Application of a data mining approach to derive operating rules for the Eleviyan irrigation reservoir

ISSN： 1040-2381

Source： Lake and Reservoir Management, Vol.28, Iss.2, 2012-06, pp. : 142-152

Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.

Previous Menu Next

Abstract

Optimum irrigation reservoir operation is a significant issue for water management decision makers. In this study, we used 4 different datasets of monthly amounts of water to be released from the Eleviyan irrigation reservoir in Iran as inputs in a data mining model; “if–conditional” operating rules were determined as outputs. Operating rules derived from data mining and rules obtained from an optimization model were in high concordance, especially between the 2 datasets from the preconstruction period, with differences in only 12 of 252 instances (252 months) used to compare the 2 methods. Operating rules determined for the preconstruction period using the data mining method were also consistent with optimum operating rules determined either by optimization or Monte Carlo simulation. Thus, we concluded that the decision tree subtechnique of data mining is an appropriate method for determining meaningful operating rules for the reservoir. [Supplemental materials are available for this article. Go to the publisher's online edition of Lake and Reservoir Management to view the supplemental file.]