Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Emitor: Jurnal Teknik Elektro
Rozali, devan
Unknown Affiliation
Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Electric Energy Measurement System For Energy Management Household With Convolutional Neural Network Method Rozali, devan; Eviningsih, Rachma Prilian; Ayub Windarko , Novie
Emitor: Jurnal Teknik Elektro Vol 25, No 2: July 2025
Publisher : Universitas Muhammadiyah Surakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23917/emitor.v25i2.10357

Abstract

Abstract- Short Term Load Forecasting (STLF) is becoming very important as the use of distributed energy sources, renewable energy, and demand side management increases. Electrical energy is one of the most widely used energy, especially in households. To avoid excessive electricity consumption, we propose a household electricity consumption forecasting system using Convolutional Neural Network (CNN) method. The input of CNN is the power of several household loads measured for one week at 10-minute intervals. This data is used to train the model and predict household electricity consumption for the next week. Forecasting results for a week show a difference in consumption of 3.623 kWh, while with the load management method the difference is 3.439 kWh. With an electricity tariff of Rp1.352/kWh, the estimated electricity cost for the following week is Rp4.892.00, and with load management, the cost drops to about Rp4.649.52 (5% savings).The testing method is done by comparing the forecasting results and actual data for one week. The results show an average difference of only 1.57W with an average error of 0.07%. The CNN method is also compared with the Long Short-Term Memory (LSTM) method. As a result, CNN has better performance with CNN RMSE value of 3.688, CNN management of 3.354, while LSTM RMSE of 12.603, and LSTM management of 13.132. CNN is proven to be more accurate for household short-term electricity load forecasting..
Co-Authors Ayub Windarko , Novie Rachma Prilian Eviningsih, Rachma Prilian