NS WWTP Fullscale Struvite Crystallizer Operations Data
Description
This dataset was first published in https://doi.org/10.1016/j.watres.2025.125122 and used to train machine learning models for prediction of a full scale struvite crystallizer with more than 5 years (2018 - 2023) of operations data. The models developed from this dataset predict P removal from solution (Conversion) and the fraction of P recovered as struvite (Yield). The raw datasets as well as the processed and engineered feature inclusive datasets are attached.
Files
Steps to reproduce
Raw Data: Operations data was requested from co-author Matt Seib PhD, a Process and Research Engineer at the Madison Metropolitan Sewerage District. Processed Data: The full data processing, feature engineering, feature selection, and model evaluation description is given in the associated journal article in the related links.