Determining the Rheological Parameters of Polymers Using Machine Learning Techniques

Introduction. The paper investigates the methodology for determining the rheological parameters of materials based on the nonlinear Maxwell-Gurevich rheological model using the stress relaxation curves. The review of the main directions of the metaheuristic approaches (local search, evolutionary algorithms) to solving the combinatorial optimization problems is presented. The metaheuristic algorithms for solving some important combinatorial optimization problems with the special emphasis on building decision trees are described. The comparative analysis of the algorithms for solving the regression problem in CatBoost Regressor is carried out. The aim of the work is to determine the rheological properties of polymers using machine learning techniques.

Materials and Methods. The objects of the study are the generated data sets obtained on the basis of the theoretical stress relaxation curves. The source data tables for model training across all samples are presented, and the statistical analysis of the source data sets characteristics is carried out. The total number of numerical experiments across all samples amounted to 346020 variations. To develop the models, the CatBoost artificial intelligence techniques were used; the regularization techniques (Weight Decay, Decoupled Weight Decay Regulation, Augmentation) were used to increase the model accuracy; and Z–Score technique was used for data normalization.

Results. As a result of the research, the intelligent models for determining the rheological parameters of polymers (initial relaxation viscosity, velocity modulus) have been developed based on the generated data sets on the example of the epoxy binder EDT-10. Based on the testing results of the models with the best parameters, the quality assessments were carried out: for the parameter 𝜂^∗₀ the range of values MAPE 0.46 — 2.72, MSE 0.15 — 1.09, RMSE 0.19 —  0.44, MAPE 0.46 — 1.27; for the parameter 𝑚^∗ — MAPE 0.07 — 0.32, MSE 0.01 —  0.13, RMSE 0.10 — 0.41, MAPE 0.58 — 2.72. The resulting metric values are permissible. The training graphs demonstrate the stability of the process.

Discussion and Conclusion. The developed intelligent models are scalable and cross-platform, have practical applied significance that ensures their implementation in a wide range of the scientific and engineering apps. 

1. Chepurnenko V., Yazyev B., Song X. Creep Calculation for a Three-Layer Beam with a Lightweight Filler. In: Proceedings of the International Conference on Modern Trends in Manufacturing Technologies and Equipment (ICMTMTE 2017). MATEC Web Conf. Vol. 129. EDP Sciences; 2017. 05009 https://doi.org/10.1051/matecconf/201712905009

2. Litvinov S.V., Yazyev B.M., Turko M.S. Effecting of Modified HDPE Composition on the Stress-Strain State of Constructions. In: Proceedings of the International Multi-Conference on Industrial Engineering and Modern technologies. IOP Conference Series: Materials Science and Engineering. Vol. 463(4). IOP Publishing; 2018. 042063 https://doi.org/10.1088/1757-899X/463/4/042063

3. Amjadi M., Fatemi A. Creep and Fatigue Behaviors of High-Density Polyethylene (HDPE): Effects of Temperature, Mean Stress, Frequency, and Processing Technique. International Journal of Fatigue. 2020;141:105871. https://doi.org/10.1016/j.ijfatigue.2020.105871

4. Xiang G., Yin D., Meng R., Lu S. Creep Model for Natural Fiber Polymer Composites (NFPCs) Based on Variable Order Fractional Derivatives: Simulation and Parameter Study. Journal of Applied Polymer Science. 2020;137(24):48796. https://doi.org/10.1002/app.48796

5. Tezel T., Kovan V., Topal E.S. Effects of the Printing Parameters on Short‐Term Creep Behaviors of ThreeDimensional Printed Polymers. Journal of Applied Polymer Science. 2019;136(21):47564. https://doi.org/10.1002/app.47564

6. Trush L., Litvinov S., Zakieva N., Bayramukov S. Optimization of the Solution of a Plane Stress Problem of a Polymeric Cylindrical Object in Thermoviscoelastic Statement. In: International Scientific Conference Energy Management of Municipal Transportation Facilities and Transport EMMFT 2017. Advances in Intelligent Systems and Computing, Vol 692. Cham: Springer; 2017. P. 885–893. https://doi.org/10.1007/978-3-319-70987-1_95

