Recently, with the increased awareness of harmful and environmental effects of conventional energy resources, new renewable energy sources (RESs) have proved better solutions and replacement candidates. Increasing the participation levels of RESs in electricity generation systems can relieve the effects of greenhouse gases and increased temperature of the earth. However, the inherent characteristics of RESs based generation systems are different from the conventional synchronous generator-based generation, particularly from the side of inertia support. Additionally, increased installations lead to having continuously reduced power systems' inertia. Thence, designing proper load frequency controllers (LFCs) represents an efficient way to face the inertial problems of RESs-based modern power grid systems. The LFC systems are subjected to different communication delays, which exist in the LFC control loops and in the generation commands. Therefore, this paper presents an optimized design methodology for LFC methods. The proposed design methodology presents the design of integral (I), proportional-integral (PI), and proportional-integral-derivative (PID) controllers for LFC applications. The manta-ray foraging optimizer (MRFO) algorithm is proposed for the design optimization process. Sensitivity analysis against the different communication delay locations and values are made in the paper to compare and evaluate the considered control method. Additionally, the effects of intermittency property of RESs are considered in the design verifications and comparisons. The obtained results show the stability issues of existing communication delays and their margin values to preserve stable operation of power grids. Additionally, the RESs fluctuations are highly affecting the frequency response and regulation of RESs-based modern power grids.