Algorithm of axial compressor stall warning based on BP neural network and fuzzy logic

Stall warning of axial compressor is very challenging and the existing warning margin is not enough. A algorithm based on BP neural network fusion fuzzy logic is proposed. Firstly, BP neural network is used for training recognition, next the identification results are fused with fuzzy logic reasoning to form the result judgment of time sequence, finally the stall early warning of axial compressor is realized. The simulation results of the experimental data show that the stall data at all speeds are at least 0.1s in advance of the early warning. Compared with other methods, this method has a better surge early warning margin performance and engineering practicability.

Stall Warning Strategy Based On Fast Wavelet Analysis in a Multi-Stage Axial Flow Compressor

As a reliable stall warning strategy, the fast wavelet method was introduced to successfully predict the aerodynamic instability of a multi-stage axial flow compressor. One single sensor installed at each stage is proved to be sufficient to predict the stability status in a three-stage axial flow compressor. The whole prediction strategy includes the dynamic pressure signal capture, disturbance extraction using decomposition and reconstruction via fast wavelet transform, and stall warning index calculation based on statistical probability distribution. On this premise, the first occurrence of the stall in this three-stage axial flow compressor is predicted to be within the first stage, which is consistent with the stall route captured by the eight transducers around the casing wall. Thereafter, the stall warning index is used to monitor the stability status during the continuous throttling process. Furthermore, the validation using tip air injection and inlet radial distortion indicated that the stall warning index decreases as the compressor's stability improves. Conversely, the deterioration of stability causes the increase of the stall warning index. Thus, experimental results demonstrate that the stall warning method based on fast wavelet analysis can predict the aerodynamic instability in actual application.

Compressor Stall Warning Using Nonlinear Feature Extraction Algorithms

Stall is a type of flow instability in compressors that sets the low-flow limit for compressor operation. During the past few decades, efforts to develop a reliable stall warning system have had limited success. This paper focuses on the small nonlinear disturbances prior to deep surge and introduces a new approach to identify these disturbances using nonlinear feature extraction algorithms including phase-reconstruction of time-series signals and evaluation of a parameter called approximate entropy. To the best of our knowledge, this is the first time approximate entropy has been used for stall warning, and thus, its definition and utility are presented in detail. The technique is applied to stall data sets from two different compressors: a high-speed centrifugal compressor that unexpectedly entered rotating stall during a speed transient and a multistage axial compressor with both modal- and spike-type stall inception. In both cases, nonlinear disturbances appear, in terms of spikes in approximate entropy, prior to surge. The presence of these presurge spikes indicates the potential of using the approximate entropy parameter for small disturbance detection and stall warning. The details of the nonlinear feature extraction algorithm, including guidelines for its application as well as results from applying the algorithm to rig-level data, are presented.

Compressor Stall Warning Using Nonlinear Feature Extraction Algorithms

Stall is a type of flow instability in compressors that sets the low-flow limit for compressor operation. During the past few decades, efforts to develop a reliable stall warning system have had limited success. This paper focuses on the small nonlinear disturbances prior to deep surge and introduces a new approach to identify these disturbances using nonlinear feature extraction algorithms including: phase-reconstruction of time-series signals and evaluation of a parameter called approximate entropy. To the best of the authors’ knowledge, this is the first time approximate entropy has been used for stall warning, and thus, its definition and utility are presented in detail. The technique is applied to stall data sets from two different compressors: a high-speed centrifugal compressor that unexpectedly entered rotating stall during a speed transient and a multi-stage axial compressor with both modal- and spike-type stall inception. In both cases, nonlinear disturbances appear, in terms of spikes in approximate entropy, prior to surge. The presence of these pre-surge spikes indicates the potential of using the approximate entropy parameter for small disturbance detection and stall warning. The details of the nonlinear feature extraction algorithm, including guidelines for its application as well as results from applying the algorithm to rig-level data, are presented.

Comparison and Sensibility Analysis of Warning Parameters for Rotating Stall Detection in an Axial Compressor

The present paper aims at evaluating the surveillance parameters used for early stall warning in axial compressors, and is based on unsteady pressure measurements at the casing of a single stage axial compressor. Two parameters—Correlation and Root Mean Square (RMS)—are first compared and their relative performances discussed. The influence of sensor locations (in both radial and axial directions) is then considered, and the role of the compressor’s geometrical irregularities in the behavior of the indicators is clearly highlighted. The influence of the throttling process is also carefully analyzed. This aspect of the experiment’s process appears to have a non-negligible impact on the stall warning parameters, despite being poorly documented in the literature. This last part of this research work allow us to get a different vision of the alert parameters compared to what is classically done in the literature, as the level of irregularity that is reflected by the magnitude of the parameters appears to be an image of a given flow rate value, and not a clear indicator of the stall inception.