Bat Inspired Algorithm (BIA) has recently been explored to develop a novel algorithm for distributed optimization and control. This paper proposes a multi-agent Model Predictive Control (MPC) of Load Frequency Control (LFC) based on BIA to enhance the damping of oscillations in a two-area power system. A two-area hydro-thermal system is considered to be equipped with multi-agent MPC. The proposed power system model considers generation rate constraint (GRC), dead band, and time delay imposed to the power system by governor-turbine, thermodynamic process, and communication channels. BIA is utilized to search for optimal controller parameters by minimizing a time-domain based objective function. The performance of the proposed controller has been evaluated with the performance of the conventional PI controller based integral square error technique , and PI controller tuned by GA in order to demonstrate the superior efficiency of the proposed multi-agent MPC tuned by BIA