Aiming at the problem that the operation of dredging equipment such as the trailing arm system is affected by waves,and the requirement for rapid prediction of waves during the operation of dredging ship,a rapid wave prediction model for the operation sea area of dredgers is established based on the ConvLSTM neural network model.The numerical simulation results of wave height and period in the dredging project sea area are used as training samples to train the wave prediction network model.Based on the analysis of the demand for rapid wave prediction in the dredging operation process,this paper proposes ConvLSTM-based neural network models for rapid wave prediction in dredging project areas.The models are trained using retrospective data from traditional wave mathematical models,enabling accurate and rapid prediction of wave height and period in dredging operation water areas.The results show that the established neural network required only 2-5 s for wave prediction in the dredging project,and the correlation of wave prediction for the forward 6 h wave height distribution in the operating sea area reaches 0.956.When the prediction step size increases to 12 h,the neural network’s prediction accuracy for waves significantly decreases to 0.849 due to the uneven training data.Therefore,this model can be used for short-term and rapid wave prediction in dredging operation areas,providing a basis for safe operation of ships.