SteelTemp

Overview

Predicting high-temperature strength of steels using machine learning.

SteelTemp is an advanced predictive tool that estimates the mechanical performance of high-strength steels exposed to elevated temperatures. By integrating cutting-edge machine learning techniques with experimental and literature-based data, SteelTemp enables engineers to obtain accurate stress reduction predictions instantly — without the need for physical testing or finite element simulations.

This software was developed based on a peer-reviewed scientific model published in Results in Engineering (Elsevier), ensuring both scientific rigor and practical reliability.

Academic Publication:
C. Yazici, F.J. Domínguez-Gutiérrez (2025). Machine learning techniques for estimating high–temperature mechanical behavior of high strength steels. Results in Engineering, 25, 104242. https://doi.org/10.1016/j.rineng.2025.104242

Core Features

Ultimate Strength Estimation: Predicts tensile strength (MPa) of steel at any given temperature and thickness.
Reduction Factor (χ) Calculation: Provides fire-design compliant degradation factors for Eurocode & AISC integration.
Multi-material Model: Trained on 455+ data points across 20+ steel types (S235, G550, HSA800, etc.).
Instant Results: No testing or simulation required. Results in under 0.2 seconds.
Accuracy Validated: Gradient Boosting model with R² > 0.98, RMSE: 0.05, MAE: 0.04.
User-friendly Interface: Enter temperature (°C) and material thickness (mm) — get real-time results and charts.