极端工况下转子运行非稳态信息处理与动态平衡分析方法研究

Operation and maintenance of rotor under extreme conditions h is one of the most important objects in developing the modern mechanical equipment. This project aims at one of the basic common key issues concerning the steady operation of this kind of rotor under extreme environment in Xinjiang - methods of non-stationary vibration information processing and dynamic balancing. Through the analysis of the characteristic of the dynamic response, this proposal tries to promote the development of non-stationary balancing analysis method. The practically detected signals of rotating speed fluctuation and run-up and shutdown stages are used as the only source of information and the experimental validation, combined with theoretical modeling analysis, which is adopted to implement the whole research scheme. The analysis of the nonlinear nature and laws of the vibration response of flexible rotors under extreme condition, as well as its influence on the balancing analysis, will be performed in according to the proposal. Through the testing and analysis of rotor vibration response for non-stationary vibration signal, method of over-complete representation and sparse decomposition based on Compressive Sensing, method of feature extraction for non-stationary signal based on order tracking holospectra, the balance association analysis on the rotor spatial precession behavior will be studied. The multi-objective fuzzy assessment model and optimal scheme for coordination problem among the three balancing objectives is constructed. The goal of this proposal is to provide some theoretical and experimental bases for the construction of a reliable nonstationary vibration processing & dynamic balancing analysis method which is suitable for flexible rotor system under extreme condition.

面向极端工况下转子的运行维护一直是机械装备发展的重要方向。本项目针对新疆极端环境下柔性转子稳定运行涉及的关键共性基础问题-非稳态信息处理及动态平衡分析方法-展开研究，探索极端工况下柔性转子的动力学响应规律，发展非稳态平衡分析方法。研究以实测系统转速波动和转子起停车升降速信号为具体分析对象，采取理论建模分析结合实验验证的方法进行。重点研究极端工况下转子振动响应非线性性质、规律及其对平衡分析带来的影响，寻求解决方案。通过对转子振动响应测试非平稳信号的分析，研究基于压缩感知的非平稳信号超完备稀疏表示方法，研究基于阶比全息谱的非平稳信号特征提取及动平衡方法，分析转子空间进动行为的平衡关联性质，研究构建转子平衡状态的多目标模糊评价模型及平衡优化方案。通过以上研究和应用，最终形成一种具备高可靠性、适用于极端工况柔性转子的非稳态信息处理和动平衡分析新方法。

面向极端工况下转子的运行维护一直是机械故障诊断发展的重要方向。本项目针对新疆极端环境下柔性转子稳定运行涉及的关键共性基础问题-非稳态信息处理及动态平衡分析方法展开研究，研究提出了转子非平稳信号的超完备稀疏表示、基于阶比全息谱的非平稳信号特征提取方法和转子平衡目标的优化策略及高效动平衡等一整套具备高可靠性的柔性转子非稳态信息处理和动态平衡响应分析新方法。主要成果如下：. 1）提出将相空间表达模型的主分量分析方法引入压缩感知理论的稀疏表示方法，实现转子振动信号的稀疏性表征，完成对具有多频率成分的转子含噪信号的压缩与重构。进一步通过研究系统运行状态信息在状态空间中的拓扑性质，实现旋转机械运行状态特征自适应提取。. 2）提出基于阶比、分数阶全息的非平稳信号特征提取方法，利用其对调频信号处理能力，将转子起停车振动信号投影至分数阶域，克服了传统傅里叶变换只能处理稳态信号的缺陷，实现转子起停车振动信号的精确分解。在此基础上依据传统全息谱理论构造阶比全息瀑布图进行转子起停车过程特征提取及故障诊断；结合证据理论在多源信息决策层融合的优势，提出了一种符号熵与证据理论集成的转子故障识别方法。. 3）提出采用残余振动平方和、残余振动最大值和残余振动极差作为衡量机组平衡状况优劣的目标，提出了基于模型残差稀疏性控制的不平衡量识别方法，进一步提出采用微分搜索优化算法，对异常值敏感的最小二乘回归中的目标函数进行优化，得到存在外界干扰情况下的最佳估值，从而实现机组轴系动平衡的多目标优化，有效提高不平衡量识别过程的效率与稳健性。. 上述研究成果为长期运行在严苛自然环境和极端复杂工况下的各类大中型关键机组的运行与维护，尤其对关键部件-转子的可靠运行与视情维护提供了坚实的理论基础和技术支撑，保障我国边疆地区特定环境下大型旋转机械安全稳定运行，具有重要的科学意义和工程应用价值。