RSK World - Energy Consumption Dataset | Time Series Analysis | Energy Forecasting | Smart Grid Analytics | RSK World

Time Series Hourly Data Forecasting Machine Learning Python Energy Analytics

This project features a comprehensive Energy Consumption dataset designed for professional time series analysis, energy demand forecasting, and smart grid applications. The dataset includes 43,800 hourly records from 5 households with seasonal patterns, peak hour identification, and consumption trends. Includes powerful Python scripts: analysis.py for data analysis, forecasting.py for machine learning models, anomaly_detection.py for outlier detection, advanced_analysis.py for time series decomposition, preprocessing.py for data preparation, and model_evaluation.py for model comparison. Also includes multiple visualization scripts and interactive demo website. The package includes interactive demo website, comprehensive README.md, and MIT License. Perfect for data scientists, researchers, students, and developers working on energy forecasting, load prediction, smart grid analytics, and time series modeling projects.

If you find this Energy Consumption Dataset useful, you can support with a small contribution.

Secure Fast Trusted

Pay via UPI QR

Scan or tap an amount to auto-generate

₹

Open UPI app

GPay PhonePe Paytm

Download Free Source Code

Dataset Overview

Complete energy consumption dataset with hourly electricity usage patterns, smart meter data, seasonal patterns, and peak hour identification for energy analytics.

Hourly consumption data from smart meters
43,800 records covering full year 2023
5 households with unique consumption patterns
365 days of hourly data
Seasonal patterns and peak hour identification
Temperature correlation data included
Time series ready format
Ready for machine learning and forecasting
CSV and JSON formats available
Multiple analysis scripts included
Perfect for energy forecasting & smart grid analytics

Dataset Structure & Files

Well-organized project structure with energy consumption data, Python scripts for time series analysis, forecasting, anomaly detection, visualization tools, and interactive demo.

energy_consumption.csv - Main dataset file (43,800 hourly records, 5 households)
energy_consumption.json - JSON format with structured data
generate_data.py - Synthetic energy consumption data generation script
analysis.py - Basic statistical analysis and insights script
visualization.py - Create charts and visualizations script
forecasting.py - Machine learning forecasting models (Linear Regression, Random Forest)
anomaly_detection.py - Multiple anomaly detection methods (IQR, Z-score, Isolation Forest, Time Series)
advanced_analysis.py - Advanced time series analysis (decomposition, autocorrelation, trend detection)
preprocessing.py - Data preprocessing and feature engineering utilities
model_evaluation.py - Comprehensive model evaluation metrics and comparison
index.html - Interactive demo website with Chart.js visualizations
README.md - Comprehensive project documentation
ADVANCED_FEATURES.md - Advanced features documentation
PROJECT_INFO.md - Project metadata and information
RELEASE_NOTES.md - Version history and updates
ERRORS_FIXED.md - Bug fixes and error resolutions
requirements.txt - Python dependencies (pandas, numpy, scikit-learn, matplotlib, seaborn, scipy)
LICENSE - MIT License file
.gitignore - Git ignore configuration
Consistent naming convention across all files
Easy to load with pandas (pd.read_csv)
Scikit-learn ready format for ML models
Time series ready format
Multiple export formats (CSV, JSON)

Energy Analytics & Machine Learning

Complete analysis pipeline with support for time series analysis, forecasting models, anomaly detection, and energy analytics.

Linear Regression Model - Fast and interpretable forecasting
Random Forest Model - Ensemble method for improved accuracy
24-hour future forecasting - Predict consumption for next 24 hours
Model performance metrics - MAE, RMSE, R², MAPE for regression
Feature importance analysis - Identify key factors affecting consumption
Time series decomposition - Trend, seasonal, and residual components
Autocorrelation analysis - Identify patterns and dependencies
Trend detection - Mann-Kendall test for trend significance
Stationarity testing - Variance ratio and mean change analysis
Seasonality analysis - Strength of hourly, daily, and monthly patterns
Anomaly detection - IQR, Z-score, Isolation Forest, Time Series methods
Model comparison - Side-by-side performance metrics
Scikit-learn integration - Ready-to-use ML pipeline
Model training and validation - Train/test split methodology
Energy analytics utilities - Custom functions for energy data analysis
Peak hour identification - Identify high consumption periods
Household comparison - Analyze consumption patterns across households
Correlation analysis - Temperature and consumption relationships
Forecast visualization - Plot predictions vs actual consumption

Multiple File Formats

Dataset available in multiple formats for maximum compatibility with different data science tools and ML frameworks.

