a passionate Data Scientist who builds data-driven software products and helps organizations transform numbers into optimal decisions.
I have analyzed highly dimensional datasets and built machine learning models, from the research stage to the implemented deployed solution. With a multidisciplinary background, I have worked in academia and industry, focusing on various topics as smart grids, wind production, risk analysis, optimal scheduling, predictive maintenance, and computer vision.
I love the coalition between machine learning modeling and operations research, sophisticated mathematical formulations and simple robust approaches, black-box modeling and interpretability, design and deployment. Thriving in dynamic environments, I enjoy dissecting complex business problems and learning new tools. I am careful, very organized, meticulous about planning always focusing on delivering the right solution.
Preferred tools: Python, Numpy, Pandas, sci-kit learn, Tensorflow, Rasterio, Gdal, Dash, R, Shiny, CPLEX, SQL, Docker, AWS, Linux, git.
Kerstin Dächert, Sauleh Siddiqui, Javier Saez-Gallego, Steven Gabriel, and Juan Miguel Morales
Networks and Spatial Economics 2019
This paper presents a corrigendum to Theorems 2 and 3 in Siddiqui S, Gabriel S (2013). In brief, we revise the claim that their L-penalty approach yields a solution satisfying complementarity for any positive value of L, in general. We also elaborate further assumptions under which the L-penalty approach yields a solution satisfying complementarity.
Luke Simmons, K. Franke, Christos Tsouknidas, Javier Saez-Gallego, E. Weyer, and Paula Gómez
TORQUE 2018
A comparative exercise for estimating the uncertainty associated with new methods for power performance measurements was coordinated by the International Energy Agency (IEA) Wind Task 32. The exercise showed significant variability among participants reflecting difficulty with the interpretation and application of the informative guidance.
Javier Saez Gallego and Juan M. Morales
In IEEE Transactions on Smart Grid 2016
This paper presents a practical solution method for generalized inverse optimization problems, which is then applied to short-term load forecasting. The price-response of the aggregation is modeled by an optimization problem that is characterized by a set of marginal utility curves and minimum and maximum power consumption limits. A case study based on the simulation of the price-response behavior of a pool of buildings equipped with a heat pump is built and analyzed.
Javier Saez Gallego, Mahdi Kohansal, Ashkan Sadeghi-Mobarakeh, and Juan M. Morales
In IEEE Transactions on Power Systems 2016
In this paper we seek to optimally operate a retailer that, on one side, aggregates a group of price-responsive loads and on the other, submits block-wise demand bids to the day-ahead and real-time markets. Such a retailer/aggregator needs to tackle uncertainty both in customer behavior and wholesale electricity markets. We derive closed-form solutions for the risk-neutral case and also provide a stochastic optimization framework to efficiently analyze the risk-averse case.
Javier Saez Gallego, Juan M. Morales, Marco Zugno, and Henrik Madsen
In IEEE Transactions on Power Systems 2016
This paper deals with the market-bidding problem of a cluster of price-responsive consumers of electricity. An inverse optimization method is used for estimating the optimal complex market bid, relative to a pool of price-responsive consumers. Results show that the price-sensitive consumption of the cluster of flexible loads can be largely captured in the form of a complex market bid, so that this could be ultimately used for the cluster to participate in the wholesale electricity market.
Javier Saez Gallego, Juan M. Morales, Henrik Madsen, and Tryggvi Jonsson
In Energy 2014
It consists of a stochastic optimization model to optimally schedule the electricity reserves in the western power system area of Denmark, based on the uncertainty related to the load, the wind power production, and to the outages of power plants.
A data-driven bidding model for a cluster of price-responsive consumers of electricity
EURO Glasgow. 27th European Conference on Operational Research & INFORMS Annual Meeting in Philadelphia
Optimal Spinning Reserve by taking advantage of probabilistic forecasting
Danish Wind Industry Annual Event, Denmark
Determining reserve requirements in DK1 area of Nord Pool using a probabilistic approach
International Conference on Computationa Management Science, Lisbon, Portugal