The intrinsic law of epidemic spread and evolution under crowd control mode is explored based on the improved Markov chain Monte Carlo method (Metropolis–Hastings method). First of all,?based on the classical SEIR?model, ?the?SEDIQR model is established, and the model?parameters?by?Markov?chain Monte Carlo method is established. At?the?same time, the convergence of Markov chain is diagnosed by Geweke method,?and?the?expression of effective regeneration number of epidemic situation?was deduced?by?using?basic regeneration matrix theory. Then, the epidemic data of South Korea is used for empirical analysis and?further analyzed?the?strength?of management and?control. The results show that the theoretical value obtained by the model is in good agreement with the actual value, and the?changing?trend of effective regeneration number can reasonably reflect the change of epidemic spreading ability.Besides,reducing the?number of initial?susceptible?population?and effective contact rate can reduce the peak value and peak time of the number of confirmed?patients?and infected?patients,?and shorten the time?needed to reduce?the?number of?effective regeneration?to one. And the number of initial susceptible?population?and effective contact rate are important sensitive parameters?affecting?the development of?epidemic, controlling the number of initially susceptible persons at the same range has a greater effect on the prevention and control of the epidemic than controlling the effective exposure rate. Under?the new crowd control?model, the prediction results of epidemic spread model?are more accurate than those predicted by previous?literature.Meanwhile,the?intrinsic?law of epidemic spread and evolution?has?important reference value for the administrative departments of government and?related?medical experts to formulate and revise epidemic prevention programs, and for the world in the prevention of similar epidemic provides a theoretical basis.