Estimation of Oxygen Exchange during Treatment Sludge Composting through Multiple Regression and Artificial Neural Networks (Estimation of Oxygen Exchange during Composting)

In general, amount of sludge will definitely increase in near future and composting processes, optimum composting conditions and compost use as fertilizer and soil amendment will then be significant research topics. The present study was conducted for O2 parameter estimation by multiple regression and artificial neural networks methods. Daily temperature, CH4, H2S, CO2 and O2 measurements were performed over three different windrows during the composting period (136 days). Three different models were developed for each windrow. Multiple regression and artificial neural network methods were then applied to these models for O2 estimations. High confidence levels were attained between the parameters of multiple regression analysis. However, correlation values in artificial neural network applications (R2 = 0.65-0.98) were even higher. Thus, artificial neural network model applied for each windrow and model was highly confident. The present results indicated that temperature, CH4, CO2 and H2S measurements performed during the composting of waste treatment sludge yielded satisfactory estimations for O2. The recommended correlation may provide significant contributions to composting processes and implementations.