基于改进模拟退火算法的火焰温度场反演计算  

Inversion Calculation of Flame Temperature Field Based on Improved Simulated Annealing Algorithm

作　　者：王旭 祝海江[1] 刘兴旺[2] 李小春[2] WANG Xu;ZHU Hai-jiang;LIU Xing-wang;LI Xiao-chun(College of Information Science&Technology,Beijing University of Chemical Technology,Beijing 100029,China;The 41st Research Institute/China Electronics Technology Instruments Co.,Ltd,Qingdao Shandong 266555,China)

机构地区：[1]北京化工大学信息科学与技术学院,北京100029 [2]中国电子科技集团公司第四十一研究所,山东青岛266555

出　　处：《计算机仿真》2022年第6期313-318,共6页Computer Simulation

基　　金：装备预先研究项目一般基金(61400030202)。

摘　　要：基于CCD摄像机采集到火焰的辐射图像的温度场纵向反演算法在温度测量等领域获得了较为广泛的应用。上述方法通常利用模拟退火算法计算图像上每个像素点在纵向上的多段温度值,其本质上是对一组函数求解最优化问题。虽然应用模拟退火算法能够很好的计算出火焰温度,但算法运行时人工设置参数的过程较为繁琐。因此研究了一种基于改进模拟退火算法的火焰温度场反演计算方法。方法利用贝叶斯优化,自动实现模拟退火算法中的超参数设置问题。实验结果表明所采用的基于贝叶斯优化的模拟退火算法能自动优化模拟退火算法中的超参数,减少人工设置参数的繁琐,并且计算的温度值也满足要求,极大地提高了计算效率。The longitudinal inversion algorithm of temperature field based on the radiation image of flame acquired by CCD camera has been widely applied in the field of temperature measurement. The simulated annealing algorithm is usually used to calculate the multi-segment temperature value of each pixel on the image in the longitudinal direction, and it is essentially an optimization problem for a set of functions. The flame temperature can be computed using simulated annealing algorithm, but some parameters need to be set manually in advance. The process of parameter setting is relatively complicated, and the parameter setting is not necessarily the best. Therefore, a flame temperature field inversion method based on the improved simulated annealing algorithm was studied in this paper. Bayesian optimization may automatically solve the hyperparameter setting problem in the simulated annealing algorithm. The experimental results show that the simulated annealing algorithm based on Bayesian optimization can automatically optimize the hyperparameters in the simulated annealing algorithm and reduce the cumbersome manual setting of parameters;the calculated temperature value also meets the requirements, which greatly improves the calculation efficiency greatly.

关 键 词：温度场重建 纵向反演计算 模拟退火算法 贝叶斯优化 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]