Machine learning as a support tool in wastewater treatment systems – a short review

Sanja Radović, Sabolc Pap, Maja Turk sekulić

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)

Abstract

Machine learning (ML) is a subset of artificial intelligence (AI). It is based on teaching computers how to learn from data and how to improve with experience. This valuable technique has been increasingly supporting different spheres of life. This includes ML application in enhancement and optimisation of many ecological and environmental engineering solutions, such as wastewater treatment systems (WWTS). Complexity of processes triggers challenges in ensuring good effluent quality by adequate response to dynamic process conditions. That is why techniques such as ML which, after being trained, have strong prediction ability, have been applied in WWTS. ML facilitates understanding of correlation between input features and output targets through a data-driven approach. Different ML models have been used for this purpose. Some of the commonly used were artificial neural network (ANN) or deep neural network (DNN) model, support vector machine (SVM) and its variation support vector regression (SVR) model, random forest (RF) model and many others. More often authors apply a few different models in order to obtain the one that most appropriately works for specific problem. In wastewater management those problems are various, and could include modelling of WWT processes, prediction of certain technology performance, optimisation of technology working parameters, optimisation of the production of the materials there are being used in WWT technology etc. For instance, there are several articles which describes ML power in optimisation of material synthesis (e.g., biochar production). Application of ML led to reduction in number of runs which were necessary for obtaining the best results by applied production procedure, which saved time and was also cost-beneficial. Indeed, ML incorporation in solving or avoiding potential problems within WWTS is a promising approach which has gained more attention in recent years due to the exponential technology development and progress in artificial intelligence application.
Original languageEnglish
Pages799-807
DOIs
Publication statusPublished - 3 Nov 2022

Fingerprint

Dive into the research topics of 'Machine learning as a support tool in wastewater treatment systems – a short review'. Together they form a unique fingerprint.

Cite this