CSV format - energy_consumption.csv (comma-separated values, UTF-8 encoded)
JSON format - energy_consumption.json with structured data
JSON Structure - Nested format with metadata, statistics, and consumption records
Pandas DataFrame ready - Direct loading with pd.read_csv()
NumPy array compatible - Easy conversion to numpy arrays for ML
Scikit-learn compatible - Ready for train_test_split, preprocessing, models
XGBoost/LightGBM ready - Compatible with gradient boosting frameworks
TensorFlow/PyTorch ready - Can be converted for deep learning models
Time series libraries ready - Compatible with statsmodels, Prophet, ARIMA
Standard data science formats - Widely supported formats
Easy to import and process - One-line data loading
Compatible with all ML libraries - Universal format support
Jupyter Notebook ready - Perfect for interactive analysis
Python pandas ready - Native pandas support
R compatible - Can be imported into R for analysis
SQL import ready - Can be imported into databases
API integration ready - JSON format for web services
Data validation support - Easy to validate and clean data
Time series ready - Proper datetime indexing for time series analysis

Analysis & Visualization

Comprehensive analysis tools with visualization capabilities and interactive energy data explorer.

Interactive Energy Data Explorer - Chart.js powered dashboard
Multiple Visualization Charts - consumption trends, hourly patterns, seasonal analysis
Consumption distribution charts - Visualize energy usage patterns
Peak hour identification charts - Identify high consumption periods
Household comparison charts - Compare consumption across households
Time series plots - Trend analysis over time
Correlation heatmap - Temperature and consumption relationships
Feature importance chart - ML model feature rankings
Forecast visualization - Predicted vs actual consumption comparison
Anomaly detection visualization - Highlight unusual consumption patterns
Seasonal pattern charts - Monthly and daily consumption patterns
Temperature vs Consumption - Scatter plot visualization
Hourly consumption heatmap - Hour of day vs day of week matrix
Dataset statistics - Comprehensive summary statistics
Interactive Chart.js charts - Doughnut, Bar, and Line charts
Real-time chart updates - Dynamic filtering and visualization
Household and time filtering - Filter data by household and time period
Performance benchmarking - Model evaluation and comparison
Model evaluation metrics - Regression metrics (MAE, RMSE, R², MAPE)
Export visualization data - Download chart data in multiple formats
Responsive design - Works on desktop, tablet, and mobile devices

Compatible Frameworks

Works with all major data science and time series analysis frameworks out of the box.

Scikit-learn ML library - Regression, clustering, preprocessing
Random Forest models - Ensemble learning for forecasting
Linear Regression - Fast and interpretable forecasting
Gradient Boosting - XGBoost, LightGBM, CatBoost support
Deep Learning - TensorFlow, PyTorch, Keras compatibility
Time series libraries - statsmodels, Prophet, ARIMA, SARIMA
NumPy numerical computing - Array operations and mathematical functions
pandas data manipulation - DataFrames, grouping, filtering, time series operations
matplotlib visualization - Static charts and graphs
seaborn statistical visualization - Advanced statistical plots
Statistical analysis (scipy) - T-tests, correlation, time series tests
Plotly interactive charts - Dynamic and interactive visualizations
Jupyter Notebook support - Interactive data science environment
Google Colab ready - Works in cloud-based notebooks
VS Code integration - Python extension support
PyCharm compatible - Full IDE support
Energy forecasting models - Custom ML models for energy data
Smart grid analytics tools - Time series analysis for grid management
Tableau/Power BI ready - Can be imported for business intelligence
R/RStudio compatible - Cross-platform data science
SQL databases - PostgreSQL, MySQL, SQLite import support

What You Get

Complete package with all files needed for professional energy analytics and time series forecasting projects.

