基于多尺度组合模型的区域农业旱灾损失率预测
|
摘要
针对旱灾系统的非平稳和非线性特征,提出了基于经验模态分解和最小二乘支持向量机的多尺度组合预测模型。为了避免分解过程中旱灾序列取值域发生改变的问题,首先采用逆Logistic变换对原始序列的取值域进行扩展;然后使用经验模态分解将旱灾序列进行平稳化,提取出旱灾序列中不同时间尺度的子序列,根据子序列的波动特征选择适用的方法进行预测;最后将预测子序列进行重组和还原。以河南省农业旱灾综合损失率为例进行3步仿真预测。预测结果表明,多尺度组合模型的预测效果和精度均好于最小二乘支持向量机模型的,说明在旱灾序列处理中应优先选用多尺度分析法。
For the non-stationary and non-linear characteristics of drought system, a multi-scale combined forecasting model based on empirical mode decomposition and least squares support vector machine was proposed. In order to avoid changing the range of drought sequence in the decomposition process, the inverse Logistic transform was used to extend the range of original sequence. Then, the empirical mode decomposition was used to smooth the drought sequence and extract the subsequences of different time scales in the drought sequence. According to the fluctuation characteristics of the subsequences, an appropriate method was selected to predict the drought sequence. Finally, the prediction subsequences were reconstructed and restored. The three-step simulation prediction was carried out based on the comprehensive loss rate of agricultural drought in Henan. The result shows that the multi-scale combined forecasting model is better than the least squares support vector machine model in forecasting effect and accuracy, which indicates that the multi-scale analysis method should be given priority in drought sequence processing.
For the non-stationary and non-linear characteristics of drought system, a multi-scale combined forecasting model based on empirical mode decomposition and least squares support vector machine was proposed. In order to avoid changing the range of drought sequence in the decomposition process, the inverse Logistic transform was used to extend the range of original sequence. Then, the empirical mode decomposition was used to smooth the drought sequence and extract the subsequences of different time scales in the drought sequence. According to the fluctuation characteristics of the subsequences, an appropriate method was selected to predict the drought sequence. Finally, the prediction subsequences were reconstructed and restored. The three-step simulation prediction was carried out based on the comprehensive loss rate of agricultural drought in Henan. The result shows that the multi-scale combined forecasting model is better than the least squares support vector machine model in forecasting effect and accuracy, which indicates that the multi-scale analysis method should be given priority in drought sequence processing.
引文
[1] 陈海涛,徐冬梅,王文川,等.河南省农业干旱风险关键技术研究[M].北京:科学出版社,2017.
[2] 国家防汛抗旱总指挥部.中国水旱灾害公报[M].北京:中国水利水电出版社,2017.
[3] 赵勇,翟家齐,蒋桂芹,等.干旱驱动机制与模拟评估[M].北京:科学出版社,2017.
[4] 金菊良,郦建强,周玉良,等.旱灾风险评估的初步理论框架[J].灾害学,2014(3):1-10.
[5] PAULO A A,FERREIRA E,COELHO C,et al.Drought class transition analysis through Markov and Loglinear models,an approach to early warning[J].Agricultural water management,2005,77(1):59-81.
[6] DURDU ? F.Application of linear stochastic models for drought forecasting in the Büyük Menderes river basin,western Turkey[J].Stochastic environmental research and risk assessment,2010,24(8):1145-1162.
[7] 姜翔程,陈森发.加权马尔可夫SCGM(1,1)c模型在农作物干旱受灾面积预测中的应用[J].系统工程理论与实践,2009(9):179-185.
[8] 张强,韩兰英,王胜,等.影响南方农业干旱灾损率的气候要素关键期特征[J].科学通报,2018,63(23):2378-2392.
[9] 罗党,王浍婷.灰色神经网络下的多变量土壤含水量预测模型[J].华北水利水电大学学报(自然科学版),2017,38(5):70-75.
[10] 刘会玉,林振山,张明阳.建国以来中国洪涝灾成灾面积变化的小波分析[J].地理科学,2005(1):43-48.
