Energy demand prediction is a relevant problem whose solution is evaluated by policy makers in order to take key decisions affecting the economy of a country. A number of previous approaches to improve the quality of this estimation have been proposed in the last decade, the majority of them applying different machine learning techniques. In this project, we will investigate the performance of a robust hybrid approach, composed of a metaheuristic procedure and a neural network. On the one hand, the metaheuristic approach is focused on obtaining the most relevant features among the set of initial ones. We will explore how different models (linear, exponential, etc.) are suitable for this problem.
n this project, we will focus on indoors and large spaces (i.e. hospitals, universities, shopping centres, educational centers, galleries and tunnels). These spaces can be structured spaces or diaphanous spaces where energy monitoring can play a key role in management of emergency situations in the SMART ENERGY program.This research will developa a platform based on Internet of Things (IoT) technologies and accessible interfaces capable of operating indoors with hostile conditions associated with an emergency situation
We are interested in got talent for research and implementation of guidance systems sensory. Systems that physiologically monitor the user to make decisions in the navigation in different spaces in emergency situations. These spaces can be structured spaces or diaphanous spaces where energy monitoring can play a key role in management of emergency situations in the SMART ENERGY program.
Analysis of the behavioral patterns of customers in transportation networks where there exist different alternatives for the recharging of electric/hybrid vehicles, and where the arrival rate of customers is variable.
Recent advances in signal processing algorithms and digital signal processors have allowed to implement many digital applications. These advances can be applied to energy management systems. These systems need to process enormous amounts of information in order to identify various problems and make management decisions more accurate.
Accelerating Application-specific Energy-related Algorithms with High Performance Computing graphic cards
In different research areas in the Energy Sector there are problems to solve that many times involve complex algorithms. Some of these algorithms can benefit from the fact that there can be a great reduction in their processing times when using the High Performance computing power of GPUS (Graphic Proccessing Units).
Smart energy management system for Secure HANs (Home Area Networks) that integrates with Smart grid architectures
Energy management system that integrates Smart Grids with Home Area Networks (HAN) that use smart appliance, home automation and Internet of Thing devices must address security and privacy features. That system must enable energy efficiency, energy service demand and save energy through monitoring and controlling energy in real time.
The systems based in spectral images has been proved as a very powerful tool in many different applications such as biotechnology, high precision agriculture, remote sensing, satellite image systems, artificial vision systems, among others. This research line is devoted to the study and development of systems for sensing spectral images based on smart sensors. The main goal is to design sensing systems appropriate for each application but with the reduction of the energy consumption in the sensing process in mind and with computing systems that can analyze the data in all spectral bands efficiently
Scientific data mining is catching attention since it can find answers in data where traditional methods are incapable. Lately, computing learning techniques are applied to new virtual scenarios such as medical and biological data. By applying datelining techniques, this project aims to find a relationship between the material and decisions made during the production of the devices to get decision support to produce high-efficient synthetic photosynthesis devices.
Energy production directly impacts on climate change. Fossil fuels combustion means close to 50% of the Spanish electricity mix. Spain is the European country with most kWh/m2 average yearly irradiance received, with also high wind resources. The promotion of the renewable energies would potentially save important amount of emissions. Developing an algorithm for energy production optimisation could save tons of carbon dioxide emissions, between other highly toxic gasses. Therefore, we aim to build an optimisation algorithm to calculate the best option for the greenest development of the energy sector which would define a new Spanish electricity mix, by proposing the alternative scenario where most of the produced electricity would come from renewable generators