Abstract
Pollution of beaches and loss of recreational water amenity caused by municipal effluent discharges from coastal primary wastewater treatment plants (WWTP) have been a problem around the world. Oil and grease (O&G) are one of the significant pollutants in the effluent that impacts beaches and recreational water quality and have stringent regulatory discharge limits. In this article, an artificial neural networks (ANN) model, based on artificial intelligence computation technique, has been developed to predict the effluent O&G from a coastal primary and chemically assisted primary WWTP, 1, 2, and 3 days in advance. Results show that the models are able to predict effluent O&G with fair reliability for both primary sedimentation (PS) and chemically assisted primary sedimentation (CAPS) WWTPs (root-mean-square error 3.6–4.5 mg/L, average absolute error 2.9–3.5 mg/L, and average absolute percentage error 10%–11.6%). The results show that ANN modeling is more effective and efficient in forecasting effluent O&G for PS WWTPs than for CAPS.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have