CN113552415B - Ultra-high harmonic wave measuring device and measuring method - Google Patents

Abstract

The invention relates to an ultra-high harmonic wave measuring device and a corresponding measuring method, and belongs to the technical field of electric energy quality detection. The analog input interface of the backboard of the device is respectively connected with the corresponding input end of the FPGA through a group of parallel filter conversion circuits, and the filter conversion circuits consist of band-pass filters and digital-to-analog converters; the digital output interface is connected with the corresponding input end of the FPGA through the backboard network interface and the network physical layer interface; the band-pass filter is composed of an analog front-end circuit connected in series and a first filter circuit and a second filter circuit connected in parallel; output end connection of D/A converter and a corresponding input port of the FPGA. The method based on the device of the invention actually realizes the correction of the final harmonic calculation result according to the actually measured amplitude-frequency characteristic, thereby improving the harmonic measurement precision. The invention can make the measurement data have repeatability and consistency, and can practically meet the requirements of IEC 61000-4-30 (version 3.0).

Description

Ultra-high harmonic wave measuring device and measuring method

Technical Field

The invention relates to a harmonic wave measuring device, in particular to an ultra-high harmonic wave measuring device, and also relates to a corresponding measuring method, belonging to the technical field of electric energy quality detection.

Background

With the wide application of distributed photovoltaic power generation, micro-grids, electric automobile charging piles and the like to a power distribution network and IGBT inverter with high switching frequency, the problem of higher harmonic waves is brought about because the switching frequency is improved to improve the electric energy quality of conventional harmonic pollution.

The higher harmonics are not precisely defined, and in the IEC 61000-4-30 (version 2.0) (or IEC 61000-4-7) (version 08) standard, the higher harmonics are defined as components of the signal with frequencies above 2KHz and 9 KHz. In IEC 61000-4-30 (version 3.0), the upper frequency limit of the higher harmonics is extended to 150KHz. From this, it is clear that the coverage of the higher harmonics changes with the expansion of the application. The higher and higher switching device frequencies make it necessary to extend the upper limit of the possible higher harmonic frequencies that the original 9KHz frequency has not been able to cover. Although the emission value of the higher harmonic itself is not high from the perspective of the harmonic source, the higher harmonic may be amplified or resonated due to reasons such as cabling of the distribution network, distributed capacitance, and the like, resulting in the higher harmonic affecting the power quality. According to the knowledge of the applicant, the existing power quality monitoring device is designed to measure according to the frequency range of harmonic waves (2-50 times), and the monitoring of the higher harmonic waves is lacked, so that the actual higher harmonic wave emission value data cannot be obtained, and the analysis and investigation when problems occur are very unfavorable.

The IEC 61000-4-30 (3.0 edition) formally issued in 2015 divides the frequency band of the higher harmonic wave 2-150KHz into two frequency bands of 2-9KHz and 9-150KHz, wherein the higher harmonic wave measurement of 2-9KHz still adopts the calculation method recommended by IEC 61000-4-7 to carry out measurement calculation, and for the higher harmonic wave of 9-150KHz, the cutting-slope measurement method is not given due to the difference in realization, the difference of application scenes, the compromise of measurement accuracy and cost, but only some suggestions are given. Therefore, measurement of ultra-high harmonics is an important subject of interest.

As can be seen from the search, chinese patent document No. 201910458946.5 discloses an ultra-high harmonic measurement method based on flexible atomic filtering, in which a plurality of FAFs are set in an ultra-high frequency band range, so that the filtering measurement bandwidth covers the entire ultra-high frequency band without overlapping; after the discrete ultra-high harmonic signals and the discrete expression of FAF are processed by inner product, the corresponding ultra-high harmonic frequency and amplitude are determined according to the calculation result. The technical scheme of the patent does not accord with the measurement method explicitly required in IEC 61000-4-30 (3.0 edition), although the measurement method can be used for single-point measurement, the statistical analysis of regional data cannot be performed because the measurement data does not have repeatability, and the statistical significance of higher harmonics on the time scale cannot be estimated through data analysis.

