Research On Monte Carlo Dynamics Simulation Method Of Particle Size Distribution Of Continues Grinding Process

Grinding process is an important part of industrial beneficiation process.To provide available mineral to the benefication process,the grinding process have to separate the mineral and gangue.The quality of the grinding product affects the quality of mineral process directly.The particle size distribution of mineral from the grinding process can be a indicator to analysis the quality of the grinding process.So to establish suitable grinding process simulation method,on the one hand it can be a model to help build the loop control method of grinding particle size distribution,on the other hand it can improve the grinding production condition by modify the grinding process control method.At present,the general grinding simulation method is the numerical approximation method,but it is difficult to control in the accuracy,and can not accurately simulate the evolution of mineral particles in grinding process.Due to the advantage of Monte Carlo algorithm is that the MC's discrete characteristics is suitable to describe the discrete phase system and can be realize by the compute program easily,so that the grinding process simulation method that based on MC algorithm is receiving increasing attention.Now the MC algorithm on the grinding process simulation is mainly used to simulate the batch grinding problem.But the truth is that the common state of ball mill working in is the continuous grinding pattern.So the reasearch on the continuous grinding process simulation based on the MC algorithm will have practical significance in engineering.Aiming at the problem that lack of high accuracy simulation method of the continuous grinding,This paper provides a continuous grinding simulation method that based on the MC algorithm As a kind of distributed parameter simulation method,this method can fill in the field of the accuracy method of the continuous grinding simulation.The main idea of the method is based on the discrete difference method,first take the ball mill in axial section,then analysis the random of each subsegment of the mill to structure a continuous probability model of the continuous grinding process,last by the sampling method to simulate the continuous grinding.The simulation results show that the MC method have a high accuracy and can reflect the randomness of the continuous grinding process.Though the precision of the MC method is higher,the efficiency is low.So the method does not apply to requirements for accuracy is not high,but need to improve the efficiency of the simulation problem.By drawing from the ?-Leap acceleraton method of the batch grinding process,this paper proposed the ?-Leap acceleration method to simulate the continuous grinding process.The experimental results show that although the ?-Leap method lost some accuracy compared with the MC method,but the simulation efficiency has been greatly improved.