Development of models for prediction of the energy content of disposed MSW from an unsecured landfill

Abstract

This research was undertaken to estimate the energy potential of municipal solid waste (MSW) disposed over the past 15 years in an unsecured landfill located in Okhla, Delhi, India. The research collected a total of 98 samples from various depths and locations, which were analyzed for physico-chemical characterization. Multiple regression models for estimating the energy content of the disposed MSW were developed to predict average gross calorific value (GCV) and net calorific value (NCV). Forward entry regression was applied to identify the variables that contributed significantly to the regression. Thus, 4-parametric models and 2-parametric models were developed for predicting both average GCV and NCV of the disposed MSW. The developed models gave a reasonably good prediction of the energy content of MSW. Both the models gave low values of average absolute relative error and high values of correlation coefficient. It was found that the 2-parametric models exhibited nearly the same prediction ability with less measurable dimensions as the 4-parametric models. These models are unique and the research is a pioneering effort since apart from being the only models based on physico-chemical characteristics of disposed MSW to be reported, they also incorporate the effect of the depth of waste as an independent variable. The developed models can be used to predict calorific values of disposed MSW having similar characteristics in other landfill sites in India and abroad.