An improved particle swarm optimization for calculating failure probability of structure with defects
The First-Order-Second-Moment (FOSM) method and Monte Carlo (Monte-Carlo) method have been applied to calculate the reliability of structures with defects,such as boilers,pressure vessels and piping systems in recent years.But FOSM will produce more error when the failure function is highly nonlinear and the random variables serve as non-normal distribution.The simulation results using Monte-Carlo are close to exact solutions,but time-consuming.So this paper proposes an improved particle swarm optimization method (IPSO) applied to calculate failure probability of structure with defects.The inertia weight of standard PSO adopts linear reduction strategy which makes the algorithm converge at the local optimum easily;furthermore this approach depends on the number of the largest iteration making the choosing of inertia weight be in blindness.The improved particle swarm optimization method (IPSO) presents the way of changing inertia weight dynamic and takes into account the disturbance of the surrounding particles,Through a case analysis by using MATLAB-the Programming Language,it is shown that the improved particle swarm optimization method is simple,efficient and accurate for reliability assessment of the structure with defects.The IPSO not only has the advantage of standard particle swarm optimization ,particularly significant to many random variables and complex functions comparing with the above reliability of calculation methods,but also fast convergence and the optimal solution does not fall into a local optimum.
Improved particle swarm optimization Structures with defects Failure probability reliability
Qingqing Wu Jianping Zhao
College of Mechanical & Power Engineering,Nanjing University of Technology,Nanjing 210009,China
国际会议
成都
英文
413-418
2009-10-16（万方平台首次上网日期，不代表论文的发表时间）