Optimizing MRT data processing via comparative analysis of SA and PSO algorithms: a simulation and numerical study.

Area of Science:

• Thermophysics
• Computational Intelligence
• Optical Engineering

Background:

• Accurate temperature measurement is crucial in high-temperature industrial processes.
• Unknown emissivity presents a significant challenge in Multispectral Radiation Thermometry (MRT) data processing.
• Optimization algorithms are needed for robust and efficient MRT data analysis.

Purpose of the Study:

• To systematically compare Particle Swarm Optimization (PSO) and Simulated Annealing (SA) algorithms for MRT data processing.
• To evaluate the performance of PSO and SA in handling unknown emissivity challenges.
• To demonstrate the applicability of PSO for accurate surface temperature measurement in demanding environments.

Main Methods:

• Comparative analysis of PSO and SA algorithms applied to simulated MRT data.
• Simulation of six hypothetical emissivity models to test algorithm robustness.
• Application of the superior algorithm to simulated rocket motor nozzle temperature data.

Main Results:

• PSO demonstrated superior accuracy, efficiency, and stability compared to SA across various emissivity models.
• PSO achieved a maximum absolute error of 16.27 K and a maximum relative error of 0.65% for rocket motor nozzle temperature.
• PSO processing time was less than 0.3 seconds, highlighting its computational efficiency.

Conclusions:

• PSO is a highly effective algorithm for accurate temperature measurement in MRT, particularly in overcoming emissivity uncertainties.
• The proposed PSO-based method is suitable for real-time industrial applications and can be extended to other multispectral systems.
• This research provides a robust computational approach for high-temperature measurements in diverse industrial settings.