基于相关度的蚁群优化算法对内热源位置的识别  

Identification of Internal Heat Source Position by Using Coalition-Based Ant Colony Optimization Algorithm

作　　者：李博汉[1] 卢玫[1] 

机构地区：[1]上海理工大学能源与动力工程学院,上海200093

出　　处：《上海理工大学学报》2015年第3期225-232,共8页Journal of University of Shanghai For Science and Technology

基　　金：国家自然科学基金资助项目(51176126)

摘　　要：为了提高寻源导热反问题的求解精度和求解速度,针对导热问题中热源位置对边界温度分布影响的特点,提出了适用于寻源导热反问题的基于相关度的蚁群优化算法.该方法分别针对热源位置的每一个坐标,运用能反映计算测点温度曲线与真实测点温度曲线相似程度的量即相关度的方法来构造其相对应的启发信息值;并对蚁群优化算法中路径选择机制、目标函数的构造进行了改进.以数值计算代替实际试验得到测点温度,并对反问题进行计算机编程试验.计算结果表明,此种启发信息值的标定方法和目标函数的构建方法能够很好地区分出路径的质量,从而提高了蚁群收敛到最好路径的速度.计算效率较不考虑相关度的蚁群算法提高了18%~60%.In order to improve the accuracy and efficiency in solving the inverse heat source problem,a coalition-based ant colony optimization method was introduced based on the features of the influence of heat source position on boundary temperature profiles. In the method, the coalition coefficient that reflects the similarity between the real temperature profile and the measured ones was selected as the heuristic information value for each coordinate of the position of heat source. The route construction rule and the building of objective in the ant colony optimization were also modified. Instead of the actual experiments, the temperature measurements at the measurement points were obtained from numerical simulations. With the help of computer programming, several times of computation were carried out to test the modified algorithm. The results show that the definition of the heuristic information value and the method of building objective functions can help the ants to identify the quality of the path, so that the ants can converge on the real route quickly. The computational efficiency is improved by 18 % -60 % compared with that of the algorithm that does not consider the coalition coefficient.

关 键 词：蚁群优化 寻源导热反问题 路径构造 目标函数 

分 类 号：TK124[动力工程及工程热物理—工程热物理]