Please use this identifier to cite or link to this item:
Application of Data Mining for Studying Dangerous Coastal Waves
tripple wave instability
|Issue Date: ||2017-01-05T08:38:52Z
|Publisher: ||第 29 屆海洋工程研討會論文集|
|Abstract: ||台灣海岸地區每年平均有數十件，因遭浪擊落海事件發生，其發生時間與地點不確定性高。目前雖然對這危險波浪 (俗稱為瘋狗浪) 的特性有些了解，但若要完全掌握其物理機制，進而用即時之波浪資料作預警還是有困難。本研究延續多年的計畫經驗，收集過去因大浪造成的落海事件記錄、與其相關之氣象與海象資料記錄，利用資料探勘(data mining)分析工具，尋找與落海事件發生的強關聯因子；並以相關因子做為預警模式主要輸入參考依據，據此建立瘋狗浪的預警模式，以茲訂定海岸安全依據，期能減少海邊活動人口的生命財產損失。本研究現階段以資料探勘尋找與發生浪擊落海事件之強關聯因子，以做為下一階段預警模式建立的重要依據。
Abstract: Many fishermen lost their lives to dangerous coastal waves each year at the coast. There has been progress in understanding the characteristics of these dangerous waves, also known as “mad-dog” waves. General knowledge in the conditions for their occurrence, such as seasons, wave conditions, locations and water depth and coastal topography, are being clarified. However, the physical mechanisms for these mad-dog waves still need more field data and research. Hence, at this stage, it is still difficult to predict the timings and locations for the occurrence of these dangerous waves. In order to prevent further loss of lives due to these dangerous waves, this study tried to use data mining techniques by analyzing wave and weather conditions when past mad-dog events occurred and determine relevent factors that are highly correlated with the occurrence of these incidents. The ultimate goal for this study is to estabish a preliminary early-warning system.
|Appears in Collections:||[海洋環境資訊系] 演講及研討會|
Files in This Item:
All items in NTOUR are protected by copyright, with all rights reserved.