Research on building an ocean wave level prediction model using convolutional neural network (CNN) for a ship steering system

The maritime industry plays a vital role in transporting international commercial goods by sea. To enhance efficiency and safety, modern ships are increasingly equipped with advanced and intelligent technologies. Among these, neural network-based systems offer significant benefits by optimizing routes and reducing fuel consumption. This paper proposes a wave level forecasting model using a Convolutional Neural Network (CNN). The model achieves a high accuracy rate of up to 95.6% in predicting current wave levels. Additionally, it is integrated into a maritime simulation system connected with a real steering system, enabling operators to receive optimal recommendations under various weather conditions. This combination enhances navigational safety and minimizes operational risks. By simulating real scenarios, the system also provides valuable support in training and decision-making. Looking ahead, as intelligent autonomous ships become more prevalent, the proposed CNN-based forecasting model holds strong potential for practical application. Its implementation can significantly improve maritime operations, contributing to the industry's modernization and overall operational quality.