An R&D project in collaboration with our partner Beia and the National Institute for Laser, Plasma and Radiation Physics (NILPR) (Romania).
iPREMAS – Intelligent PREdictive Maintenance for Aquaculture Systems – is a research project with the goal to improve the performance of aquaculture farms by introducing a novel platform and service for Intelligent predictive maintenance. The platform is based on innovative monitoring systems and smart infrastructure, relying on machine learning and artificial intelligence techniques. The platform measures key parameters in real time introducing innovative multisensory gauges which feed a chain of ML models for Time Series Forecasting (TSF), Anomaly Detection (AD), Fault Classification (FC) and Remaining Useful Life (RUL) estimation. The measurements give the current health status of the farm site while the forecasts provide a glimpse of future status; analysis of the predictions allow to identify the potential need of preventive/corrective maintenance. A cloud-based integration of the different components of the platform allows to improve connectivity and optimize the business process which lets the farmers benefit from a tailor-made Software as a Service solution. The SaaS approach empowers the aquaculture farmers by providing a digital twin of their facilities ‘in their pocket ‘The new service gives farmers real time access to the current health status of the farm and facilitates them planning activities and measures based on the forecasted status. In the end the iPREMAS pursues cutting O&M costs for the farmer and give additional tools for reducing negative effects to the environment in case of calamity.
The project’s objectives are three-fold: technological (iPREMAS platform and new sensors development), scientific (collection and large-scale availability of valuable data from multiple types of farms and various environmental conditions) and socio-economic (use of data to support stakeholders’ decision-making process regarding aquaculture management activities).
- Leading: preparation, setting the scene, managing and coordinating the project.
- Digitalisation: development of protocols for data-analysis and dashboarding.
- Data Science: development of ML models for Time Series Forecasting, Anomaly Detection, Fault Classification and Remaining Useful Life estimation.
- Design: SMART infrastructure, where monitoring is part of the structural design.
- Testing, failing, learning and developing