Research On Ship Trim Optimization Method For Energy Consumption Management

To fulfill the responsibilities of international marine environmental protection and respond to the national "dual carbon" strategic policy,as well as to cope with the rising fuel prices,the related researches on ship energy saving and emission reduction has gradually attracted the attention of the International Maritime Organization(IMO),the maritime departments of various countries,the shipowners,and other relevant organizations.Under the background of shipping big data,effectively using the monitoring data of ship energy consumption to accurately predict the operating energy consumption of ships,and then combining the operation optimization methods to reduce the energy consumption of ships,is an effective way for ships to save energy and reduce emissions.The prediction and optimization technology of ship energy consumption are the key technologies to reduce ship energy consumption.However,the operating energy consumption of ships is affected by many factors,such as sailing state,draft,weather and sea conditions,etc.,resulting in many challenges for accurate prediction and optimization of ship energy consumption.Based on the above background,a number of real ships(container ship,oil tanker,bulk carrier)are taken as the research object,and their energy consumption is collected,and the data processing method is designed.Then,a real-time prediction model of energy consumption with high accuracy and strong robustness is established.Finally,the methods of ship energy saving and emission reduction based on trim optimization is proposed.The main research contents and innovations of this study are as follows:(1)To solve the problem of "remodeling method and ignoring data processing" in current research on ship energy consumption,a universal data processing method for ship energy consumption was proposed.By combining domain knowledge and statistical methods,firstly,the collection frequency synchronization,feature fusion and transformation of multi-source ship energy consumption data are carried out;Secondly,the related algorithm is designed to deal with the null value,noise value and outliers of the feature data;Finally,the feature selection is carried out using the Person correlation coefficient,and then a universal energy consumption data processing method is formed.The experimental results show that the proposed data processing method can effectively improve the quality of ship energy consumption data.It can provide the reliable data sets for subsequent energy consumption prediction and optimization studies.(2)In view of the problem that the existing energy consumption prediction research is based on a single-model established by a certain algorithm,which is difficult to analyze the energy consumption data from multiple perspectives or structures,resulting in poor accuracy or robustness of the energy consumption prediction model.Therefore,a real-time and hybrid prediction method of ship energy consumption based on model fusion is proposed.Firstly,some mainstream single energy consumption prediction models(based on neural networks,support vector machine,linear regression,and tree models)were analyzed;Secondly,based on the above analysis,a variety of single models with better performance are integrated,and then a real-time and hybrid prediction model of ship energy consumption is developed;Finally,the model performance is quantitatively evaluated in terms of both accuracy and robustness.The experimental results show that the proposed hybrid energy consumption prediction model is superior to the single models in both accuracy and robustness,and meets the real-time prediction requirements.It can provide theoretical methods and technical support for subsequent energy consumption optimization research.(3)The current research on ship energy consumption optimization mainly focuses on route and speed optimization methods,and rarely involves trim optimization research.In addition,the trim optimization method based on pool and simulation experiments fails to effectively consider the marine environmental factors(meteorology and sea conditions).Therefore,a data-driven ship trim optimization method is proposed.Firstly,based on the proposed hybrid energy consumption prediction model,combined with the ship’s cargo stowage and ballast water adjustment methods,two data-driven trim optimization models,static and dynamic,were established;Secondly,the influence of marine environmental factors on trim optimization is verified;Finally,based on the established dynamic trim optimization model,the inherent evolution law of the "speed-average draft-optimal trim value" is deeply studied.The simulation results show that both the static and dynamic trim optimization models of ships can effectively reduce the energy consumption of ships,thus contributing to the refined management of ship energy consumption.Moreover,considering the marine environment helps to accurately excavate the optimal trim value,so as to guide the crew to drive the ship with the best trim state.In addition,in the specific marine environment,the overall trend of the evolution of the optimal trim value of the ship is that it gradually decreases with the increase of the speed and the average draft,and finally tends to be stable.This research can provide a theoretical method reference for real-time monitoring and refined management of ship energy consumption.In addition,it can also provide technical support for energy conservation and emission reduction in the shipping industry,and promote the green and low-carbon development of the shipping industry.