Artificial Neural Networks and Microtremor Measurements in Estimating Peak Ground Acceleration at Main Lines of Kaohsiung Mass Rapid Transit, Taiwan

Peak ground acceleration is a very important factor, which must be considered in construction site for analyzing the potential damage resulting from earthquake. The actual records by seismometer at stations related to the site may be taken as a basis, but a reliable estimating method may be useful for providing more detailed information of the strong motion characteristics. Therefore, the purpose of this study is by using back-propagation neural networks to develop a model for estimating peak ground acceleration at two main lines of Kaohsiung Mass Rapid Transit in Taiwan. In addition, the microtremor measurements with Nakamura transformation technique are taken to further validate the estimations. Three neural networks models with different inputs including epicentral distance, focal depth and magnitude of the earthquake records are trained and the output results are compared with available nonlinear regression analysis. The comparisons showed that the present neural networks model has a better performance than that of the other methods, as the calculation results are more reasonable and closer to the actual seismicrecords. Besides, the distributions of estimating peak ground acceleration from both of computations and measurements may provide valuable information from theoretical and practical standpoints.