Distribution rate prediction of gravity separation based on the arctangent model

ABSTRACT

To accurately predict the product yield during coal preparation plant design, an arctangent model is proposed, and the derivation process of parameters in the model is described. To determine the adequacy and reliability of the model, a statistical analysis approach is used, and the accuracy of the model was verified using the prediction data of usual approximate formula method (AFM) and the actual data. Compared with the actual distribution rate, the error of distribution rate predicted by the model is larger near the separation density. These results are similar to those obtained using AFM. The distribution curves of predicted data fitting based on arctangent model are closer to the actual distribution curve than that fitted based on the approximate formula predicted data. Compared with the actual clean coal yield, the absolute error predicted by the model is between –1.09 and 0.50 for the test sample, while the absolute error of AFM is between –1.91 and 1.04 for the test sample. Thus, it is verified to be better and more reasonable than the usual AFM. The model is helpful to improve the design level of coal preparation plant.

KEYWORDS:

• Arctangent model
• gravity separation
• distribution curve
• product prediction
• approximate formula method

Additional information

Funding

The research work is financially supported by the National Natural Science Foundation of China (Nos. 51904301, 51904298, U1903132, U1508210), Guizhou Provincial Education Department fund project (qianjiaohekyzi [2019] 140), the Natural Science Foundation of Jiangsu Province (BK20180650), the China Postdoctoral Science Foundation (2020M671652), the Academician Workstation of Liupanshui Normal University (qiankehepingtairencai [2019]5604 hao), a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions , Guizhou Key Laboratory of Coal Clean Utilization (qiankehepingtairencai [2020]2001) and Liupanshui Key Laboratory (Nos. 52020-2019-05-04, 52020-2019-05-06).