Solar Forecast Email System (Minimal)

This is a minimal branch containing only the code required for the solar forecast email workflow.

Features

• Solar production forecasting for CRIPVSOL ENERGY SRL - Unirea (2.9 MW AC, 2.916 MW DC)
• Excel report generation with 15-minute and hourly data
• Automated email delivery via Zoho Mail

Quick Start

Using UV (Recommended)

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

# Run the complete workflow
uv run python scripts/run_forecast_and_email.py

Using Traditional Python

# Install dependencies
pip install -r requirements.txt

# Run the complete workflow
python scripts/run_forecast_and_email.py

Individual Steps

# Step 1: Generate forecast
uv run python scripts/run_intraday_cm.py

# Step 2: Create Excel report
uv run python scripts/export_forecast_to_excel.py

# Step 3: Send email
uv run python scripts/send_forecast_zoho.py

Docker Deployment

# Build and run
docker-compose up -d

# Run once
docker-compose --profile manual up solar-forecast-once

# Check logs
docker-compose logs -f solar-forecast-email

Configuration

1. Ensure scripts/email_config_zoho_working.json exists with valid Zoho credentials
2. The system is configured for CRIPVSOL ENERGY SRL (45°06'41.76"N 27°47'04.56"E / 45.1116°N, 27.7846°E, 2.9 MW AC)
   • Panels: Canadian Solar CS6W-540 540W x 5,400 panels
   • Inverters: Huawei SUN2000-215KTL-H0 185kW x 16 units
   • Location: Unirea, Brăila, Romania

Output Files

• Forecasts: data_output/intraday/cm_forecast_*.csv
• Excel Reports: data_output/intraday/cripvsol_unirea_forecast_*.xlsx
• System State: data_output/intraday/cm_forecast_system_state.json

Email Recipients

• bordeivlad@gmail.com
• vpinteay@gmail.com

Minimal Dependencies

• pandas, numpy, pytz
• catboost, scikit-learn
• requests, pvlib
• openpyxl (Excel export)
• ephem (solar calculations)
• schedule (scheduling)