首页|基于FCM和EO-SVM水轮机尾水管压力脉动特征识别

基于FCM和EO-SVM水轮机尾水管压力脉动特征识别

扫码查看
为有效识别水轮机尾水管压力脉动特征,提出了 一种基于模糊C均值聚类、平衡优化器算法与支持向量机的识别方法.该方法首先采用平衡优化器算法优化SVM的惩罚因子和核函数以获得更好的SVM参数组合,构建EO-SVM识别模型以实现其在水轮机尾水管压力脉动特征识别中的应用.然后采用模糊C均值聚类算法将待分类的压力脉动特征进行初始聚类,将其分为四类,并依据聚类结果选择最靠近每类中心的样本作为EO-SVM模型的训练样本.将SVM和EO-SVM两种模型的识别分类结果进行比较,验证了所提EO-SVM模型的有效性.
Characteristics Identification of Pressure Fluctuation in Draft Tube of Hydraulic Turbine Based on FCM and EO-SVM
In order to effectively analyze the characteristics of pressure fluctuation in draft tube of hydraulic turbine,an identification method based on fuzzy C-means clustering,equilibrium optimizer algorithm,and support vector machine was proposed.Firstly,the equilibrium optimizer algorithm was used to optimize the penalty factor and kernel function of SVM to obtain a better combination of SVM parameters,and the EO-SVM identification model was constructed to realize its application in the identification of pressure pulsation characteristics in the draft tube of hydraulic turbine.Then the fuzzy C-means clustering algorithm was used to initially cluster the pressure fluctuation characteristics to classify into four categories.According to the clustering results,the samples closest to the center of each category were selected as the training samples of EO-SVM model.The identification and classification results of the SVM and EO-SVM were com-pared to verify the effectiveness of the proposed EO-SVM model.

pressure fluctuationwavelet packet analysisfuzzy C-means clusteringequilibrium optimizer algo-rithmsupport vector machine

刘茜媛、王利英、张路遥、曹庆皎

展开 >

河北工程大学水利水电学院,河北邯郸 056038

压力脉动 小波包分析 模糊C均值聚类 平衡优化器算法 支持向量机

国家自然科学基金项目

11972144

2024

水电能源科学
中国水力发电工程学会 华中科技大学 武汉国测三联水电设备有限公司

水电能源科学

CSTPCD北大核心
影响因子:0.525
ISSN:1000-7709
年,卷(期):2024.42(1)
  • 3