论文部分内容阅读
一定区域的森林火灾发生具有周期性和趋势性特征,运用ARMA模型对我国森林火灾的受害面积进行了模拟和预测,通过阶数初步判定、模型选择、参数估计、模拟检验、预测分析,所选取的ARMA(1,2,3,11,13,15)、(1,2,3,4)模型最大拟合良度可以达到0.885 3,根据模拟和预测发现,我国的森林火灾呈现出13a左右的爆发周期,林火发生受邻近2、3、4、5a的发展趋势影响较大。2015-2020年森林火灾受害面积有增加的趋势。
The occurrence of forest fires in a certain area has periodic and trend characteristics. The ARMA model is used to simulate and predict the area affected by forest fires. Through the preliminary determination of order, model selection, parameter estimation, simulation test and prediction analysis, Of ARMA (1, 2, 3, 11, 13, 15) and (1, 2, 3, 4) models can reach a maximum of 0.885 3. According to the simulation and forecasting, Of the outbreak cycle, the occurrence of forest fire by the neighboring 2,3,4,5 a greater impact on the development trend. 2015-2020 forest fire victims have an increasing trend.