Search

Briquetting is considered one of the pre-treatment methods available to produce raw materials of uniform size and moisture content that are easy to process, transport, and store. The quality of briquettes in terms of density and strength depends on the physical and chemical properties of the raw material and the briquetting conditions. However, determining briquette quality is difficult, very costly, and requires long laboratory studies. In this paper, an easy, inexpensive, and fast methodology based on machine learning for the determination of quality parameters of briquette samples is presented. Compressive resistance, one of the most important briquette quality parameters, was estimated by machine learning methods, considering particle size, material moisture, applied pressure value, briquette density, shatter index, and tumbler index. Extra Trees, Random Forest, and Light Gradient Boosting regression models were used. The best estimate is seen in the Extra Trees regression model. The R2 and MAPE values are 0.76 and 0.0799, respectively. [ABSTRACT FROM AUTHOR]

Briquetting is considered one of the pre-treatment methods available to produce raw materials of uniform size and moisture content that are easy to process, transport, and store. The quality of briquettes in terms of density and strength depends on the physical and chemical properties of the raw material and the briquetting conditions. However, determining briquette quality is difficult, very costly, and requires long laboratory studies. In this paper, an easy, inexpensive, and fast methodology based on machine learning for the determination of quality parameters of briquette samples is presented. Compressive resistance, one of the most important briquette quality parameters, was estimated by machine learning methods, considering particle size, material moisture, applied pressure value, briquette density, shatter index, and tumbler index. Extra Trees, Random Forest, and Light Gradient Boosting regression models were used. The best estimate is seen in the Extra Trees regression model. The R2 and MAPE values are 0.76 and 0.0799, respectively.