论文部分内容阅读
为了研究高速多级轴流压气机中复杂气动失稳现象的检测问题,提出了一种基于D-S(Dempster-Shafer)证据融合的通用型检测算法。在时域中,采用短时能量表达气动失稳过程中动态压力的脉动幅值特征;在频域中,选择频谱相关系数表达信号频谱变化的特征。将这两种信号特征分别作为气动失稳现象的证据。根据统计规律设计了各证据的mass分配函数。采用Dempster合成规则计算联合证据的mass函数。通过分别比较单一证据及联合证据的mass函数值与其门限值,判决压气机处于气动失稳、失速、喘振或正常状态。进一步地,将该方法扩展为气动失稳检测的多传感器融合模型。该方法计算量小,适用于在线检测系统。采用压气机台架试验实测数据验证,可在气动失稳数毫秒内发出报警信号。
In order to study the detection of complex aerodynamic instability in high-speed multi-stage axial flow compressors, a universal detection algorithm based on D-S (Dempster-Shafer) evidence fusion is proposed. In the time domain, the short-time energy is used to express the characteristics of the fluctuating amplitude of dynamic pressure in the process of aerodynamic instability. In the frequency domain, the spectral correlation coefficient is used to express the characteristics of the signal spectrum changes. The two signal characteristics are taken as evidence of the phenomenon of aerodynamic instability. According to the law of statistics, the mass distribution function of each evidence is designed. The mass function of the joint evidence is calculated using the Dempster synthesis rule. Comparing the mass function value of the single evidence and the joint evidence respectively with the threshold values, the compressor is judged to be in aerodynamic instability, stall, surge or normal condition. Further, the method is extended to multi-sensor fusion model of aerodynamic instability detection. The method is small in calculation and suitable for on-line inspection system. Using compressor bench test data validation, can be issued within a few milliseconds aerodynamic instability alarm signal.