Monte Carlo Simulation with a new method of Adaptive Importance Sampling
Monte Carlo Simulation with a new method of Adaptive Importance Sampling
Anahita Mokhtaari1 Majid Mahjoor Lotf Abadi2
1) Tabari University of Babol, Iran,
IOI Code
:
XYBH-AZDHF
Publication :
International Conference on Civil Engineering, Architecture and Cityscape(iccacs.com)
Abstract :
Monte Carlo simulation is a suitable method for structural reliability analysis. In this simulation method, in order to calculate the small failure probabilities a large number of simulations are required. So far, a wide range of methods have been presented such as Importance Sampling in order to reduce computational efficiency. In this paper, a new method of Importance Sampling is proposed to reduce the number of limit state function computations. On the other hand, the proposed algorithm does not need to know the position of the design point or shape of limit state function. The main idea of the proposed algorithm is that the average amount of sampling probability density function is changing in the simulation process. In fact, in process of generating of random points, each point with lower limit state function and closer distance of problem space center is considered as an average sampling density function. Accordingly, the concentration of sampling will be on the areas of problem space’s importance.
Keywords : monte carlo simulation importance sampling reliability analysis random variable