The existence of the principal directions of the ground motion based on Arias intensity is well-known. These principal directions do not necessarily coincide with the orientations of recording sensors or with the orie...The existence of the principal directions of the ground motion based on Arias intensity is well-known. These principal directions do not necessarily coincide with the orientations of recording sensors or with the orientations along which the ground motion parameters such as the peak ground acceleration and the pseudo-spectral acceleration （PSA） are maximum. This is evidenced by the fact that the maximum PSA at different natural vibration periods for horizontal excitations do not correspond to the same orientation. A recent analysis carried out for California earthquake records suggests that an orientation-dependent ground motion measurement for horizontal excitations can be developed. The main objective of this study is to investigate and provide seismic ground motion measurements in the horizontal plane, including bidirectional horizontal ground motions, for Mexican interplate and inslab earthquake records. Extensive statistical analyses of PSA are conducted for the assessment. The analysis results suggest that similar to the case of California records, the average behavior of the ratio of the PSA to the maximum resulting PSA can be approximated by a quarter of an ellipse in one quadrant; and that the ratio can be considered to be independent of the value of the maximum resulting PSA, earthquake magnitude, earthquake distance and the focal depth. Sets of response ratios and attenuation relationships that can be used to represent a bidirectional horizontal ground motion measurement for Mexican interplate and inslab earthquakes were also developed.展开更多
Application of the artificial neural network （ANN） to predict pseudospectral acceleration or peak ground acceleration is explored in the study. The training of ANN model is carried out using feed-forward backpropaga...Application of the artificial neural network （ANN） to predict pseudospectral acceleration or peak ground acceleration is explored in the study. The training of ANN model is carried out using feed-forward backpropagation method and about 600 records from 39 California earthquakes. The statistics of the residuals or modeling error for the trained ANN-based models are almost the same as those for the parametric ground motion prediction equations, derived through regression analysis; the residual or modeling error can be modeled as a normal variate. The similarity and differences between the predictions by these two approaches are shown. The trained ANN-based models, however, are not robust because the models with almost identical mean square errors do not always lead to the same predictions. This undesirable behaviour for predicting the ground motion measures has not been shown or discussed in the literature; the presented results, at least, serve to raise questions and caution on this problem. A practical approach to ameliorate this problem, perhaps, is to consider several trained ANN models, and to take the average of the predicted values from the trained ANN models as the predicted ground motion measure.展开更多
基金Natural Science and Engineering Research Council of Canada, the University of Western Ontario and the National Council of Science and Technology (CONACyT) of Mexico
文摘The existence of the principal directions of the ground motion based on Arias intensity is well-known. These principal directions do not necessarily coincide with the orientations of recording sensors or with the orientations along which the ground motion parameters such as the peak ground acceleration and the pseudo-spectral acceleration （PSA） are maximum. This is evidenced by the fact that the maximum PSA at different natural vibration periods for horizontal excitations do not correspond to the same orientation. A recent analysis carried out for California earthquake records suggests that an orientation-dependent ground motion measurement for horizontal excitations can be developed. The main objective of this study is to investigate and provide seismic ground motion measurements in the horizontal plane, including bidirectional horizontal ground motions, for Mexican interplate and inslab earthquake records. Extensive statistical analyses of PSA are conducted for the assessment. The analysis results suggest that similar to the case of California records, the average behavior of the ratio of the PSA to the maximum resulting PSA can be approximated by a quarter of an ellipse in one quadrant; and that the ratio can be considered to be independent of the value of the maximum resulting PSA, earthquake magnitude, earthquake distance and the focal depth. Sets of response ratios and attenuation relationships that can be used to represent a bidirectional horizontal ground motion measurement for Mexican interplate and inslab earthquakes were also developed.
基金The financial support received from the Natural Science and Engineering Research Council of Canadathe University of Western Ontario
文摘Application of the artificial neural network （ANN） to predict pseudospectral acceleration or peak ground acceleration is explored in the study. The training of ANN model is carried out using feed-forward backpropagation method and about 600 records from 39 California earthquakes. The statistics of the residuals or modeling error for the trained ANN-based models are almost the same as those for the parametric ground motion prediction equations, derived through regression analysis; the residual or modeling error can be modeled as a normal variate. The similarity and differences between the predictions by these two approaches are shown. The trained ANN-based models, however, are not robust because the models with almost identical mean square errors do not always lead to the same predictions. This undesirable behaviour for predicting the ground motion measures has not been shown or discussed in the literature; the presented results, at least, serve to raise questions and caution on this problem. A practical approach to ameliorate this problem, perhaps, is to consider several trained ANN models, and to take the average of the predicted values from the trained ANN models as the predicted ground motion measure.
