针对海水泵的复杂多激励源难以准确识别问题，提出一种基于变分模态分解（VMD）、鲸鱼优化算法（WOA）、多尺度熵（MSE）和支持向量机（SVM）相结合的海水泵激励源识别方法。基于海水泵振动信号，首先采用VMD和WOA算法获取分解层数K与惩罚系数α两个重要参数，并对信号进行分解重组；然后提取重组信号的多尺度熵，作为WOA-SVM模型的输入特征向量，并对SVM的惩罚因子c和核系数g两个重要参数进行寻优，最后将得到的参数代入SVM模型进行训练与激励源识别。通过实船海水泵激励源识别及对比分析，验证识别方法的有效性。研究结果表明提出的WOA-VMD-MSE-SVM算法满足海水泵激励源识别准确要求。
Aiming at the problem that it is difficult to identify the complex multiple excitations of the sea water pump accurately, a method based on the combination of variational modal decomposition(VMD) , whale optimisation algorithm (WOA) ,multi-scale entropy (MSE) and support vector machine (SVM) is proposed to identify the excitation source of the sea water pump. Based on the vibration signal of the sea water pump, the WOA algorithm and VMD algorithm are firstly used to optimise the two important parameters , the number of decomposition layers K and the penalty coefficient α,then the signal is decomposed and reorganised. The multiscale entropy of the restructured signal is extracted and input into the WOA-SVM model as a feature vector to perform parameter optimisation on the two important parameters of the SVM, the penalty factor c and the kernel coefficient g. Finally, the obtained coefficients are brought into the SVM model for training and excitation source identification. Through the real ship seawater pump excitation source identification and comparative analysis, verified that the identification method is effective. The research results surface that the proposed WOA-VMD-MSE-SVM algorithm meets the accurate requirements for the identification of excitation sources of seawater pumps.
2024,46(18): 44-48 收稿日期：2023-10-11
DOI：10.3404/j.issn.1672-7649.2024.18.007
分类号：U661.44;TH165+.3
基金项目：基础产品创新科研资助项目（0207024）
作者简介：滕佳篷(1998-),男,硕士研究生,研究方向为海水泵激励源识别
参考文献：
[1] RIZWAN-UL-H, LI C, LIU Y. Online dynamic security assessment of wind integrated power system using SDAE with SVM ensemble boosting learner[J]. International Journal of Electrical Power and Energy Systems, 2021, 125: 106429.
[2] MINSHUI H, YONGZHI L , XIFAN L, et al. Damage identification of bridge structures considering temperature variations-Based SVM and MFO[J]. Journal of Aerospace Engineering, 2021, 34(2): 04020113.
[3] 吕石磊, 王宏炜, 李震, 等. 基于BWO-SVM算法的广陈皮陈化年份高光谱鉴别模型[J]. 光谱学与光谱分析, 2023, 43(9): 2894–2901.
[4] 任晨曦, 王黎明, 韩星程, 等. 基于联合神经网络的水声目标识别方法[J]. 舰船科学技术, 2022, 44(1): 136-141.
REN Chenxi, WANG Liming, HAN Xingcheng, et al. Underwater acoustic target recognition method based on joint neural network[J]. Ship Science and Technology, 2022, 44(1): 136-141.
[5] 马玉鹏, 郑茂, 吴勇, 等. 基于神经网络特征提取的, SSA-SVM船舶目标识别方法[J]. 舰船科学技术, 2023, 45(8): 158-164.
MA Yupeng, ZHENG Mao, WU Yong, et al. Based on neural network feature extraction research on SSA-SVM ship target identification method[J]. Ship Science and Technology, 2023, 45(8): 158-164.
[6] HANG J, ZHANG J, CHENG M. Application of multi-class fuzzy support vector machine classifier for fault diagnosis of wind turbine[J]. Fuzzy Sets and Systems, 2016, 297(15): 128–140.
[7] 李一鸣. 基于小波包多尺度模糊熵和加权KL散度的煤岩智能识别[J]. 工矿自动化, 2023, 49(4): 92-98.
[8] 赵江平, 张雪莹, 侯刚. 基于多源信息融合和WOA-CNN-LSTM的外脚手架隐患分类预警研究[J/OL]. 安全与环境学报: 1–10[2023-09-26].
[9] 左思雨, 赵强, 张冰, 等. 基于VMD-SSA-GRU的船舶运动姿态预测[J]. 舰船科学技术, 2022, 44(23): 60-65+69.
ZUO Siyu, ZHAO Qiang, ZHANG Bing, et al. Prediction of ship motion attitude based on VMD-SSA-GRU mode l[J]. Ship Science and Technology, 2022, 44(23): 60-65+69.
[10] 刘强, 赵荣珍, 杨泽本. K-VMD融合包络熵与SVM滚动轴承故障识别方法研究[J]. 噪声与振动控制, 2022, 42(3): 92-97+121.
[11] 张龙, 彭小明, 熊国良, 等. 基于MSE与PSO-SVM的机车轮对轴承智能诊断方法[J]. 铁道科学与工程学报, 2021, 18(9): 2408-2417.
