基于遗传算法的土壤墒情传感器优化布局策略

Layout optimization policy of soil moisture sensors with genetic algorithm

针对用于节水灌溉的土壤墒情传感器布局问题,该文提出一种结合了基于全局优化遗传算法和改进的加权圆集布局理论的人机交互优化布局策略。该策略综合考虑了传感器覆盖精度、重叠限制等约束条件。通过在人机交互中引入专家知识和在遗传算法中采用十进制编码,策略易于与其它算法和附加参数协作以进行升级或移植于其它的应用场合。在给定传感器成本的仿真实验中,该文算法比四边形方案节约成本17.5%,比六边形方案节约成本34.0%。

To optimize the layout of soil moisture sensors in water-saving irrigation system,we proposed an interactive policy which was combined with global optimization genetic algorithms and improved weighted round layout theory.This approach was designed based on constraints of sensors performance such as the accuracy of coverage and overlay restrictions.With the help of bringing expert knowledge into the interaction and adopting decimal codes within the genetic algorithm,this interactive policy was easy to be updated and transplanted in other areas by cooperating with other algorithms or by adding extra parameters.The experimental results indicated the optimal approach proposed in this paper decreased the cost by 17.5% than quadrangle layout approach and by 34.0% than hexagon layout approach.