多智能体系统支持下的耕地非农化流转决策优化研究

Decision Optimization on Non-Agriculturalization of Arable Land Based on Multi-Agent System Model

耕地非农化是全球十分关注的一个重大问题,提升其流转的决策水平是解决该问题的有效手段。然而不同主体之间的竞争性、互动性及复杂性,成为制约耕地非农化流转决策技术推广应用的主要瓶颈。本文拟建立基于决策主体交互影响的耕地非农化流转时空决策模型,结合多智能体系统(multi-agents system, MAS)、微粒群优化算法(particle swarm optimization, PSO)在时间和空间上对耕地非农化流转进行空间决策研究,为区域耕地资源有效利用提供技术支持。研究结果表明:考虑政府、土地开发企业和农户的MASPSO决策模型,能合理地对区域耕地流转过程中的土地利用结构和空间布局进行合理配置,不仅能满足区域耕地量的要求,又能提高耕地质量,以实现资源节约与环境友好的可持续发展目标。

Non-agriculturalization of arable land is a major problem which is concerned by all over the world. Key to solve this problem is to improve its decision level of cultivated land transfer. However, there are a lot of bottlenecks which restrict the application of decision making technique including competition, interaction and complexity between different decision makers. This paper establishes space decisions model based on multi-agents system and particle swarm optimization in time and space. Research shows that: decision-making model of farmland non-agriculturalization considered the factors of government, land development enterprises and farmers, and it not only can rationally allocate the land structure and spatial layout in the process of regional arable land circulation, but also can meet the requirements of regional arable land farmland quality. And the model can achieve the goals of resource conservation and environment-friendly sustainable development.