Linear regression-based correlations for estimation of high heating values of Pakistani lignite coals

This study was carried out to develop a correlation for the prediction of calorific values of high ash lignite coals. The multiple linear regression models were developed for estimation of calorific values of high ash Pakistani lignite coals. Three models were developed based on proximate components such volatile, ash, and fixed carbon, on as-received, moist free coal datasets from Khushab, Mianwali, Jhelum, and Chakwal regions. The linear regression models were developed, and tested using analysis of variance, an absolute error estimate, and compared with earlier studies available in the literature. According to the findings, all the three models predicted the response, within acceptable limits of error. The two models developed in this study involving i) ash, volatile matter, and fixed carbon and ii) ash and fixed carbon as independent variables were recommended for the prediction of high ash lignite coals from this study.