The Chinese patent document with the application number 201810826416.7 discloses an ultra-high harmonic detection device and a detection method based on compressed sensing, a sensing circuit is arranged at a measurement end of a power distribution network, signals output by the sensing circuit firstly pass through a high-pass filter, then are mixed with a pseudo-random sequence output by a D/A module of a multifunctional data acquisition card and then are sent to the low-pass filter, low-speed sampling of the signals is realized through the A/D module of the multifunctional data acquisition card, and finally the signals are uploaded to an upper computer through a USB bus to finish reconstruction and display of the signals. The high-pass filter is designed to be adjustable in gain, and can realize signal conditioning besides filtering power frequency and other low-frequency signal interference. The low-pass filter is designed to be adjustable in cut-off frequency, so that variable compression ratio detection can be realized. The synchronous triggering module of the multifunctional data acquisition card can realize D/A and A/D synchronization, an observation matrix with the phase deviation of 0 is constructed, and the reconstruction accuracy is improved. According to the technical scheme, although the measurement result of the higher harmonic can be obtained under the condition of reducing the sampling frequency and the sampling point number through a complex circuit structure, the control strategy is complex, the actual analysis data source is reconstruction data and is not original signal data, and the measurement method does not completely meet the requirements in IEC 61000-4-30 (3.0 edition).

The Chinese patent document with the application number 202010395443.0 discloses an ultra-high harmonic measurement method based on fixed-frequency asynchronous sampling, wherein the sampling frequency of a measuring instrument is 409.6 KHz, the measured voltage and current signals are subjected to filtering and frequency division treatment, and harmonic components with the frequency lower than 1.5 KHz and higher than 64KHz are filtered; taking 200ms as a basic measurement window, extracting data of 0-20ms, 80-100ms and 160-180ms, respectively performing discrete Fourier analysis and averaging to obtain an output spectrum analysis result; aggregating the results by using a bandwidth of 2KHz, and outputting an aggregated spectrum signal; and carrying out time window aggregation on the aggregated spectrum signal every 3s without gaps, solving the root mean square value of the aggregated spectrum signal, and outputting a time-frequency domain processing result. Although the technical scheme of the patent refers to the processing mode and the calculation method of 2 KHz-9 KHz in the IEC 61000-4-30 (2.0 edition) of the measurement range and the calculation method of the higher harmonic wave, the requirements of the IEC 61000-4-30 (3.0 edition) are not met at all, because the IEC 61000-4-30 (3.0 edition) defines standardized measurement and data processing specifications, and only the measurement results of different devices aiming at the same measurement point have consistency under the premise, thereby laying a foundation for regional statistics.

Disclosure of Invention

The invention aims at: aiming at the defects existing in the prior art, an ultra-high harmonic measuring device which not only has repeatability and consistency of measured data, but also has enough measuring precision is provided, and a corresponding measuring method is provided, so that the requirements of IEC 61000-4-30 (3.0 edition) are practically met.

In order to achieve the above purpose, the basic technical scheme of the ultra-high harmonic measuring device of the invention is as follows: the device comprises a backboard with a network interface, a network physical layer interface, an analog input interface and a digital output interface;

the analog input interface is respectively connected with corresponding input ends of the FPGA through a group of parallel filter conversion circuits, and the filter conversion circuits consist of band-pass filters and digital-to-analog converters;

The digital output interface is connected with the corresponding input end of the FPGA through the backboard network interface and the network physical layer interface;

The band-pass filter is composed of an analog front-end circuit connected in series and a first filter circuit and a second filter circuit connected in parallel;

The analog front-end circuit is composed of two following loops which convert a single input into two original inputs and is used for respectively transmitting signals of 2-9 KHz and 9-150 KHz to corresponding filter circuits;

The first filter circuit consists of a first low-pass filter loop and a first high-pass filter loop which are connected in series and is used for extracting signals of 9KHz to 150KHz in an original input signal;

the second filter circuit consists of a second low-pass filter circuit and a second high-pass filter circuit which are connected in series and is used for extracting signals of more than 2KHz and less than 9KHz from an original input signal;

the first filter circuit and the second filter circuit are respectively connected with the input ends of the corresponding digital-to-analog converters through corresponding signal modulation circuits and are used for amplifying and lifting the filtered signals to preset sampling precision and resolution;

And the output end of the digital-to-analog converter is connected with a corresponding input port of the FPGA.

The ultra-high harmonic measurement method is carried out by an FPGA according to the following steps:

dividing the preset number of the frequency data into time slices with equal intervals;

Step two, taking a continuous preset number of sampling values in each time slice as a calculation time window, and performing DFT or FFT conversion to obtain the required frequency resolution;

Thirdly, taking a preset number of frequency components, taking the obtained frequency resolution as an interval, and respectively removing a preset number of lowest results and highest results;

And fourthly, carrying out operations of taking the maximum value, taking the minimum value and taking the average value from the reserved result to obtain the maximum value, the minimum value and the average value of each higher harmonic.

The invention is further perfected as follows: the amplitude-frequency characteristics of the first filter circuit and the second filter circuit of the band-pass filter at a preset frequency point are measured in advance to obtain a segmented filter coefficient; and obtaining amplitude-frequency characteristic filter coefficients of the 9-150KHz full frequency band by using the amplitude-frequency characteristics obtained through actual measurement and adopting an interpolation method to correct signals filtered by the first filter circuit and the second filter circuit.

