基于无偏灰色马尔科夫链的稻瘟病预测研究(英文)

• 作 者: 魏代俊 曾艳敏 邹迎春
• 机 构: 湖北民族学院数学系,恩施州农业科学院,
• 出 处: Plant Diseases and Pests
• 发表日期: 2010年06期
• 关键词: 无偏 五点 马尔科夫预测 马尔科夫链 五点滑动法 模型预测结果 稻瘟病 湖北省恩施州 动法 blast

采用五点滑动法优化原始数据,建立新的灰色马尔科夫预测模型。在分析利用湖北省恩施州1995～2004年稻瘟病历史资料的基础上,分别应用五点滑动优化无偏和无偏GM(1,1)模型预测2005～2010年稻瘟病发生面积。预测结果表明:五点滑动优化无偏GM(1,1)模型预测精度较高,应用于稻瘟病预测切实可行。结合五点滑动优化无偏GM(1,1)模型预测结果,笔者给恩施州2010年稻瘟病的防治提供了政策建议。

The five-point sliding method is used to optimize the original data and a new gray Markov forecasting model is established. Based on the historical data of rice blast in Enshi Prefecture of Hubei Province from 1995 to 2004, the five-point sliding optimization unbiased and unbiased GM (1,1) model was used to predict the occurrence area of ​​rice blast from 2005 to 2010 respectively. The prediction results show that the prediction accuracy of the five-point sliding optimization unbiased GM (1,1) model is high, and it is feasible to forecast the rice blast. Combined with the five-point sliding optimization of unbiased GM (1,1) model prediction results, the author provided a policy suggestion to the 2010 prevention and treatment of rice blast in Enshi Prefecture.