Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants
碩士 === 佛光大學 === 資訊學系 === 95 === ABSTRACT People regard the quality and quantity of water supply as highly important. Due to the difficulties of water exploitation and deployment, water resource management must be allocated well. The key point of the best operation and management depends on forecasti...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2007
|
Online Access: | http://ndltd.ncl.edu.tw/handle/99016997529244534603 |
id |
ndltd-TW-095FGU00585008 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-095FGU005850082015-10-13T10:42:08Z http://ndltd.ncl.edu.tw/handle/99016997529244534603 Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants 設計以類神經網路為基礎之智慧型淨水場時出水量預測與即時調控機制 LIN, CHIH HSIEN 林志憲 碩士 佛光大學 資訊學系 95 ABSTRACT People regard the quality and quantity of water supply as highly important. Due to the difficulties of water exploitation and deployment, water resource management must be allocated well. The key point of the best operation and management depends on forecasting the short-term water requirement accurately. Unfortunately the traditional pump control used today does not calculate the water level precisely setting, making it difficult to form an accurate economic strategy. Because the global energy cost is climbing continuously, reducing the cost of production becomes the important subject of water treatment plants. This research designs an intelligent model, which is includes two stages: forecast and control. The first is the Real-time Water output Forecasting Module (RWFM). The RWFM is modeled by an Artificial Neural Network, where data-sets are calculated by the time the window shifts. This research uses the trial and error method in various kinds of network structures with parameters to find the best efficiency on real-time water output forecasting. The second one is the Real-time Pump Scheduling Module (RPSM). The purpose of the RPSM is to plan optimization of the pump combinations according to pump performance. The time window pump scheduling is arranged by the real-time prediction of water output. The RPSM requires a high-speed computer and is effective in finding the most economic pump scheduling with the optimization approach on real-time. This research model is combined the RWFM with the RPSM have been proven to have high accurate forecasting ability. By using the traditional pump control by water level setting, the cost of pump power can effectively be reduced. An intelligent and effective solution to water control has been achieved that has real-time self control, lower costs and raises economic worth. These satisfy the modernized demand. 駱至中 2007 學位論文 ; thesis 103 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 佛光大學 === 資訊學系 === 95 === ABSTRACT
People regard the quality and quantity of water supply as highly important. Due to the difficulties of water exploitation and deployment, water resource management must be allocated well. The key point of the best operation and management depends on forecasting the short-term water requirement accurately. Unfortunately the traditional pump control used today does not calculate the water level precisely setting, making it difficult to form an accurate economic strategy. Because the global energy cost is climbing continuously, reducing the cost of production becomes the important subject of water treatment plants.
This research designs an intelligent model, which is includes two stages: forecast and control. The first is the Real-time Water output Forecasting Module (RWFM). The RWFM is modeled by an Artificial Neural Network, where data-sets are calculated by the time the window shifts. This research uses the trial and error method in various kinds of network structures with parameters to find the best efficiency on real-time water output forecasting.
The second one is the Real-time Pump Scheduling Module (RPSM). The purpose of the RPSM is to plan optimization of the pump combinations according to pump performance. The time window pump scheduling is arranged by the real-time prediction of water output. The RPSM requires a high-speed computer and is effective in finding the most economic pump scheduling with the optimization approach on real-time.
This research model is combined the RWFM with the RPSM have been proven to have high accurate forecasting ability. By using the traditional pump control by water level setting, the cost of pump power can effectively be reduced. An intelligent and effective solution to water control has been achieved that has real-time self control, lower costs and raises economic worth. These satisfy the modernized demand.
|
author2 |
駱至中 |
author_facet |
駱至中 LIN, CHIH HSIEN 林志憲 |
author |
LIN, CHIH HSIEN 林志憲 |
spellingShingle |
LIN, CHIH HSIEN 林志憲 Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
author_sort |
LIN, CHIH HSIEN |
title |
Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
title_short |
Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
title_full |
Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
title_fullStr |
Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
title_full_unstemmed |
Building an ANN-Based Intelligent Mechanism for the Real-Time Control of Hourly Water Output in Water Treatment Plants |
title_sort |
building an ann-based intelligent mechanism for the real-time control of hourly water output in water treatment plants |
publishDate |
2007 |
url |
http://ndltd.ncl.edu.tw/handle/99016997529244534603 |
work_keys_str_mv |
AT linchihhsien buildinganannbasedintelligentmechanismfortherealtimecontrolofhourlywateroutputinwatertreatmentplants AT línzhìxiàn buildinganannbasedintelligentmechanismfortherealtimecontrolofhourlywateroutputinwatertreatmentplants AT linchihhsien shèjìyǐlèishénjīngwǎnglùwèijīchǔzhīzhìhuìxíngjìngshuǐchǎngshíchūshuǐliàngyùcèyǔjíshídiàokòngjīzhì AT línzhìxiàn shèjìyǐlèishénjīngwǎnglùwèijīchǔzhīzhìhuìxíngjìngshuǐchǎngshíchūshuǐliàngyùcèyǔjíshídiàokòngjīzhì |
_version_ |
1716832355561242624 |