本研究藉由李顯智博士於2019年所提出的最強地震圓弧雙交叉概念，以及其研發的MATLAB程式ASICAE，分析中央氣象局所公布的台灣地區1981年~2016年間的地震，並將這些歷史地震事件分為30、70、110公里三個震源深度，搭配不同的參數設定，在搜尋週期為4年的設定下，找出強震預測命中率較高且預測天數較短的參數。 本研究發現，在地震事件搜尋深度為0-70公里時，精度條件設定弧向誤差參數 TOLde 為 0.13~ 0.14 且徑向誤差參數 TOLRe 為 0.014 ~ 0.016 之組合，具有較佳的強震預測命中率，其中雙凸交叉的70天內強震預測命中率高達80%，並且在三種不同歷史地震事件震源深度設定下預測後續70天內發生強震的命中率，皆高於強震發生背景值，藉此足以佐證最強地震圓弧雙交叉的概念並非簡單的地震事件幾何排列，而是具有力學意義的。 ;The theory of the strongest double intersection of circular arcs of earthquakes (SDICAE) proposed by Hin-Chi Lei in 2019 is applied to analyze the records of the earthquake events from 1981 to 2016 provided by the Central Weather Bureau of Taiwan. The earthquake events are classified into several groups according to different ranges of depths such as 0-30Km, 0-70Km, and 0-110Km. The searching period for the events is set to be four years. The efficiencies of the predictions of strong earthquakes by using the SDICAEs searched by adopting different ranges of depths are evaluated. We try to find out the best choice of the range of depth which can give us the highest hit rate of the prediction of strong earthquake occurred within 70days after the formation of the SDICAE. It is found that, when the range of depth is 0-70Km, the arc-direction precision parameter TOLde is 0.13 and 0.14, and the radial direction precision parameter TOLRe is 0.014, 0.015, and 0.016, the prediction efficiency is the best, and the hit rate of the prediction of strong earthquake within 70 days reaches 80%. Moreover, the hit rates of the predictions of strong earthquakes within 70 days corresponding to all the three choice of ranges of depths are higher than the background value of the strong earthquake occurrence rate. This shows that the formation of the SDICAE in the crust is mechanically meaningful.
