Abstract: Diagnosis and early warning of compressor instability in aero engines are among the current research hotspots and challenges in the field of aero engines. Addressing challenges such as the difficulty in capturing non-integer order frequency compo-nents in acoustic signals, the singularity of feature dimensions, the complexity of instability mechanisms, and the difficulty in quantifying evolutionary trajectories of compressor instability, this paper proposes a compressor aerodynamic instability early warning method based on phase-locked averaging filter and moment function neural network, using a multi-stage high-speed compressor as the research object. The method first employs phase-locked averaging filter to extract non-integer order frequency disturbance features under high-load conditions. Subsequently, an instability early warning model based on mo-ment function neural network is constructed, which utilizes moment functions to capture global statistical features of instabil-ity and local detail features of asymmetric separation and intermittent pulses in early-stage weak signals. Next, Box-Cox transformation is introduced to eliminate heterogeneity among higher-order moment features, and multi-layer perceptron network layers are adopted to achieve early warning and diagnosis of compressor instability. Finally, the effectiveness of the proposed method is validated through experimental data at different rotational speeds. Results demonstrate that the method accurately characterizes the evolutionary laws of higher-order moment feature spaces, enabling efficient visualization, identi-fication, and separation of instability precursors and steady-state data, with high accuracy and robustness in instability detec-tion.

Key words: compressor instability, fault early warning, moment function, phase-locked averaging filter, neural network