<p>Mining-induced water-conducting fractures are the root cause of water disaster in coal mines. However, the traditional prediction for the height of the water-conducting fracture under complex stress environments remains challenging due to low accuracy and poor applicability. In this study, the conventional prediction parameters for the height of the water-conducting fracture were used, the stress growth coefficient of the overburden was introduced, and a dynamic all-dimensional prediction model for the height of the water-conducting fracture was constructed using the APSO algorithm. Subsequently, a dynamic all-dimensional APSO algorithm prediction method for the height of the water-conducting fracture in complex stress environments was proposed, and prediction software was developed based on the proposed method. The prediction results show that 90% of the predicted data exhibit an accuracy of up to 95%. The maximum deviation in on-site measurement points remains within a tolerance of 2%, and the absolute error in predicting the height of water conduction crack is less than 1.5&#xa0;m. The research results are of great significance for the prevention and control of water disaster in the coal mine.</p>