摘要
Abstract
The increasing scale and complexity of complex systems have raised the challenge of multi-objective optimization in the design and implementation of national large-scale projects and complex military systems.In this paper,a multi-objective optimization method for complex systems based on knowledge representation is proposed,which portrays the association relationship between the elements of complex systems by constructing a knowledge map,and combines the knowledge representation learning and optimization algorithms to achieve a more scientific and accurate multi-objective optimization of complex systems.The method includes four key steps:ontology design,graph construction,knowledge representation and optimization solution.The experimental results show that the method can quickly generate a multi-objective optimization scheme that meets the constraints,and the optimization scheme is reasonable and balanced,avoiding extreme configuration,and realizing the effective balance of different objectives by adjusting the weight coefficients,which can provide an auxiliary support for the decision-making of system design.关键词
复杂体系/知识图谱/知识表示/多目标优化/TransE算法Key words
complex system/knowledge graph/knowledge representation/multi-objective optimization/TransE algorithm分类
信息技术与安全科学
