This project aims to optimize solar power generation by analyzing environmental factors that influence the efficiency of solar panels. A key focus is on the impact of precipitation on generated solar power. By understanding this relationship, we can better predict and optimize solar power generation under varying weather conditions.
The analysis is based on two primary datasets:
- Solar power generation data (
solar-panel-data.csv): Contains historical data on generated power and various environmental parameters. (https://www.kaggle.com/datasets/stucom/solar-energy-power-generation-dataset) - Weather data from INMET (
inmet-weather2023.csv): Provides detailed meteorological data, including precipitation measurements at various stations. (https://www.kaggle.com/datasets/gregoryoliveira/brazil-weather-information-by-inmet/data)
The approach involves several key steps:
-
Data Preprocessing: Cleaning and transforming data for analysis. This includes:
- Filtering and renaming columns.
- Handling missing values.
- Converting precipitation measurements from millimeters to inches.
-
Data Analysis and Visualization: Exploring the relationship between total precipitation and solar power generation using scatter plots and statistical analysis.
-
Predictive Modeling: Implementing a Linear Regression model to predict generated power based on precipitation data. The model was trained on historical data and tested on new, unseen weather data.
-
Model Application: Applying the trained model to predict generated power for specific weather conditions, focusing on data from the INMET weather station 'A771 - São Paulo - Interlagos'.