Fetching Electricity Prices from OMIE with Python

Spanish electricity prices swing from €10 to €200 per MWh within a single week. With a simple Python function, you can fetch this data directly from OMIE—no API key required—and use it for energy analytics, battery optimization, or cost forecasting.

In this article, you’ll build a reusable function to download any date range of 15-minute resolution prices for Spain and Portugal.

Questions

1. Where does OMIE publish electricity price data?
2. How do you download and parse a single day’s prices?
3. How do you generalize this into a reusable function?
4. How do you fetch multiple days efficiently?
5. What are the common pitfalls to avoid?

Implementation

Finding OMIE data

OMIE publishes market data at omie.es/en/file-access-list. Navigate to:

1. Day-ahead market
2. Day-ahead market prices in Spain

The files follow a predictable naming pattern: marginalpdbc_YYYYMMDD.1

For example, the latest day-ahead prices might be at: https://www.omie.es/en/file-download?parents=marginalpdbc&filename=marginalpdbc_20260112.1

Download a single day

Let’s start by downloading yesterday’s prices directly:

import pandas as pd
import requests
from datetime import datetime, timedelta

yesterday = datetime.now() - timedelta(days=1)
date_str = yesterday.strftime('%Y%m%d')

url = f"https://www.omie.es/en/file-download?parents=marginalpdbc&filename=marginalpdbc_{date_str}.1"

Download: https://www.omie.es/en/file-download?parents=marginalpdbc&filename=marginalpdbc_20260111.1

Let’s look at the raw content:

MARGINALPDBC; 2026;01;11;1;108.2;108.2; 2026;01;11;2;97.72;97.72; 2026;01;11;3;93.7;93.7; 2026;01;11;4;87.99;87.99; 2026;01;11;5;90;90;

This is a semicolon-separated file with a header row (“MARGINALPDBC;”). The file contains 96 rows - one for each 15-minute period of the day. The columns are year, month, day, period (1-96), price Portugal, and price Spain. Pandas can read it directly from the URL:

df_raw = pd.read_csv(
  url, sep=';', skiprows=1, header=None,
  names=['year', 'month', 'day', 'period', 'price_portugal', 'price_spain', '_']
)

df_raw = df_raw.dropna(subset=['period'])

Parse the data

Each period represents a 15-minute interval (period 1 = 00:00, period 2 = 00:15, etc.):

df_prices = df_raw[['period', 'price_spain', 'price_portugal']].copy()
df_prices['datetime'] = pd.to_datetime(yesterday.date()) + pd.to_timedelta((df_prices['period'] - 1) * 15, unit='m')
df_prices = df_prices.set_index('datetime')

Price summary for 2026-01-11: - Minimum: 50.25 €/MWh - Maximum: 108.20 €/MWh - Spread: 57.95 €/MWh

Visualize the daily profile

Create a reusable function

Now let’s wrap this into a function that can fetch any date:

def omie_day_ahead(date):
    """
    Fetch day-ahead electricity prices from OMIE for a specific date.

    Parameters:
        date: datetime object

    Returns:
        DataFrame with columns: period, price_spain, price_portugal (96 rows per day)
        Returns None if data unavailable
    """
    date_str = date.strftime('%Y%m%d')
    url = f"https://www.omie.es/en/file-download?parents=marginalpdbc&filename=marginalpdbc_{date_str}.1"

    try:
        df = pd.read_csv(url, sep=';', skiprows=1, header=None,
                         names=['year', 'month', 'day', 'period', 'price_portugal', 'price_spain', '_'])
        df = df.dropna(subset=['period'])
        df['datetime'] = pd.to_datetime(date.date()) + pd.to_timedelta((df['period'] - 1) * 15, unit='m')
        return df.set_index('datetime')[['period', 'price_spain', 'price_portugal']]
    except Exception:
        return None

Test with a few different dates:

df_yesterday = omie_day_ahead(datetime.now() - timedelta(days=1))
df_week_ago = omie_day_ahead(datetime.now() - timedelta(days=7))
df_future = omie_day_ahead(datetime.now() + timedelta(days=7))

• Yesterday: 96 periods (96 = 24 hours × 4)
• Week ago: 96 periods
• Future date: 0 periods (expected - data not yet available)

Fetch multiple days

For analysis spanning weeks or months, we need to fetch a date range:

import time

def omie_day_ahead_range(start_date, end_date, delay=0.1):
    """
    Fetch OMIE prices for a date range.

    Parameters:
        start_date: First date to fetch
        end_date: Last date to fetch (inclusive)
        delay: Seconds between requests (be nice to the server)

    Returns:
        DataFrame with all available data
    """
    all_data = []
    current = start_date

    while current <= end_date:
        df = omie_day_ahead(current)
        if df is not None:
            all_data.append(df)
        current += timedelta(days=1)
        time.sleep(delay)

    if not all_data:
        return None

    return pd.concat(all_data)

end = datetime.now() - timedelta(days=1)
start = end - timedelta(days=6)

df_week = omie_day_ahead_range(start, end)

Fetched 672 periods across 7 days.

Common pitfalls

1. Period numbering: OMIE uses periods 1-96 (15-minute intervals). Period 1 = 00:00, period 96 = 23:45.
2. Data availability: Day D prices are published around 13:00 CET on day D-1. Don’t fetch today’s prices before they exist.
3. Rate limiting: Add delays between requests when fetching many days. OMIE may block aggressive scrapers.
4. DST transitions: On daylight saving time change days, you’ll get 92 or 100 periods instead of 96.

Conclusions

OMIE electricity price data is freely accessible via predictable URLs:

• Single day: https://www.omie.es/en/file-download?parents=marginalpdbc&filename=marginalpdbc_YYYYMMDD.1
• Format: Semicolon-separated, 96 periods per day (15-minute resolution)
• Coverage: Spain and Portugal day-ahead prices

With this data, you can build energy analytics applications:

• Electricity arbitrage opportunity analysis
• Multi-day arbitrage patterns
• Industrial battery optimization