CLEARDBP
Description
CLEARDBP is an R&D project aimed at developing an advanced digital solution to optimize the operation of water treatment plants (WTPs), minimizing the formation of disinfection byproducts (THMs, HAAs, etc.) through the use of artificial intelligence, innovative sensors, and predictive modeling.
The project addresses the challenges of climate change, stricter health regulations, and the need to digitize the water cycle.
Objective
To design an intelligent tool to support real-time decision-making for WTP operators, based on flexible and adaptive models that integrate water quality data, operational parameters, and advanced sensors. This tool will reduce disinfection byproducts, improve drinking water quality, ensure compliance with current legislation, and reduce energy consumption.
Actions
- Research and development of contactless sensors based on radar technology
- Generation of predictive models using AutoML techniques
- Design of a modular pipeline for refining ML models
- Application of Active Learning and Conditional Generation techniques to guide experiments.
- Integration of expert knowledge through Human-in-the-Loop and Causal Inference.
- Validation of the solution in a real-world environment (CATABB plant).
- Dissemination of results and national and international commercialization strategy.