Thermal-Acoustic Correlation

Thermal-Acoustic Correlation exploits the relationship between temperature and acoustic properties in molten metals within EAFs.

Fundamental Physics

As temperature changes in molten metals, several physical parameters that affect acoustic behavior also change:

• Density decreases with rising temperature
• Viscosity decreases significantly
• Speed of sound varies (typically decreasing with higher temperatures in molten metals)
• Thermal expansion and contraction generate acoustic events

Working Mechanism

This methodology establishes mathematical relationships between acoustic parameters and temperature through:

1. Empirical calibration using reference measurements
2. Physical modeling of wave propagation in molten metal
3. Statistical correlation between acoustic features and thermal states

Practical Example

In an operating EAF, a 50°C temperature increase in molten steel typically causes a measurable 2-5% decrease in the central frequency of arc-generated acoustic emissions. By tracking this frequency shift, the system can accurately estimate temperature with precision comparable to immersion thermocouples (±7°C), but without interrupting the process or risking sensor damage.

Technical Implementation

Advanced systems utilize multiple acoustic parameters simultaneously to improve accuracy:

• Amplitude attenuation rates across different frequencies
• Phase velocity of specific frequency components
• Resonant frequency shifts of the molten bath
• Statistical moments of the frequency spectrum (mean, variance, skewness)

These parameters are fed into multivariate regression models or neural networks trained on thousands of temperature-acoustic data pairs to achieve accurate temperature estimation.

Related Topics

• Acoustic Emission Spectroscopy
• Phase Transition Detection
• Core Technologies