High-precision multi-spectral radiation thermometry method based on the improved grey wolf optimization algorithm inversion.

In this paper, in order to rapidly measure the temperature of a high-temperature target in real time without emissivity data, a high-precision multispectral radiation temperature measurement method based on the improved grey wolf optimization (IGWO) algorithm is proposed. The method can automatically identify the emissivity models of different trends and realize the simultaneous estimation of temperature and emissivity without the emissivity hypothesis model. The IGWO algorithm is applied to the temperature test of a silicon carbide and tungsten material. The temperature test results show that the absolute and relative errors of the silicon carbide (the tungsten) are less than 3 K (4.5 K) and 0.25% (0.18%), respectively. The average time of the algorithm is 0.28 s. The IGWO algorithm can be expected to be applied to some high-precision temperature measurement scenarios.

Data processing for simultaneous inversion of emissivity and temperature using improved CABCSMA and target-to-best DE algorithms in multispectral radiation thermometry (MRT).

In this paper, what we believe to be, a new combined algorithm of artificial bee colony and slime mould algorithm (CABCSMA) and a differential evolution (DE) algorithm using target-to-best variation strategy are proposed to process the data based on Planck's radiation law and the mathematical model of reference temperature. The material model with 6 different emissivity trends is simulated. Simulation results show that the average relative error of CABCSMA algorithm is less than 0.68%, and the average calculation time is 0.44s. The average relative error of DE algorithm is less than 0.43%, and the average calculation time is only 0.06s. The two algorithms were applied to the temperature test of silicon carbide sample, tungsten material and rocket engine nozzle. The experimental results show that the relative error of silicon carbide experimental temperature is less than 0.41% and 0.28%, and the relative error of tungsten material experimental temperature is less than 0.31% and 0.3%. The relative errors of rocket engine nozzle temperature experiments are within 0.68% and 0.52%, respectively. The results show that these two algorithms are expected to be applied in practical measurement scenarios.