论文部分内容阅读
针对非线性隐式极限状态方程的可靠度指标计算,将支持向量机和粒子群优化算法相结合,提出了一种结构可靠度算法.首先结合支持向量机不受样本点限制的优点,将历次迭代产生样本点加入本次迭代样本点中,采用支持向量机对样本点进行训练,然后引入粒子群优化算法计算可靠度指标,解决迭代过程中支持向量机回归模型可靠度指标计算震荡不收敛的情况,最后根据可靠度指标收敛得到的支持向量机回归模型,采用重要抽样法计算失效概率.计算结果表明:该方法得出的失效概率具有较好的精度,特别是针对迭代过程中可靠度指标不收敛的情况具有良好的适用性.
Aiming at the reliability index calculation of nonlinear implicit limit state equation, a combination of support vector machine and particle swarm optimization algorithm is proposed, and a structural reliability algorithm is proposed.Firstly, based on the advantages of support vector machine without sample points limitation, The iterative sample points are added into this iterative sample point, the sample points are trained by using SVM, then the particle swarm optimization algorithm is introduced to calculate the reliability index, and the reliability index of the support vector machine regression model is calculated in the iterative process. The concussion is not convergent Finally, according to the support vector machine regression model converged by the reliability index, the probability of failure is calculated by the important sampling method.The calculation results show that the failure probability obtained by the method has better accuracy, especially for the reliability index The case of non-convergence has good applicability.