43,800 hourly records from 5 households (full year 2023 data)
8 Python analysis scripts - generate_data.py, analysis.py, visualization.py, forecasting.py, anomaly_detection.py, advanced_analysis.py, preprocessing.py, model_evaluation.py
analysis.py - Basic statistical analysis with statistics and insights
visualization.py - Create charts and visualizations
forecasting.py - Machine learning forecasting models (Linear Regression, Random Forest)
anomaly_detection.py - Multiple anomaly detection methods (IQR, Z-score, Isolation Forest, Time Series)
advanced_analysis.py - Advanced time series analysis (decomposition, autocorrelation, trend detection)
preprocessing.py - Data preprocessing and feature engineering utilities
model_evaluation.py - Comprehensive model evaluation metrics and comparison
generate_data.py - Synthetic energy consumption data generation
energy_consumption.csv - Main dataset file (43,800 records, UTF-8 encoded)
energy_consumption.json - JSON format with structured data
index.html - Interactive demo website with Chart.js visualizations
Multiple Visualization charts - Time series plots, consumption patterns, forecasts, anomaly detection
Complete documentation - README.md, ADVANCED_FEATURES.md, PROJECT_INFO.md, RELEASE_NOTES.md
Documentation files - Comprehensive guides and project information
requirements.txt - All Python dependencies listed and versioned
LICENSE - MIT License (free for commercial and non-commercial use)
ERRORS_FIXED.md - Bug fixes and error resolutions
Ready-to-use code examples - Copy and run scripts immediately

Interactive Demo Website

Beautiful demo website with energy data explorer, energy analytics dashboard, and comprehensive guide.

Modern animated design - Smooth transitions and visual effects
Interactive Energy Data Explorer - Filter and search consumption records
Energy Analytics Dashboard - Real-time statistics and metrics
Chart.js Integration - Interactive charts (Doughnut, Bar, Line charts)
Consumption Metrics - Visual representation of energy usage patterns
Filter by household and time period - Dynamic data filtering
Energy feature visualization - Hourly consumption, temperature, trends visualization
Consumption distribution charts - Hourly, daily, monthly breakdown
Dataset statistics display - Total records, households, date ranges
Real-time chart updates - Charts update when filters change
Step-by-step usage guide - Comprehensive instructions
Dark theme with gradients - Modern, professional appearance
Fully responsive layout - Mobile, tablet, and desktop support
Data export options - Download CSV, JSON formats
Python scripts download - Access to all analysis scripts
Interactive filters - Filter by household, date range, time period
Consumption detail view - Individual record display
Statistics summary - Quick overview of dataset metrics
No backend required - Pure HTML, CSS, JavaScript
Cross-browser compatible - Works on Chrome, Firefox, Safari, Edge

Python Scripts Included

Professional Python scripts for energy analysis, time series forecasting, anomaly detection, and data visualization.

analysis.py - Basic statistical analysis with statistics, insights, and reports
visualization.py - Create charts and visualizations for energy data
forecasting.py - Machine learning forecasting models (Linear Regression, Random Forest)
anomaly_detection.py - Multiple anomaly detection methods (IQR, Z-score, Isolation Forest, Time Series)
advanced_analysis.py - Advanced time series analysis (decomposition, autocorrelation, trend detection)
preprocessing.py - Data preprocessing and feature engineering utilities
model_evaluation.py - Comprehensive model evaluation metrics and comparison
generate_data.py - Synthetic energy consumption data generation
Time series analysis integration - Decomposition, autocorrelation, stationarity tests
Energy analytics utilities - Custom functions for energy data processing
Dataset verification - Data type checking, missing value detection, range validation
Batch processing support - Process multiple files or datasets efficiently
ML model utilities - Model training, evaluation, and prediction functions
Visualization generation - Automatic chart creation and saving
Report generation - Text-based analysis reports with insights
Error handling - Comprehensive error checking and informative messages
Logging support - Track script execution and debugging
Configurable parameters - Customize analysis settings easily
Code comments and documentation - Well-documented code for learning
Complete code examples - Ready-to-run scripts with examples
Modular design - Reusable functions and classes
Best practices - Follows Python coding standards (PEP 8)