With the development of mariculture moving to the deep sea, mariculture equipment will face worse sea conditions. To ensure the safety of the equipment, it is necessary to evaluate the safety of its anchoring system. Using the traditional hydrodynamic calculation and analysis method, the accurate mooring system tension can be obtained, but the calculation time is very long. To quickly obtain the tension distribution of the mooring system of the aquaculture platform under bad sea conditions, a calculation model for the prediction and evaluation of the mooring force of the aquaculture platform (ssa-bp model) is constructed based on the machine learning method. The sparrow search algorithm is introduced into the model to optimize the weight and threshold of the BP neural network, which improves the prediction performance of the model. The model takes the regular wave height, period, and velocity as the model input index and the mooring line force as the model output index to train the BP prediction model. The sparrow search algorithm is used to optimize and train the BP model (ssa-bp model), and the prediction results of the improved ssa-bp model are compared with those of the traditional BP model. Through comparative analysis, it is found that the overall indicators of ssa-bp model are lower than the BP model, and the prediction error of the mooring force of ssa-bp model under various working conditions is also lower than the BP model and closer to the real value. Finally, it is concluded that the new ssa-bp model can give more accurate prediction results.