HI IBERIA INGENIERIA Y PROYECTOS S.L.
Imagen
1
Imagen
2
Imagen
3
Imagen
4
Pic legend 1: HI Iberia IA projects presentation
Pic legend 2: DONES Infographics
Pic legend 3: HI Infographics
Pic legend 4: Green Infographics
General information
-
Company nameHI IBERIA INGENIERIA Y PROYECTOS S.L.
-
AdressC/ Juan Hurtado de Mendoza, 14
-
Turnover3.80 million EUR in year 2023
-
Employees108 in year 2023
-
SMEYES
-
Contact Info:
-
Phone91 458 51 19
-
Emaildones@hi-iberia.es
Activity and Skills
At HI Iberia (HIB), we bring over two decades of expertise in AI development tailored to meet the rigorous demands of Big Science facilities. Our competencies are anchored in deploying advanced AI, machine learning, and deep learning techniques to enhance operational efficiency, predictive modeling, and real-time data analytics in complex scientific environments.
Capabilities for Big Science Facilities
1. Advanced Data Analytics and Machine Learning: HIB applies AI for predictive maintenance and fault detection in large-scale scientific equipment, ensuring uptime and efficiency. Projects like MAPRE and DONES-FLUX showcase their expertise in handling complex data streams and optimizing critical infrastructure, including particle accelerators.
2.AI-Driven Materials Discovery: HIB accelerates material discovery with AI, using deep learning and reinforcement learning. Projects like SMART MATERIAL are essential for rapid material innovation in fusion reactors and accelerator technologies.
3. Simulation and Predictive Modeling: HIB uses AI techniques like Deep Learning Surrogate Models (DLSMs) and Fourier Neural Operators (FNOs) to enhance predictive modeling for fusion facilities, improving efficiency and accuracy in physical processes while reducing computational costs.
4. Computer Vision and Image Analysis: HIB has developed real-time computer vision solutions for monitoring and analyzing visual data, as demonstrated in the LIFEonLive project, which are crucial for anomaly detection in Big Science facilities.
5. AI for Energy Optimization: HIB optimizes renewable energy integration and energy flow within fusion research facilities. Projects like ENIGMA and DONES-FLUX focus on maximizing energy efficiency and reducing costs in large-scale scientific environments.
6. Intelligent Infrastructure Management: HIB creates AI-based models for managing and optimizing infrastructure in complex scientific environments, as demonstrated in DONES-MAGIA and DONES-FLUX, enhancing reliability and maintainability.
7. Natural Language Processing (NLP) and Conversational AI: HIB develops NLP and conversational AI tools for better human-machine interaction, assisting researchers and operators in managing complex datasets in Big Science environments.
8. Cybersecurity and Edge Computing: HIB integrates cutting-edge cybersecurity and edge computing solutions to ensure data security and low-latency processing, vital for the secure and efficient operation of scientific facilities, especially under extreme conditions like those in IFMIF-DONES.
Contracts for Big Science facilities
No registered contracts
Relevant R&D projects
[MISIONES CDTI
] -Industrial technology research project aimed at optimizing the efficiency of a large scientific fusion facility, such as IFMIF-DONES.
(DONES-FLUX
)
(2022 - 2024)
Industrial technology research project aimed at optimizing the efficiency of a large scientific fusion facility, such as IFMIF-DONES. The project's main objective is to optimize the different flows within IFMIF-DONES, eliminating inefficiencies and reducing risk in the facility's operation and maintenance.
> Grid power flow: An intelligent system with buffering capacity and intelligent demand management will be developed to control the power flow in the grid.
> Deuteron flow: Deuteron flow control will be optimized through extraction systems, radiofrequency cavities and AI strategies, improving efficiency and reducing maintenance.
> Lithium flow: Through the development of optical systems to measure lithium curtain parameters and argon pressure in the target chamber, together with predictive assembly and control.
> Neutron flux: Implementation of a predictive control system to optimize neutron flux characteristics in the irradiation zone, enhancing operational safety and efficiency.
[MISIONES CDTI
] -Integrated digital platform for Operation and Maintenance of floating wind farms
(ePROA
)
(2022 - 2024)
Advanced technologies for the Operation and Maintenance (O&M) of floating wind farms. Its focus is on improving the functionality, sustainability and circular economy of these facilities, thus contributing to the modernization of the Spanish shipbuilding sector. One of e-PROA's key achievements is the design of an advanced Digital Twin for floating wind facilities. The integration of surrogate models with deep learning, including PiNNs (Physics informed Neural Networks) and NOs (Neural Operators), has allowed to significantly improve the accuracy and speed in the estimation of the lifetime of the structures, anomaly detection and failure prediction. The developed Digital Twin focuses on several critical components of floating wind farms, such as mooring lines, tower and float. For each of these components, specific surrogate models have been generated to facilitate load monitoring, fatigue estimation and anomaly detection.
[I+D DGAM Ministerio de Defensa
] -Predictive On-board Maintenance
(MAPRE
)
(2022 - 2024)
The MAPRE project has focused on developing an advanced predictive maintenance platform, based on Artificial Intelligence, for the Spanish Navy ships. The project has integrated two main components: A Predictor Ashore (PET) and a Predictor On Board (PAB). The PET has been designed to process data and train AI models, while the PAB, equipped with previously trained models, processes data in real time to assess the condition of the assets and predict their remaining useful life. This strategy enables predictive maintenance to be planned in advance, improving operational efficiency and reducing downtime. The MAPRE platform, with its innovative combination of AI, real-time monitoring, and Airflow architecture on Kubernetes, has achieved a significant advance in predictive maintenance for the Spanish Navy, optimizing resources and improving the operational efficiency of its ships.
