Generalized inverse matrix normalization algorithm to extract high-temperature data from multiwavelength pyrometry.

Area of Science:

• Thermophysics
• Optical Measurement Techniques
• Materials Science

Background:

• Accurate high-temperature surface measurement is crucial for non-gray materials.
• Unknown spectral emissivity poses a significant challenge in multiwavelength pyrometry (MWP).
• Existing methods struggle to effectively process temperature inversion data with variable emissivity.

Purpose of the Study:

• To develop a direct and fast data processing algorithm for MWP to minimize effects of unknown emissivity.
• To improve the accuracy and efficiency of high-temperature measurements using MWP.
• To provide a robust method for transient high-temperature field measurements.

Main Methods:

• Proposed a Generalized Inverse Matrix normalization (GIM-NOR) algorithm based on GIM theory for underdetermined equations.
• Utilized emissivity distribution shape to select optimal channels for temperature inversion.
• Implemented a NOR compensation method to further enhance inversion accuracy.

Main Results:

• Simulations with six materials at 1800 K showed an average relative temperature error of 0.63% within 8 ms.
• The algorithm maintained accuracy and efficiency with up to 5% random noise.
• Experimental data from rocket nozzle measurements yielded a relative temperature error below 0.50% within 9 ms.

Conclusions:

• The GIM-NOR algorithm effectively overcomes emissivity challenges in MWP for accurate high-temperature measurements.
• The method demonstrates high precision and computational efficiency for transient temperature fields.
• This algorithm has broad potential applications in measuring high-temperature fields using MWP.