To predict stall and surge in advance that make the aero-engine compressor operate safely, a stall prediction model based on deep learning theory is established in the current study. The Long Short-Term Memory (LSTM) originating from the recurrent neural network is used, and a set of measured dynamic pressure datasets including the stall process is used to learn what determines the weight of neural network nodes. Subsequently, the structure and function hyperparameters in the model are deeply optimized, and a set of measured pressure data is used to verify the prediction effects of the model. On this basis of the above good predictive capability, stall in lowand high-speed compressor are predicted by using the established model. When a period of non-stall pressure data is used as input in the model, the model can quickly complete the prediction of subsequent time series data through the self-learning and prediction mechanism. Comparison with the real-time measured pressure data demonstrates that the starting point of the predicted stall is basically the same as that of the measured stall, and the stall can be predicted more than 1 s in advance so that the occurrence of stall can be avoided. The model of stall prediction in the current study can make up for the uncertainty of threshold selection of the existing stall warning methods based on measured data signal processing. It has a great application potential to predict the stall occurrence of aero-engine compressor in advance and avoid the accidents.

To assess the aerodynamic performance and vibration characteristics of rotor blades during rotation, a study of unsteady blade surface forces is conducted in a low-speed axial flow compressor under a rotating coordinate system. The capture, modulation, and acquisition of unsteady blade surface forces are achieved by using pressure sensors and strain gauges attached to the rotor blades, in conjunction with a wireless telemetry system. Based on the measurement reliability verification, this approach allows for the determination of the static pressure distribution on rotor blade surfaces, enabling the quantitative description of loadability at different spanwise positions along the blade chord. Effects caused by the factors such as Tip Leakage Flow (TLF) and flow separation can be perceived and reflected in the trends of static pressure on the blade surfaces. Simultaneously, the dynamic characteristics of unsteady pressure and stress on the blade surfaces are analyzed. The results indicate that only the pressure signals measured at the mid-chord of the blade tip can distinctly detect the unsteady frequency of TLF due to the oscillation of the low-pressure spot on the pressure surface. Subsequently, with the help of one-dimensional continuous wavelet analysis method, it can be inferred that as the compressor enters stall, the sensors are capable of capturing stall cell frequency under a rotating coordinate system. Furthermore, the stress at the blade root is higher than that at the blade tip, and the frequency band of the vibration can also be measured by the pressure sensors fixed on the casing wall in a stationary frame. While the compressor stalls, the stress at the blade root can be higher, which can provide valuable guidance for monitoring the lifecycle of compressor blades.