投影寻踪模型在水稻灌溉制度评价中的应用

优化水稻灌溉制度是解决粮食安全与水资源短缺两大紧迫问题的关键。从水稻灌溉制度评价方法研究出发,针对传统的水稻灌溉制度评价中指标权重赋值客观性差的问题,提出了基于粒子群优化投影寻踪模型的水稻灌溉制度评价方法,该方法通过对水稻主要生育阶段的多个状态变量建立投影寻踪模型并优化投影方向,从而根据状态变量在最佳投影方向线型投影得到的投影值大小进行水稻灌溉制度的综合评价,避免了专家赋权的人为干扰。选取水稻分蘖及灌浆2个重要生育阶段,分别建立模型并对4种灌溉模式进行综合评价,其评判结果与实际相符,验证了该方法的准确性与可行性。

The Application of Projection Pursuit Model in the Evaluation of Rice Irrigation System

Optimizing the rice irrigation system is the key to solve the two pressing problems of the food security and the water shortage.In this paper, starting from the evaluation method of rice irrigation system research, in order to solve the problem of poor objectivity of index weight assignment for traditional rice irrigation system evaluation, the projection pursuit model optimized by particle swarm optimization algorithm is proposed and applied to evaluate the rice irrigation system.In this method, the projection pursuit model is set up with a plurality of state variables, the projection direction is optimized, and the size of the projected value is compared to evaluate the rice irrigation system to avoid human disturbance.This paper selects two important reproductive stages, the rice tillering and grouting, respectively to set up a model and evaluate four irrigation modes.The evaluation result is consistent with the actual situation, which verifying the feasibility and accuracy of the method.