Optimal Gaussian Nonlinear Stochastic Control of HIV-۱ InfectionBased on Particle swarm optimization Method

Using nonlinear stochastic state-space model of HIV-۱ infection, having as state variables the concentration of healthy and infected cells and the concentration of virions (free virus particles), utilized for design a control method. In this paper, a new optimal nonlinear stochastic controller is presented based on a particle swarm optimization (PSO) method to decrease the number of infected cells in presence of stochastic parameters of HIV dynamic. Particle swarm optimization gaussian nonlinear control (PSO-GNC) is a novel nonlinear robust optimal method that can control the biological characteristics of nonlinear stochastic HIV dynamic by drug dosage management. The PSO should optimize this kind of controller included three parameters. The proposed control method searches the best controller parameters domain subject to minimize a stochastic expected value of cost function. Simulation results show that the proposed PSO-GNC scheme does improve the treatment performance in compare to other control methods. For comparison with PSO-GNC method, a modified PID controller is chosen as controller structure.

Keywords:

Nonlinear stochastic control , HIV-۱ infection , Particle Swarm optimization (PSO) method , Gaussian controller (convex optimization)