The invention is further perfected as follows: in the second step, a sampling value calculation time window of a preset number is gradually increased according to the power of n of 2, DFT or FFT is respectively carried out, and a group of frequency resolutions are obtained; and determining the frequency resolution corresponding to the most suitable data window according to the hardware computing capacity as the required frequency resolution.

The common harmonic measurement in the prior art requires 2-50 times (100 HZ-2.5 KHz), the ultra-high harmonic measurement is divided into 2-9KHz harmonic measurement and 9-150KHz harmonic measurement, the long-term practice is that a first-order filter is adopted in 100 Hz-2.5 KHz, a second-order filter is adopted in 2-9KHz, and an 8-order filter is required to be arranged for realizing full filtering in 9-150Khz, which becomes a difficult point. The invention takes the actual analysis data source as the original signal data, reasonably designs the second-order filter, and has two characteristics compared with the prior art: firstly, the prior art adopts the bandwidth of 2KHz for gathering the output of 2-9KHz and 9-150KHz, which is not in accordance with the requirements that the bandwidth of 200Hz is adopted for gathering the 2-9KHz signal and the bandwidth of 2KHz is adopted for gathering the 9-150KHz signal in IEC 61000-4-30; the algorithm is strictly formulated according to IEC61000-4-30 (3.0 edition), so that the requirements are mutually consistent, the monitoring results of monitoring equipment of different factories/brands under the same signal source condition can be compared, and the repeatability and consistency of regional monitoring are realized. Secondly, in the prior art, aiming at 9-150KHz signal filtering, an ideal filtering effect is achieved through simulation, and an 8-order hardware filter is required to be arranged; the invention adopts 2-order hardware filter and reasonable filtering process flow to ensure the measurement accuracy while greatly reducing the hardware requirement. Therefore, the invention can ensure that the measured data has repeatability and consistency and can practically meet the requirements of IEC61000-4-30 (version 3.0).

Drawings

The invention is further described below with reference to the accompanying drawings.

FIG. 1 is a block diagram of the apparatus of one embodiment of the present invention.

Fig. 2 is a block diagram of the signal processing procedure in the embodiment of fig. 1.

Fig. 3 is an analog front end circuit diagram of fig. 2.

Fig. 4 is a low pass filter circuit diagram of fig. 2.

Fig. 5 is a high pass filter circuit diagram of fig. 2.

Fig. 6 is a circuit diagram of the signal modulation in fig. 2.

Fig. 7 is a diagram of the analog-to-digital conversion circuit in fig. 2.

Fig. 8 is a graph showing comparison of actual measurement filtering effects of the present invention.

Detailed Description

Example 1

The ultra-high harmonic measurement device of this embodiment is configured as shown in fig. 1 (see fig. 2-6), and the back plate has a network interface and a network physical layer interface, and an analog input interface and a digital output interface. The analog input interface is respectively connected with corresponding input ends of the field programmable gate array (Field Programmable GATE ARRAY) through a group of parallel filter conversion circuits, and the filter conversion circuits consist of band-pass filters and digital-to-analog converters; the digital output interface is connected with the corresponding input end of the field programmable gate array through the backboard network interface and the network physical layer interface; the band-pass filter is composed of an analog front-end circuit connected in series and a first filter circuit and a second filter circuit connected in parallel.

The analog front-end circuit is composed of two following loops U1A, U B for converting a single input into two original inputs, as shown in FIG. 3, and is used for transmitting signals of 2-9 KHz and 9-150 KHz to corresponding first and second filter circuits respectively.

As shown in fig. 4 and 5, the first filter circuit is composed of a U2A first low-pass filter circuit and a U3A first high-pass filter circuit connected in series, and the extraction of signals below 9KHz and below 150KHz in the original input signal can be realized only by selecting a resistor-capacitor component with proper parameters. The second filter circuit consists of a U2B second low-pass filter circuit and a U4B second high-pass filter circuit which are connected in series, and the signals above 2KHz and above 9KHz in the original input signals can be extracted just by selecting a resistor-capacitor part with proper parameters. Because the transition frequency band of 9KHz-150KHz is narrower, the traditional band-pass filter is not adopted in the embodiment, and the band-pass function is realized by adopting a mode of connecting a low-pass filter loop and a high-pass filter loop in series.

As shown in fig. 6, the first filtering circuit and the second filtering circuit are respectively connected with the input end of the corresponding digital-to-analog converter through a signal modulation circuit formed by U4A, U B, and amplify and raise the filtered signal to a predetermined sampling precision and resolution. Finally, referring to fig. 7, the output of the digital-to-analog converter is connected to the corresponding input port of the field programmable gate array.

