Sensor drift detection framework for water systems with probabilistic machine learning

Smart sensors and meters have facilitated the control and optimization of water and wastewater treatment processes to achieve an acceptable effluent with cost efficiency. However, sensor drift is a challenge for plant operators as it can induce significant uncertainty in process control and thus red...

Full description

Saved in:
Bibliographic Details
Main Author: Hoang, Thu Minh
Other Authors: Law Wing-Keung, Adrian
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/167026
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-167026
record_format dspace
spelling sg-ntu-dr.10356-1670262023-05-19T15:34:16Z Sensor drift detection framework for water systems with probabilistic machine learning Hoang, Thu Minh Law Wing-Keung, Adrian School of Civil and Environmental Engineering CWKLAW@ntu.edu.sg Engineering::Environmental engineering::Water treatment Smart sensors and meters have facilitated the control and optimization of water and wastewater treatment processes to achieve an acceptable effluent with cost efficiency. However, sensor drift is a challenge for plant operators as it can induce significant uncertainty in process control and thus reduce productivity. Current mitigation methods are tedious and time-consuming and require external intervention in the treatment process. Anomaly detection via advanced Machine Learning (ML) models available nowadays can be a solution to this sensor drift problem because the models are fast, accurate, and able to give long-term prediction. The focus of this project is to develop a real-time framework for drift detection using probabilistic ML models with anomaly detection. Experiments were conducted to obtain pH and DO data for model development. With the available measurements, four anomaly detection models (i.e., Moving LOF, NGBoost, XGBoost Distribution and Prophet) were examined in this project, and the Prophet model showed the best performance and was not affected by data limitations. Subsequently, the Prophet model was further used to construct the drift determination procedure that could send warnings before sensor drift developed significantly. Bachelor of Engineering (Environmental Engineering) 2023-05-15T02:34:18Z 2023-05-15T02:34:18Z 2023 Final Year Project (FYP) Hoang, T. M. (2023). Sensor drift detection framework for water systems with probabilistic machine learning. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/167026 https://hdl.handle.net/10356/167026 en EN-30 application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Environmental engineering::Water treatment
spellingShingle Engineering::Environmental engineering::Water treatment
Hoang, Thu Minh
Sensor drift detection framework for water systems with probabilistic machine learning
description Smart sensors and meters have facilitated the control and optimization of water and wastewater treatment processes to achieve an acceptable effluent with cost efficiency. However, sensor drift is a challenge for plant operators as it can induce significant uncertainty in process control and thus reduce productivity. Current mitigation methods are tedious and time-consuming and require external intervention in the treatment process. Anomaly detection via advanced Machine Learning (ML) models available nowadays can be a solution to this sensor drift problem because the models are fast, accurate, and able to give long-term prediction. The focus of this project is to develop a real-time framework for drift detection using probabilistic ML models with anomaly detection. Experiments were conducted to obtain pH and DO data for model development. With the available measurements, four anomaly detection models (i.e., Moving LOF, NGBoost, XGBoost Distribution and Prophet) were examined in this project, and the Prophet model showed the best performance and was not affected by data limitations. Subsequently, the Prophet model was further used to construct the drift determination procedure that could send warnings before sensor drift developed significantly.
author2 Law Wing-Keung, Adrian
author_facet Law Wing-Keung, Adrian
Hoang, Thu Minh
format Final Year Project
author Hoang, Thu Minh
author_sort Hoang, Thu Minh
title Sensor drift detection framework for water systems with probabilistic machine learning
title_short Sensor drift detection framework for water systems with probabilistic machine learning
title_full Sensor drift detection framework for water systems with probabilistic machine learning
title_fullStr Sensor drift detection framework for water systems with probabilistic machine learning
title_full_unstemmed Sensor drift detection framework for water systems with probabilistic machine learning
title_sort sensor drift detection framework for water systems with probabilistic machine learning
publisher Nanyang Technological University
publishDate 2023
url https://hdl.handle.net/10356/167026
_version_ 1772828966897319936