中文摘要：

基于极限分析上限定理，假定破坏面为任意滑动面，提出了一种用于评价加筋土石坝坡抗震稳定性的方法。该方法将素土和筋材的内能耗散率分开考虑，计算各滑动土条的外力功率与内能耗散率，并通过功能平衡条件，利用优化算法确定加筋土石坝的极限抗震能力，所得解物理意义明确、理论基础严格，能够很好反映加筋后土石坝处于极限状态时抗震能力的提高。通过一简单均质加筋边坡的算例分析，结果表明，该方法所得与已有研究成果有较好的一致性，且当水平条分数增加到一定数量时，解答趋于稳定，确定的任意滑动面能够很好地模拟加筋结构临界失稳时的破坏面。同时，通过对坝坡滑动体的水平条分，克服了以往竖向条分对拟静力地震荷载计算不精确的问题。应用该方法对一典型加筋心墙土石坝进行坝坡稳定分析。计算结果表明，坝坡加筋后，土石坝的抗震稳定性有了明显提高，其极限抗震能力较未加筋时提高了19％．21％，且加筋长度对土石坝的极限抗震能力有较大的影响。在实际工程中建议进行合理的计算分析以确定最佳加筋长度，对于本算例，最佳的加筋长度为30-40m。

英文摘要：

A new approach based on the limit analysis upper bound theorem is proposed to assess the seismic stability of reinforced slopes of earth-rock dams. First, the dam slope is divided into horizontal slices with regarding the failure surface as an arbitrary surface. Under the assumption that the energy dissipation rate during failure due to reinforcement is caused by the effect of tensile forces, the internal work rate and the dissipation rate of each soil slice are then calculated. With the conditions of the energy-work balance equation, the maximum anti-seismic capability of dams can finally be optimized by intelligent algorithm. From the contrastive analysis, it is shown that the solutions presented here is in good agreement with those published previously in the literature. The results also indicate that if the quantity of horizontal slices is enough, the solutions obtained arc stable and the failure surfaces are well predicted. Based on the limit analysis, the solutions are proved to be very rigorous. Moreover, a more precise calculation of the pseu- do-static seismic loads could be obtained by the horizontal slice approach than the conventional vertical slice approach. Using the proposed method, the seismic stability of a typical reinforced rockfill dam with core wall is analyzed. The results show that the maximum anti-seismic capability of dams is increased by 19%-21% when the slope is reinforced. Meanwhile, the length of reinforcement has a great influence on the seismic stability of dams. Therefore, it is suggested that the optimum length of reinforcement should be investigated for engineering practice. For this numerical example, the optimum length of reinforcement is 30-40m.