基于MAS与PSO算法的土地资源决策优化配置研究

Optimized Allocation of Land Resource Based on MAS and PSO Model

摘要: 土地资源优化配置对实现土地资源可持续利用具有重要意义，其难点是实现不同主体之间利益间的协调统一。本文以社会、经济和生态效益最大化为目标，采取多智能体系统(MAS)与微粒群算法(PSO)相结合方法，综合考量不能智能体之间的相互关系，得出多个土地资源优化配置方案，并对不同方案进行了综合评价。研究结果表明：1) MAS与PSO相结合的方法能够能够将规划期内的不同土地利用类型以多目标约束的方式配置到不同的空间单元格中，从而实现区域土地资源的空间优化配置；2) 采用灰色关联投影法与灰色多目标决策进行优化方案选择，其中方案2的投影值最大，达到0.951，结果证明其为最优方案。

Abstract: The optimal allocation of land resources is very important to achieve sustainable utilization of land resources, but it is very difficult to realize the harmonization of multi-interest between different subjects. This paper takes the maximization of social, economic and ecological benefits as the goal, and adopts the method of multi-agent system (MAS) and particle swarm optimization (PSO), comprehensively considers the interrelationship between agents, and obtains a number of land resources optimization configuration program, and a comprehensive evaluation of different projects. Results show that: 1) The combination of MAS and PSO can be used to allocate different land use types into different spatial units in the multi-objective constraints in the planning period, so as to realize the optimal allocation of regional land resources. 2) Gray-related method and gray multi-objective decision making method are used to evaluate the different projects, and results show that scheme 2 is most valuable solution, and its value reaches 0.951, so it has been proved to be the optimal scheme.