A crucial role is played by automatic generation control (AGC) in effective performance of electrical power systems. In order to compensate for the difference between the power generated and the demand for power, an AGC system automatically restores the generator's set point with the help of a suitable governing strategy (primary controller). In this article, a novel Artificial Protozoa Optimizer-based cascaded Fuzzy PI-ID controller is proposed for automatic generation control of power systems. A two-area (PV-Reheat thermal power system and PV-Nuclear power system) is taken into account, and an optimally tuned cascaded Fuzzy PI-ID controller is implemented for AGC. A comparative analysis of optimally tuned cascaded Fuzzy PI-ID controller, fuzzy PI, and PID has been performed. For the comparison, three indicators of performance were used: undershoot, overshoot, and settling time for the frequency deviation and tie line power fluctuations. APO-based Cascaded Fuzzy PI-ID has overshoot (Hz), undershoot (Hz), and settling time (sec) of 0.128, 0, and 6.780, 0.095, 0, and 7.251 for frequency fluctuations in area-1 and area-2 respectively whereas the tie line power fluctuations in terms of overshoot (pu), undershoot (pu) and settling time (sec) are 0.017, 0, and 4.670. In terms of overshoot, undershoot, and settling time, the proposed APO-based Fuzzy PI-ID controller proves to be superior in comparison to other techniques. The MATLAB/SIMULINK environment is used for carrying out simulation.