Artificial Intelligence-Assisted Prediction of Effluent Phosphorus in a Full-Scale Wastewater Treatment Plant with Missing Phosphorus Input and Removal Data

Abstract

Although artificial intelligence (AI) such as machine learning (ML) and deep learning (DL) has been recognized as an emerging and promising tool, its application becomes challenging with incomplete data collection. Herein, in the absence of the influent phosphorus load and chemical dosage data for phosphorus removal, we employed ML/DL models to predict effluent phosphorus using nine-year data from a small-scale wastewater treatment plant. Attempts were made to select essential model input features from 42 variables by using Pearson correlation analysis to reveal internal correlations among variables. First, five ML regression models were used to predict the effluent phosphorus load, and a maximum coefficient of determination (R 2) of 0.637 was achieved with the support vector machine model. Then, the DL model named long short-term memory could predict phosphorus load in one-day advance with an R 2 value of 0.496. Finally, on the basis of the historical data, an anomaly alarm design was proposed to minimize the chance of exceeding the discharge permit and achieved a maximum accuracy of 79.7% to predict the phosphorus concentration after comparing seven ML classification models. This study provides an example of applying AI for process improvement and potential cost reduction with incomplete data sets.

Prediction of biogas production in anaerobic digestion of a full‐scale wastewater treatment plant using ensembled machine learning models

“…Pearson correlation analysis measures the strength and direction of the linear relationship between features. The Pearson correlation coefficient (r) can be calculated using the formula shown in Equation (), where $$$ {x}_i $$$ (or $$$ {y}_i $$$) represents the observed values of the two features, and $$$ \overline{x} $$$ (or $$$ \overline{y} $$$) represents the mean values of the data for each feature (Xu et al, 2023). The value of correlation coefficient is between −1 and 1; strong positive correlation is 1, strong negative correlation is −1, and no correlation is 0.…”

“…The SHAP interpretation method calculates the Shapley value according to the alliance game theory. The analysis of interpretability through SHAP regression values aims to assess the contribution of input features to the predictions made by ML prediction models (Xu et al, 2023). Significant features and driving forces are identified when the SHAP plot shows a higher mean of absolute Shapley values, which indicates a stronger driving force from the input features to the output features.…”

Nature-Based Solutions and Climate Resilience: A Bibliographic Perspective through Science Mapping Analysis

“…Connections with outdoor recreation • designing multifunctional areas for climate resilience and recreation [27] • the role of NBSs in shaping multifunctional land use [28] • multifunctional outdoor recreation and flood management in flood-prone areas [29] Sustainability • ways in which water management can improve sustainability [30] • using ecosystem services for greater sustainability [31] • using NBSs for economic sustainability in cities [32] Ecosystem services • using technology to expand ecosystem services [20] • using ecosystem services to support nutrient cycling food supply and resource allocation [33] • ways in which ecosystem services can contribute to sustainability [31] • using ecosystem services for cultural purposes and nature recreation [34] • using ecosystem services for wastewater management [35] • using technology to improve ecosystem services and resilience [36] • using urban and spatial planning to improve or extend the resilience of urban ecosystems (how can we improve ecosystem resilience through urban and spatial planning strategies?) [37] Water management • improving wastewater treatment by using the functions of NBSs [38] • forecasting the quantity of refuse and developing an intelligent system within water treatment facilities to facilitate immediate anticipatory management of sewage treatment [39] • creating an intelligent microgrid for waste management [40] • leveraging the Internet of Things (IoT) for the purpose of effectively handling domestic waste [41] • using ecosystem services for wastewater treatment [42] Mitigate and absorb carbon dioxide • investigating crucial elements in the cultivation of algal biomass and lipids for the generation of sustainable energy sources [43] • predicting future biomass yields of crops [44] • investigating the impact of climate change on carbon flux as a major driver of algal biofuel production [45] • technology potential for carbon uptake from the air and other resources [46] • prediction of renewable energy production [47] • accurate prediction of CO2 emissions [48] Flooding • the floating city and the use of NBSs to improve performance [49] • the role of ecosystem services in mitigating or preventing flooding…”

Advancements in Biochar Modification for Enhanced Phosphorus Utilization in Agriculture

“…In wastewater remediation, another realm where biochar holds promise, AI techniques such as ML and DL come into play. These techniques can predict effluent P levels even when data are scarce, facilitating compliance with regulatory standards while potentially reducing costs [116].…”