No abstract

No abstract

No abstract

No abstract

The article shows the application of a neural network for modeling coke quality indicators Coke Reactivity Index (CRI) and Coke Strength after Reaction (CSR). Two optimization methods were used to train the neural network. The influence of the number of neurons on the simulation results was studied. The difference between experimental and calculated data on average does not exceed 2 %. The conclusion is made about the prospects of using a neural network to predict the values of CRI and CSR of coke. Keywords:artificial neural network, coke, coke reactivity index, coke strength after reaction IntroductionAn important task of metallurgical coke production is to obtain coke of a given quality.It is known (see, for example, [1]) that metallurgical coke can be characterized by two parameters: CRI and CSR. At present, there is no reliable model for calculating CRI and CSR based on the characteristics of charge materials and coking mode. In this regard, studies in the direction of establishing the dependence of coke quality on the characteristics of the charge are relevant.In studies in this direction, regression and correlation analysis methods are often used (see, for example, [2]). In [2], a linear regression dependence between CRI and CSR was obtained with a correlation coefficient of 0.97; the determination coefficients between CRI and the volume fraction of internite, vitrinite reflection index, hygroscopic humidity are in the range 0.26 -0.27, which indicates a complex mutual influence of the factors considered.It can be concluded that the problem of determining the characteristics of the quality of coke CRI and CSR can be attributed to an insufficiently formalized problem in which there are many influencing factors (for example, petrographic composition, degree of metamorphism, etc.), which are not always strictly possible to take into account.