Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Biogas Production Enhancement from a Cold Region Municipal Wastewater Anaerobic Digestion
https://orcid.org/http://orcid.org/0000-0002-5305-8332
https://orcid.org/http://orcid.org/0009-0002-0866-1375
https://orcid.org/http://orcid.org/0000-0002-4323-0063
https://orcid.org/http://orcid.org/0000-0001-9718-6793
Biogas is a clean and renewable source of energy, and its production from anaerobic digestion (AD) systems at municipal wastewater treatment plants (MWTPs) has become of greater interest to municipal environmental authorities and decision-makers given recent carbon pricing initiatives. Saskatoon Wastewater Treatment Plant (SWTP) is an advanced facility subjected to highly varying regional temperatures due to its location in Saskatchewan, Canada. Its AD systems stabilize sludges from thickened waste-activated sludge (TWAS), fermenter sludge (FS), and primary sludge (PS). Therefore, this study aimed to determine the optimum mixing ratios of TWAS, FS, and PS at the SWTP to enhance biogas production during AD processes. Extreme gradient boosting (XGBoost) was used to develop a data-driven-based model estimating biogas production rates using the input variables of TWAS, FS, and PS flow rates into digestors. The model performance was examined using root mean square error (RMSE) and coefficient of determination (R2). The developed model was then coupled with the Bayesian Optimization Algorithm (BOA) to determine optimum mixing ratios. Reasonable XGBoost model performance for biogas production estimation using all three inputs was observed (R2 > 0.93, and RMSE < 0.53). Feature important analysis indicated that the model’s accuracy significantly relied on TWAS and was followed by FS and PS. However, the developed model using two inputs of TWAS and FS showed weaker performances. BOA then solved the XGBoost fitness function, and results showed the optimal digestor influent ranges were 6.4–6.8 L/s for TWAS, 3.0–6.0 L/s for FS, and below 4 L/s for PS translating to an approximate TWAS: FS: PS ratio of 48%: 23%: 29% per volume. Overall, data-driven modelling of the AD system and its optimization provides valuable, cost-effective, and accurate information, useful for more efficient AD MWTP design and management.
Biogas Production Enhancement from a Cold Region Municipal Wastewater Anaerobic Digestion
https://orcid.org/http://orcid.org/0000-0002-5305-8332
https://orcid.org/http://orcid.org/0009-0002-0866-1375
https://orcid.org/http://orcid.org/0000-0002-4323-0063
https://orcid.org/http://orcid.org/0000-0001-9718-6793
Biogas is a clean and renewable source of energy, and its production from anaerobic digestion (AD) systems at municipal wastewater treatment plants (MWTPs) has become of greater interest to municipal environmental authorities and decision-makers given recent carbon pricing initiatives. Saskatoon Wastewater Treatment Plant (SWTP) is an advanced facility subjected to highly varying regional temperatures due to its location in Saskatchewan, Canada. Its AD systems stabilize sludges from thickened waste-activated sludge (TWAS), fermenter sludge (FS), and primary sludge (PS). Therefore, this study aimed to determine the optimum mixing ratios of TWAS, FS, and PS at the SWTP to enhance biogas production during AD processes. Extreme gradient boosting (XGBoost) was used to develop a data-driven-based model estimating biogas production rates using the input variables of TWAS, FS, and PS flow rates into digestors. The model performance was examined using root mean square error (RMSE) and coefficient of determination (R2). The developed model was then coupled with the Bayesian Optimization Algorithm (BOA) to determine optimum mixing ratios. Reasonable XGBoost model performance for biogas production estimation using all three inputs was observed (R2 > 0.93, and RMSE < 0.53). Feature important analysis indicated that the model’s accuracy significantly relied on TWAS and was followed by FS and PS. However, the developed model using two inputs of TWAS and FS showed weaker performances. BOA then solved the XGBoost fitness function, and results showed the optimal digestor influent ranges were 6.4–6.8 L/s for TWAS, 3.0–6.0 L/s for FS, and below 4 L/s for PS translating to an approximate TWAS: FS: PS ratio of 48%: 23%: 29% per volume. Overall, data-driven modelling of the AD system and its optimization provides valuable, cost-effective, and accurate information, useful for more efficient AD MWTP design and management.
Biogas Production Enhancement from a Cold Region Municipal Wastewater Anaerobic Digestion
https://orcid.org/http://orcid.org/0000-0002-5305-8332
https://orcid.org/http://orcid.org/0009-0002-0866-1375
https://orcid.org/http://orcid.org/0000-0002-4323-0063
https://orcid.org/http://orcid.org/0000-0001-9718-6793
Biogas is a clean and renewable source of energy, and its production from anaerobic digestion (AD) systems at municipal wastewater treatment plants (MWTPs) has become of greater interest to municipal environmental authorities and decision-makers given recent carbon pricing initiatives. Saskatoon Wastewater Treatment Plant (SWTP) is an advanced facility subjected to highly varying regional temperatures due to its location in Saskatchewan, Canada. Its AD systems stabilize sludges from thickened waste-activated sludge (TWAS), fermenter sludge (FS), and primary sludge (PS). Therefore, this study aimed to determine the optimum mixing ratios of TWAS, FS, and PS at the SWTP to enhance biogas production during AD processes. Extreme gradient boosting (XGBoost) was used to develop a data-driven-based model estimating biogas production rates using the input variables of TWAS, FS, and PS flow rates into digestors. The model performance was examined using root mean square error (RMSE) and coefficient of determination (R2). The developed model was then coupled with the Bayesian Optimization Algorithm (BOA) to determine optimum mixing ratios. Reasonable XGBoost model performance for biogas production estimation using all three inputs was observed (R2 > 0.93, and RMSE < 0.53). Feature important analysis indicated that the model’s accuracy significantly relied on TWAS and was followed by FS and PS. However, the developed model using two inputs of TWAS and FS showed weaker performances. BOA then solved the XGBoost fitness function, and results showed the optimal digestor influent ranges were 6.4–6.8 L/s for TWAS, 3.0–6.0 L/s for FS, and below 4 L/s for PS translating to an approximate TWAS: FS: PS ratio of 48%: 23%: 29% per volume. Overall, data-driven modelling of the AD system and its optimization provides valuable, cost-effective, and accurate information, useful for more efficient AD MWTP design and management.
Biogas Production Enhancement from a Cold Region Municipal Wastewater Anaerobic Digestion
Lecture Notes in Civil Engineering
Desjardins, Serge (Herausgeber:in) / Poitras, Gérard J. (Herausgeber:in) / Ng, Kelvin Tsun Wai (Herausgeber:in) / Asadi, Mohsen (Autor:in) / Zeynali, Rahman (Autor:in) / Soltan, Jafar (Autor:in) / McPhedran, Kerry (Autor:in)
Canadian Society of Civil Engineering Annual Conference ; 2023 ; Moncton, NB, Canada
26.09.2024
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
| British Library Conference Proceedings | 2012
NaOH–urea pretreatment for biogas enhancement from algal biomass anaerobic digestion
| American Institute of Physics | 2021
| American Chemical Society | 2024
| American Institute of Physics | 2013