IEEJ Transactions on Industry Applications
Online ISSN : 1348-8163
Print ISSN : 0913-6339
ISSN-L : 0913-6339
Special Issue Paper
Knowledge Acquisition from In-Operation Data for Water Supply Systems using PLS Regression and K-Means Method
Hiroshi MatsukiYasutaka Fujimoto
Author information
JOURNAL FREE ACCESS

2014 Volume 134 Issue 3 Pages 301-307

Details
Abstract

This study investigates the scheduling of water flow between water treatment plants and distribution reservoirs. We aim to reproduce the performance of an experienced operator for a water supply system. The steering groups of water supply systems face problems due to the decreasing number of experienced operators. Without the expertise of experienced operators, it is difficult to carry out operations in a safe and stable manner. For this purpose, regression analysis using Partial Least Squares was applied to reproduce the performance of an experienced operator. In addition, the k-means method was applied to cluster predictors for regression analysis. Experimental results show that the root-mean-square values for the reservoir water level obtained using the proposed method were smaller than those obtained using a conventional method and our previously proposed method. Thus, we confirmed that the proposed method is able to acquire operation knowledge from experienced operators, and un-skilled operators can use this data to regenerate an operation plan. This method is expected to compensate for the decrease in the number of experienced operators.

Content from these authors
© 2014 by the Institute of Electrical Engineers of Japan
Previous article Next article
feedback
Top