Fast data processing method for multispectral radiation thermometry based on Euclidean distance optimization.

Area of Science:

• Materials Science
• Thermodynamics
• Optical Engineering

Background:

• Accurate temperature measurement of materials like steel is crucial for industrial processes.
• Traditional radiation thermometry often relies on known emissivity, which can be difficult to obtain accurately.
• Multispectral radiation thermometry offers potential for improved accuracy but requires sophisticated data processing.

Purpose of the Study:

• To develop a fast and accurate data processing method for multispectral radiation thermometry.
• To enable true temperature measurement of steel materials without a priori emissivity models.
• To simultaneously determine the true temperature and emissivity of the object.

Main Methods:

• A novel data processing technique involving temperature matrix generation and reference vector selection.
• Utilizing Euclidean distance minimization between rearranged temperature matrices and a reference vector.
• Employing an unconstrained optimization technique to solve for temperature and emissivity concurrently.

Main Results:

• Simulations show relative temperature errors within 0.229% and average computation times under 112.301 ms.
• Experimental validation demonstrates a maximum temperature error of 0.813% within the 1.4–2.5 µm band and 873–1173 K range.
• The method successfully measures true temperature and emissivity simultaneously without prior emissivity data.

Conclusions:

• The developed method provides a feasible and efficient solution for real-time temperature measurement of steel.
• This approach overcomes limitations of traditional methods by not requiring a priori emissivity models.
• The high accuracy and speed make it suitable for demanding industrial applications.