Q-learning[C]//Proc. of the 24th Annual Conference on Neural Information Processing Systems, 2010.
34	CAO J W , ZHAO S C , GUO C Q , et al. Study on the path planning algorithm based on dueling deep Q-network[J]. Journal of Physics: Conference Series, 2021, 1920 (1): 012084. doi: 10.1088/1742-6596/1920/1/012084
35	LI J X , CHEN Y T , ZHAO X N , et al. An improved DQN path planning algorithm[J]. Journal of Super Computing, 2022, 78 (1): 616- 639. doi: 10.1007/s11227-021-03878-2?utm_source=xmol&utm_content=meta

装备类型	装备主要功能
侦察监视预警类装备	有目标侦察、情报获取和战场监视功能
通信与指挥控制类装备	对干扰类实体进行信息处理与分析, 协助决策, 执行命令与控制功能
联合火力打击和干扰类装备	精确打击、火力毁伤和电子干扰功能

变量	符号定义	说明
装备数量	n∈Ν⁺	待发展的装备总量
发展成本	$E=\left[e_i\right]_{1 \times n}, e_i \in R^{+}$	每个待发展装备的成本
预计发展时间	$\mathrm{PY}=\left[\mathrm{py}_i\right]_{1 \times n}, \mathrm{py}_i \in N^{+}$	根据装备属性和技术条件确定的装备预计发展时间, 以年为单位
已发展年限	$\mathrm{AY}=\left[\mathrm{ay}_i\right]_{1 \times n}, \mathrm{ay}_i \in N^{+}$	装备组合发展规划中已发展时间
发展情况	$S=\left[s_i\right]_{1 \times n}, s_i \in\{0, 1\}$	装备最终发展成功与否
预期能力值	$C=\left[c_{i j}\right]_{m \times n}, c_{i j} \in N^{+}, \in[1, 9]$	根据装备属性获取的统一量纲后的装备能力值
发展结束后的能力	$\mathrm{AC}=\left[\mathrm{ac}_i\right]_{1 \times_n}, \mathrm{ac}_i \in N^{+}, \in[0, 9]$	装备发展成功则为预期能力值, 否则为0
发展阶段	$\operatorname{stage}_i, i=1, 2, \cdots, t \quad t \in N^{+}, >1$	装备多阶段发展规划的不同阶段
每阶段投资	$\mathrm{SI}=\left[\mathrm{si}_i\right]_{1 \times t}, \mathrm{si}_i \in R^{+}$	每阶段的投资总金额
发展方案	$X=\left[x_{i j}\right]_{n \times t}, x_{i j} \in\{0, 1\}$	多阶段装备发展规划方案

装备属性	发展成本/千万元	预计发展年限/年	发展成功后的能力值
S1	31	4	9
S2	40	1	6
S3	66	3	4
D1	59	4	4
D2	51	1	8
D3	42	4	7
I1	50	4	6
I2	60	1	3
I3	65	2	8
I4	75	3	9

功能链序号	装备组成			功能链能力值
功能链序号	侦察监视预警类装备	通信与指挥控制决策类装备	联合火力打击和干扰类装备	功能链能力值
1	S1	D1	I1	11
2	S1	D1	I2	13
3	S1	D2	I1	15
4	S1	D2	I2	16
5	S2	D1	I1	18
6	S2	D1	I2	18
7	S2	D2	I1	19
8	S2	D2	I2	19
9	S3	D1	I1	21
10	S3	D1	I2	21
11	S3	D2	I1	22
12	S3	D2	I2	22

[1]	唐开强, 傅汇乔, 刘佳生, 邓归洲, 陈春林. 基于深度强化学习的带约束车辆路径分层优化研究[J]. 系统工程与电子技术, 2025, 47(3): 827-841.
[2]	张庭瑜, 曾颖, 李楠, 黄洪钟. 基于深度强化学习的航天器功率-信号复合网络优化算法[J]. 系统工程与电子技术, 2024, 46(9): 3060-3069.
[3]	夏雨奇, 黄炎焱, 陈恰. 基于深度Q网络的无人车侦察路径规划[J]. 系统工程与电子技术, 2024, 46(9): 3070-3081.
[4]	杨志鹏, 陈子浩, 曾长, 林松, 毛金娣, 张凯. 复杂环境下的飞行器在线航路规划决策方法[J]. 系统工程与电子技术, 2024, 46(9): 3166-3175.
[5]	郭宏达, 娄静涛, 徐友春, 叶鹏, 李永乐, 陈晋生. 基于MADDPG的多无人车协同事件触发通信[J]. 系统工程与电子技术, 2024, 46(7): 2525-2533.
[6]	张梦钰, 豆亚杰, 陈子夷, 姜江, 杨克巍, 葛冰峰. 深度强化学习及其在军事领域中的应用综述[J]. 系统工程与电子技术, 2024, 46(4): 1297-1308.
[7]	李彦铃, 罗飞舟, 葛致磊. 基于鲁棒观测器的深度强化学习垂直起降运载器姿态稳定研究[J]. 系统工程与电子技术, 2024, 46(3): 1038-1047.
[8]	吴冯国, 陶伟, 李辉, 张建伟, 郑成辰. 基于深度强化学习算法的无人机智能规避决策[J]. 系统工程与电子技术, 2023, 45(6): 1702-1711.
[9]	唐进, 梁彦刚, 白志会, 黎克波. 基于DQN的旋翼无人机着陆控制算法[J]. 系统工程与电子技术, 2023, 45(5): 1451-1460.
[10]	唐斯琪, 潘志松, 胡谷雨, 吴炀, 李云波. 深度强化学习在天基信息网络中的应用——现状与前景[J]. 系统工程与电子技术, 2023, 45(3): 886-901.
[11]	陈子夷, 豆亚杰, 徐向前, 谭跃进, 杨克巍, 姜江. 共建共享双层策略驱动的复杂装备组合优化求解[J]. 系统工程与电子技术, 2023, 45(2): 431-443.
[12]	李信, 李勇军, 赵尚弘. 基于深度强化学习的卫星光网络波长路由算法[J]. 系统工程与电子技术, 2023, 45(1): 264-270.
[13]	王冠, 茹海忠, 张大力, 马广程, 夏红伟. 弹性高超声速飞行器智能控制系统设计[J]. 系统工程与电子技术, 2022, 44(7): 2276-2285.
[14]	孟泠宇, 郭秉礼, 杨雯, 张欣伟, 赵柞青, 黄善国. 基于深度强化学习的网络路由优化方法[J]. 系统工程与电子技术, 2022, 44(7): 2311-2318.
[15]	杨清清, 高盈盈, 郭玙, 夏博远, 杨克巍. 基于深度强化学习的海战场目标搜寻路径规划[J]. 系统工程与电子技术, 2022, 44(11): 3486-3495.