Power Consumption & Weather Data Analysis

This project analyzes power consumption patterns alongside weather data for various cities in Andhra Pradesh, India. Using Python and Jupyter Notebook, it demonstrates how to preprocess, merge, and analyze time-series power and weather datasets, visualize their correlations, and draw actionable insights for energy demand forecasting.

Table of Contents

• Project Overview
• Live Demo
• Data Sources
• Installation
• Usage
• Workflow & Features
• Visualizations
• Key Insights
• File Structure
• Contributing
• License

Project Overview

The notebook guides you through:

• Loading and cleaning daily power consumption and weather datasets.
• Merging weather data from 5 major cities.
• Resampling, filtering, and handling missing data.
• Visualizing trends, seasonality, and correlations.
• Extracting insights for energy demand modeling.

Live Demo

Check out the deployed application here:

👉 Click here to view the Live Demo

Direct Link: https://cv-p07.streamlit.app/

This application is hosted on Streamlit and provides an interactive interface for the analysis.

Data Sources

• Power Consumption: aplsdc_forecast_discom.csv
• Weather Data: Visual Crossing API and city-specific CSVs (Vijayawada, Guntur, Vishakapatnam, Rajahmundry, Kurnool).
• Merged Data: Cleaned and concatenated CSVs.

Installation

1. Clone the repository:

   git clone https://github.com/tejaj2005/Internvita_PV-07.git
   cd Internvita_PV-07

2. Install dependencies:

   • Python 3.x
   • Jupyter Notebook or Google Colab
   • Required libraries:

     pip install pandas matplotlib seaborn

Usage

• Run power_project.ipynb in Jupyter Notebook or Google Colab.

  Google Colab Links:

  • Notebook Part 1
  • Notebook Part 2

• Ensure all CSV data files are available in /content (for Colab) or the working directory.

• Follow the step-wise code cells for data loading, cleaning, merging, analysis, and visualization.

Workflow & Features

1. Data Loading & Cleaning:

   • Load power and weather datasets.
   • Handle missing values, drop or fill NaNs.
   • Convert date columns to datetime formats.

2. Resampling & Aggregation:

   • Resample to hourly/daily frequencies.
   • Summarize and export cleaned datasets.

3. Merging Data:

   • Merge power consumption with weather features.
   • Concatenate data from multiple cities.

4. Visualization:

   • Correlation heatmaps.
   • Scatter plots (power vs. temperature, humidity, wind speed).
   • Time series plots for trends and seasonality.

5. Analysis:

   • Identify key predictors for energy demand.
   • Explore seasonal effects and weather dependencies.

Visualizations

• Correlation Matrix: Visualizes relationships between all numeric variables.
• Scatter Plots: Power consumption vs. temperature, humidity, wind speed.
• Time Series: Trends in consumption and weather variables over time.

Key Insights

• Temperature shows strong positive correlation with power consumption.
• Humidity has moderate positive influence.
• Wind speed and pressure have weak correlations.
• Seasonality is evident in power consumption, aligning with weather trends.

File Structure

power_project.ipynb       # Main analysis notebook
aplsdc_forecast_discom.csv # Power data
Guntur_weather_data.csv    # City weather data
Vishakapatnam_2023-02-01_to_2025-02-26.csv
rajahmundry_weather_data.csv
vijayawada_weather_data_2023-2025.csv
kurnool_2023-02-01_to_2025-02-28.csv
5_city_weather_data.csv    # Merged weather data
powergrid_data_20260215_115900.csv # Power Grid data
README.md                  # Project documentation

Contributing

Contributions are welcome! Please open issues or submit pull requests for improvements, additional analyses, or new datasets.

License

This project is licensed under the MIT License.

---

Author: tejaj2005,Prashanth, Naga Vineela, Shanmukha, Nandhini
Contact: GitHub Profile