[11] 宋新山,王宇晖,严登华,等.基于分形特征的我国干旱灾害小波分析[J].系统工程理论与实践,2013,33(11):2986-2992.
[12] 刘玉,潘瑜春,高秉博,等.中国农作物成灾面积的演化趋势及波动特征研究[J].自然灾害学报,2015(4):15-23.
[13] 马建琴,韩曦.甘肃省近48年春旱时空变化趋势分析[J].华北水利水电学院学报,2013,34(3):40-43.
[14] 李治国,郭志富,张延伟,等.近36年河南农作物干旱灾情时空规律研究[J].干旱区资源与环境,2015,29(4):119-124.
[15] 朱业玉,程炳岩,王记芳.河南旱涝灾害的演变特征分析[J].灾害学,2006,21(3):93-97.
[16] HUANG N E,SHEN Z,LONG S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the Royal Society of London series A,1998,454:903-995.
[17] 唐杰明,刘俊勇,杨可,等.基于灰色模型和最小二乘支持向量机的电力短期负荷组合预测[J].电网技术,2009,33(3):63-68.
[18] 国家防汛抗旱总指挥部办公室.干旱评估标准(试行)[S].北京:中华人民共和国水利部,2006.
[2] 国家防汛抗旱总指挥部.中国水旱灾害公报[M].北京:中国水利水电出版社,2017.
[3] 赵勇,翟家齐,蒋桂芹,等.干旱驱动机制与模拟评估[M].北京:科学出版社,2017.
[4] 金菊良,郦建强,周玉良,等.旱灾风险评估的初步理论框架[J].灾害学,2014(3):1-10.
[5] PAULO A A,FERREIRA E,COELHO C,et al.Drought class transition analysis through Markov and Loglinear models,an approach to early warning[J].Agricultural water management,2005,77(1):59-81.
[6] DURDU ? F.Application of linear stochastic models for drought forecasting in the Büyük Menderes river basin,western Turkey[J].Stochastic environmental research and risk assessment,2010,24(8):1145-1162.
[7] 姜翔程,陈森发.加权马尔可夫SCGM(1,1)c模型在农作物干旱受灾面积预测中的应用[J].系统工程理论与实践,2009(9):179-185.
[8] 张强,韩兰英,王胜,等.影响南方农业干旱灾损率的气候要素关键期特征[J].科学通报,2018,63(23):2378-2392.
[9] 罗党,王浍婷.灰色神经网络下的多变量土壤含水量预测模型[J].华北水利水电大学学报(自然科学版),2017,38(5):70-75.
[10] 刘会玉,林振山,张明阳.建国以来中国洪涝灾成灾面积变化的小波分析[J].地理科学,2005(1):43-48.
[11] 宋新山,王宇晖,严登华,等.基于分形特征的我国干旱灾害小波分析[J].系统工程理论与实践,2013,33(11):2986-2992.
[12] 刘玉,潘瑜春,高秉博,等.中国农作物成灾面积的演化趋势及波动特征研究[J].自然灾害学报,2015(4):15-23.
[13] 马建琴,韩曦.甘肃省近48年春旱时空变化趋势分析[J].华北水利水电学院学报,2013,34(3):40-43.
[14] 李治国,郭志富,张延伟,等.近36年河南农作物干旱灾情时空规律研究[J].干旱区资源与环境,2015,29(4):119-124.
[15] 朱业玉,程炳岩,王记芳.河南旱涝灾害的演变特征分析[J].灾害学,2006,21(3):93-97.
[16] HUANG N E,SHEN Z,LONG S R,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proceedings of the Royal Society of London series A,1998,454:903-995.
[17] 唐杰明,刘俊勇,杨可,等.基于灰色模型和最小二乘支持向量机的电力短期负荷组合预测[J].电网技术,2009,33(3):63-68.
[18] 国家防汛抗旱总指挥部办公室.干旱评估标准(试行)[S].北京:中华人民共和国水利部,2006.