Renewable Energy Consumption Analysis in the U.S.

Overview

This project analyzes renewable energy consumption patterns in the United States using data from Kaggle. The analysis explores trends across different energy sources, sectors, and temporal patterns to understand the growth and composition of renewable energy adoption.

Dataset

Source: Kaggle - Renewable Energy Consumption in the U.S.

The dataset contains historical data on various renewable energy sources including:

• Hydroelectric Power (Conventional)
• Geothermal Energy
• Solar Energy
• Wind Energy
• Biomass Energy (Wood, Waste, Biofuels)
• And other renewable sources

Requirements

pandas
numpy
matplotlib
seaborn
statsmodels
kagglehub

Install dependencies:

pip install pandas numpy matplotlib seaborn statsmodels kagglehub

Project Structure

1. Data Loading and Exploration

• Download dataset using Kaggle API
• Initial data inspection and understanding
• Exploratory Data Analysis (EDA)

2. Key Analyses

Temporal Trends Analysis

• Year-over-Year Consumption: Tracks the evolution of different renewable energy types over time
• Total Renewable Energy Growth: Demonstrates consistent growth in total renewable energy consumption

Energy Source Breakdown

The analysis focuses on five main renewable energy sources:

• Conventional Hydroelectric Power
• Geothermal Energy
• Solar Energy
• Wind Energy
• Biomass (Adjusted)

Note: Biomass Adjusted = Biomass Energy - Biomass Losses and Co-products + Waste Energy

Sector Analysis

Examines renewable energy consumption across different economic sectors to identify primary consumers and adoption patterns.

Seasonal Patterns

Monthly trend analysis reveals interesting seasonal variations:

• Solar Energy: Slight increase during summer months
• Wind Energy: Higher production in colder months due to increased wind speeds
• Hydroelectric Power: Peaks during warming periods (snow melt) and decreases during freezing temperatures

3. Correlation Analysis

Uses one-hot encoding for categorical variables (Sector) to explore relationships between different energy sources and consumption patterns.

4. Forecasting

Implements Exponential Smoothing (Holt-Winters method) to forecast total renewable energy production for the next 5 years (60 months):

• Trend component: Additive
• Seasonal component: Additive
• Seasonal period: 12 months

Key Findings

1. Consistent Growth: Total renewable energy consumption shows steady upward growth over the analyzed period

2. Dominant Sources: The analysis identifies which renewable energy sources contribute most significantly to overall growth

3. Seasonal Influence: Energy production varies by season, with each source showing distinct monthly patterns influenced by weather and climate

4. Sector Variation: Different economic sectors show varying levels of renewable energy adoption

Visualizations

The project includes multiple visualizations:

• Line plots for temporal trends
• Correlation heatmap
• Bar charts for sector comparison
• Seasonal pattern analysis
• Forecast plot with historical data comparison

Usage

# Run the complete analysis
python renewable_energy_analysis.py

The script will:

1. Download the dataset from Kaggle
2. Perform exploratory data analysis
3. Generate visualizations
4. Create a 5-year forecast

Future Improvements

• Add more sophisticated forecasting models (ARIMA, Prophet, LSTM)
• Include economic and policy factors in the analysis
• Compare renewable vs. non-renewable energy consumption
• Regional breakdown analysis
• Cost-benefit analysis of different energy sources