In order to improve the accuracy and speed of solving the inverse problem of source-seeking heat conduction, the paper proposes a correlation-based ant colony optimization algorithm for the inverse problem of source-seeking heat conduction based on the characteristics of the influence of the heat source position on the boundary temperature distribution in the heat conduction problem. This method is used to construct the corresponding heuristic information value for each co-ordinate of the heat source location, which can reflect the degree of similarity between the temperature curve of the calculated measuring point and the temperature curve of the real measuring point, namely the correlation degree. The ant colony optimization algorithm the medium path selection mechanism and the structure of the objective function have been improved. The paper replaces the actual experiment with numerical calculation obtain the temperature of the measuring point, and performs computer programming experiment on the inverse problem. The calculation results show that the calibration method of this heuristic information value and the objective function the construction method can distinguish the quality of the path well, thereby increasing the speed of the ant colony converging to the best path. The computational efficiency is improved by 18-60% compared with the ant colony algorithm that does not consider the correlation.