【作者】 高盼盼； 郄志红； 孔祥斌； 吴鑫淼； 李楠楠； 王建；

【Author】 GAO Pan-pan~1,QIE Zhi-hong~1,KONG Xiang-bin~2,WU Xin-miao~1,LI Nan-nan~1,WANG Jian~3 (1.College of Urban and Rural Construction,Agricultural University of Hebei,Baoding 071001,China; 2.College of Resources and Environmental Science,China Agricultural University,Beijing 100083,China; 3.Xinjiang Survey and Design Institute of Water Resources and Hydropower Ministry of Water Resource, Wulumuqi 830000,China)

【机构】 河北农业大学 城乡建设学院； 中国农业大学 资源与环境学院； 新疆维吾尔自治区水利水电勘测设计研究院 石河子分院；

【摘要】 针对土壤养分含量变化的随机性和实际监测数据少的问题,提出一种用于区域土壤养分含量动态变化模拟预测的Markov链模型。将每种养分含量水平划分为4个级别,采用灰色聚类方法建立样本分级的模糊隶属函数和分级模型,并针对土壤养分监测次数少的问题,提出基于混合微粒群优化(HPSO)求解转移矩阵的新方法。在河北省某县进行模拟预测,预测结果与实际变化规律和趋势吻合,预测的绝对误差较小。更多还原

【Abstract】 Aiming at the problems of random change of soil nutrient content and limited monitoring data in practice,a fuzzy Markov chain model was proposed to simulate and forecast the dynamic change of the region soil nutrient content. The content of soil nutrient was divided into four levels by using grey clustering method in order to establish the grading model and fuzzy membership function of the sample classification.In the view of limited monitoring data of soil nutrient, a new method based on hybrid particle swarm optimization(HPSO) was proposed to optimize the transition matrix.A county in Hebei province was taken as sample area,the simulation and forecast of dynamic change of soil nutrient was achieved.The analysis results show that the simulation suits well with the practical change trend,and the absolute error of simulation is relatively small.更多还原