Dr. Juan Gómez-Romero will be giving a course on Deep Reinforcement Learning at the Escuela de Ciencias Informáticas) of the University of Buenos Aires (Argentina) this summer.
The paper A Probabilistic Algorithm for Predictive Control with Full-Complexity Models in Non-Residential Buildings , developed in collaboration with Acciona and the partners of the Energy IN TIME project, has been published in the IEEE Access journal.
This paper presents a novel control algorithm that uses a full-complexity simulation model to optimize HVAC operation in non-residential buildings. The system was implemented and tested in an office building in Helsinki, both in a simulated environment and in the real building, yielding energy savings around 35% during the intermediate winter season and 20% in the whole winter season with respect to the current operation of the heating equipment.
The paper Comfort as a Service: A new paradigm for residential Environmental Quality control has been published in the Sustainability journal.
This paper introduces the concept of Comfort as a Service (CaaS), a new energy supply paradigm for providing comfort to residential customers which disrupts the traditional model of paying per consumed energy in favor of a model of paying per provided comfort.
The paper Visualizing large knowledge graphs: A performance analysis, developed in collaboration with the Data Science Institute of Imperial College London, has been published in the Future Generation Computer Systems journal.
In this paper, we present a Spark-based framework for the generation of graph visualizations, and we study the performance of the different stages of the process. Our software achieves high performance and low execution time even without the use of graph optimization strategies.
Dr. Juan Gómez-Romero will be giving a keynote at the Artificial Intelligence International Conference (A2IC) on the results of the BIGFUSE project.
The paper 'Data science for building energy management: A review', developed in collaboration with the Data Science Institute of Imperial College London, has been published in the Renewable and Sustainable Energy Reviews journal.
In this work, we review how Data Science has been applied to address the most difficult problems faced in the field of Energy Management, especially in the building sector. We also discuss the challenges and opportunities that will arise with the advent of fully connected devices and new computational technologies. The main conclusion is that, despite the significant advances achieved in the last years, there is still considerable room for improvement, since most approaches only explore the general-purpose techniques already implemented in common statistical and data analysis packages.
Paper accepted in the International Journal of Approximate Reasoning on fusing data with higher-level fuzzy association rules.
Paper accepted in the Applied Soft Computing journal on the use of association rules to extract knowledge from multiple databases.