Multicriteria model of the process of crushing rock

Abstract

The article deals with the modernization and adjustment of the fine chalk grinding process. The crushing process is an energy-consuming procedure, annually spent about 5% of all energy produced on Earth, including the energy of internal combustion engines. This indicates its great importance. In addition to the cost of electricity, large expenses go to repair the equipment. The greatest replacements are made on the main working parts of machines. In the course of substitutions a lot of time is spent, in order not to spend this rather important resource, it is necessary to approach this procedure from a scientific point of view. The organization and conduct of research on the replacement of the main working parts of crushers and mills will increase the productivity of the main equipment, improve the quality of the finished product and reduce production costs in terms of energy saving. Modernization and adjustment of technological equipment in order to improve the production process of fine chalk significantly increase the service life of the main equipment. For this purpose, it is proposed to conduct an active experiment. Before carrying out the experiment, it is necessary to set the model. The classical regression analysis is based on the assumption that the model type is a priori specified with accuracy to the parameters, and that an experiment has already been implemented that supplies the initial data for the regression construction. Hence, the problem is to choose the best method of data processing. In this paper, we propose a fundamentally new approach-automatic evaluation of the model options on a set of indicators, the calculation of which is based on a set of pareto-optimal variants of the model.The proposed method made it possible to identify two best alternatives out of 16384. Obviously, this approach can be easily modified for any other set of regression model quality criteria.