Robust decentralized model predictive load-frequency control design for time-delay renewable power systems

A robust decentralized model predictive control (DMPC) design is proposed for frequency stability of hybrid renewable power systems considering high renewables energy penetration and nonlinearity effects. The Egyptian power system (EPS) considered as a test system comprises both traditional power stations (i.e., steam, gas, combined cycle, and hydraulic power plants) and renewable energy sources (RESs). Where the considered RESs contain both the wind power generated from Zafarana and Gabel El-Zeit wind farms and the solar power generated from Benban solar park, which is considered one of the world's largest photovoltaic (PV) plants. To obtain an accurate insight into a real modern power system, this research takes into account the effects of the important nonlinearity such as generation rate constraints (GRCs), governor deadband (GDB), and communication time delay (CTD). The designed control is set based on the DMPC for each subsystem independently to ensure the frequency stability of the whole system as each subsystem has different characteristics and operating constraints than the others. Moreover, the decentralized control scheme has become imperative for large power systems due to the high cost of transmitting data over long distances and the probability of error occurrence with the centralized control scheme. To verify the effectiveness and robustness of the proposed DMPC for the EPS, it is compared with the centralized MPC (CMPC) scheme in different operating conditions. The simulation results, which are conducted using MATLAB/SIMULINK (R) software, emphasized that the proposed DMPC scheme can effectively handle several load disturbances, high uncertainty in the system parameters, and random communication delays. Hence, it can regulate the grid frequency and ensure the robust performance of the studied renewable power system with high RESs penetration and maximum communication delays in the system.