Optimization of pelleting parameters for producing composite pellets using agricultural and agro-processing wastes by taguchi-grey relational analysis

Abstract Agricultural and agro-processing sector produces large stream of by-products that are either discarded or are underutilized. Lignocellulosic biomass especially crop residues’ potential to green energy development has been investigated. Material conditions as process variable was made a centerpiece of investigation in the present study since several parameters in the feedstock mixture come into interplay during pelletization. Moisture, binders, blending ratio, particle size, and principle feedstock material were the parameters of our interest in finding their effect on physical and thermochemical characteristics. It is also an endeavor to investigate low cost binder options in producing composite pellets of high quality. A Taguchi L9 orthogonal array experimental design was employed and Taguchi based Grey relational analysis was performed to determine the best combinatorial optimization of factor levels settings for producing best pellets. The optimal parameter level settings for pelletization was determined to be 3mm wheat feedstock material blended with pine shavings at 60/40 ratio. An analysis of variance (ANOVA) was conducted to identify the contribution of each parameter. The magnitude of the significant impacts of parameters on performance characteristics were in the order of binders > blend > grind > feedstock. Results also showed that the physico-chemical characteristics of the pellets were concomitant with the proportion of ingredients or recipes in a blend.