結構物的耐風設計通常需要經由風洞實驗，取得風壓頻譜的實驗數據，其過程相當耗時且費用昂貴。使用回歸公式來整理分析實驗數據，常無法得到準確的風壓頻譜值，因此，如何更有效的利用風洞實驗氣動力資料庫是一個重要的課題。 淡江大學風工程研究中心利用類神經網路對於高層建築建立風力估算模式已有成熟的發展，但在半圓頂型與大跨徑結構卻沒有應用類神經網路的完整研究，只有研究助理鐘欣潔在2011年參與過的大跨度計畫案，曾應用類神經網路訓練、預測單一尺寸比例半圓頂型結構的風力頻譜，討論其自相關風力頻譜的軸圈關係。然而，就不同半圓頂比例而言，其估算模式仍有很大的發展空間。 本研究利用羅元隆博士於日本東京大學研究期間所建立的半圓頂型結構風壓資料庫，相較於鐘欣潔所使用的風力資料，更注重於半圓頂模型其曲率與結構高度的變化，對於子午線上風壓頻譜的影響，利用隨機選取法撰寫RBFNN類神經網路程式，在訓練、驗證與測試網路的過程中，尋找符合理論且準確的估算模式，與前人之回歸公式做進一步的比較探討，最後將程式應用於網路平台，建立簡易使用者介面，只需輸入簡單的參數即可透過伺服器運算，得到風壓頻譜的類神經網路估算值，可供相關研究進行試驗量測前的初步評估。 Wind resistant design of buildings often needs to acquire wind spectra from wind tunnel tests. Using regression formulas to process and analyze experimental data of wind spectra usually is not very accurate. Therefore, one of the most important issue is how to use experimental wind load aerodynamic database more effectively. The development of wind load estimation models for high-rise buildings using artificial neural networks (ANNs) has already been studied by the Wind Engineering Research Center of Tamkang University (WERC-TKU) for a long time. However, no complete research about dome structures using ANNS has been conducted. Only a large-span research project in 2011 conducted by research assistant Hsin-Chieh Chung trained ANNs for the predictions of wind spectra of fixed shape dome, which examined the axis and circle relation to coherence wind spectra. Never the less, there are a lot of rooms for further development of the estimation models for different shapes of domes. In this study, the wind pressure database of dome models that established by Dr. Yuan-Lung, Lo in University of Tokyo was used. Comparing with the data that used by Hsin-Chieh Chung, the focus is more on the differences of wind pressure spectra on the meridian with the change of curvature and height. Random center selection method was used to write RBFNN program to train, validate and test the ANNs. The estimation models found not only accurate but also theoretically consistent. Models were also compared with previous regression formula. At the end, the ANN models were applied to a network platform and a simple web browser user interface was built. Wind pressure spectra calculated by the server can be easily obtained with simple parameter inputs, which can be used as preliminary estimations before wind tunnel tests.