7. Tsybin N.Y., Turusov R.A., Andreev V.I. Comparison of Creep in Free Polymer Rod and Creep in Polymer Layer of the Layered Composite. Procedia Engineering. 2016;153:51–58. https://doi.org/10.1016/j.proeng.2016.08.079

8. Andreev VI, Sereda SA. Calculation of an Inhomogeneous Polymer Thick-Walled Cylindrical Shell Taking into Account Creep under the Action of Temperature Load. In: Proceedings of IOP Conference Series: Materials Science and Engineering. Vol. 1015(1). XXIX R-P-S Seminar 2020. Wroclaw, Poland: IOP Publishing; 2021. 012002. https://doi.org/10.1088/1757-899X/1015/1/012002

9. Litvinov S.V., Klimenko E.S., Kulinich I.I., Yazyeva S.B. Longitudinal Bending of Polymer Rods with Account Taken of Creep Strains and Initial Imperfections. International Polymer Science and Technology. 2015;42(2):23–26. https://doi.org/10.1177/0307174X1504200206

10. Litvinov S.V., Trush L.I., Yazyev S.B. Flat Axisymmetrical Problem of Thermal Creepage for Thick-Walled Cylinder Made of Recyclable PVC. Procedia Engineering. 2016;150:1686–1693. https://doi.org/10.1016/j.proeng.2016.07.156

11. Yazyev B.M., Chepurnenko A.S., Savchenko A.A. Calculation of Three-Layer Panels with Polyurethane Foam Filler Taking into Account the Rheological Properties of the Middle Layer. Materials Science Forum. 2018;935:144–149. https://doi.org/10.4028/www.scientific.net/MSF.935.144

12. Dudnik A.E., Chepurnenko A.S., Litvinov S.V. Determining the Rheological Parameters of Polyvinyl Chloride, with Change in Temperature Taken into Account. International Polymer Science and Technology. 2017;44(1):43–48. https://doi.org/10.1177/0307174X1704400109

13. Litvinov S, Yazyev S, Chepurnenko A, Yazyev B. Determination of Rheological Parameters of Polymer Materials Using Nonlinear Optimization Methods. In: Proceedings of the XIII International Scientific Conference on Architecture and Construction 2020. Singapore: Springer; 2021. P. 587–594. https://doi.org/10.1007/978-981-33-6208-6_58

14. Chepurnenko A. Determining the Rheological Parameters of Polymers Using Artificial Neural Networks. Polymers. 2022;14(19):3977. https://doi.org/10.3390/polym14193977

15. Muhammad W., Brahme A.P., Ibragimova O., Kang J., Inal K. A machine learning framework to predict local strain distribution and the evolution of plastic anisotropy & fracture in additively manufactured alloys. International Journal of Plasticity. 2021;136:102867. https://doi.org/10.1016/j.ijplas.2020.102867

16. Oh W.B., Yun T.J., Lee B.R., Kim C.G., Liang Z.L. A Study on Intelligent Algorithm to Control Welding Parameters for Lap-joint. Procedia Manufacturing. 2019;30:48–55. https://doi.org/10.1016/j.promfg.2019.02.008

17. Ramos-Figueroa O., Quiroz-Castellanos M., Mezura-Montes E., Schütze O. Metaheuristics to solve grouping problems: a review and a case study. Swarm and Evolutionary Computation. 2020;53:100643. https://doi.org/10.1016/j.swevo.2019.100643

18. Kondratieva T, Prianishnikova L, Razveeva I. Machine Learning For Algorithmic Trading. In: Proceedings of E3S Web Conf. Topical Problems of Agriculture, Civil and Environmental Engineering (TPACEE 2020). Vol. 224. E3S Sciences; 2020. 01019. https://doi.org/10.1051/e3sconf/202022401019

19. Krause A, Fairbank M. Baseline Win Rates for Neural-Network Based Trading Algorithms. In: 2020 International Joint Conference on Neural Networks (IJCNN). Glasgow, UK: IEEE; 2020. P. 1–6. https://doi.org/10.1109/IJCNN48605.2020.9207649