基于模糊点估计法的水闸渗流可靠度分析

针对水闸观测数据偏少、水闸渗流分析中会出现大量不确定因素等问题,对水闸渗流进行模糊可靠度研究。模糊点估计法能同时考虑参数的模糊性和随机性,因此将该方法应用于水闸渗流可靠度分析。考虑闸基土质参数的模糊随机性,对闸前水位、闸基土质的土粒相对密度、孔隙率、各土层的渗透系数等模糊随机变量取不同水平截集,将同一水平截集得到的参数进行组合,通过统计矩法计算了水闸的渗流模糊可靠指标。选取运行多年的水闸为例,考虑隶属度函数的不同取值界限对可靠指标的影响,建议取值界限取3。通过该方法得到止水完好时水闸渗流安全系数为2.70,可靠指标为5.597;止水破坏时,安全系数为0.98,模糊可靠指标为-0.151。其计算结果与拉丁超立方抽样法(LHS)得到的可靠度指标比较,两者基本一致,说明该方法适用于水闸渗流可靠度分析。

Fuzzy point estimation method-based analysis on reliability of sluice seepage

Aiming at the problems of less observation data of sluice and large amount of uncertain factors occurring during analysis of sluice seepage，a study on the fuzzy reliability of sluice seepage is made herein. As the fuzzy point estimation method has the characteristics of that the fuzziness and randomness of parameter can be simultaneously taken into account by it，the method is applied to the reliability analysis of sluice seepage. Under the consideration of the fuzzy randomness of the sluice foundation soil parameter，the various level cut sets are taken for the fuzzy random variables，such as the water level in front of sluice，the specific gravity of solid particles and porosity of sluice foundation soil，the permeability coefficients of all the soil layers，etc. ， and then the parameters obtained from the same level cut set are combined and the fuzzy reliable index of sluice seepage is calculated through then method of statistic moment. Taking a sluice that operates for many years as the study case，it is suggested that 3 is taken for the value range under the consideration of the influence from the membership function on the reliable index. Through this method，it is obtained that the safety coefficient of the sluice seepage is 2. 70 and the reliable index is 5. 597 when the sluice is well water-sealed，while the safety coefficient of the sluice seepage is 0. 98 and the reliable index is － 0. 151 when the water seal is damaged. If the calculation result is compared with the reliable index got from the Latin hypercube sampling method ( LHS) ，both of them are basically coincided，thus it is indicated that this method is adaptable for the reliability analysis of sluice seepage.