Co-combustion of binary and ternary blends of industrial sludge, lignite and pine sawdust via thermogravimetric analysis: Thermal behaviors, interaction effects, kinetics evaluation, and artificial neural network modeling

Abstract

Co-combustion behaviors, performances, and kinetic parameters of binary and ternary blends of industrial sludge (IS), lignite (HL), and pine sawdust (SD) were investigated by thermogravimetric analysis, as well as interaction effects and artificial neural network (ANN) modeling. The measure of blending HL and SD into the IS would boost the ignition temperature, while decreased burnout temperature. For the binary blends of IS and HL, the combustion performance would be best with 30% HL ratio, and the combustible index (Ci), combustion stability index (G) and comprehensive combustibility index (CCI) were separately 7.57 × 10−5 %/min·°C2, 3.22 × 10−5 %/min·°C2, 1.39 × 10−7 %2/min2·°C3. With the SD ratio increasing, the combustion performance of the IS and SD blends, and ternary blends improved. It was observed that the interaction effects did exist during the co-combustion of the binary and ternary blends, which were beneficial for the IS combustion disposal. The combination of Coats-Redfern and Malek methods was utilized to the combustion kinetics analysis of blends. The combustion process of binary and ternary blends of IS, HL, and SD could be effectively predicted by the ANN model, and the regression coefficients of the training, validation, testing, and all of the ANN model were all 0.99996.