Study Of Adsorption Phase Recognition And The Property Of Adsorption Of Polymer Based On Monte Carlo Simulation

As an important statistical simulation method,Monte Carlo method is widely used in the simulation of polymer science,statistical physics,biomedicine,finance and other fields.The Monte Carlo method can be used to simulate various microscopic states of polymers and construct polymer conformations with different probability distributions.Monte Carlo(MCMC)sampling based on Markov chain is a common sampling method,which is combined with a specific method to solve the problems that are difficult to achieve by ergodicity.The Monte Carlo method is based on probability and statistics,and solves theoretical and experimental problems that are difficult to solve through a large number of sampling.As one of the branches of machine learning,deep learning has achieved impressive results in image recognization,and a lot of projects based on deep learning have also emerged.Convolutional neural networks(CNN)are more brilliant in the field of image recognition because of its special feature extraction methods.The data objects processed by the convolutional neural network(CNN)are generally pictures,and attempts to apply them to the polymer field are still few.In this paper,Monte Carlo method is used to simulate the movement of polymer chains on the surface with adsorption.With the influence of different temperatures,the Force of polymer monomers and adsorption surfaces will change correspondingly.Therefore,it has a unique conformational property at various temperature.With the simulated annealing algorithm,the linear chain presents two-state and three-state phase changes on the homogeneous-surface and the stripe-surface corresponding.The conformation and temperature information are subjected to special processing,and the neural network is used to identify the critical phase transition of the linear chain,obtaining a higher recognition rate.The recognization of mixed samples with different states under different stripe-width surface will show different recognition difficulties,with various cross temperature.There are many differences in terms of the adsorption property of the circular chain on the homogeneous surface compared with the linear chain.The special ring structure makes its conformation structure more compact,that increases the chance it contact the surface,so the circular chain has a smaller radius of gyration and a smaller critical phase transition point relative to the linear chain.But on the scale index,they have the same trend and range at different temperatures.