Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Forecasting household monthly electricity consumption using the similar pattern algorithm
This article discusses forecasting the monthly electricity consumption time series of household subscribers using the similar pattern algorithm, which uses each subscriber’s unique consumption pattern. The primary goal of forecasting the monthly consumption of household subscribers is to issue monthly bills and cluster subscribers for consumption management planning. The study’s results demonstrate the efficiency of the proposed algorithm compared to traditional statistics and machine learning methods. Out of the 72,000 predictions made using the proposed algorithm, the mean absolute percentage error is 16.9%. This shows that the algorithm successfully forecasts the monthly consumption of subscribers. The similar pattern algorithm maximizes the use of information in the billing database and takes advantage of expert opinions, which play a crucial role in avoiding unusual forecasts. The household subscribers involved in this study are located in two climatic regions: temperate and tropical. As this study has shown a significant dependency of household electricity consumption on the two major climate types, the obtained results have the potential to be generalized to other climatic regions as well.
Forecasting household monthly electricity consumption using the similar pattern algorithm
This article discusses forecasting the monthly electricity consumption time series of household subscribers using the similar pattern algorithm, which uses each subscriber’s unique consumption pattern. The primary goal of forecasting the monthly consumption of household subscribers is to issue monthly bills and cluster subscribers for consumption management planning. The study’s results demonstrate the efficiency of the proposed algorithm compared to traditional statistics and machine learning methods. Out of the 72,000 predictions made using the proposed algorithm, the mean absolute percentage error is 16.9%. This shows that the algorithm successfully forecasts the monthly consumption of subscribers. The similar pattern algorithm maximizes the use of information in the billing database and takes advantage of expert opinions, which play a crucial role in avoiding unusual forecasts. The household subscribers involved in this study are located in two climatic regions: temperate and tropical. As this study has shown a significant dependency of household electricity consumption on the two major climate types, the obtained results have the potential to be generalized to other climatic regions as well.
Forecasting household monthly electricity consumption using the similar pattern algorithm
Bistoon Hosseini (Autor:in)
2025
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
Metadata by DOAJ is licensed under CC BY-SA 1.0
District household electricity consumption pattern analysis based on auto-encoder algorithm
| BASE | 2019
Characterization of Household Electricity Consumption in Uruguay
| TIBKAT | 2024