基于自适应滑动均值和小波阈值的叶尖间隙信号降噪方法

Denoising Method of Blade Tip Clearance Signal Based on Adaptive Moving Average and Wavelet Threshold

动叶片与发动机机匣之间的叶尖间隙参数是反映航空发动机工作性能和运行安全的关键状态参数之一。提高叶尖间隙信号信噪比是实现高精度叶尖间隙测量的关键,为此提出基于自适应滑动均值和小波阈值的混合叶尖间隙信号实时降噪方法。首先根据叶片转速等信息,估算叶尖间隙信号带宽大小,然后通过动态改变滑动均值滤波的滑动点数来实现自适应低通滤波,最后结合小波阈值降噪时频分析能力,进一步降低噪声干扰。通过仿真实验确定采用db5小波基函数进行6层小波阈值分解,仿真结果表明该方法在各项降噪评价指标方面全面优于现有的有限脉冲响应和固定点滑动均值滤波方法。1000 rpm~4000 rpm转速范围内实际测试表明,该方法最大测量误差为16μm,有效提高叶尖间隙测量精度。

Blade tip clearance(BTC)between rotor blade and engine casing is one of the key state parameters to reflect performance and safety of aero-engine.Improving SNR of BTC signal is the key factor to realize high BTC measurement accuracy.A real-time denoising method based on adaptive moving average and wavelet threshold is proposed.After the signal bandwidth is estimated according to blade rotate speed and other parameters,the adaptive low-pass filter is realized by dynamically changing the point number of moving average.With the strong time-frequency analysis ability,wavelet threshold further reduces noise jamming.The db5 wavelet basis was selected to carry out 6-layer decomposition for BTC signal.Simulation results show that the method is superior to the existing FIR and fixed point MAF method.At 1000 rpm-4000 rpm,the maximum BTC measurement error of the proposed method is 16μm,which effectively improves the BTC measurement accuracy.