단면보강 형상에 따른 제방의 동적거동 특성평가에 관한 연구

Abstract

In this study, for the accurate examination of the reinforcing section of the embankment in the seismic design of the reservoir, the 1g shaking table test, one-dimensional ground response analysis, and two-dimensional ground dynamic analysis were conducted. As an experimental procedures, the height of the reservoir embankment used for the model test was 30m, the slope was 1:2.0, and the shape of the sandstone reinforcement section was classified into three cases (no, partial, and full reinforcement).The Ofunato and Hachinohe seismic waves for short-period and long-period waves, respectively, were used as the input seismic waves and seismic waveforms observed for Gyeongju and Pohang earthquakes that have recently occurred in Korea were also applied. The results of 1g shaking table tests showed that the PGA increment in the conditions of partial and full reinforcements were 10 and 19% for upstream slope, 13 and 20% for middle part, and 9 and 12% for downstream slope, respectively. The effect of reinforcement was expected to be stronger in the slope part than in the middle part, but the test results showed that the effect was stronger in middle part. This is perhaps because the shallower ground has correlation with the lower constraint pressure condition due to the geologic stress of the ground. Numerical analysis results showed that partial and full reinforcement using sandstone on both slopes of the model embankment increases the load pressing the embankment, resulting in decrease of acceleration increment interval. The acceleration increment interval trend was analyzed against spectral acceleration (SA) and PGA obtained both from the 1g shaking table test and numerical analysis. The results showed that it was generally consistent with the incremental interval trend of SA, suggesting that it has been verified also through numerical analysis. It was found that, regardless of the presence of reinforcement, the PGA value increased as the height of the ground increased in all the sections. The decreasing trend of the PGA increment interval was found to be 10∼15% higher on average for the full reinforcement than in partial one. In addition, as a result of numerical calculation and comparative analysis of the reliability between the 1g shaking table test and the numerical analysis for the amplification coefficient, the average reliability was 79.8% for non-reinforced section, 79.3% for some reinforced section, and 82.5% for front reinforced. This means that the dynamic analysis results according to the two-dimensional nonlinear time history predicted the results of the 1g shaking table test band experiment very similarly, and it is judged that 1g shaking table test for the model embankment is sufficiently valid. Based on these results, it is recommended that acceleration amplification analysis for dynamic loads should also be performed at the same time when designing cross-sections with reinforcement of sandstone through existing static analysis. In addition, since the analysis results of this study were able to obtain reliable results, it is judged that it will be helpful in determining the cross-section of sandstone reinforcement when designing seismic reinforcement.