Applications of machine learning algorithms for biological wastewater treatment: Updates and perspectives

Biological wastewater treatment using algae-bacteria consortia for nutrient uptake and resource recovery is a 'paradigm shift' from the mainstream wastewater treatment process to mitigate pollution and promote circular economy. The symbiotic relationship between algae and bacteria is complex in open or closed biological wastewater treatment systems. In this regard, machine learning algorithms (MLAs) have found to be advantageous to predict the uncertain performances of the treatment processes. MLAs have shown satisfactory results for effective real-time monitoring, optimization, prediction of uncertainties and fault detection of complex environmental systems. By incorporating these algorithms with online sensors, the transient operating conditions during the treatment process including disruptions or failures due to leaking pipelines, malfunctioning of bioreactors, unexpected fluctuations of organic loadings, flow rate, and temperature can be forecasted efficiently. This paper reviews the state-of-the-art MLA approaches for the integrated operation of biological wastewater treatment systems combining algal biomass production and nutrient recovery from municipal wastewater. [GRAPHICS] .

Automated summary

This study highlights the importance and advantages of using algae-bacteria consortia for wastewater treatment to achieve nutrient uptake and resource recovery. Machine learning algorithms play a crucial role in predicting uncertainties in the treatment process and provide reliable support for real-time monitoring, optimization, and fault detection in wastewater treatment systems.