板壳结构作为航空航天、船舶、建筑等工程领域的关键承载组件，其屈曲稳定性直接决定整体结构的安全性与可靠性。然而，材料属性离散性等因素显著影响屈曲临界载荷的分布特性，传统确定性分析方法难以准确量化此类随机影响。为此，提出一种径向基函数（RBF）增强的直接概率积分法（DPIM），用于高效求解板壳结构在多重随机变量作用下的屈曲临界载荷概率特性，为随机屈曲不确定性量化评估提供理论依据。通过RBF构建屈曲临界载荷与随机变量之间的高精度显式代理模型，有效减少原始直接概率积分法中实施耗时的屈曲有限元分析的计算代价。数值算例中将径向基函数增强的直接概率积分法与原始直接概率积分法、蒙特卡洛模拟方法进行对比，结果表明提出的方法在保持精度损失可控的同时，能显著提升计算效率。

As critical load-bearing components in aerospace， marine， and construction engineering fields， the buckling stability of plate and shell structures directly determines the safety and reliability of entire systems. However， factors such as material property variability significantly influence the distribution characteristics of critical buckling loads， the traditional deterministic analysis methods are hard to quantify the stochastic influence. The Radial Basis Function （RBF）-enhanced Direct Probability Integral Method （DPIM） is proposed to efficiently determine the probabilistic characteristics of buckling critical loads in plate and shell structures under multiple random variables， which provides theoretical foundations for stochastic buckling uncertainty quantification. By constructing a high-precision explicit surrogate model between critical buckling loads and random variables using RBF， this method effectively reduces the computational cost for implementing the time-consuming buckling finite element analysis in the original direct probability integral method. Comparative analyses with traditional DPIM and Monte Carlo simulation methods in numerical examples demonstrate that the proposed RBF-enhanced DPIM achieves remarkable computational efficiency improvements while maintaining controlled accuracy loss.