Controlling Hepatitis B Virus Infection using PID like Neural Network

In this paper, an artificial neural network-based controller is used for controlling HBV system behavior. The designed controller imitates the PID controller behavior, which is known as PID like Neural Network Controller (PIDNNC). In addition to have a PID controller characteristics, our designed controller is adaptive, like neural network controllers. The considered HBV infection system has one input (drug dosage) and three outputs (uninfected cells, infected cells and free virions). The main objective of this paper is minimizing the population of free virions and infected cells. The proposed structure of the PIDNNC contains two layers. The first layer performs proportional-integral-derivative operation and the second layer is the sum of the neuron outputs of the first layer. Combining these layers results in the PID-like controller behavior. For demonstrating the performance of the proposed controller, Hepatitis B Virus (HBV) system is controlled under ideal conditions. In addition, parametric and initial condition uncertainties are considered. Moreover, noise is added to the studied system so as to investigate the controller performance under more realistic conditions.