中小學校舍耐震評估－以演化式支持向量機推論系統為評估方法

臺灣位於環太平洋地震帶，地震頻繁，歷年來之地震常造成中小學校舍極大的損傷。由於校舍除了做為教育場所外，大地震過後亦常當作災民的臨時收容所，地位極為重要。故如何評估現有校舍的耐震能力，以發揮校舍的功能，乃一件刻不容緩的工作。性能目標地表加速度為國家地震工程研究中心所提出的理論，代表校舍結構可否滿足耐震能力的一個重要指標，其值之計算繁複且耗時，非一般工程人員可勝任。本研究為解決上述問題，以人工智慧方法發展一演化式支持向量機推論系統（ESIS），此系統乃結合支持向量機（SVM）以及快速混雜基因演算法（fmGA）兩種理論而成，可快速評估性能目標地表加速度。本研究以異於傳統的研究方法切入耐震評估領域，以新的思維探討校舍耐震能力。文末以臺灣地區525棟典型校舍做為研究範例，並將樣本區分為訓練範例及測試範例。從研究結果中可知，訓練範例之誤差均方根值（RMSE）介於0.06464與0.08758之間；測試範例的RMSE則介於0.01329與0.02876之間，顯示結果十分良好，可提供學術界及工程界的參考。

關鍵字

中小學校舍；性能目標地表加速度；支持向量機；快速混雜基因演算法；耐震評估

並列摘要

School buildings in Taiwan are designed to serve not only as places of education but also as temporary shelters in the aftermath of major earthquakes. Effective evaluation of the seismic resistance of current school buildings is a critical issue that deserves further exploration. The National Center for Research on Earthquake Engineering (NCREE) currently employs performance-target ground acceleration (Ap) as the key index to evaluate school structure compliance with seismic resistance requirements. However, computational processes are complicated, time consuming, and require the input of many experts. To address this problem, this research developed an evolutionary support vector machine inference system (ESIS) that integrated two AI techniques, namely, the Support Vector Machine (SVM) and Fast Messy Genetic Algorithm (fmGA). Based on training results, the developed system can predict the Ap of a school building in a significantly shorter time base, thus increasing evaluation efficiency significantly. Samples of 525 typical school buildings in Taiwan were used in this research. Divide them as training cases and testing cases, which were used to calculate the root mean square error (RMSE). According to the results, the RMSE of the training cases are between 0.06464 and 0.08758, while the testing cases are 0.01329 and 0.02876. Another aim of this research is to retain and apply expert knowledge and relevant experience to the solution of similar problems in the future.