Efficiently and accurately solving the failure credibility of a structural system plays a significant role in the engineering field under fuzzy environment. A more precise failure credibility can help assess the safety degree of the system, and then reduce the occurrence of security accidents. At present, the extended fuzzy first-order and second-moment method (EFFOSM) is effective to analyze failure credibility for some occasions. However, when it comes to complicated performance functions and fuzzy inputs, the accuracy of EFFOSM is greatly reduced and its efficiency also needs to be improved. To overcome the above shortcomings, this paper proposes two types of novel fuzzy simulation algorithms, namely uniform discretization algorithm (UDA) and bisection simulation algorithm (BSA). For the system involving frequently-encountered continuous and strictly monotone performance functions of regular LR fuzzy interval inputs, these two algorithms are designed to estimate failure credibility with higher efficiency and accuracy. Subsequently, with the aid of the linearization and regularization procedures in EFFOSM, the application of UDA and BSA is extended to non-monotone performance functions of irregular LR fuzzy intervals. To evaluate and verify the performance of the proposed two algorithms, their comparisons with EFFOSM are conducted through some numerical examples and practical problems. The results show that the proposed two algorithms outperform EFFOSM in terms of accuracy and efficiency, and also have wider application range for estimating the failure credibility of strictly monotone performance functions involving regular LR fuzzy interval inputs. Meanwhile, BSA is slightly better than UDA for the less runtime.

中文翻译：

---

用于估计模糊环境下工程结构失效可信度的高效模糊仿真


在模糊环境下的工程领域，高效、准确地解决结构系统的失效可信度具有重要作用。更精确的故障可信度有助于评估系统的安全程度，进而减少安全事故的发生。目前，扩展模糊一阶和第二矩法 （EFFOSM） 在某些场合可以有效地分析失效可信度。然而，当涉及到复杂的性能函数和模糊输入时，EFFOSM 的精度大大降低，其效率也需要提高。为了克服上述缺点，该文提出了两种新颖的模糊仿真算法，即均匀离散化算法 （UDA） 和等分仿真算法 （BSA）。对于涉及常规 LR 模糊区间输入的频繁遇到的连续和严格单调性能函数的系统，旨在以更高的效率和准确性估计失效可信度。随后，借助 EFFOSM 中的线性化和正则化程序，UDA 和 BSA 的应用扩展到不规则 LR 模糊区间的非单调性能函数。为了评估和验证所提出的两种算法的性能，通过一些数值示例和实际问题将它们与 EFFOSM 进行了比较。结果表明，所提出的两种算法在准确性和效率方面优于 EFFOSM，并且在估计涉及常规 LR 模糊区间输入的严格单调性能函数的失效可信度方面也具有更广泛的应用范围。同时，BSA 的运行时间比 UDA 略好。