Multi-wavelength pyrometry based on robust statistics and cross-validation of emissivity model

Area of Science:

• Thermometry
• Optical Engineering
• Materials Science

Background:

• Accurate surface temperature measurement is crucial in various scientific and industrial applications.
• Traditional multi-wavelength pyrometry methods often struggle with unknown surface emissivity.
• Developing robust and automated techniques for temperature determination is an ongoing challenge.

Purpose of the Study:

• To develop a systematic and automated procedure for calculating surface temperature from radiance measurements.
• To quantify the accuracy and precision of the proposed method using statistical analysis.
• To retrieve both temperature and emissivity accurately, even with unknown surface emissivity.

Main Methods:

• A cross-validated procedure testing multiple emissivity candidates on random subsets of radiance measurements.
• Utilizing an emissivity model to compute temperature from the average of all possible two-wavelength ratios.
• Selecting the emissivity model that minimizes the coefficient of dispersion.
• Applying statistical methods to quantify accuracy and precision.

Main Results:

• The method accurately determines surface temperature and retrieves true emissivity.
• Precision increases with the number of wavelengths used.
• Optimal precision is achieved when the ratio of minimum to maximum wavelength is 2.46.
• Excellent agreement was observed when applied to both numerical and experimental data.

Conclusions:

• The proposed automated procedure offers an accurate and precise approach to surface temperature determination.
• The method effectively handles surfaces with unknown emissivity.
• The findings have significant implications for non-contact thermometry and material characterization.