学术文献库

Hepatitis type B is one of the most common infectious disease worldwide that can pose severe threats to human health up to the point that may contribute to severe liver damage or cancer. Over the past two decades a large number of dynamic models have been presented based on experimental data to predict the HBV infection behavior. Besides, several kinds of controllers have been employed to obtain effective solutions from the HBV treatment. In this essay we consider the nonlinear HBV dynamic model which subjected to both parametric and non-parametric uncertainties without using any linearization. In previous control methods three HBV dynamic states should be measured virus safe, and infected cells. However in most of the biological systems, the amount of virus is experimentally measured. Accordingly, the necessity to seek a method that can estimate the amount of required drug by receiving the virus data emerges. An ANFIS method is developed in this work to provide an intelligent controller for the drug dosage based on the number of viruses together with an estimator for the amount of infected and uninfected cells. This controller is trained first using the data provided from a previous adaptive control strategy. After that to improve the closed-loop system capabilities two unmeasured state variables of fundamental dynamics are estimated through the training phase of the ANFIS observer. The results of simulations demonstrated that the accuracy of the proposed intelligent controller is high in the tracking of the desired descending virus population.