应用希尔伯特-黄变换对铝电解过程阳极电流的分析

The application of HHT to analyzing the anode current for aluminum reduction process

铝电解过程阳极导杆等距压降信号是反映电解槽状态变化的重要参数之一,所以对该参数的检测与分析尤为重要。本文设计的铝电解槽多参数采集与分析系统是以PCI-8602数据采集卡为核心的数据采集与分析系统。阳极导杆等距压降信号的分析,采用了希尔伯特-黄变换法(HHT),实现信号中的噪声高频成分和真实低频成分的逐层分离,清晰得到信号主频。分析过程包括:经验模态分解,Hilbert变换以及积分求边际谱。与傅里叶变换谱图进行对比,希尔伯特-黄变换法在分析铝电解槽阳极导杆等距压降信号时分辨率较高、能获取信号时频特征。结合铝电解工艺特点,分析了信号频率存在的合理性,和用于电解槽故障诊断的有效性。系统软件由Visual C++6.0、MATLAB、Microsoft Access 2003数据库联合编写,其中MATLAB软件用于数据分析,Access数据库完成数据存储,VC软件以友好的界面实现了数据采集与实时显示、处理结果显示等功能。

Anode voltage signal of aluminum reduction is one of the most important parameters to reflect cell status accurately. As a result, detection and analysis for this anode voltage signal is rather valuable and significant. Multi - parameter acquisition and analysis system for aluminum reduction cell in this paper takes PCI - 8602 data acquisition card as the center. Hilbert - Huang Transform method is used to analyze the non - stationary aluminum reduc- tion cell anode voltage signal. The expected finding is the graded separation of the high - frequency noise composition. Analysis include ： EMD, HHT and integrate to get marginal spectrum. To contrast with the FFT spectrogram, the Hilbert - Huang Transform method is effective for high resolution and time frequency characteristics in extracting the frequency feature of the signal. Combining with the characteristics of aluminum reduction process, the rationality of the signal frequency and the availability of fault diagnose for aluminum reduction cell. The system software consists of Visual C＋＋ and MATLAB6.0, Microsoft Access 2003 database. MATLAB software is used for data analysis and Access database is for data storage. Visual C＋＋ , as the upper computer software, realizes the data acquisition, real - time display and processing with a friendly interface.