一種基于ML-PMRF的復(fù)雜仿真系統(tǒng)可信度智能分配方法

打開文本圖片集

中圖分類號：TP391.9 文獻(xiàn)標(biāo)志碼：A DOI：10.12305/j.issn.1001-506X.2025.05.14

Abstract：In order to ensure that the complex simulation system can meet the credibility requirements，the credibility of each simulation sub-system should be determined at the beginning of the construction of the complex simulation system to shorten the development cycle.Therefore，a complex simulation system intellgent credibilityalocation method is proposed，whichcanobtain thecredibilityallocationresultsof each simulation sub-system under the condition of the known whole credibilityof the complex simulation system. According to thecomposition and structure of the complex simulation system，the credibility allocation model of thecomplex simulation system is proposed based on the multi-layer pairwise Markov random field（MLPMRF）.Based on the maximum a posteriori inference and the discrete glowworm swarm optimization，an intellgent inference method forML-PMRF is proposed.The effectivenessand rationalityof the proposed method are verified by practical examples and comparative experiments.

Keywords：complex simulation system；credibility allocation；multi-layer pairwise markov random field （ML-PMRF）；intelligence inference

0 引言

第3種認(rèn)識、改造客觀世界的重要手段[1-2]。（剩余19352字）