https://github.com/drisskhattabi6/energy-consumption-analysis-and-prediction

This Repo contains Energy Consumption Analysis and Prediction using LSTM
https://github.com/drisskhattabi6/energy-consumption-analysis-and-prediction

deep-learning energy-consumption energy-consumption-data lstm lstm-neural-networks prediction predictive-analytics time-series time-series-analysis time-series-prediction

Last synced: 7 months ago
JSON representation

This Repo contains Energy Consumption Analysis and Prediction using LSTM

• Host: GitHub
• URL: https://github.com/drisskhattabi6/energy-consumption-analysis-and-prediction
• Owner: drisskhattabi6
• Created: 2024-11-02T14:51:48.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-11-02T15:00:37.000Z (12 months ago)
• Last Synced: 2025-01-26T14:49:01.775Z (9 months ago)
• Topics: deep-learning, energy-consumption, energy-consumption-data, lstm, lstm-neural-networks, prediction, predictive-analytics, time-series, time-series-analysis, time-series-prediction
• Language: Jupyter Notebook
• Homepage:
• Size: 3.05 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Energy Consumption Analysis and Prediction using LSTM

## Overview

This project implements a Long Short-Term Memory (LSTM) neural network to analyze and predict energy consumption patterns. The model uses historical energy usage data to forecast future consumption levels, leveraging deep learning techniques for time series prediction. 

![](imgs/img1.jpg)

## Features

- Seasonal pattern recognition

- Time series analysis of energy consumption patterns

- Data preprocessing and normalization of energy metrics

- LSTM model implementation using PyTorch

- Energy consumption prediction

- Performance visualization and metrics

- Model evaluation with MSE and R-squared metrics

- Peak usage detection

- Tuning the Parameters

- Comparison between LSTM and BiLSTM

## Installation

1. Clone the repository:

```bash

git clone https://github.com/drisskhattabi6/Energy-Consumption-Analysis-and-Prediction.git

cd Energy-Consumption-Analysis-and-Prediction

```

## Model Architecture

The LSTM model consists of:

- Input layer (1 features: previous consumption)

- 3 LSTM layers with 150 hidden units

- Dropout layers (0.2) for regularization

- Fully connected output layer

## Results

- Mean Squared Error (MSE): 103.4072

- R-squared (R²) score: 0.9024

- Visualization of actual vs predicted consumption:

![](imgs/img2.png)