Recap of three profiles (ContourGlobal/Arrubal, TotalEnergies/Castejón, Castleton/Amorebieta) and the unifying thesis: location determines whether a CCGT participates in Operación Reforzada redispatch at all; within a participating cluster, the owner decides whether you win or lose. Renewable-saturated nodes select; the rest of the market is a footnote.
Third of the post-blackout winners series. Amorebieta (Vizcaya, owned by Castleton Commodities, represented in OMIE by Axpo) operated at saturation — 744 hours/month, the theoretical max — for 5 consecutive months after the blackout. The outlier in the winners club: not 'hibernation → base' like Arrubal, but 'active → 100% all summer → back to normal'. And next door, Bahía de Bizkaia collapses.
Spain's midday solar glut makes noon the cheapest hour of the day, and batteries are scaling fast to soak it up. But the market where they earn most — secondary-regulation band (aFRR) — stopped being reported per unit on 2024-11-20, exactly as the storage boom began. The only battery revenue still visible per unit is energy arbitrage: profitable, but the minority stream.
Almost no "chat with your data" demo answers the key question: whose permissions does the agent query with? How we built a production data agent for a Spanish energy company where every question travels with the user's identity all the way to the lakehouse — one MCP server, three clients (M365 Copilot, Claude, VSCode), and OAuth 2.0 On-Behalf-Of preserving RLS end to end.
In most companies, answering a business question with data takes days: ticket, analyst, query, Excel. How a Spanish energy company cut it to minutes — any employee asks in their own language, from Teams or Claude, and gets the answer with their own permissions. No new passwords, no bypassing governance.
Second of the post-blackout winners series. TotalEnergies' Castejón 3 (Navarra) was a rare peaker — under 1 GWh per month at €600+/MWh emergency prices. Post-blackout it runs 27-43 GWh/month at €250/MWh, in line with Arrubal and Escatrón. Three plants, three owners, same grid signal.
First in a series profiling the winners of Spain's post-blackout redispatch. Arrubal 1+2 (800 MW, La Rioja) operated near hibernation pre-blackout — under 1 GWh/month average. Twelve months on, the same plant runs 24-34 GWh/month at €190/MWh. The plant didn't change. The grid did.
Two Spanish CCGT complexes share the same Endesa/Naturgy ownership split — and produce opposite results under Operación Reforzada. In Barcelona, Naturgy's unit wins; in Cádiz, Endesa's dominates. The difference is geography: mixed clusters reward the operator, uniform clusters reward the location.
Twelve months after the Iberian blackout, the pay-as-bid Spain pays its CCGT fleet for solving day-ahead technical constraints (consumer-side RT3) is €3,870M per year. Three groups absorb 62% of the flow. Iberdrola is the marginal winner (+€152M post vs pre); Endesa, the only major loser (-€100M).
Drilldown to the plant level: post-blackout pay-as-bid for solving technical constraints does not reward all combined-cycles equally. Within the Besós complex in Barcelona, unit 4 (Naturgy) gains €28M while unit 5 (Endesa) loses €71M — same site, two owners, opposite trajectories. The cluster pattern: south (Cádiz/Huelva) all winners, north interior all winners, Mediterranean coast and SE mixed.
Spain's technical-constraints cost (compodem RT3) has multiplied tenfold since 2020. Quantitative analysis over 148 months: Pearson correlation +0.76 with the share of variable generation (PV + wind), controlled for OMIE wholesale price. Corroborated by PwC's February 2026 report on the post-blackout Reinforced Operation.
365-day pre vs 365-day post-blackout comparison across five dimensions: the Iberian power system absorbs the same renewable share, but the cost of technical constraints (compodem RT3) has multiplied by 2.5 and negative-price hours by 4.4 — driven by REE's post-blackout Reinforced Operation.
Install the library, write a SQL query, get a DataFrame. No file downloads, no Excel parsing — just clean I90 data from a daily-updated ClickHouse backend.
Endesa, Iberdrola, Repsol, Gas Natural — each company has a different generation portfolio. I90 settlement data reveals their strategies across market programs.
PDBF, PDVP, PHF1, RR, BT — each market program has its own logic, actors, and economics. I90 data shows who participates in each and how much they earn.
Follow Algeciras 3 through all 8 market programs in a single day — from day-ahead scheduling to real-time balancing — with real I90 settlement data.
Same hour, five technologies — nuclear, hydro, gas, solar, and wind — each with a radically different market fingerprint. Real I90 settlement data reveals how each technology type participates in Spain's market cascade.
Data analysis of how nuclear and solar PV exploit the PDVP mechanism in Spain's electricity market — who wins, who loses, and why the system design creates perverse incentives.
Install the python-entsoe CLI and drop a skill file so Claude Code can query prices, load, generation, and transmission data across 30+ European countries.
Install the python-esios CLI and drop a skill file so Claude Code can search, fetch, and analyze ESIOS indicators from the terminal.
