The Big Data Industry of Offshore Wind Energy in Taiwan: Development and Applications ( IV)

瀏覽統計 Email 通知 RSS Feed

簡歷

基本資料

Project title

Code/計畫編號

MOST109-2218-E001-003

Translated Name/計畫中文名

台灣離岸風能大數據產業之發展與創新應用（IV）

Funding Organization/主管機關

National Science and Technology Council

Co-Investigator(s)/共同執行人

郭志禹(計畫主持人)
張文鎰
杜佳穎

Department/Unit

The Research Center of Applied Sciences

Website

https://www.grb.gov.tw/search/planDetail?id=13531030

Year

2020

Start date/計畫起

01-06-2020

Expected Completion/計畫迄

31-05-2021

Co-Investigator(s)

Tai-Wen Hsu

Bugetid/研究經費

6440千元

ResearchField/研究領域

資訊科學--軟體

綠能為政府五大科技創新產業研發計畫,規劃 2025 年達總能源 20%,離岸風電目標為 3GW,將建立離岸風機 600 座,投資金額超過 4800 億。本計畫前三年著重於與永傳能源公司合作,以彰化外海福海示範風場及其周圍風場為對象,以「台灣離岸風能大數據產業之發展與創新應用」為研究主題,巨量與多樣性海氣象大數據為研究標的,以具備模擬颱風劇烈天氣能力之高解析海氣象模式進行預測,並使用深度學習大數據分析方法,建立測風塔觀測值與模式預測值關連模式,模式預測與大數據深度學習分析修正預測雙重機制演算法,可創造離岸風電場海氣象數據最高精準預測的可能。除此之外，亦完成離岸風電場海氣象數值預報平台建置及歷史資料庫, 進行了離岸風場海上施工決策輔助分析及風機群組營運效益等相關項目之研發。具體而言，本計畫完成以下資訊計算平台：1.) 開發整合高效能計算資源，進行需要大量計算資源的海氣象數值預報作業，以及 2.) 開發一大數據資料整合分析及機器深度學習平台，稱之為GreenDancer 3+1系統，具有平行分散式計算環境, 可整合數值預報與環境觀測資料等大數據，進行離岸風電相關工程應用。因此在本計畫第四年度，將著重在強化利用前述兩平台，進行數值預報關鍵技術精進，使其可進行高穩定預報作業; 再者強化落地推廣GreenDancer平台，利用其具備之大數據機器深度學習模組，培養工程領域相關技術人才，創造離岸風能大數據數位經濟; 最後利用我們所掌握的數值預報、大數據深度學習分析修正技術，進一步嘗試與離岸結構物監測的邊緣計算應用結合，探討未來新型態監測技術。Green energy research and development plan is one of the government's five major innovation industries of science and technology. It is planned the green energy in 2025 will be 20% of total energy used in Taiwan. While the offshore wind power is targeted for 3GW in 2025, that is required to install 600 offshore wind turbines, equivalent to an investment amount of 480 billion NTD. In the first three years of execution of the research project "The Big Data Industry of Offshore Wind Energy in Taiwan: Development and Applications", we selected the Fuhai demonstrative wind farm and its neighboring wind farms in the Changhua offshore area as the study site, collaborated with the Taiwan Generations Corp. and achieved essential milestones. These milestones include the the deployments of high-resolution marine and atmospheric models, the capability of simulating extreme typhoon scenarios, the integration of the forecast and offshore observed data by using big data and deep machine learning analyses. The combination was demonstrated to be capable of providing highly accurate forecasts for the wind industry. In addition, high performance computing facilities are constructed and historical forecast database have been established. Studies have also been performed for aiding decision making strategies for offshore construction and operational efficiency of wind farms. Explicitly, we constructed the following main information and computation platforms in the project: 1.) Developed and integrated high performance computational resources for numerical marine and weather forecasts, and 2.) Developed a big data, machine learning oriented clustered information platform, called GreenDancer 3+1 system. This system has parallel and distributed computational environment which can be used to integrate numerical forecasts and environmental big data observations for offshore wind energy applications. In the fourth year extension of the project, the focus is shift to strengthen applications of the aforementioned computation and information platforms. On the one hand, we will continue to improving the key technology of numerical forecast and aiming at providing stable and robust forecasts. On the other hand, we will promote the GreenDancer platform to the public, incubate technical talents in related engineering fields with its integrated big data and machine learning software modules, and assist in creating offshore-wind-energy big data economy. Last, with the numerical forecast, big data and machine learning techniques, we propose to combine the edge computing capability and design into the next generation of monitoring technique

Keyword(s)

大數據平台
大數據分析
深度學習
離岸風力發電
氣象預報
海象預報
邊緣計算離岸結構監測
Big Data Platform
Big Data Analytics
Deep Learning
Offshore Wind Energy
Weather Forecast
Marine Forecast
Edge Computing for Offshore Structural Monitoring

DSpace CRIS

The Big Data Industry of Offshore Wind Energy in Taiwan: Development and Applications ( IV)

基本資料

Description