Abstract: In order to accurately evaluate the complexity of the mine geological structure, taking Huangling No.1 coal mine as the research object, based on the analysis of the development characteristics and laws of the mine geological structure, 11 evaluation indicators that can reflect and affect the complexity of the mine geological structure are selected. The well field is divided into 185 evaluation units according to the specification of 1 km×1 km, the evaluation index value of each evaluation unit is calculated, and each evaluation index value is divided into 4 categories according to the orderly quality optimal segmentation analysis, corresponding to the geological structure of simple, medium, complex, and extremely complex. The training samples of BP neural network are obtained by using the segment interpolation method. In order to overcome the defects of the simple BP neural network program, which lacks the global optimization of the hidden layer neuron structure, slows the convergence speed and is easy to fall into the local minimum value, this paper tries to use the BP neural network algorithm（ IGA-BP） based on the immune geneticalgorithm （IGA） for optimization to comprehensively evaluate the complexity of the mine geological structure. With the help of the established training samples, the global optimization of the hidden layer structure of the BP network and the training of the BP neural network were successfully realized. Finally, the trained IGA-BP network was used to comprehensively evaluate the complexity of the geological structure of the unknown evaluation unit, and the complex degree zoning map of mine structure is drawn. The results show that the structurally simple area is located in the north, northeast and south of the study area, the structurally complex area is located in the west of the middle of the study area, and the structurally moderate area is distributed on the north and south sides of the tectonic complex area in the middle of the study area. Compared with the GA-BP and BP neural network methods, the evaluation results based on IGA-BP are more consistent with the actual situation of the mine, and the IGA-BP evaluation method does not need to consider the correlation and weight between the evaluation indicators, which provides a useful tool for mine structure evaluation. A new evaluation method, the evaluation results can guide the reasonable mining deployment of the mine.