Multispectral thermometry data processing algorithm combining augmented Lagrangian method and modified equilibrium...

Area of Science:

• Physics
• Engineering
• Materials Science

Background:

• Accurate temperature measurement is critical in high-temperature industrial processes.
• Traditional methods face limitations at extreme temperatures (1500-3000 K).
• Multispectral thermometry offers a promising approach for non-contact temperature determination.

Purpose of the Study:

• To develop an advanced data processing algorithm for multispectral thermometry at high temperatures.
• To enhance the accuracy and robustness of temperature inversion methods.
• To provide a real-time temperature measurement solution for extreme conditions.

Main Methods:

• The proposed algorithm transforms the temperature inversion problem into an equality-constrained optimization problem.
• It utilizes Planck's law and the augmented Lagrangian method (ALM).
• A modified equilibrium optimizer is introduced to provide a robust initial emissivity point for ALM, forming the modified equilibrium optimizer-augmented Lagrangian method.

Main Results:

• Simulation experiments with hypothetical targets, tungsten, and rocket engine plumes validated the algorithm's effectiveness.
• The modified equilibrium optimizer-augmented Lagrangian method demonstrated superiority over standard ALM.
• Practical application experiments on 2219 aluminum alloy confirmed the algorithm's real-world applicability.

Conclusions:

• The developed algorithm offers a feasible and efficient solution for online, real-time multispectral high-temperature measurement.
• It overcomes limitations associated with initial iteration points in ALM.
• The method shows significant potential for industrial applications requiring precise high-temperature monitoring.