文献类型：期刊文献

英文题名：Drought prediction in the Yunnan-Guizhou Plateau of China by coupling the estimation of distribution algorithm and the extreme learning machine

作者：Li, Qiongfang Du, Yao Liu, Zhennan Zhou, Zhengmo Lu, Guobin Chen, Qihui

第一作者：Li, Qiongfang

通信作者：Li, QF[1];Du, Y[1];Li, QF[2]

机构：[1]Hohai Univ, Coll Hydrol & Water Resources, Nanjing 210098, Peoples R China;[2]Hohai Univ, Yangtze Inst Conservat & Dev, Nanjing 210098, Peoples R China;[3]Guizhou Univ, Sch Civil Engn, Guizhou Inst Technol, Guiyang 550000, Peoples R China;[4]Hohai Univ, Nanjing 210098, Peoples R China

第一机构：Hohai Univ, Coll Hydrol & Water Resources, Nanjing 210098, Peoples R China

通信机构：corresponding author), Hohai Univ, Coll Hydrol & Water Resources, Nanjing 210098, Peoples R China;corresponding author), Hohai Univ, Yangtze Inst Conservat & Dev, Nanjing 210098, Peoples R China.

年份：0

外文期刊名：NATURAL HAZARDS

收录：;Scopus(收录号：2-s2.0-85129768865);WOS:【SCI-EXPANDED(收录号:WOS:000792998400001)】；

基金：Financial support is gratefully acknowledged from the National Natural Science Foundation Commission of China under Grant numbers 51879069 and 41961134003, Guizhou Science Foundation-ZK [2021] General 295 and the Jiangsu Provincial Collaborative Innovation Center of World Water Valley and Water ecological civilization, China.

语种：英文

外文关键词：EDA-ELM model; GA-ELM model; ANFIS model; Drought prediction; The Yunnan-Guizhou Plateau

摘要：Drought prediction is a critical non-engineering approach to mitigate their significant threats to water availability, food safety, and ecosystem health. Therefore, to improve the efficiency and accuracy of drought prediction, a novel drought prediction model was proposed by optimizing the extreme learning machine (ELM) using the estimation of distribution algorithm (EDA) (EDA-ELM) and evaluated by the comparison with the genetic algorithm-optimized ELM (GA-ELM) model, standard ELM model, and adaptive network-based fuzzy inference system (ANFIS) in drought prediction for Yunnan-Guizhou Plateau (YGP). The standardized precipitation evapotranspiration index (SPEI) in 3/6/12-month time scales was treated as the dependent variable and the primary drought driving factors as predictor variables. The results revealed that the EDA-ELM model performed best in multiscalar SPEI prediction, followed by GA-ELM, ANFIS, and standard ELM models, while the model execution time was descended by EDA-ELM, GA-ELM, ANFIS, and standard ELM models, varying from 100 to 700 s. The outputs could provide a novel approach to drought prediction and benefit drought prevention and mitigation.