基于小波降噪的转动惯量测试仪周期计算改进算法

Improved algorithm of periodic computation for moment of inertia tester based on wavelet noise-reduction

摘要:

转动惯量是弹药静态特征量的重要参数之一。在试验中发现，85mm样柱的转动惯量偏差值达到了0.83%，无法满足任务需求。经过对信号检查，确认为光电传感器采集的信号混有噪声信号所致。为解决脉冲电压序列信号在截止电压处抖动造成测量周期偏大的问题，采用小波分析的方法，利用sym8小波进行信号分解、阈值选择算法对信号进行降噪，使脉冲电压序列曲线得到平滑。分别对比haar小波、db4小波、coif1小波系、Bior1.5小波、sym2小波、dmey小波等小波母函数的信号降噪效果，确认dmey小波母函数处理信号信噪比高。改进了周期提取算法，提高了测量数据的一致性与精度。实验结果表明，本文算法优于原有的算法，具有精度高、容错能力好等优点，在转动惯量测量领域具有良好的应用价值。

Abstract:

The moment of inertia is one of the important parameters of the static characteristic of ammunition.It was found in the test that the deviation value of the rotaional inertia of 85mm spline reached 0.83% and the result cannot meet the task requirements.After checking the sigal,it is confirmed that the signal collected by the photoelertric sensor is mixed with noise signal.To solve the problem of the voltage pulse sequence signal jittering at the cut-off voltage which cause bigger measured periodic deviation, decomposing signal in sym8 wavelet way and de-nosing signal by the threshold selection algorithm make the pulse voltage sequence curve more smooth.De-nosing effect of haar wavelet,db3 wavelet,coif wavelet,Bior wavelet,sym 2wavelet,dmey wavelet is compared,and it shows that dmey wavelet is more effective. A new method is adopted to calculate the period,and the consistency and precision of measured data is improved.The result shows that new algorithm is better than existing one,and it has high precision and better fault-tolerant ability.The new algorithm has high application value in measurement for moment of inertia.