基于数据驱动的汽轮发电机组非线性振动故障诊断技术研究进展  

Research Progress of Nonlinear Vibration Fault Diagnosis Technology based on Data-Driven for Turbo-Generator Units

作　　者：王颖[1] 李录平[1] 陈尚年 樊昂 欧阳敏南 WANG Ying;LI Lu-ping;CHEN Shang-nian(School of Energy and Power Engineering,Changsha University of Science and Technology)

机构地区：[1]长沙理工大学能源与动力工程学院,湖南长沙410014

出　　处：《电站系统工程》2023年第1期1-6,共6页Power System Engineering

摘　　要：汽轮发电机组的精准、快速地故障诊断对提高其安全性、经济性具有重要意义。随着汽轮发电机组运行数据的积累与数据驱动技术的发展,基于数据驱动的汽轮发电机组非线性振动故障诊断技术成为了该领域内的研究重点。首先统计了汽轮发电机组振动故障的概率分布,阐述了主要的非线性振动故障机理;其次综述了振动故障的非线性特征提取技术、基于数据驱动的非线性振动故障诊断方法的研究进展;最后展望了汽轮发电机组非线性振动故障诊断技术的发展方向:将虚拟现实、数字孪生、人工智能、数据挖掘深度融合,实现汽轮发电机组非线性振动故障诊断。The accurate and fast fault diagnosis of turbo-generator units is of significance to improve their safety and economy. With the accumulation of operating data and the development of data-driven technology, a nonlinear vibration fault diagnosis technology of turbo-generator units based on data-driven has become the focus of research in this field.Firstly, the probability distribution of vibration faults for turbine-generator sets is calculated, and the main nonlinear vibration fault mechanisms are described. Secondly, the research progress of the extraction technology of nonlinear feature for vibration fault and fault diagnosis methods of nonlinear vibration fault for turbo-generator sets based on data-driven are reviewed. Finally, the development direction of turbine-generator nonlinear vibration fault diagnosis technology is foreseen: making the virtual reality, digital twin, artificial intelligence, data mining deep integration to realize nonlinear vibration fault diagnosis.

关 键 词：汽轮发电机组 非线性振动 特征提取 故障诊断 机器学习 

分 类 号：TK267[动力工程及工程热物理—动力机械及工程]