The process of development of a mathematical model of the process of homogenization of dairy products is considered in the work. The theory of Markov's chains was used in the development of the mathematical model, Markov's chain with discrete states and continuous parameter for which the homogenisation pressure is taken, being the basis for the model structure. Machine realization of the model is implemented in the medium of structural modeling MathWorks Simulink™. Identification of the model parameters was carried out by minimizing the standard deviation calculated from the experimental data for each fraction of dairy products fat phase. As the set of experimental data processing results of the micrographic images of fat globules of whole milk samples distribution which were subjected to homogenization at different pressures were used. Pattern Search method was used as optimization method with the Latin Hypercube search algorithm from Global Optimization Тoolbox library. The accuracy of calculations averaged over all fractions of 0.88% (the relative share of units), the maximum relative error was 3.7% with the homogenization pressure of 30 MPa, which may be due to the very abrupt change in properties from the original milk in the particle size distribution at the beginning of the homogenization process and the lack of experimental data at homogenization pressures of below the specified value. The mathematical model proposed allows to calculate the profile of volume and mass distribution of the fat phase (fat globules) in the product, depending on the homogenization pressure and can be used in the laboratory and research of dairy products composition, as well as in the calculation, design and modeling of the process equipment of the dairy industry enterprises.

1. 1 Dairy Processing Handbook. Lausanne, Tetrapak, 2015.

2. 2 Nuzhin E.V., Gladushnyak A.K. Gomogenizatsiya i gomogenizatory [Homogenization and blenders]. Odessa, Pechatnyi dom, 2007. (In Russ.).

3. 3 Petrachkov B. V. Razrabotka vikhrevogo gomogenizatora na osnove teoreticheskikh i eksperimental'nykh issledovanii proctsessa nizkotemperaturnoi kavitatsionnoi gomogenizatsii: Avtoreferat kand. tekh. nauk. [The development of a vortex homogenizer on the basis of theoretical and experimental studies of low-temperature cavitation homogenizing process. Abstr. diss. cand. tech. sci.]. Moscow, 2006. (In Russ.).

4. 4 Oreshina M. N. Impul’snoe dispergirovanie mnogokomponentnykh pishchevykh system i ego apparaturnaya realizatsiya: autoref. diss. doc. tekh. nauk. [Pulse dispersion of multicomponent food systems and its hardware implementation]. Moscow, 2010. (In Russ.).

5. 5 Ventsel' E.S. Issledovanie operatsii. Zadachi, printsipy, metodologiya [Operations research. Objectives, principles, methodology]. Moscow, KNORUS, 2013. (In Russ.).

6. 6 D'yakonov V.P. MATLAB I Simulink dlya radioinzhenerov [MATLAB and Simulink for radio]. Moscow, DMK-Press, 2015. (In Russ.).

7. 7 Martinez W.L, Martinez A.R. Computational Statistics Handbook with MATLAB®. New York, Chapman & Hall/CRC, 2002.

8. 8 Bityukov V.K., Khvostov A.A., Rebrikov D.I., Merzlikin V.E. Automation of processing of dairy products micrographs using ImageJ and Statistica. Vestnik Voronezhskogo gosudarstvennogo universiteta inzhenernyh tekhnologij. [Proceedings of the Voronezh state university of engineering technologies], 2014, no. 4(62), pp. 58-63. (In Russ.).

9. 9 Khvostov A.A., Rebrikov D.I., Merzlikin V.E. Identification of the mass distributions of the fat phase in milk using universal distributions Pearson. Vestnik Voronezhskogo gosudarstvennogo universiteta inzhenernyh tekhnologij. [Proceedings of the Voronezh state university of engineering technologies], 2015, no. 2(64), pp. 66-72. (In Russ.).