基于插件技术的人工免疫智能空间优化平台研究

A study on artificial immune computing platform for spatial optimization based on plugin technique

采用人工智能方法求解空间优化问题已成为当前地理学研究领域的热点之一。研究目的在于通过设计一个开放的、可扩展的人工免疫空间优化模型框架,开发基于插件技术的优化工具软件,丰富空间优化决策支持的方法和技术体系。研究从空间优化问题的基本特点和计算需求出发,定义了面向空间优化问题求解的克隆选择优化模型框架。在此基础上,采用"插件-宿主平台"结构设计了人工免疫智能空间优化平台的总体架构,定义免疫算子插件与空间优化问题应用插件开发机制。实验结果表明,平台的架构和功能设计充分考虑了计算平台的开放性和扩展性,通过平台标准接口的功能扩展可实现对不同类型空间优化问题的求解。空间优化问题研究者或政府决策部门无需关系人工免疫优化方法的具体内涵即可基于本平台完成其具体空间优化决策支持任务。

Solving the spatial optimization problems can be a complex and difficult task, since it has to handle some high-dimension, nonlinear, complicated relationships. Many efforts have been made with regard to this specific issue, and the strong ability of artificial immune system algorithms has been proven in previous studies. In this study, a novel framework used for spatial optimization based on clonal selection algorithms, which are the most popular immune algorithms in geoscience, is proposed. Then, the spatial optimization platform was designed based on the architecture of ＂ plugins - platform＂, and some key technologies about the developing of immune operator plugins and spatial optimization application plugins were described. Based on the standard APIs provided by this platform, researchers can develop their own problem-specific application plugins to solve the practical problems or to implement some advanced immune operators into the platform to improve the performance of the algorithm. Finally, the functionality, reusability and extensibility of platform were tested by using the Traveling Salesman Problem as a benchmark testing. Experiments show that, the platform is capable of solving various optimization problems, and it is expected to bridging the gap between the immune algorithm researchers, geographers and decision makers.