Author(s): Cumhur Ozbey; Serhat Kucukali

Linked Author(s): Serhat Kucukali

Keywords: Wastewater treatment plant; Multivariate statistics; Analysis of variance; Biogas production; Biochemical oxygen demand

Abstract: Within the scope of this study, Statgraphics Software Version XVII is used to apply the multivariate statistical methods for one-year raw operational data of Seyhan WWTP which is located in Adana, Turkey. The collected data are first analyzed in terms of descriptive statistics to express the central tendency and dispersion characteristics of the entire dataset and the corresponding Spearman rank correlations reveal the apparent interdependencies among COD, SS, and BOD variables at both inlet and outlet sections of the plant. Then, a two-way ANOVA is performed for the daily biogas production considering the BOD and COD parameters and it can be inferred from the P-values that there is no statistically significant relationship between the influent COD and the daily biogas production, resulting in a P-value of 0.85. On the contrary, the influent BOD has a P-value of 0.0001 (≤ 0.05) indicating a strong relationship between BOD and the daily biogas production within the plant. This finding is also verified by examining the correlation coefficients of the monthly scatter plots of COD and BOD parameters and it is found that influent BOD is up to ten times more correlated with the biogas yield compared to the influent COD. Finally, daily biogas production is examined by performing a one-way ANOVA to investigate the seasonality factor. The output of the analysis extracted a P-value of zero, showing that there is a statistically significant relationship between seasonal variations and daily biogas production in the plant.

DOI: https://doi.org/10.3850/978-90-833476-1-5_iahr40wc-p1419-cd

Year: 2023