Environment and Ecology Research Vol. 11(3), pp. 456 - 466
DOI: 10.13189/eer.2023.110305
Reprint (PDF) (941Kb)

Application of ARIMA Imputation Model and RNN Forecasting Model – WWTPs Performances Optimization

A. Chaoui ^1,*, W. Elkhoumsi ¹, M. Laaouan ², R. Bourziza ¹, K. Sebari ^1
1 Hassan II Institute for Agronomy and Veterinary Medicine, Rabat, Morocco
² International Institute for Water and Sanitation, Rabat, Morocco

ABSTRACT

Machine learning forecasting has been widely used in order to increase wastewater treatment plants (WWTPs) performance and provide support to WWTPs management. Forecasting wastewater influent quality parameters is not only beneficial for operators and the plant itself, but it is also important environmentally, economically, and socially. In this paper, Chefchaouen's WWTP BOD5 variable will be used as a case example. The current paper applies ARIMA imputation model to have a complete time series variable and complete the dataset that will enable conducting a machine learning forecasting model that is Recurring Neural Networks (RNN) in order to provide accurate predictions. The aim of this paper is to assess the impact of the application of these models on providing support to the plants management and control. In addition to that, analyses will also assess the impact of implementing these models on the use of energy and the injection of oxygen. The models used are statistically correct. The forecasted BOD5 values were close to the actual provided values. BOD5 predictions were converted in order to suggest the total energy consumption per day as well as the total oxygen to be injected. Energy consumption could have decreased in the period of assessment by a percentage of 37% while the oxygen injected could have decreased by a percentage of 90%. Finally, this paper concludes with a discussion as well as the limitations of this work.

KEYWORDS
BOD5, Forecasting, Wastewater, Machine Learning, ARIMA, RNN, Energy Consumption, Oxygen Injection

Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] A. Chaoui , W. Elkhoumsi , M. Laaouan , R. Bourziza , K. Sebari , "Application of ARIMA Imputation Model and RNN Forecasting Model – WWTPs Performances Optimization," Environment and Ecology Research, Vol. 11, No. 3, pp. 456 - 466, 2023. DOI: 10.13189/eer.2023.110305.

(b). APA Format:
A. Chaoui , W. Elkhoumsi , M. Laaouan , R. Bourziza , K. Sebari (2023). Application of ARIMA Imputation Model and RNN Forecasting Model – WWTPs Performances Optimization. Environment and Ecology Research, 11(3), 456 - 466. DOI: 10.13189/eer.2023.110305.