AbstractTri-flo cyclone, as a dense-medium separation device, is one of the most typical environmentally friendly industrial techniques in the coal washery plants. Surprisingly, no detailed investigation has been conducted to explore the effectiveness of tri-flo cyclone operating parameters on their representative metallurgical responses (yield and recovery). To fill this gap, this work for the first time in the coal processing sector is going to introduce a type of advanced intelligent method (boosted-neural network “BNN”) which is able to linearly and nonlinearly assess multivariable correlations among all variables, rank them based on their effectiveness and model their produced responses. These assessments and modeling were considered a new concept called “Conscious Laboratory (CL)”. CL can markedly decrease the number of laboratory experiments, reduce cost, save time, remove scaling up risks, expand maintaining processes, and significantly improve our knowledge about the modeled system. In this study, a robust monitoring database from the Tabas coal plant was prepared to cover various conditions for building a CL for coal tri-flo separators. Well-known machine learning methods, random forest, and support vector regression were developed to validate BNN outcomes. The comparisons indicated the accuracy and strength of BNN over the examined traditional modeling methods. In a sentence, generating a novel BNN within the CL concept can apply in various energy and coal processing areas, fill gaps in our knowledge about possible interactions, and open a new window for plants' fully automotive process.
Quantitative Evaluation of Osteocyte Morphology and Bone Anisotropic Extracellular Matrix in Rat Femur
