基于工况聚类优化的船舶辐射噪声评估方法  

Ship-radiated noise evaluation method based on optimized working-conditions clustering

作　　者：李瑞彪[1,2] 何琳[1,2] 卜文俊[1,2] LI Ruibiao;HE Lin;BU Wenjun(Institute of Noise and Vibration,Naval University of Engineering,Wuhan 430033,China;National Key Laboratory on Ship Vibration and Noise,Naval University of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学振动与噪声研究所,湖北武汉430033 [2]海军工程大学船舶振动噪声重点实验室,湖北武汉430033

出　　处：《华中科技大学学报（自然科学版）》2020年第10期63-68,103,共7页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(11904407);湖北省自然科学基金资助项目(2019CFB247)。

摘　　要：针对现有船舶工况数据输入不能满足工况传递路径分析(OTPA)辐射噪声评估方法线性系统条件,从而导致部分频段辐射噪声评估误差较大的问题,提出基于聚类优化的船舶辐射噪声评估方法.该方法通过改进的双向二维主成分分析(2DD-PCA)法提取工况聚类特征,并利用K-means进行工况聚类,以聚类后的工况作为OTPA输入并评估辐射噪声.试验结果表明:基于聚类优化的船舶辐射噪声评估方法评估辐射噪声,1/3倍频程评估误差最大为2.1 dB,绝大部分频带小于1 dB,较OTPA方法下降约6.8 dB,验证了该方法的有效性.While evaluating ship-radiated noise with operational transfer path analysis(OTPA),a question arises that the input operating condition data cannot meet the linear system requirement of the method,which results in a large estimated error of radiated noise in several frequencies.Therefore,an optimized OTPA based on operating condition clustering for ship-radiated noise evaluation was proposed herein.An optimized method of two-directional-two-dimensional principal component analysis(2DD-PCA)to extract operating condition clustering feature was proposed,and K-means was used to cluster working-conditions based on the clustering feature.The classified working-conditions were taken as the input of OTPA to evaluate radiated noise.Experimental results show that the maximum estimated error of 1/3 octave by the optimized OTPA based on operating condition clustering is 2.1 dB,and less than 1 dB in most frequencies,which is reduced by approximately 6.8 dB compared with that of OTPA,verifying the feasibility of the proposed method.

关 键 词：辐射噪声 工况传递路径分析 工况聚类 主成分分析 K-MEANS 

分 类 号：U661.44[交通运输工程—船舶及航道工程] TB532[交通运输工程—船舶与海洋工程]