Predictive Modelling of Advanced Wastewater Treatment Technologies Using Artificial Intelligence
Bibliographic record
Abstract
Traditional mathematical models have many limitations, and current machine learning models are black-box type approaches with little insight into the process dynamics. The need for reliable predictive tools to avoid expensive operation interruptions at wastewater treatment plants is growing. This dissertation presents predictive models for aerobic granular sludge (AGS), and biological nutrient removal (BNR) activated sludge processes using machine learning. The main objective of this thesis is to develop and present models that can be used on-site for long-term operation and provide enough information about the process for the detection of faults before they occur. The data for this thesis were collected from laboratory and pilot-scale reactors for the AGS model and historical operational data of the Pine Creek Wastewater Treatment Plant (Calgary, Alberta) for the BNR model.The data were cleaned by removing outliers and filling gaps, and features were selected using multicollinearity reduction and relative parameter importances. A multi-stage model structure was developed where outputs are predicted in the sequence of the actual process progression, considering the cause-effect factor in the process. Multi-layer artificial neural networks, adaptive neuro-fuzzy inference systems, and support vector regression were applied individually and in ensembles as alternative algorithms. The performance of each of the three individual algorithms was compared to each other, and the best model was used to make final predictions. The ensemble techniques used were artificial neural networks, adaptive neuro-fuzzy inference systems, support vector regression, arithmetic mean, and weighted average.The model simulated the AGS process by predicting the biomass concentrations, settling properties, granulation ratio, granule size, and effluent quality with R2 between 89% and 99.9%. It was also able to track predicted abnormalities to their potential cause, utilizing the multi-stage feature in the model. The model was also able to simulate the full-scale BNR process at the Pine Creek WWTP with some parameter modifications, predicting 15 outputs representing the state of the biomass, the waste and return sludge amounts, and the effluent quality, with R2 up to 82%.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".