Prediction of Future Electric Energy Consumption using Machine Learning Framework
Juveria Khan1, Jyoti Rao2, Pramod Patil3
1Juveria Khan*, Department of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Savitribai Phule Pune University, Pune, India.
2Dr. Jyoti Rao, Department of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Savitribai Phule Pune University, Pune, India.
3Prof (Dr.) Pramod Patil, Department of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Savitribai Phule Pune University, Pune, India.
Manuscript received on January 26, 2020. | Revised Manuscript received on February 05, 2020. | Manuscript published on February 30, 2020. | PP: 3347-3350 | Volume-9 Issue-3, February 2020. | Retrieval Number: C5829029320/2020©BEIESP | DOI: 10.35940/ijeat.C5829.029320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the last few years, the expanding energy utilization has imposed the formation of solutions for saving electricity. Of many solutions, one is generating a power saving policies which is defined as prediction of energy in smart environments. This model is built, based on the idea that the building residences are provided with smart meters to monitor energy consumption and can be managed accordingly. Recent prediction models focuses on performance of the prediction, but for developing a reliable energy system, it is required to predict the demand taking into account different scenarios. In this paper we propose a model for predicting future demand for energy according to different conditions using advanced machine learning framework. In this system we have a projector that builds proper state for a particular condition and using that defined state a future power demand is forecasted by the predictor. The proposed model generates utilization predictions for every 2 hours. Demonstrating the electricity consumption data for 5 years, the proposed system achieves a better performance.
Keywords: Electric energy; energy management system; energy consumption forecasting; energy prediction; machine learning.