基于压缩感知的异步电动机信号重构方法研究

Research on Signal Reconstruction Method of Asynchronous Motor Based on Compressed Sensing

压缩感知理论为数据采集技术带来了革命性的突破,其应用越来越广泛。压缩感知采用了一种非自适应线性投影技术,完整的将信号的原始结构保留下来,能够通过数据最优化的方式准确的将原始信号进行重构。根据这一特点,提出一种基于压缩感知的异步电动机信号重构方法,采用非自适应线性投影用来维持信号的原始结构,对可以进行压缩的信号采用随机采样的方式,在远低于耐奎斯特频率的标准下进行数据采集,通过数值组优化将原始信号进行重构。实验结果表明,该方法能够更加精确的将异步电动机信号重构。

The theory of compressed sensing brings a revolutionary breakthrough in data acquisition technology, and its application is more and more extensive. Compressed sensing uses a non-adaptive linear projection technique that preserves the original structure of the signal and reconstructs the original signal accurately in a data-optimized manner. According to this feature, a method of signal reconstruction based on compressed sensing is proposed. The non-adaptive linear projection is used to maintain the original structure of the signal. The signal that can be compressed is randomly sampled. The data is collected under the standard of the Quest frequency, and the original signal is reconstructed by numerical group optimization. Experimental results show that this method can reconstruct the asynchronous motor signal more accurately.