[Red.ES
] -Collaborative intelligence for sustainable cities
(Green
)
(2022 - 2024)
The GREEN project is an artificial intelligence platform that seeks to develop an industrial solution to exploit IoT data by combining technologies such as Federated Learning, Blockchain and Smart Contract. Its main objective is to optimize and accelerate electric vehicle charging service transactions in a decentralized, secure and transparent way.
The platform allows to generate a predictive model of the energy demand required in charging stations (RE). In this way, charging station managers (RMS) can negotiate a price adjusted to their needs with the Electricity Grid Provider (ERP), without compromising customer data and maintaining fair competition.
[ISR12205 INARIM I+D DGAM Ministerio de Defensa
] -Maduration of the SEDA Technology Demonstrator
(MADS
)
(2022 - 2024)
The MADS project seeks to improve the SEDA technology demonstrator to provide research and software development services in the field of Artificial Intelligence systems. New satellite data sources and AI techniques will be incorporated to improve automatic processing and expand the capabilities of the SEDA system. It also focuses on additional technological objectives such as the implementation of a REST API and the definition of interfaces for the use of already trained AI algorithms in other operational contexts.
With the integration of Edge Computing technology, it will be possible to carry out data processing and analysis tasks at the place where they are generated, which will optimize decision making and improve real-time information management. SEDA will be fully parameterizable so that users can configure the location to be monitored, the area to be covered and the type of actions, and allow the configuration of model retraining actions by qualified technical personnel.
[Grid 2030 RED.es
] -ElectrIc Grid AI
(ENIGMA
)
(2019 - 2022)
The integration of emerging renewable energies into the electricity system requires the optimization of resources to address the problem of frequency stabilization. HI-Iberia in collaboration with the companies Prysma and Ingelectus proposes a paradigm shift in the control of the electricity system through the use of artificial intelligence methodologies.
The same Reinforcement Learning techniques that have been successful in defeating the world champions in chess or Go are generalized to the field of power system control. ENIGMA proposes the design of intelligent agents capable of acting as controllers in power generation plants. Replacing the current control system with a more optimized solution capable of adapting intelligently to the variability of scenarios that arise in the power grid.
[COINCIDENTE 2018
] -SatEllite Data AI
(SEDA
)
(2019 - 2022)
With the exponential increase in the number of satellites and their derived data, it is essential to have a tool to manage this information in order to make it useful. The potential of Artificial Intelligence (AI) is combined here with data processing and advances in the field of Data Fusion for the automatic analysis of satellite information.
This platform is based on a set of AI engines that allow, on the one hand, the collection, fusion and analysis of satellite data (mainly radar images) from private channels and open sources and, on the other hand, the automatic detection in the images (Remote Sensing) derived from them, of “anomalies” or objects whose identification is key when making decisions that may compromise security both nationally and internationally.
[ITEA3 Call4 cluster EUREKA
] -SeCuRe and Agile connected Things
(SCRATCh
)
(2018 - 2022)
HI-Iberia was the National Coordinator and the leader of package 1 where the Requirements and Architecture of the platform are defined. In addition, HI-Iberia's goal was to improve our methodology of work in the implementation of platforms, for this one of its main tasks in SCRATCh wasthe development of methodologies for the implementation of platforms that consider security and agility throughout the development cycle and the deployment of the same taking advantage of the proposed DevOps approach. As an additional application of the developed methodologies, through the SCRATCh project HI-Iberia expanded its technology and application of the data fusion module from different heterogeneous sources in the smart grid area.
[PENTA programme
] -MUlti-level Secur Ity for Critical Services
(MuSiC
)
(2017 - 2020)
Multi-level Security for Critical Services is a PENTA project working on the provision of scalable and certification-oriented solutions for embedded systems with ARM architectures and with mixed levels of security and criticality. A reference architecture with two separate execution levels (secure and non-secure) on the same hardware is proposed and applications will be developed to demonstrate the validity of the solution in various domains. The European project is led by ST Microelectronics France and has a duration of 36 months.
In the project, HI iberia is working on the development of tools for runtime adaptability of secure embedded scene and people recognition solutions with links to evolutions of the LIFEonLive product.
[MISIONES CDTI
] -Industrial research on strategic materials for cost-optimized, high energy density lithium-ion batteries in sustainable electro-mobility
(LiON-HD
)
(2015 - 2020)
Main objective: significantly improving the energy density of lithium-ion batteries, reducing their costs and increasing their sustainability. HI-Iberia focused on the development of AI models capable of proposing cathode materials that meet the desired characteristics. Within the framework of the project, Reinforcement Learning and generative models capable of generating materials with chosen properties were combined. This approach was based on the interaction between a generative model, which proposed materials, and a predictive one, which inferred the properties and feasibility of the suggested materials. These two models allowed a targeted and efficient search in the materials search space, with the Instituto de Ciencia de Materiales de Madrid (ICMM) experimentally testing the suggested materials. This comprehensive approach, combining advanced AI algorithms and state-of-the-art experimental techniques, enabled a massive and intelligent search for disruptive materials.
[FP7-SECURITY
] -Advanced pRotection of critical buildinGs by Overall anticipating System
(ARGOS
)
(2014 - 2015)
ARGOS aims to design and develop an innovative early warning security system for critical infrastructure protection, which is able to anticipate potential threats by improving the effectiveness of current security systems in preventing physical intrusions by expanding the “security zone” beyond the infrastructure perimeter, and to react by sharing information in a timely manner.