Simulation method and system for particle transport

Abstract

The invention provides a simulation method and system for particle transport. The simulation method for particle transport comprises the following steps that a simulation object, an interest area, a beam restriction device and a source are arranged; source particles are generated by sampling, and information of the generated source particles comprises the weight, direction, position and type of the source particles and comprises one of the energy and the speed of the source particles; information of the source particles and secondary particles penetrating through the beam restriction device is acquired; the secondary particles are generated by performing sampling in the interest area, and the transport process of the source particles and the secondary particles is simulated; a transport result is output, wherein a super uniform random number sampling method is adopted as the method for generating the source particles by sampling and generating the secondary particles by sampling in the interest area, and the super uniform random number is a sequence obtained through the steps that an interval is equally divided to obtain multiple numbers and the order of the numbers is disrupted.

Description

technical field [0001] The invention relates to the field of particle transport simulation, in particular to a particle transport simulation method and system. Background technique [0002] In the field of particle transport simulation, Monte Carlo (MC, also known as stochastic simulation method) is widely used to simulate the particle transport process. There are two main factors to consider in the simulation process: the simulation accuracy and the time it takes. Therefore, improving the accuracy of Monte Carlo simulation and reducing the simulation time are the long-term goals of particle transport simulation. [0003] The first way to improve the simulation accuracy is to increase the number of simulated particles, but the accuracy improvement brought by the increase in the number of particles is not obvious, but the computational resources consumed are unbearable. Most of the time spent in the Monte Carlo simulation process comes from the time spent in the random samp...

AI Technical Summary

Problems solved by technology

However, the quasi-random number technology in the prior art has natural defects, and it is only applicable to regular geometric shapes, especially geometric shapes with symmetry

[0007] It can be seen that the use of pseudo-random or true random numbers in Monte Carlo simulation is not affected by the geometric shape, but its uniformity is poor, and the uncertainty of the simulation is proportional to the -1 / 2 power of the number of particles. The uniformity of random numbers is relatively good, and the uncertainty of the simulation can reach the -1 power of the number of particles. Under the same number of particles, the uncertainty of simulation with pseudo-random numbers is relatively higher than that of pseudo-random numbers or true random numbers. The random number should be reduced a lot, that is, if the same uncertainty requirements are met, the use of quasi-random numbers for simulation can greatly reduce the number of particles used, but quasi-random numbers are only suitable for regular geometric shapes, so the scope of use is very wide. largely restricted

Images