| 作成者 |
|
|
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
|
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 号 |
|
| 開始ページ |
|
| 終了ページ |
|
| 出版タイプ |
|
| アクセス権 |
|
| 関連DOI |
|
|
|
|
|
| 関連URI |
|
|
|
|
|
| 関連情報 |
|
|
|
|
|
| 概要 |
The water distribution regulation system of Fukuoka City is a system in which motor valves are operated by remote control while pressure gauges and flow meters attached to distribution pipes are monit...ored. One of the main objectives of this system is to regulate pressure in all the network nodes between lower and upper target values. With the ever-increasing complexity of the city-wide distribution pipe network, motor valve operations to regulate pressure and flow came to depend more and more on the experience and skills of operators. For this reason, an improvement of valve operations support should be done based on valve operation planning for flow and pressure regulation and the operation knowledge database which is constructed on the basis of past experience in order to prevent the events of pressure regulation falls outside the required target pressure range and also to reduce the effort of operators. Therefore in this study and with the assistance of three years of telemetry data, an analysis of the existing valve operation of a certain block within the city district is performed using both correlation analysis and an unsupervised class of Artificial Neural Networks (ANN) named Self-Organizing Maps (SOM). Results show that correlation analysis has successfully classified the operation types of different valves attached to the studied network into three categories while SOM is an efficient tool in clustering the different complicated operational patterns of valves, visualizing the huge amount of telemetry valves data, detecting the patterns which need future improvement and could present good alternative solutions for improving future valve operational support.続きを見る
|
Mendeley出力
KJ00004262221(64-1-5)