Development of Regional Environment Weather Forecasting Service System (Rewfss) for Construction Industry

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST103-2622-M019-001-CC3

Translated Name/計畫中文名

開發區域環境氣候預測服務系統於營建業之應用

Project Coordinator/計畫主持人

Chih-Chiang Wei

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Marine Environmental Informatics

Website

https://www.grb.gov.tw/search/planDetail?id=8392019

Year

2014

Start date/計畫起

01-11-2014

Expected Completion/計畫迄

01-10-2015

Bugetid/研究經費

356千元

ResearchField/研究領域

大氣科學
土木水利工程
資訊科學--軟體

"氣候因素常導致營造建築業施工過程中必須面臨停工，因為惡劣氣候不利建築工法施行，另外亦需保障工人施工過程的安全性，尤其是在工安日益高漲的今日更形重要。然後對業者來說，一件工程在合約中已明訂工期，一旦未依工期如期完工往往面臨受處罰，如罰款、扣點數等。對營造建築業而言，由於工程施作過程係在戶外居多，常因氣氛因素影響下，例如強風、豪雨等，導致無法施工的命運（如混凝土灌漿、硬化過程），雖然極端氣候（如颱風），中央氣象局會依地區發佈停班停課消息，然而有時並不符合需求性，例如颱風登陸前風速已達不利施工條件、颱風過後施工地點嚴重積水等。因此如此建立一套評估氣候因素對營造業建築施工，將能協助業界及早預測氣候重要因素（如風速、雨量），以便發揮最大施工效益和施工品質並兼顧施工安全。為了達成自然科學理論與產學合作實務目標，本研究將開發一套系統以利營建業在面對自然氣候因素干擾下，減少施工期間之自然環境之不確定因素。本期計畫擬先針對颱風時期之風速因素進行研究（若本期推動順利，未來將再納入其他重要氣候因子，如降雨量、溫度、濕度等）。本期初步選定該合作企業所在地台南為研究場址，另外，為彰顯系統之廣用性，將再擇定不同場址（如台中和離島澎湖地區等）。本計畫所研發之區域環境氣候預測服務系統（Regional Environment Weather Forecasting Service System, REWFSS）主要有五個模組，包括：氣候資料輸入模組（Weather Data Import Submodel, WDI）、資料前處理模組（Data Preprocessing Submodel, DP）、區域氣候預測模組（Regional Climate Forecasting Submodel, RCF）、預測值指標分析模組（Predictions Performance Analysis Submodel, PPA）以及工作條件評估模組（Working Day Evaluation Submodel, EWD）。上述中，主要工作可分為兩部分： • 一為區域環境氣候預測（學術理論）：此部分由計畫主持人負責建模、驗證、維護、測試等，本年度將以風力作為指標性氣候參數，同時模式開發過程中將依颱風路徑分類分別進行建模。由於颱風期間風場受台灣地形因素影響甚劇，因此開發預測模式時，將不同颱風路徑分別進行建模工作，並探討不同颱風路徑之各類模式受到地形影響下，評估模式之表現優劣。本計畫開發模式係採人工智慧之新近演算法，人工智慧是指由人工製造出來的系統所表現出來的智能，如能將其具強效之決策分析能力應用於預測颱風風速及風向，將獲得具體之價值與貢獻。 • 二為工事氣候條件評估（實務應用）：此部分為上述預測模式發展成熟後，將再配合營造業各項工程進度所適用的條件建立評估模式。本計畫透過合作企業人力、物力與財力之支援下，可以得到實際工程經驗與學理應用上之相互驗證。希望能結合合作企業相關工程施工經驗，配合本產學計畫所研發之氣候環境預測服務系統(REWFSS)，以將研發成果應用於工程界，以期業界施工期間降低自然環境所造成之不確定性。因此，期望藉由此次產學計畫產生良好合作成效。" "The construction industry is statistically one of the most hazardous industries in many countries. Besides causing human tragedy, construction accidents also delay project progress, increase costs, and damage the reputation of the contractors. For construction industry, timely and reliable information on recent, current, and future wind speeds are thus vital for enabling forecasters to make accurate and timely forecasts and to be operated under construction appropriately. Because construction often fails during their construction stage under winds, a useful scheme for wind speed forecasts during typhoon periods is highly desired. Hence, the study on typhoon wind speeds during typhoons is of great significance in the field of construction industry and operation management. The purpose of this study is to develop a Regional Environment Weather Forecasting Service System (REWFSS) for construction industry when a structure is under construction stage. The major functions will comprise: (1) forecasting the hourly typhoon wind velocity and direction based on the classified typhoon tracks, (2) understanding the effects of the CMR terrain based on typhoon tracks, and (3) evaluating the weather whether suitable to work or not for contractors. In this study, adaptive network-based fuzzy inference system (ANFIS) will be used as the forecasting technique to predict the wind velocity and direction on the study sites based on the classified typhoon tracks. Neurofuzzy systems are currently among the most widely studied hybrid systems because of the advantages of two highly popular modeling techniques: ANNs and fuzzy logic. Moreover, we will compare them with a benchmark MLPNN (multilayer perceptron neural networks). In addition, the effects of the CMR terrain will be discussed in-depth because the CMR significantly influences variations in wind field when typhoons approach Taiwan. To ensure the accuracy of the neural-based models (ANFIS and MLPNN), traditional regressions will be used as the benchmarks. Moreover, this study also compares the above data-driven statistical models with the Holland parametric wind model. In general, a forecasting horizon ranging from 1 to 48 h can satisfy the requirements of real-time disaster prevention operations for construction industry."

Keyword(s)

氣候
風力
工程
系統
預測
颱風
Climate
Wind
Engineering
System
Prediction
Typhoon

DSpace CRIS

Development of Regional Environment Weather Forecasting Service System (Rewfss) for Construction Industry

Details

Description