System Performance Evaluation Method Based on Multi-source Prior Data

doi:10.16182/j.issn1004731x.joss.21-FZ0713

Abstract

Abstract: When using the Bayes method to evaluate the performance of the system with multi-source prior data, the multi-source prior data is fused, the posterior distribution is calculated by synthesizing the fused prior distribution and test data. The parameters of posterior distribution are estimated to obtain the performance evaluation results. A weighted fusion method of multi-source prior data based on Kullback-Leibler divergence is proposed, which can effectively integrate the multi-source prior data. The commonly used Markov Chain Monte Carlo method is used to estimate the parameters of Bayes posterior distribution. The influence of different proposal distributions on the sampling results is compared, and an adaptive construction method for low-dimensional proposal distributions is proposed, which can effectively select a proper proposal distribution and improve the efficiency of sampling.

Key words: performance evaluation, bayes theory, multi-source, prior distribution, Markov Chain Monte Carlo method

CLC Number:

TP302.7

Liu Haozhe, Li Wei, Ma Ping, Yang Ming. System Performance Evaluation Method Based on Multi-source Prior Data[J]. Journal of System Simulation, 2021, 33(11): 2673-2680.

References

[1] 吴琳丽, 潘光, 黄桥高. 基于ML-II的小子样复杂系统多源信息融合方法[J]. 计算机工程与应用, 2014, 50(15): 220-222.
Wu Linli, Pan Guang, Huang Qianggao.Method of Multiple Sources Information Fusion for Complex System with Small Sample Test Based on ML-II Theory[J]. Computer Engineering and Applications, 2014, 50(15): 220-222.
[2] 陈飞, 王鑫. 数据融合技术在导弹武器系统抗干扰性能评估中的应用[J]. 航天电子对抗, 2017, 33(3): 1-4.
Chen Fei, Wang Xin.Application of Data Fusion in Anti-jamming Performance Evaluation of Missile Weapon Equipment System[J]. Aerospace Electronic Warfare, 2017, 33(3): 1-4.
[3] 闫志强. 装备试验评估中的变动统计方法研究[D]. 长沙: 国防科学技术大学, 2010.
Yan Zhiqiang.Statistical Inference for Dynamic Population in Equipments Test Evaluation[D]. Changsha: National University of Defense Technology, 2010.
[4] Roberts G O, Gelman A, Gilks W R.Weak Convergence and Optimal Scaling of Random Walk Metropolis Algorithms[J]. Annals of Applied Probability (S1050-5164), 1997, 7(1): 110-120.
[5] Roberts G O, Rosenthal J S.Optimal Scaling for Various Metropolis-Hastings Algorithms[J]. Statistical Science (S0883-4237), 2001, 16(4): 351-367.
[6] Papaioannou I, Betz W, Zwirglmaier K, et al.MCMC Algorithms for Subset Simulation[J]. Probabilistic Engineering Mechanics (S0266-8920), 2015, 41(Jul.): 89-103.
[7] 陈璇, 李奇, 李博文, 等. 基于多源数据可信度的制导精度融合评估[J]. 空天防御, 2020, 3(4): 14-20.
Chen Xuan, Li Qi, Li Bowen, et al.Guidance Precision Fusion Evaluation Based on Multi-Source Data's Credibility[J]. Air & Space Defense, 2020, 3(4): 14-20.
[8] 刘琦, 冯静, 周经伦. 基于专家信息的先验分布融合方法[J]. 中国空间科学技术, 2004(3): 68-71.
Liu Qi, Feng Jing, Zhou Jinglun.The Fusion Method for Prior Distribution Based on Expert's Information[J]. Chinese Space Science and Technology, 2004(3): 68-71.
[9] 孙锦, 李国林, 许诚. 基于仿真信息的反舰导弹小子样靶场试验综合评定方法[J]. 系统工程理论与实践, 2014, 34(4): 1077-1088.
Sun Jin, Li Guolin, Xu Cheng.Synthetical Evaluation Method of Anti-ship Missile in Range Trial in Condition of Small Sample Based on Simulation Information[J]. Systems Engineering-Theory & Parctice, 2014, 34(4): 1077-1088
[10] Kullback S, Leibler R.On Information and Sufficiency[J]. Annals of Mathematical Statistics (S0003-4851), 1951, 22: 79-86.
[11] Rosenbluth A, Rosenbluth M, et al.Equation of State Calculations by Fast Computing Machines[J]. The Journal of Chemical Physics (S0021-9606), 1953, 21(6): 1087-1092.
[12] 陈平, 徐若曦. Metropolis-Hastings自适应算法及其应用[J]. 系统工程理论与实践, 2008(1): 100-108.
Chen Ping, Xu Ruoxi.Metropolis-Hastings Adaptive Algorithm and Its Application[J]. System Engineering Theory and Practice, 2008(1): 100-108.