Cost-Optimized Household Energy Management Considering Variable Electricity Tariffs and Local Energy Storage
Abstract
The drive for sustainable and green energy systems is increasing globally to combat climate change and achieve a net-zero energy system. As a result, the proliferation of renewable energy and electrified households is rising. Variability in electricity prices is also increasing due to supply-demand imbalances and more renewable energy in the grid. With increased household electrification, including battery and electric vehicle (EV) installations, households can now shift their electricity consumption to better align with renewable energy production and lower prices. This enhances sustainability, reduces costs, and supports the grid power balance. A home energy management system (HEMS) optimally distributes and uses energy within a household to achieve economic, sustainability, and efficiency goals. HEMS can significantly benefit both customers and the wider energy system by automatically adjusting household energy consumption to align with renewable energy production and low-price periods.
This thesis designs and compares two novel HEMS algorithms that control the EV and battery in an electrified household. The objective is to minimize household electricity costs while considering variable electricity prices. It is crucial for algorithms in consumer products to have short implementation and computation time, and adaptability to changing system conditions. Existing HEMS algorithms often overlook these aspects, so the new algorithms address these gaps. Additionally, since full battery control is not always possible, each algorithm is evaluated using two battery control methods: active control and point of common coupling (PCC) zero control, commonly used in consumer products.
The proposed algorithms are simulated and tested on a model of an average electrified household in Germany. The new HEMS algorithms achieve superior cost savings compared to existing algorithms. The performance is robust across different seasonal conditions, EV driving behaviors, and prediction accuracies. In addition, the algorithm implementation and computation times are minimized and the impact of PCC-zero controlled batteries on HEMS performance is evaluated for the first time.