Sari95/Transformer-Architecture-for-Energy-Consumption-Prediction

Transformer-Architecture for Energy Consumption Prediction

Description

This model applies Transformer architecture to predict energy consumption over a 48-hour period using historical energy usage and weather data from 2021 to 2023.

Model Details

Model Type: Transformer
Data Period: 2021-2023
Variables Used:

1. Transformer with Energy consumption data and weather data
2. Transformer with Energy consumption data and two additional variables: 'Lastgang_Moving_Average' and 'Lastgang_First_Difference'

Features

The model uses a sequence length of 192 (48 hours) to create input sequences for training and testing. Positional encodings are added to the sequences to provide temporal information to the Transformer model.

Installation and Execution

To run this model, you need Python along with the following libraries:

• pandas
• numpy
• matplotlib
• scikit-learn
• torch
• gputil
• psutil
• torchsummary

Steps to Execute the Model:

1. Install Required Packages

2. Load your Data

3. Preprocess the data according to the specifications

4. Run the Script