农业气象灾害等级优化预测仿真研究

发布时间：2018-06-23 10:30

本文选题：农业气象灾害 + 等级预测　；参考：《计算机仿真》2017年06期

【摘要】：对农业气象灾害等级进行准确预测,可对灾害进行有效防御,提高农业生产量。进行灾害等级预测过程应建立在利用农业气象状态观测矩阵,计算农业气象灾害预测加权系数的基础上,而传统方法通过获取灾害关键属性的聚类中心和隶属度矩阵,组建农业气象灾害等级判断准则完成预测,但不能准确组建气象状态观测矩阵,无法精确计算灾害预测加权系数,导致预测误差大的问题。提出基于改进混沌理论的农业气象灾害等级预测优化方法。先融合于混沌理论对农业气象灾害发生等级时间序列数据进行相空间重构,计算气象数据混沌相空间的嵌入维数和嵌入延迟,以卡尔曼滤波方法为基础组建农业气象状态观测矩阵,计算出农业气象灾害预测加权系数,以计算的结果为依据组建农业气象灾害等级预测模型,利用上述模型完成对农业气象灾害等级预测,弥补了传统方法无法预测隐含的灾害发生规律的弊端。仿真结果表明,所提方法可以有效地预测可能即将发生的农业气象灾害及其等级,具有较高的预测精度和良好的适用性。
[Abstract]:Accurate prediction of the grade of agrometeorological disasters can effectively prevent disasters and increase agricultural production. The process of disaster grade prediction should be based on the calculation of the weighted coefficient of agrometeorological disaster prediction by using the agrometeorological state observation matrix, and the traditional method can obtain the clustering center and membership matrix of the key attributes of the disaster. The criterion of grade judgment of agricultural meteorological disasters is established, but the observation matrix of meteorological state can not be set up accurately, and the weighted coefficient of disaster prediction can not be calculated accurately, which leads to the problem of large prediction error. Based on the improved chaos theory, an optimization method of agrometeorological disaster grade prediction is proposed. Firstly, the phase space reconstruction of agricultural meteorological disaster occurrence grade time series data is carried out by using chaos theory, and the embedding dimension and embedding delay of chaotic phase space of meteorological data are calculated. Based on the Kalman filter method, the agrometeorological state observation matrix is established, and the weighted coefficient of agrometeorological disaster prediction is calculated. Based on the calculated results, a prediction model of agrometeorological disaster grade is established. The above model is used to forecast the grade of agrometeorological disasters, which makes up for the disadvantages of the traditional method which can not predict the hidden disaster occurrence law. The simulation results show that the proposed method can effectively predict the possible agricultural meteorological disasters and their grades, and has a high prediction accuracy and good applicability.
【作者单位】：山西农业大学信息科学与工程学院;山西农业大学软件学院;
【分类号】：S42

【相似文献】