TY - CONF A1 - Franke, Ariane A2 - Uhl, Christian A2 - Walter, Michael A2 - Stiehl, Annika T1 - Electricity load forecasting for industrial microgrid and load management T2 - Applied Research Conference 2022 Conference Proceedings: 4th July 2022, University of Applied Sciences Ansbach (Onsite Conference) N2 - As more energy systems from renewable sources and the electricity market becomes more volatile, new solutions to ensure the security of supply are tested. Microgrids offer a possibility to prevent construction downtime. In order to use the energy sources and storage facilities of the microgrid effectively, load forecasting algorithms are essential. Thus, in this study a short term load forecasting model for a construction company is designed, that will be used for managing a microgrid as well as the general load consumption. This paper proposes the use of the XGBoost algorithm for the 36-hour forecast including predictors based on past measurements as well as information extracted from the timestamp. Moreover, it is shown that including load profiles attained by conventional methods has the potential to improve the accuracy of the model. KW - load forecast KW - XGBoost KW - load profile Y1 - 2022 UR - https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/6823 UR - https://nbn-resolving.org/urn:nbn:de:bvb:898-opus4-68231 SP - 405 EP - 409 PB - Hochschule Ansbach CY - Ansbach ER -