A low frequency sine wave fast response full parameter estimation method and device

Abstract

The invention discloses a low-frequency sine wave fast-response full-parameter estimation method and device. The method includes the steps of: taking the sampling frequency f 0 Sampling the original waveform; taking 4 continuous sampling points from any moment as input, calculating and outputting the DC component, amplitude, frequency and phase of the original waveform; each time a new sampling point is obtained, the new The sampling points are updated rollingly in a first-in-first-out manner, and the updated 4 consecutive sampling points are used as input to calculate and output the DC component, amplitude, frequency and phase of a new round of the original waveform. The present invention utilizes four continuous sampling points from any time as input to perform full parameter estimation, requires few parameters to be set, and has low application requirements.

Description

technical field [0001] The present invention relates to the technical field of industrial control automation, in particular to a low-frequency sine wave fast response full parameter estimation method and device. Background technique [0002] In the fields of power frequency signal detection, vibration table feedback, vibration object monitoring, engine monitoring and other fields, it is necessary to estimate the parameters of low-frequency sinusoidal signals. In order to make feedback adjustments in time, sometimes it is hoped that the response of parameter estimation can quickly reflect the original signal at this moment. The real situation is not affected by multiple historical waveforms. At the same time, due to the consideration of cost and response speed, it is not desirable to add pre-processing links such as filters. [0003] The full parameters of the sine wave include four parameters: DC component, amplitude, frequency, and phase. Since different parameter estimatio...

AI Technical Summary

Problems solved by technology

For example, the least squares method can be used to estimate three parameters except the frequency at the same time. Because the frequency cannot be estimated directly, it is necessary to know the range of the frequency, and perform segmented scanning within this range. The number of segments affects the final calculation accuracy. The minimum The square method is a statistical method that needs to store and calculate more data. It is more suitable for high-precision estimation of time-invariant signals, and its response speed is not as good as the single-shot algorithm.

[0009] Using the method of discrete Fourier transform, the amplitude-frequency characteristics and phase-frequency characteristics of the sinusoidal signal can be obtained at the same time, but because the result obtained by this calculation is about frequency discrete, only the discrete frequency with the largest amplitude can be obtained as an output by scanning, and the accuracy is affected by the transformation. The number of points is limited. If you use the collection-distributed Fourier transform of the classic algorithm, you can use the dichotomy method to further search for the precise frequency after obtaining the maximum frequency initially. The results on the Nth power fixed frequency points cannot be further refined

[0010] Through the analysis of the existing technology, the parameter estimation method based on the digital sampling method and the single parameter estimation method that do not require peripheral devices are generally not designed for sine waves. They are versatile but poor in accuracy. Some multi-parameter estimation methods need to be Frequency is pre-measured, some need to scan the frequency, some need to obtain a large number of sampling points, and obtain full parameter estimation through statistical methods in a longer time window, but lack of full parameter with fast response ability for low-frequency sinusoidal signals estimated method

Images