Harmonic wave measurement method based on spectral energy interpolation

Abstract

The invention discloses a harmonic wave measurement method based on spectral energy interpolation. According to ten cyclic wave detection environments meeting the IEC standard, concrete parameters of a certain component are detected out accurately, and an IEC method only can roughly detect out the amplitude of signals within the certain frequency band; from the point of view of spectral energy in a signal frequency domain for the first time, transformation and interpolation of an energy spectrum are carried out by multiplying the frequency spectrum by a rotation phase factor, influence of overlaid leakage of sidelobes of the frequency spectrum is eliminated by mutual offset of energy of interference components, and the detection precision of the frequency, the amplitude and the phase position is improved; when the components in signals are likely submerged in noise signals, main signals are highlighted through transformation of the spectrum energy, and therefore usable signals are extracted from the noise signals effectively. The method is based on a DFT / FFT algorithm recommended by the IEC standard, the principle is simple, the effect and stability are good, and the method has certain engineering application value and good application prospects.

Description

technical field [0001] The invention relates to a harmonic measurement method based on spectral line energy interpolation, and belongs to the technical field of power quality analysis. Background technique [0002] The quality of electric energy has a crucial impact on the stable and strong operation of the power grid, and the harmonic and interharmonic parameters are an important parameter to measure the quality of power supply. With the rapid development of high-speed rail, electric vehicles, wind power and For the access of new energy sources such as photovoltaic power generation, whether the harmonic and interharmonic parameters can be accurately detected has a great impact on its evaluation. The standards IEC61000-4-7 and IEC61000-4-30 issued by the International Electrotechnical Commission (International Electrotechnical Commission, IEC) stipulate the use of DFT / FFT for spectrum analysis, and stipulate that the analysis window length is 10 cycles (50Hz system), IEC Th...

AI Technical Summary

Problems solved by technology

The standards IEC61000-4-7 and IEC61000-4-30 issued by the International Electrotechnical Commission (International Electrotechnical Commission, IEC) stipulate the use of DFT / FFT for spectrum analysis, and stipulate that the analysis window length is 10 cycles (50Hz system), IEC The standard stipulates the spectrum grouping detection method of 10-cycle rectangular window during synchronous sampling. The advantage is that it can quickly detect the harmonic and interharmonic content in different frequency bands. The disadvantage is that it cannot measure the specific parameters of a certain frequency component, and in non- There is a large measurement error when sampling synchronously

[0003] In response to the defects in the IEC standard, there are currently many improved algorithms based on DFT / FFT. The two commonly used algorithms are the window interpolation algorithm and the phase difference algorithm. The window interpolation algorithm generally uses the largest and second largest near the component frequency The amplitude of the spectral line is interpolated, and its accuracy has a great relationship with the type of window added. The better the side lobe suppression performance of the window function, the higher the detection accuracy, but at the same time the larger the main lobe width, the current method Using the window interpolation algorithm will increase the length of the data, so that the length of the analysis window reaches dozens of cycles, which cannot meet the requirements of the IEC standard; the phase difference algorithm uses two pieces of data at different times for spectrum analysis. The parameters of the components are obtained by phase difference, so the sampling length must be at least 20 cycles to meet the 5Hz resolution standard of the IEC standard. At the same time, the side lobe leakage caused by asynchronous sampling has a greater impact on the detection results. The above two algorithms are in The side lobe leakage interference of other components is ignored when calculating the spectral value of the corresponding spectral line. When the frequency components are close together or the leakage is serious, a large error will occur

Images