Modeling of Unburned Carbon in Pulverized Coal-Fired Supercritical Steam Generators
Pawan BajpaiTanuja SheoreyD. Santhosh KumarAnil Kumar SinghJai Prakash Singha Bharat Heavy Electricals Limited,PSNR TSX,Noida,Indiab PDPM Indian Institute of Information Technology. Design & Manufacturing,Mechanical Engineering,Jabalpur,Indiac Research and Development,Bharat Heavy Electricals Limited,Tiruchirappalli,India
DOI: https://doi.org/10.1080/00102202.2024.2326861
IF: 1.9
2024-05-17
Combustion Science and Technology
Abstract:In coal-fired boilers, one of the most important metrics for optimizing combustion is the proportion of unburned carbon (UC) in the ash. In most Indian power plants, it is still assessed offline through manual sample collection and laboratory analysis. The laborious processes of sample collection, preparation, analysis, and reporting require at least five to eight hours of work. The delayed results may not be useful to optimize combustion in the boiler furnace, as the boiler operating conditions may be completely different from the conditions when the samples were collected. Instead, the use of prediction algorithms may bring timely corrective action if the UC goes high. The present experimental study is taken up to obtain a correlation between UC and corresponding relevant boiler operational parameters and on the residence time of coal particles in the boiler furnace. Multiple regression analysis is performed using ANOVA tools. One hundred and ninety-four sets of coal, ash samples and boiler operating parameters are collected from 07 supercritical units. Coal proximate, coal fineness and ash UC analysis are carried out. The analyzed results and collected boiler operational data are statistically analyzed using Microsoft Excel's ANOVA tool to understand the dependent behavior of UC on the boiler operational parameters and fired coal characteristics. The empirical equations are proposed for the prediction of UC in bottom and fly ash for supercritical boilers. Additionally, fresh 20 sets of coal, ash samples, and related operating data are gathered for the validation of the proposed empirical equations. For additionally collected samples, the predicted results are compared with the offline measured laboratory results. The maximum deviation between the predicted and laboratory results are within the range of ± 0.40% and ± 0.24% for bottom and fly ash samples, respectively.
energy & fuels,engineering, chemical,thermodynamics, multidisciplinary