发动机喘振故障检测的神经网络免疫识别模型

PDF(1467 KB)

振动与冲击 ›› 2010, Vol. 29 ›› Issue (1) : 170-172.

论文

发动机喘振故障检测的神经网络免疫识别模型

侯胜利1;王威1;胡金海2;史霄霈1;周根娜1

作者信息 +

Neural networks based Immune Recognition Model of Aeroengine Surge Detection

HOU Shengli1; WANG Wei1;HU Jinhai2; SHI Xiaopei1; ZHOU Genna1

Author information +

文章历史 +

摘要

提出了一种航空发动机喘振故障检测的神经网络免疫识别模型。该模型利用人工免疫系统的反面选择原理来构建神经网络检测器，通过训练将失速压力信号的模式特征存储在分布的检测器中。检测器用于捕获信号的失速模式特征，当检测器与特征样本匹配时则激活该检测器，根据检测器的激活情况来发现失速点。对某型涡喷发动机压力测量信号的分析结果表明，该方法对由失速气团造成的压力信号突变具有较强的分辨力，可以用于发动机喘振的早期检测。

Abstract

A model of aeroengine surge detection based on neural networks immune recognition is proposed. In this model, neural networks-based detectors are constructed based on negative selection principle of artificial immune system. Through neural network training, the pattern features of stall pressure signal are stored in the distributed neural networks-based detectors. These detectors are used to capture the stall pattern features. When a detector is matched up with a feature sample, the detector is activated. A stall point can be found out through the relevant activated detectors. When applied to a certain type of turbo engine, the result shows that the proposed model has high resolution for locations of pressure signal singularities, which are caused by stall air mass in the compressor. The method can be used in the early detection of surge signal.

导出引用

侯胜利;王威;胡金海;史霄霈;周根娜. 发动机喘振故障检测的神经网络免疫识别模型[J]. 振动与冲击, 2010, 29(1): 170-172

HOU Shengli;WANG Wei;HU Jinhai;SHI Xiaopei;ZHOU Genna. Neural networks based Immune Recognition Model of Aeroengine Surge Detection[J]. Journal of Vibration and Shock, 2010, 29(1): 170-172