Input-Output Dynamical Systems Modelling and Control for Integrated Renewable Energy Systems

Abstract

This thesis focuses on modeling and optimal control of integrated renewable energy systems (IRES). Mathematical models for three renewable energy sources and an integrated input-output dynamical systems model are developed and implemented in MATLAB/Simulink. The system combines wind turbines, solar photo-voltaic, and biomass in an IRES design. Energy storage is realized via battery technology. Model predictive control (MPC) is considered to optimize energy management and minimize the cost of energy. The cost functions utilize the output energy of the three integrated renewable energy technologies and link to a realistic load profile. A complete MATLAB/Simulink model of the controlled IRES is presented, and case study simulations considering different scenarios explore MPC/optimization performance. The MPC approach is considered due to its advantages such as a closed-loop topology for better performance for a relatively long simulation period. The thesis further explores the application of particle swarm optimization (PSO) to enhance the performance of MPC. While the MPC objective function optimizes for different objectives of the IRES simultaneously, multi-objective PSO, a nature-inspired optimization technique, is employed to search for optimal weight coefficients that minimize respective cost functions considering factors like generation, grid interactions, and battery storage. The proposed approach harnesses the collective intelligence of particles within the swarm to iteratively refine the control parameters, thus enabling the MPC to adapt to varying conditions and uncertainties inherent in renewable energy generation.