MindSpore tutorial implements DRSN for noisy vibration signals
A tutorial presents a MindSpore implementation of the Deep Residual Shrinkage Network for vibration-signal feature learning in strong-noise conditions. It covers channel-shared and channel-independent learned thresholds, soft thresholding, residual blocks, and code for both DRSN-CS and DRSN-CW variants.
The article provides reusable MindSpore code for applying learned soft thresholding to vibration signals, illustrating a concrete implementation of noise-focused feature learning.
深度残差收缩网络(Deep Residual Shrinkage Network,DRSN)是一种专为强噪声背景下的振动信号特征学习而设计的深度学习架构,通过在残差网络中引入软阈值化作为非线性变换层来消除噪声相关特征。该网络提出于发表在《IEEE Transactions on Industrial Informatics》期刊的论文《Deep Residual Shrinkage Networks for Fault Diagnosis》中,包含通道共享阈值(DRSN-CS)和通道独立阈值(DRSN-CW)两种网络变体,并利用专门设计的子网络自动学习阈值,无需人工干预或信号处理先验知识。
一、深度残差收缩网络结构解析
图4 (a) RSBU-CS基本单元;(b) DRSN-CS整体结构;(c) RSBU-CW基本单元;(d) DRSN-CW整体结构
1. 软阈值化(Soft Thresholding)
软阈值化是经典信号降噪中的核心收缩函数,用于将绝对值接近零的冗余或噪声特征置零,同时保留幅值较大的正负有效特征。其数学表达式为: