Energy Efficiency Prediction Model of Suction Hopper Dredger Based on Correlation Analysis and Convolutional Neural Network

Suction hopper dredgers are the main type of ship used in dredging projects. Due to the complex dynamic characteristics of the dredging process and the fact that the power consumption characteristics of dredgers are affected by many factors such as soil, meteorological environment, and hydrology, it is difficult to obtain accurate prediction results using traditional methods. This paper proposes a data extraction method based on correlation analysis to filter out the most critical influencing features from a large amount of data that affects the working efficiency of dredgers. An energy efficiency prediction model is established using convolutional neural network method to estimate the actual production efficiency of bucket-wheel dredgers and applied to actual production cases. The results show that the energy efficiency prediction model established using this method can effectively predict the power consumption of suction hopper dredgers.