Journal of Propulsion Technology ›› 2013, Vol. 34 ›› Issue (2): 263-268.

Previous Articles     Next Articles

Improved Fuzzy Membership Method in FSVM for Aeroengine Vibration Performance Fusion Analysis

  

  1. AVIC Shenyang Engine Design and Research Institute, Shenyang 110015, China ;School of Jet Propulsion, Beijing University of Aeronautics and Astronautics, Beijing 100191, China;School of Jet Propulsion, Beijing University of Aeronautics and Astronautics, Beijing 100191, China;School of Jet Propulsion, Beijing University of Aeronautics and Astronautics, Beijing 100191, China
  • Published:2021-08-15

基于FSVM改进隶属度的发动机振动性能分析

郭秩维1,2,费成巍2,白广忱2   

  1. 中航工业沈阳发动机设计研究所,辽宁 沈阳 110015;北京航空航天大学 能源与动力工程学院,北京 100191;北京航空航天大学 能源与动力工程学院,北京 100191;北京航空航天大学 能源与动力工程学院,北京 100191
  • 作者简介:郭秩维(1982—),男,博士生,研究领域为航空发动机结构可靠性设计与优化、振动分析等。通讯作者:费成巍(1983—),男,博士生,研究领域为航空发动机可靠性设计与多学科优化、故障分析等。E-mail:Feicw544@163.com
  • 基金资助:
    国家自然科学基金(51175017);北京航空航天大学博士研究生创新基金(YWF-12-RBYJ-008)

Abstract: To master more effectively the impact factors on an aeroengine whole-body vibration performance, the improved Fuzzy Support Vector Machine (FSVM) information entropy technique was proposed. Firstly, the computing model of multi-class fuzzy membership was established based on the improved fuzzy membership of FSVM and the information entropy theory. And secondly, this method was applied to the aeroengine vibration performance evaluation, and the multi-parameter vibration performance analysis model was developed and the relationship between fault modes and fault causes was determined.Thus, aeroengine overall vibration performance was quantitatively analyzed and a quantitative reference index was provided for aeroengine vibration control. Finally, the validity and feasibility of this method in aeroengine whole-body vibration performance analysis were validated by mean of the fusion analysis of example.

Key words: Aeroengine; Fuzzy support vector machine; Information entropy; Fuzzy membership; Performance analysis

摘要: 为了更有效地掌握航空发动机振动性能的影响因素,提出了改进FSVM信息熵的融合定量分析方法。首先,对模糊支持向量机(FSVM)的模糊隶属度函数进行改进,建立多类模糊隶属度计算模型。再将该方法应用到航空发动机整机振动性能评估,计算出振动故障模式与故障原因之间的权值,建立了一个多参数的发动机振动性能分析模型;并对各类振动原因对发动机整体性能的影响进行定量分析,为发动机的振动抑制提供量化参考指标。最后,通过与实际经验作比较,验证了该方法是可行和有效的。

关键词: 航空发动机;模糊支持向量机;信息熵;模糊隶属度;性能分析