This thesis presents the development of an algorithm that combines particle swarm optimization (PSO) and the basic ideas of model predictive control (MPC) with the aim of minimizing household electricity costs while ensuring energy system stability. Although the algorithm uses a rolling horizon approach similar to MPC control, in this thesis, we do not use a classic MPC controller, but rather schedule optimization assuming known future data. The algorithm optimizes the time intervals of battery storage charging and discharging, electric vehicle charge scheduling, and heat pump power modulation based on predefined system parameters and realistic input data such as household consumption, generated electricity, electricity price, outdoor temperature, and consumption and presence of an electric vehicle.
The algorithm was implemented in the MATLAB environment and tested on short-term (48 hours) and long-term (30 days) time horizons. Simulation results demonstrate that the developed algorithm, compared to a simple rule-based control method, achieves up to about 10 % lower monthly costs. Further analysis shows the effects of individual parameters and input variables such as battery storage capacity, solar power plant size, electricity price, seasonal changes, and other relevant factors.
The results confirm that the optimization-based algorithm enables more efficient use of self-generated energy, reduces grid dependency, and serves as a useful tool for the design of household energy systems.
|
