The aim of this paper is to study the dynamics of a new chronic HBV infection model that includes spatial diffusion, three time delays and a general incidence function. First, we analyze the well-posedness of the initial value problem of the model in the bounded domain. Then, we define a threshold parameter \begin{document}$R_{0}$\end{document} called the basic reproduction number and show that our model admits two possible equilibria, namely the infection-free equilibrium \begin{document}$E_{1}$\end{document} as well as the chronic infection equilibrium \begin{document}$E_{2}$\end{document} . Further, by constructing two appropriate Lyapunov functionals, we prove that \begin{document}$E_{1}$\end{document} is globally asymptotically stable when \begin{document}$R_{0} , corresponding to the viruses are cleared and the disease dies out; if \begin{document}$R_{0}>1$\end{document} , then \begin{document}$E_{1}$\end{document} becomes unstable and the equilibrium point \begin{document}$E_{2}$\end{document} appears and is globally asymptotically stable, which means that the viruses persist in the host and the infection becomes chronic. An application is provided to confirm the main theoretical results. Additionally, it is worth saying that, our results suggest theoretically useful method to control HBV infection and these results can be applied to a variety of possible incidence functions presented in a series of other papers.