Compressor Stall Warning Using Nonlinear Feature Extraction Algorithms

Abstract

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.

A stall diagnosis method based on entropy feature identification in axial compressors

“…The effect of the key parameters like data size N , embedding dimension m , and similarity radius r on the feasibility of approximate entropy has been experimentally investigated in Lou and Key, 28 which presents a comparative study that shows that these parameters do affect the approximate entropy method. For example, approximate entropy decreases when disturbance emerges with an embedding dimension m of 4.…”

“…Approximate entropy was used as a stall warning approach and its algorithm has been described in detail in Lou and Key 28 26 and fuzzy approximate entropy 27 were proposed.…”

“…Approximate entropy was used as a stall warning approach and its algorithm has been described in detail in Lou and Key. 28 To improve the approximate entropy algorithm, other entropy concepts like sample entropy 26 and fuzzy approximate entropy 27 were proposed. The comparative studies and the obtained results indicate that sample entropy does not rely on the length of the time series, and the parameters have minimal effect on the sample entropy results.…”

“…The algorithms of approximate entropy were improved, namely, sample 26 and fuzzy approximate entropy, 27 and were widely adopted in fault diagnosis. Lou and Key 28 applied approximate entropy for stall warning in compressors using an innovative approach, providing a new perspective of stall diagnosis. The approximate entropy value considerably increases when disturbances appear.…”

A method of stall recognition using nonlinear feature extraction from the compressor outlet pressure

“…Based on the fractal dimension of the pressure signal, they were able to identify the stall. This work showed a new direction in stall warning research, which was further developed by Lou [25] . The author used a parameter called approximate entropy to identify small non-linear disturbances in the pressure signal preceding a deep surge.…”