A Python library that wraps the ESIOS API into simple method calls — search indicators, fetch historical data, and get DataFrames with automatic caching.
Download, parse, and analyze I90 files from REE using Python — curtailment trends, generation mix, and price correlations for the Spanish electricity market.
Install the python-eia CLI and drop a skill file so Claude Code can explore, fetch, and analyze EIA data from the terminal.
A Python library that wraps the EIA API v2 into simple method calls — get generation mix, demand, interchange, and natural gas prices as DataFrames.
A Python library to download electricity prices, generation, and cross-border flows from ENTSO-E. Install with pip and start analyzing European energy markets.
Simulate a full year of constrained battery operation to understand profit variability and realistic annual expectations.
Analyze electricity prices to identify profitable charge/discharge windows for battery storage systems.
Access real-time and historical day-ahead electricity prices from the Spanish electricity market operator—no API key required.
"Kernel died" when using pandas; that's when polars comes in handy.
Having a database that needs to be updated with new data published every day, how to automate script execution with GitHub Actions?
Understand the process of creating a normalized database to query historical information of energy programming units in the Spanish electrical system with practical examples.
Build robust database systems tailored to your business needs for seamless data storage, retrieval, and scalability.
Understand the steps to automate the preprocessing of I90 Excel files, detailing the operation of energy facilities according to daily generation programs.
Transform your Excel calculations into robust, scalable Python scripts and pipelines with Excel Automations.
Explore how AI Agents can automate script organization, handle inputs/outputs seamlessly, and execute commands through a chatbot interface, making your workflow more efficient and effortless.
New edition of the course "AI Trading Agents Integrated with News and ML Models".
New edition of the course "Applied Python: Data Manipulation and Visualization".
An in-depth analysis of the FAERS Q1 2024 data, focusing on drug frequency and adverse reactions.
New edition of the course "Applied Statistical Programming: Academic Projects", in collaboration with the Salamanca Association of Pharmacy Students (ASEF).
Learn how to access the European Medicines Agency's (EMA) European Union Drug Safety Surveillance (EUDRA) data through quarterly reports and APIs for post-market drug surveillance.
Learn how to access the FDA’s Adverse Event Reporting System (FAERS) data through quarterly reports and APIs for post-market drug surveillance.
Saturday, October 5, 2024.
This year, I will be collaborating with student associations from universities to organize live courses on the application of Data Analysis, Machine Learning and AI models using Python/R programming.
An quick example on how to use ChatGPT programmatically with Python.
Housing prices in Spain by Autonomous Communities since 2020 according to the National Statistics Institute (INE). Includes indices, quarterly variations, annual variations, and year-to-date changes.
Learn how to calculate the components of a solar self-consumption simulation to find the optimal power of solar panels to install.
Discover how to analyze and visualize MIBGAS contract prices using Pandas and Matplotlib. This tutorial guides you through data filtering, date handling, and interpolation techniques.
Weekly report (July 15 - July 21, 2024) of the market price spread in OMIE markets.
If you've invested valuable time developing your Machine Learning model, why not take advantage of it and put it to work calculating predictions in a web calculator?
How to differentiate technical professionals who truly understand from those who just grasp concepts: charlatans.
Which library should you use for data visualization in Python? Matplotlib, Seaborn, or Plotly? Learn the main differences between them and when to use each one.
Learn how to process multiple financial assets and include them in a report highlighting each asset's annual performance.
Using Bollinger Bands, we can create a trading strategy to buy and sell stocks based on the market's moving volatility.
Discover how to explain the influence of explanatory variables in a Machine Learning model that detects anomalies using SHAP values.
Learn how to automate trading strategies by programming buy and sell orders with Python using the Alpaca API.
Learn to statistically analyze the stock performance of a company with Python.
Automate your European energy analyses with Python and the ENTSO-E API. We explain step by step with practical examples.
Learn to integrate a Machine Learning model into an investment strategy and evaluate its performance using the backtesting.py library with Python.
Leverage the properties of DatetimeIndex in Pandas for more efficient time series analysis, from formatting the column to creating reports with pivot tables.
Learn to detect anomalies in time series with Python, using advanced techniques and Machine Learning algorithms.
Python tutorial to unstack the row categories into columns (long to wide table) to later create a heat matrix.
Learn how to highlight the most valuable cells in a Pandas pivot table that summarizes information on billionaires by country and industry.
Step by step, you'll learn how to download, preprocess, and visualize data from the Spanish Electric Grid's API using Python.
Understand the structure of the EIA API and learn how to use it with Python to automate the downloads and exports into an Excel.
A step-by-step guide to automatically download, export and visualize economic data from the St. Louis Fed using their API with Python.
Processing a new variable to distinguish the periods of the 2008 financial crisis improves the statistical relationship between Mortgage Rates (MR) and Inflation (CPI).
Analyzing the balance between potential and actual energy production of Spain's energy portfolio.
We use cookies to improve your experience and analyze site usage. You can manage your preferences at any time.