基于改进POT模型的土石坝渗流监控指标拟定方法

Method based on improved POT model for determining seepage monitoring index of earth-rock dam

中文关键词: 土石坝  渗流监测  POT模型  麻雀搜索算法  参数优化

英文关键词: earth-rock dam  seepage monitoring  POT model  sparrow search algorithm  parameter optimization

中文摘要:

      针对现有土石坝渗流监控指标拟定方法存在主观性较强和精度较低的不足，基于智能算法改进的超阈值(peaks over threshold，POT)模型，提出了优化的土石坝渗流监控指标拟定方法．以3σ准则为确定最优阈值的理论基础，采用基于混沌映射、结合Levy飞行和逆向学习的动态选择策略改进的麻雀搜索算法(improved chaos sparrow search algorithm，ICSSA)，对POT模型中阈值的选取方法进行优化．建立了ICSSA-POT模型，实现对监测资料尾部数据的拟合，从而得到更为合理的土石坝渗流监控指标．研究表明，相比于传统方法，所提方法可有效避免主观性与随机误差，得到的监测资料尾部数据的拟合决定系数提高了5%，具有更高的计算精度，拟定的渗流监控指标更偏于安全，对防范土石坝渗流破坏、确保土石坝安全长效运行具有较强的指导意义．

英文摘要:

      In view of the shortcomings of the existing methods for determining seepage monitoring index of earth-rock dam, which are subjective and low in accuracy, an optimization method for determining seepage monitoring index of earth-rock dam is proposed based on the improved peaks over threshold (POT) model using intelligent algorithm. Using 3σ criterion as the principle of determining the most reasonable threshold, the improved chaos sparrow search algorithm (ICSSA), which is based on chaotic mapping and dynamic selection strategy combined with Levy flight and reverse learning, is adopted to optimize the selection method for threshold of POT model. The ICSSA-POT model is established to realize the fitting of the tail data of monitoring data, so as to obtain a more reasonable seepage monitoring index of earth-rock dam. Research shows that compared with traditional methods, the proposed method can effectively avoid subjectivity and random errors, and the fitting determination coefficient of the obtained tail data of monitoring data is increased by 5%, which has higher calculation accuracy. The proposed seepage monitoring index is more safety oriented, and has strong guiding significance for preventing seepage damage of earth-rock dam and ensuring safe and long-term operation of earth-rock dam.