- Overview
- Objectives
- Key Features
- How it Works
- Installation and Setup
- How to Use
- Dependencies
- File Structure
- Hosted Version
- This project is centered around load forecasting for electricity demand in a power grid.
- The data used for this project was sourced from the Delhi State Load Dispatch Centre.
- The main objective was to implement an LSTM model to generate short-term load forecasts, predicting demand 24 hours into the future.
- Short-term load forecasting using LSTM model
- Data visualization using Streamlit
- High accuracy with an RSME (Root Mean Square Error) less than 1%
- The LSTM model is trained using the dataset scraped from the Delhi State Load Dispatch Centre (https://www.delhisldc.org/Loaddata.aspx?mode=28/05/2024)
- The model learns to predict the electricity demand for the next 24 hours based on the previous 10 days.
- The predictions are then visualized using the Streamlit app.
-
Clone the repository:
git clone https://github.com/CubeStar1/LoadPredictor.git cd LoadPredictor -
Create a virtual environment:
python -m venv venv
-
Activate the virtual environment:
- On Windows:
.\venv\Scripts\activate
- On Unix or MacOS:
source venv/bin/activate
- On Windows:
-
Install dependencies:
pip install -r requirements.txt
-
Run the Streamlit app:
streamlit run app.py
-
Use the Jupyter Notebook to train the model and perform data analysis.
- After setting up the project, you can use the Streamlit app to visualize the electricity demand and the model's predictions.
- You can also use the Jupyter Notebook to train the model and perform data analysis.
- Python
- Pandas
- NumPy
- Matplotlib
- Scikit-learn
- TensorFlow
- Keras
- Streamlit
- BeautifulSoup
- Jupyter Notebook
app.py: The Streamlit app for data visualizationutilities/datasets/: The dataset used for training the LSTM modelscripts/data-scraping.py: Used to scrape data from the Delhi State Load Dispatch Centreutilities/jupyter-notebook/: Jupyter Notebook used for data analysis and model training
- The Streamlit app is hosted on and can be accessed here: https://loadpredictor.streamlit.app/
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent