Short-Term Forecasting of Loads and Wind Power for Latvian Power System: Accuracy and Capacity of the Developed Tools
and
May 20, 2016
About this article
Published Online: May 20, 2016
Page range: 3 - 13
DOI: https://doi.org/10.1515/lpts-2016-0008
Keywords
© 2016 V. Radziukynas et al., published by De Gruyter Open
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
The paper analyses the performance results of the recently developed short-term forecasting suit for the Latvian power system. The system load and wind power are forecasted using ANN and ARIMA models, respectively, and the forecasting accuracy is evaluated in terms of errors, mean absolute errors and mean absolute percentage errors. The investigation of influence of additional input variables on load forecasting errors is performed. The interplay of hourly loads and wind power forecasting errors is also evaluated for the Latvian power system with historical loads (the year 2011) and planned wind power capacities (the year 2023).