Renmin Zhujiang (Jan 2024)
Application of Improved Fisher Optimal Segmentation Method Based on Combined Weight Method in Reservoir Flood Season Staging
Abstract
In Fisher optimal segmentation method,multi-index weighting only considers a single objective weighting method.To overcome the shortcoming,this paper introduces the entropy weight method based on information theory,the coefficient of variation method based on statistical principle,and the CRITIC weighting method in the calculation of index weight.It proposes the Fisher optimal segmentation method based on combined weight.Taking Shuibuya Reservoir as an example,five indicators are selected,including the average flow in ten days,the number of occurrences of the maximum peak flood flow,the maximum one-day flood volume in ten days,the maximum three-day flood volume in ten days,and the maximum seven-day flood volume in ten days.Then the Fisher optimal segmentation method is applied to divide the flood season,and the staging results are compared with the staging results by a single objective weighting method.It is determined that the flood season is divided into three optimal periods.The pre-flood season is from May 1 to June 10,the main flood season from June 11 to July 31,and the post-flood season from August 1 to September 30.The results show that the combined weight method takes into account the advantages of the entropy weight method,the coefficient of variation method,and the CRITIC weighting method.The method can intuitively distinguish the influence degree of different indexes on flood season staging and is suitable for determining multi-index weight in the Fisher optimal segmentation method.
