SEA GROWTH: HPC Optimizing Aquaculture Productivity

Presentation of the problem and objective of the experiment

In aquaculture, lots of parameters can influence the survival and growth of fish (temperature, salt, oxygen, nutrients…). The lack of control and knowledge of the information that influences growth means the loss of millions of dollars and specimens. In aquaculture, just a few hours can make or break fish health.

Artificial Intelligence applied to data analytics empowers the understanding of what influences fish behavior, wellness and growth.

Short description of the experiment

The experiment we propose aims to cut expenses and increase the revenues of inefficient operations in aquaculture facilities.  The project proposes the use of AI-based simulations on HPC resources to improve the productivity of fish farming.

DEICOM will experiment with HPC services to create a high-performance AI-based simulation tool for providing reliable long-term predictions, to scale up ACUATIA (solution developed by Deicom for aquaculture) business. Concretely:

  • Neural network training
  • Simulation and 2- year forecast
  • Neural network hyperparameter tuning

Sea Growth targets to experiment with advanced HPC services to train the DEICOM data predictive analytics Artificial Intelligence models for assisting aquaculture processes.

Organisations involved: 

End Users: Nueva Pescanova- Insuiña and Geneaqua
Technology provider: DEICOM
HPC provider: CESGA

Partner CESGA is part of the NCC Spain.