In recent years, safety issues respecting polar ship navigation in the presence of ice have become a research hotspot. The accurate prediction of propulsion power plays an important role in ensuring safe ship navigation and evaluating ship navigation ability, and deep learning has been widely applied in the field of shipping, of which the artificial neural network (ANN) is a common method. This study combines the scientific problems of ice resistance and propulsion power for polar ship design, focusing on the design of an ANN model for predicting the propulsion power of polar ships. Reference is made to the traditional propulsion power requirements of various classification societies, as well as ship model test and full-scale test data, to select appropriate input features and a training dataset. Three prediction methods are considered: building a radial basis function–particle swarm optimization algorithm (RBF-PSO) model to directly predict the propulsion power; based on the full-scale test and model test data, calculating the propulsion power using the Finnish–Swedish Ice Class Rules (FSICR) formula; using an ice resistance artificial neural network model (ANN-IR) to predict the ice resistance and calculate the propulsion power using the FSICR formula. Prediction errors are determined, and a sensitivity analysis is carried out with respect to the relevant parameters of propulsion power based on the above methods. This study shows that the RBF-PSO model based on nine feature inputs has a reasonable generalization effect. Compared with the data of the ship model test and full-scale test, the average error is about 14%, which shows that the method has high accuracy and can be used as a propulsion power prediction tool.

When navigating ships in cold regions, sea ice concentration plays a crucial role in determining a ship’s navigability. However, automatically extracting the sea ice concentration and floe size distribution remains challenging, due to the difficulty in detecting all the ice floes from the images captured in complex polar environments, particularly those that include both ships and sea ice. In this paper, we propose using the YOLACT network to address this issue. Cameras installed on the ship collect images during transit and an image dataset is constructed to train a model that can intelligently identify all the targets in the image and remove any noisy targets. To overcome the challenge of identifying seemingly connected ice floes, the non-maximum suppression (NMS) in YOLACT is improved. Binarization is then applied to process the detection results, with the aim of obtaining an accurate sea ice concentration. We present a color map and histogram of the associated floe size distribution based on the ice size. The speed of calculating the sea ice density of each image reaches 21 FPS and the results show that sea ice concentration and floe size distribution can be accurately measured. We provide a case study to demonstrate the effectiveness of the proposed approach.