Multiparameter design optimisation of a mud-based natural draft biomass cookstove using response surface methodology and desirability function

This work focuses on the design and development of an improved mud cookstove by optimizing 5 key design parameters: primary air area ratio, secondary air inlet area, L/D ratio, pot gap and grate height, and understanding their effect on the performance. Multiple field-scale prototypes were fabricated at various combinations of design parameters and tested for responses: thermal efficiency, CO and PM_2.5 emissions per MJ of energy delivered to the cooking pot. The results of the responses were then modelled using the RSM (Response Surface Methodology) technique to fit second-order quadratic models for predicting cookstove performance. Results showed that the improved mud cookstove can achieve Tier-3 performance for thermal efficiency and Tier-4 and Tier-5 performance for PM_2.5 and CO, respectively. The RSM models were then used with a desirability function for robust parameter optimization under different model weights to maximize efficiency and minimize emissions, which resulted in the range of responses for cookstove performance, i.e., η = 26.21-33.08%, CO = 1.72-2.59 g/MJ_d and PM = 95-252 mg/MJ_d obtained at different factor levels. Conformity trials for validating the modelled results showed < 16% average error, suggesting that fractional factorial design, RSM, and desirability functions are useful tools optimizing the performance of a biomass cookstove.