基于BSO-BP的船舶油耗预测模型  

Prediction model of ship fuel consumption based on BSO-BP

作　　者：乔磊[1,2,3] 尹奇志 姚昌宏[4] 钱巍文 赵福芹 QIAO Lei;YIN Qizhi;YAO Changhong;QIAN Weiwen;ZHAO Fuqin(School of Transportation and Logistics Engineering,Wuhan University of Technology,Wuhan 430063,Hubei,China;Reliability Engineering Institute,Wuhan University of Technology,Wuhan 430063,Hubei,China;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063,Hubei,China;Weichai Heavy Machinery Co.,Ltd.,Weifang 261061,Shandong,China;Weichai Power Co.,Ltd.,Weifang 261061,Shandong,China)

机构地区：[1]武汉理工大学交通与物流工程学院,湖北武汉430063 [2]武汉理工大学可靠性工程研究所,湖北武汉430063 [3]武汉理工大学国家水运安全工程技术研究中心,湖北武汉430063 [4]潍柴重机股份有限公司,山东潍坊261061 [5]潍柴动力股份有限公司,山东潍坊261061

出　　处：《上海海事大学学报》2024年第2期29-34,共6页Journal of Shanghai Maritime University

基　　金：工业和信息化部绿色智能内河船舶创新专项(MC-202002-C03);潍柴动力股份有限公司技术项目(WCDL-GH-2021-0050)。

摘　　要：为解决基于传统反向传播(back propagation,BP)神经网络的船舶油耗预测模型易陷入极小值和误差较大的问题,提出一种利用头脑风暴优化(brain storm optimization,BSO)算法优化BP神经网络的船舶油耗预测模型(简称BSO-BP模型)。以“维多利亚凯娅”号内河游船为研究对象,将BSO-BP模型的预测结果与采用传统BP神经网络以及模拟退火(simulated annealing,SA)算法、遗传算法(genetic algorithm,GA)、粒子群优化(particle swarm optimization,PSO)算法优化的BP神经网络的船舶油耗预测模型的预测结果进行对比分析。结果表明:与传统BP神经网络模型的预测结果相比,BSO-BP模型预测结果的可决系数R^(2)提高了0.003 9,均方误差、均方根误差、平均相对误差、平均绝对误差分别降低了0.034 4、0.154 1、0.010 2、0.017 8,说明在船舶油耗预测中BSO算法对BP神经网络的预测精度有显著的提升作用;BSO-BP模型预测结果的各项评价指标在所对比的5种模型中均表现最好,说明与SA算法、GA和PSO算法相比,BSO算法对BP神经网络的提升效果更好。The ship fuel consumption prediction model based on the traditional back propagation(BP) neural network is easy to fall into the minimum value and its error is large.In order to solve these problems,a ship fuel consumption prediction model using the brain storm optimization(BSO) algorithm to optimize BP neural network,called BSO-BP model,is proposed.Taking “Victoria Keje” inland river cruise ship as the research object,the prediction results of the BSO-BP model are compared with those of the models using the traditional BP neural network and the BP neural networks optimized by the simulated annealing(SA) algorithm,the genetic algorithm(GA),the particle swarm optimization(PSO) algorithm,respectively.The results show that,compared with the prediction results of the traditional BP neural network model,the determination coefficient R^(2) of the prediction results of the BSO-BP prediction model increases by 0.003 9,and its mean square error,root mean square error,mean relative error and mean absolute error decrease by 0.034 4,0.154 1,0.010 2,0.017 8,respectively,which indicates that the BSO algorithm can significantly improve the prediction accuracy of the BP neural network in the ship fuel consumption prediction;the evaluation indicators of the prediction results of the BSO-BP model are the best among five models,which indicates that BSO algorithm is of better improvement effect on BP neural network than SA algorithm,GA and PSO algorithm.

关 键 词：船舶油耗预测模型 头脑风暴优化(BSO) 反向传播(BP)神经网络 

分 类 号：U676.3[交通运输工程—船舶及航道工程]