科技广场
科技廣場
과기엄장
SCIENCE TECHNOLOGY PLAZA
2014年
1期
10-14
,共5页
水资源可持续利用%评价%SVM%宁夏
水資源可持續利用%評價%SVM%寧夏
수자원가지속이용%평개%SVM%저하
Sustainable Utilization of Water Resources%Evaluation%Support Vector Machine%Ningxia
本文在吸收前人研究成果的基础上，首先对水资源可持续利用评价指标体系进行了分析，确立了适合宁夏区情的初始指标，其次利用聚类法从初始指标里筛选得到15个评价指标，采用SVM对宁夏2002-2011年的水资源可持续利用状况进行了评价，并与传统的BP-ANN的评价结果加以比较，结果表明：两者评价结论基本一致；与ANN相比，SVM模型简单，泛化性强，分类精度高。该结果对宁夏水资源的调控和分配具有一定的意义。
本文在吸收前人研究成果的基礎上，首先對水資源可持續利用評價指標體繫進行瞭分析，確立瞭適閤寧夏區情的初始指標，其次利用聚類法從初始指標裏篩選得到15箇評價指標，採用SVM對寧夏2002-2011年的水資源可持續利用狀況進行瞭評價，併與傳統的BP-ANN的評價結果加以比較，結果錶明：兩者評價結論基本一緻；與ANN相比，SVM模型簡單，汎化性彊，分類精度高。該結果對寧夏水資源的調控和分配具有一定的意義。
본문재흡수전인연구성과적기출상，수선대수자원가지속이용평개지표체계진행료분석，학립료괄합저하구정적초시지표，기차이용취류법종초시지표리사선득도15개평개지표，채용SVM대저하2002-2011년적수자원가지속이용상황진행료평개，병여전통적BP-ANN적평개결과가이비교，결과표명：량자평개결론기본일치；여ANN상비，SVM모형간단，범화성강，분류정도고。해결과대저하수자원적조공화분배구유일정적의의。
In this paper, on the basis of fully absorbing predecessors research results, first,we analyzed the sustainable utilization of water resources and established the initial indicator, and then we use the clustering analy-sis-entropy weight method to screen out 15 most representative evaluation indicators from the initial indicators. The support vector machine model is used to evaluate the situation of sustainable utilization of water resources in Ningxia Hui Autonomous Region in 2002-2011. To compare with BP-ANN, the results show that:the two have ba-sically the same evaluation results and SVM model is simple and has stronger generalization ability and higher classification accuracy, which has practical significance for Ningxia water resources regulation and distribution.
