Artificial immune algorithm convergence rate estimation based on stochastic process theory

The stochastic process theory is utilized to replace the traditional eigenvalue estimation method of the state transition matrix so as to study the convergence rate of the real coded artificial immune algorithm. On the premise of satisfying the necessary condition for probability strong convergence of artificial immune algorithm, this method regards it as the sufficient condition for the general artificial immune algorithm and proposes a new method for estimating the exponential velocity probability strong convergence of real coded artificial immune algorithm. This method takes the final convergence of the optimum antibody in the population and avoids the deficiency of the conservative property of traditional estimation method. It can estimate the convergence property and convergence rate of a class of artificial immune algorithm. In the practical application of artificial immune algorithm, it has a certain theoretical reference value in optimizing the convergence rate.