Design And Implementation Of Real-time Multi-mode Dynamic Signal Processing

The dynamic signal analyzer is a general signal analysis instrument.It can perform real-time analysis in the time and frequency domains of various signals such as low frequency signals,acoustic signals,and rotating machinery vibration signals.It is widely used in electronic design,vibration analysis,acoustic testing,noise analysis and other fields.Thesis mainly studies the processing algorithms and implementation of different dynamic signals under various analysis modes of the dynamic signal analyzer.Thesis firstly studies the characteristics and processing requirements of various dynamic signals,and focuses on the analysis of the characteristics,problems and requirements of dynamic signals,acoustic signals,and rotating machinery vibration signals.Combined with the overall demonds of real-time processing and multifunction,the design and implementation of a dynamic signal processing system based on FPGA is completed.The specific research contents of thesis are as follows:1.After analyzing the problems such as the fence effect,energy leakage,limitation in frequency resolution,thesis indicate the requirements of band selection analysis and data windowing.Combined with the various test scenarios in the spectrum analysis mode,based on the Zoom FFT algorithm,thesis designs a spectrum analysis algorithm scheme with adjustable parameters such as analysis bandwidth,center frequency,resolution,window function type,etc.,based on the resource of FPGA,this article completes the realization of the algorithm;2.Acoustic signal is a kind of signal with wide frequency range.Thesis mainly studies the octave analysis of the signal in the audible sound range of 20Hz-20 k Hz,and also studies the relevant national standards.Octave analysis of acoustic signals is a technique for dividing the audible sound range according to the constant percentage bandwidth and performing spectral analysis.Unlike the constant frequency resolution within the FFT analysis bandwidth,octave analysis requires that the frequency resolution increases as the analysis bandwidth decreases.Aiming at this characteristic,thesis designs an octave analysis algorithm based on multi-level frequency resolution system,and describes the key technology for specific implementation;3.Since it is closely related to the motion of the rotating shaft,the vibration signal of the rotating machinery contains a lot of information on the status and fault types of the equipment.Thus,the vibration signal has non-uniform and non-stationary characteristics during the acceleration or deceleration of the rotating shaft,which makes the frequency spectrum appear confusion.Aiming at this phenomenon,thesis introduces the concept of order analysis.The order analysis takes the rotational speed as the fundamental frequency,and extract the resonance of the vibration signal.After comparing several implementations of order analysis and analyzing their deficiencies such as calculation amount and accuracy,thesis proposes an order analysis algorithm based on time-domain stretching,and completes the scheme design of the algorithm and the implementation of FPGA digital logic,solved the confusion problem of rotating machinery vibration signal..Finally,based on the dynamic signal analyzer test platform,test of the real-time,multi-mode dynamic signal processing system is completed,and the functional test and index analysis of the above algorithm are carried out.