Optimization of the Synthesis of Superconducting YBCO Samples Using the Random Forest Algorithm and the Taguchi Method

Authors

  • Д.Р. Джураев Бухарский государственный университет
  • А.А. Тураев Бухарский государственный университет
  • О.Г. Тураев Бухарский государственный университет

DOI:

https://doi.org/10.52304/.v27i4.611

Keywords:

YBCO, critical temperature, Taguchi method, Random Forest regression model

Abstract

This paper presents the results of a study on modeling and predicting the critical superconducting
temperature (Tc) of YBa2Cu3O7δ (YBCO) samples synthesized via the solid-state reaction
method, using a multiparameter machine approach. The influence of synthesis parameters such
as calcination temperature and duration (Tcal, τ1, τ2), sintering temperature (Tsin), pressing pressure
(P), and cooling time (τ3) was investigated. To model the complex, uncertain, and multifactor
relationships between synthesis conditions and Tc values, a Random Forest regression
model was employed. The Random Forest model achieved high accuracy: R2 = 0.993, MAE =
0.0106 K, and RMSE = 0.0125 K, indicating excellent agreement between predicted and experimental
Tc values. Feature importance analysis revealed that τ2 (sintering time) and Tsin (sintering
temperature), τ3 (cooling time), and τ1 (calcination time) have the greatest impact on Tc.
This hierarchy of importance was further confirmed using SHAP analysis (Shapley values).
This study demonstrates the potential for effective development of high-performance superconducting
materials through the integration of regression modeling and experimental design.

Published

2026-02-18

How to Cite

Джураев, Д., Тураев, А., & Тураев, О. . (2026). Optimization of the Synthesis of Superconducting YBCO Samples Using the Random Forest Algorithm and the Taguchi Method. “Uzbek Journal of Physics”, 27(4). https://doi.org/10.52304/.v27i4.611

Issue

Section

08. Materials Science

Most read articles by the same author(s)