Electrical efficiency study of alternating current and direct current power topologies in office building with integrated photovoltaics in Matlab-Simulink environment

Abstract

Electrical efficiency and the energy balance of an office building (Espoo, Finland: 60.11°N, 24.49°E) with an integrated photovoltaic system were studied using a Matlab-Simulink model. The model consisted of measured on-site photovoltaic data, measured luminaire data, and consumption data of computers and ventilation system that was obtained from online sources. Similar model was studied with four different voltages: 12 V DC, 24 V DC, 48 V DC, and 230 V AC, and the system components were selected accordingly. All components selected for the simulations were commercially available. Light-emitting diode (LED)- and fluorescent-lamp lighting were used in the simulation, and they were compared to each other. Based on the simulations, the annual energy balance of each simulated option was calculated, and the annual grid electricity costs were estimated. The results were calculated for similar systems with and without battery backup system. The energy balance of the building was studied in different months, as the seasonal differences in the available solar energy vary considerably in the location of the simulated building. The total energy consumption of the building was lowest when 230 V AC was used as the system voltage. This was mainly due to the fact that computer power supplies designed for smaller DC voltages were considerably less energy efficient, than those designed for 230 V AC. When only the energy efficiency of the lighting was considered, 48 V DC provided best results with LED lighting and 24 V DC with the fluorescent lamp lighting. The LED lighting was more energy efficient than the fluorescent lamp lighting in all studied voltages. 12 V DC system required thicker wires than the other options, and even then, the wire losses were more prominent. Battery backup system reduced the amount of bought and sold grid energy in all tested situations. Whether this is economically sensible depends on the price of the sold and bought grid energy. In this study, the initial assumption was that the price was the same regardless of the direction of energy flow, and in that case, the battery backup provided no savings, in terms of direct energy costs. The seasonal differences in the grid energy requirement were considerable. In July, the building was self sufficient in electricity production on a daily level, in April, about 80% of the electricity was produced locally, and in December, only 2%–5% of the needed electricity was produced locally.