论文部分内容阅读
针对压气机气动失速预警存在无法兼顾对渐进型和突尖型失速先兆有效识别的不足,以及提取失速团幅值需要多路传感器信号的局限。分析了压气机失速发展过程中的信号特征;根据压气机失速信号的旋转特性,提出基于单路传感器信号重构周向多路失速先兆信号的方法,讨论了信号重构步长的取值;基于模态波理论,通过对重构信号进行空间傅里叶变换,得到失速扰动信号各阶模态幅值,根据低阶模态的幅值变化实现对渐进型与突尖型失速先兆的识别。以某压气机为例进行仿真,结果表明:当信号重构的步长小于突尖波时间尺度的一半时,重构的多路信号能够复现压气机周向位置失速发展过程,且步长越小,越有利于失速先兆的识别;由模态分解得到的1阶模态幅值的变化能够有效反映失速发展过程;通过合理设置模态幅值的阀值,可以在失速发展前期对失速先兆进行准确识别,从而,基于单路传感器失速信号即可实现对渐进型失速和突尖型失速均能进行预警。
Aiming at the shortcomings of the early warning of the aerodynamic stall of the compressor, it can not take into account both the lack of recognition of progressive and sudden sharp-type stall precursors and the limitation of multi-sensor signals to extract the amplitude of the stalls. According to the rotation characteristics of the compressor stall signal, a method of reconstructing the multi-path stalling precession signal based on the single sensor signal is proposed, and the value of signal reconstruction step is discussed. Based on the modal wave theory, the modal amplitudes of the stalled disturbing signals are obtained by Fourier transform of the reconstructed signals, and the identification of progressive and sudden apexes is realized according to the amplitude changes of the low order modes . Taking a compressor as an example, the simulation results show that when the signal reconstruction step size is less than half of the sudden spike time scale, the reconstruction of the multi-channel signal can reproduce the compressor circumferential position of the stall development process, and the step The smaller the value, the more favorable to the identification of stalling precursors. The variation of the first-order modal amplitude obtained from the modal decomposition can effectively reflect the stalling development process. By setting the modal amplitude threshold reasonably, The aura is accurately identified so that early warning of both progressive and sudden sharp stalls can be achieved based on a single sensor stall signal.