Real-time electricity price prediction
Having the ability to predict future electricity price proposes an interesting strategy to electricity consumption. One can increase his usage during time of low prices and reduce the usage when prices are high to achieve the optimal cost efficiency. However, the lack of correlation of electricity p...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/138607 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Having the ability to predict future electricity price proposes an interesting strategy to electricity consumption. One can increase his usage during time of low prices and reduce the usage when prices are high to achieve the optimal cost efficiency. However, the lack of correlation of electricity prices in Singapore has made predicting it using other known factors a difficult problem. Singapore has only recently opened its electricity retail market to everyone in 2018 and most research done on this market has been using statistical methods. In this project, we will be utilising the Multilayer Perceptron to model the electricity price market and try to forecast the price of the next 10 days while comparing it to other statistical methods. Experiment was done to find the most optimised parameters in building the neural network using machine learning libraries in Python. Our neural network model was able to successfully predict the trend of the future price, but more experimentation must be done to detect outliers and predict a more accurate price value. |
|---|