TY - JOUR U1 - Zeitschriftenartikel, wissenschaftlich - begutachtet (reviewed) A1 - Chaaraoui, Samer A1 - Bebber, Matthias A1 - Meilinger, Stefanie A1 - Rummeny, Silvan A1 - Schneiders, Thorsten A1 - Sawadogo, Windmanagda A1 - Kunstmann, Harald T1 - Day-Ahead Electric Load Forecast for a Ghanaian Health Facility Using Different Algorithms JF - Energies N2 - Ghana suffers from frequent power outages, which can be compensated by off-grid energysolutions. Photovoltaic-hybrid systems become more and more important for rural electrificationdue to their potential to offer a clean and cost-effective energy supply. However, uncertainties relatedto the prediction of electrical loads and solar irradiance result in inefficient system control and canlead to an unstable electricity supply, which is vital for the high reliability required for applicationswithin the health sector. Model predictive control (MPC) algorithms present a viable option to tacklethose uncertainties compared to rule-based methods, but strongly rely on the quality of the forecasts.This study tests and evaluates (a) a seasonal autoregressive integrated moving average (SARIMA)algorithm, (b) an incremental linear regression (ILR) algorithm, (c) a long short-term memory (LSTM)model, and (d) a customized statistical approach for electrical load forecasting on real load data of aGhanaian health facility, considering initially limited knowledge of load and pattern changes throughthe implementation of incremental learning. The correlation of the electrical load with exogenousvariables was determined to map out possible enhancements within the algorithms. Results showthat all algorithms show high accuracies with a median normalized root mean square error (nRMSE)<0.1 and differing robustness towards load-shifting events, gradients, and noise. While the SARIMAalgorithm and the linear regression model show extreme error outliers of nRMSE >1, methods viathe LSTM model and the customized statistical approaches perform better with a median nRMSE of0.061 and stable error distribution with a maximum nRMSE of <0.255. The conclusion of this study isa favoring towards the LSTM model and the statistical approach, with regard to MPC applicationswithin photovoltaic-hybrid system solutions in the Ghanaian health sector. KW - Westafrika KW - West Africa KW - Ghanaian health sector KW - Load forecasting KW - LSTM KW - Neural network KW - SARIMA KW - Ghana KW - Medizinische Einrichtung KW - Elektrizitätsversorgung Y1 - 2020 UN - https://nbn-resolving.org/urn:nbn:de:hbz:832-epub4-16547 SN - 1996-1073 SS - 1996-1073 U6 - https://doi.org/10.3390/en14020409 DO - https://doi.org/10.3390/en14020409 VL - 14 IS - 2 SP - 22 S1 - 22 PB - MDPI ER -