摘要
风电机组齿轮箱故障一直是风电场主要机械故障之一,其故障信息多是混有噪声的非平稳信号。为避免陷入对复杂的非线性信号求解,提出多种算法融合下的数字信息频率筛查处理方法。首先利用小波分析对高频信号的敏感性进行消噪,然后充分利用Hilbert-Huang变换对非平稳信号的分解和时频变换能力进行信号特征挖掘,在被干扰的非平稳信号里面找出故障对应频率。充分利用多种算法融合有效去除与主频率不相关的高频噪声,找出时域变换的边际谱。避免了直接分解含有高频信号的不确定性,降低了信号分解过程中回流成分,避免发生反射现象,完成对风电机组齿轮箱故障的非平稳信号频率筛查比对。通过对某机组进行故障实验,验证数字信息频率筛查方法的有效性。
The fault ofWind Turbine Gearbox is always one of the main mechanical faults in wind farm,and its fault characteristic information is mostly non-stationary signal mixed with noise.In order to avoid falling into complex nonlinear solving problems,a digital information frequency screening processing method under the fusion of multiple algorithms is proposed.First,use wavelet analysis to denoise the sensitivity of high-frequency signals,and then make full use of the Hilbert-Huang transform to decompose non-stationary signals and time-frequency transformation capabilities for signal feature mining,and find faults in non-stationary signals under interference Corresponding frequency.Make full use of a variety of algorithm fusion to effectively remove high-frequency noise that is not related to the main frequency,and find the marginal spectrum of the time domain transform.It avoids the uncertainty of direct decomposition of high-frequency signals,reduces the return components in the signal decomposition process,avoids reflection phenomena,and completes the non-stationary signal frequency screening and comparison of the wind turbine gearbox failure.Through a failure experiment on a certain unit,the effectiveness of the digital information frequency screening method is verified.
作者
宋建
束洪春
董俊
常勇
SONG Jian;SHU Hong-chun;DONG Jun;CHANG Yong(Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Yunnan Kunming 650500,China;Facuity of Electric Power Engineering,Kunming University of Science and Technoligy,Yunnan Kunming 650500,China)
出处
《机械设计与制造》
北大核心
2022年第1期15-19,共5页
Machinery Design & Manufacture
基金
国家自然科学基金项目(51667010)
云南省重大科技专项计划(202002AF080001)
昆明理工大学虚拟仿真实验教改项目。
