This study seeks to create a smart system that has the
potential to monitor and manage electricity use in homes by the Internet of
Things (IoT) technology. Energy usage within homes is increasing due to the use
of more appliances daily. This system attempts to reduce wasted power
consumption and save users money by monitoring the power consumption of all
home appliances in real time. The system makes use of smart meters and sensors
located in real homes, and captures the data for the energy consumption of
households and sends the information from the home to a central computer or to
a Cloud server. The dataset that we used for this work was a combination of
real-time IoT sensor readings measured within homes and previously publicly
available smart energy datasets so as to be able to train and test the
forecasting model. This study used machine learning forecasting generative
models' long short-term memory' (LSTM) and Extreme Gradient Boost (XGBoost) to
forecast power usage, where the results achieved a mean absolute error (MAE) of
2.8% and root mean square error (RMSE) of 3.5%. As stated in these results
highlighted the proposed approach were effective in forecasting near-term and
long-term consumption trends compared to the ground-truth. The great value of
the platform includes a friendly user dashboard showing the energy consumption
trends, peak usage hours, and alerts, should anything alarming happen to the
household electric supply. Furthermore, the platform can provide useful
information, such as optimal times to use heavy appliances to avoid
unexpectedly high bills. The platform provides full support for renewable energy
sources such as solar and wind energy, and can operate irrespective of the
homeowners' internet speeds.
Author(s) Details
Rajendra M Jotawar
Department of MCA, Acharya Institute of Technology, Bangalore-560107,
India.
Pragathi K N
Department of MCA, Acharya Institute of Technology, Bangalore-560107,
India.
Vathsala M R
Department of MCA, Acharya Institute of Technology, Bangalore-560107,
India.
Please see the book here :- https://doi.org/10.9734/bpi/mono/978-93-88417-94-5/CH5
No comments:
Post a Comment