数据驱动的可调螺距桨船舶油耗模型及航速优化  

Data-driven Fuel Consumption Model and Speed Optimization of Ships with Controllable Pitch Propeller

作　　者：姚凤翔 王鸿东[1,2] 张海华[3] 王检耀 楼建坤 YAO Fengxiang;WANG Hongdong;ZHANG Haihua;WANG Jianyao;LOU Jiankun(State Key Laboratory of Ocean Engineering,Shanghai Jiao Tong University,Shanghai 200240,China;Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education,Shanghai Jiao Tong University,Shanghai 200240,China;Shanghai Branch,China Ship Scientific Research Center,Shanghai 200011,China)

机构地区：[1]上海交通大学海洋工程国家重点实验室,上海200240 [2]上海交通大学海洋智能装备与系统教育部重点实验室,上海200240 [3]中国船舶科学研究中心上海分部,上海200011

出　　处：《中国造船》2023年第2期226-239,共14页Shipbuilding of China

基　　金：国家自然科学基金资助项目(51909162)。

摘　　要：船舶油耗预测是实现船舶航速优化的基础。基于可调螺距桨船舶的历史航行数据,结合极限学习机(ELM)算法,开展数据驱动的船舶油耗预测模型研究。结合可调螺距桨螺距可调的特性,提出基于螺距调节的航速优化方法。以船舶燃油消耗量最小为优化目标,各时段内桨的螺距为优化变量构建船舶航速优化模型,并用改进粒子群算法求解。仿真结果表明:经螺距优化后,船舶单次航程的燃油消耗量减少约3%,进一步验证了通过螺距优化实现船舶节能航行的可行性。Prediction of ship fuel consumption is the basis of optimizing ship speed.Based on historical navigation data of a ship with a controllable pitch propeller(CPP)and combined with the algorithm of extreme learning machine,a data-driven model predicting fuel consumption is proposed.Combined with adjustable pitch characteristics of the CPP,a pitch adjustment-based method for speed optimization is proposed.Taking fuel consumption of the ship as the optimization objective and pitch values of the ship in each period as optimization variables,a model of speed optimization is constructed and solved numerically by using the optimization algorithm of improved particle swarm.The simulation result shows that fuel consumption of a single voyage of the ship with CPP is reduced by about 3%after the pitch optimization,which verifies the feasibility of optimizing the pitch to achieve energy-saving navigation.

关 键 词：数据驱动 油耗模型 可调螺距桨船舶 航速优化 改进粒子群算法 

分 类 号：U676.3[交通运输工程—船舶及航道工程]