An essential tool for WRRF modelling: a realistic and complete influent generator for flow rate and water quality...

Area of Science:

• Environmental Engineering
• Wastewater Treatment
• Data-driven Modelling

Background:

• Water resource recovery facility (WRRF) optimization relies on accurate influent data.
• Existing influent generators often require extensive data collection.
• There is a need for data-efficient models for WRRF influent dynamics.

Purpose of the Study:

• To develop a novel data-driven influent generator (IG) model.
• To utilize routine operational data and weather information exclusively.
• To provide a reliable dataset for various WRRF modelling applications.

Main Methods:

• Construction of an artificial neural network (ANN) model.
• Integration with multivariate regression for pollutant time-series generation.
• Generation of flowrate and quality data (TSS, COD, nutrients) at multiple time scales and resolutions.

Main Results:

• The proposed IG model successfully generates reliable time-series data.
• Model performance was validated using statistical criteria.
• The model demonstrated capability in producing daily or hourly datasets.

Conclusions:

• The data-driven IG model offers a reliable and efficient solution for simulating WRRF influent.
• It reduces the need for additional data collection, making it practical for optimization.
• The model supports diverse WRRF modelling applications with accurate flow and quality predictions.