This article introduces a method based on a BP neural network to predict the total force and moment of a ship caused by irregular waves in the time domain. A large bulk carrier is modeled and analyzed hydrodynamically, and a significant volume of data on ship acceleration response and total force and moment are obtained by changing the angle of wave incidence. These data are used as the basis for training and testing the neural network model. In this study, an improved BP neural network is utilized to fit the relationship between the ship acceleration response and total force and moment under multiple wave incidence angles. In order to evaluate the generalization ability of the neural network, types of wave incidence angles that have not appeared in the training set are introduced into the test set data. Empirical results validate the effectiveness and feasibility of this approach, which is anticipated to enhance ship design efficiency and provide valuable information for offshore safety assessments.