The FPGA in the device of the embodiment realizes ultra-high harmonic measurement according to the following steps:

Dividing the preset number of the frequency data into time slices with equal intervals; specifically, the whole 10-cycle data is divided into equally-spaced time slices, such as: 32, each time slice contains a number of data points of 204800/32=6400 points.

Step two, taking a continuous preset number of sampling values in each time slice as a calculation time window, and performing DFT or FFT conversion to obtain the required frequency resolution; specifically, 512 consecutive sampling values are taken as a calculated time window in each time slice, and DFT (discrete fourier transform (Discrete Fourier Transform) or FFT (fast fourier transform) calculation is performed, so that the frequency resolution at this time is (204800/512) ×5hz=2khz.

Thirdly, taking a preset number of frequency components, taking the obtained frequency resolution as an interval, and respectively removing a preset number of lowest results and highest results; specifically, after 256 frequency components are obtained, the lowest 4 results and the highest 181 results are removed at intervals of 2KHz, and the rest 71 amplitude values are components from 8KHz to 150 KHz.

And fourthly, carrying out operations of taking the maximum value, taking the minimum value and taking the average value from the reserved result to obtain the maximum value, the minimum value and the average value of each higher harmonic. Specifically, the operation of taking the maximum value, the minimum value or the average value from 32 results can obtain the maximum value, the minimum value and the average value of each higher harmonic (8-150 KHz) in 200ms calculation time, and simultaneously, the maximum value (all 71 values) can be output.

In the first step, the frequency resolution can be increased by the power of 2 to the power of n on the data window of each time slice, and each time the frequency resolution is increased by 2 times, and finally the most suitable data window and the most suitable frequency resolution are determined according to the computing capability of hardware, so that the data output can be increased by times, and the data output needs to be evaluated and determined as appropriate.

In addition, the amplitude frequency characteristics of the hardware filter circuit, namely the first filter circuit and the second filter circuit of the band-pass filter, at fixed frequency points (10 KHz, 20KHz … KHz are taken as intervals) are measured in advance to form a segmented filter coefficient, the measured amplitude frequency characteristics are utilized to calculate the amplitude frequency characteristic filter coefficient of the 9-150KHz full frequency band by adopting an interpolation algorithm, software filtering is carried out on signals acquired by hardware according to the amplitude frequency characteristic filter coefficient of the full frequency band, the signals filtered by the first filter circuit and the second filter circuit are substantially corrected, and frequency band leakage caused by hardware 2-order filtering is restrained, so that the measuring precision of harmonic waves can be further improved.

FIG. 8 is a graph showing the actual filtering effect, wherein harmonic signals with the frequency of 9-150KHz and the amplitude of 3V are added to the device through a standard source, the curve with the larger blue curvature is the amplitude frequency characteristic of 9-150kHz obtained by only two-order hardware filtering and not by software filtering, the filtering effect is still not ideal, the curve with the smaller curvature is the amplitude frequency characteristic obtained by software filtering under the same condition, and the filtering effect is further improved.

When a continuous 512 sampling value data window is taken in each time slice to perform DFT calculation, the frequency resolution of a calculation result is 2KHz, when the data window is increased to 2 times, namely 1024 sampling values, the frequency resolution of the calculation result can reach 1KHz, but the calculation amount can be increased by about 2 times, so that the larger the data window is, the smaller the frequency resolution is, and finally, the proper data window is determined according to the calculation capacity of hardware and the optimization program of DFT.

Experiments show that the embodiment adopts a 2-order hardware filter and a reasonable filtering processing flow to be organically combined, so that the measured data has repeatability and consistency, and the measurement precision of ultra-high harmonic waves is obviously improved.

In addition to the embodiments described above, other embodiments of the invention are possible. For example, FPGAs may be replaced with other functionally-close smart devices. All technical schemes formed by equivalent substitution or equivalent transformation fall within the protection scope of the invention.

Claims (3)

1. The measuring method of the ultra-high harmonic measuring device comprises a backboard with a network interface, a network physical layer interface, an analog input interface and a digital input/output interface; the method is characterized in that: the analog input interface is respectively connected with corresponding input ends of the FPGA through a group of parallel filter conversion circuits, and the filter conversion circuits consist of band-pass filters and digital-to-analog converters; the digital input/output interface is connected with the corresponding input end of the FPGA through the backboard network interface and the network physical layer interface; the band-pass filter is composed of an analog front-end circuit connected in series and a first filter circuit and a second filter circuit connected in parallel; the analog front-end circuit is composed of two following loops which convert a single input into two original inputs and is used for respectively transmitting signals of 2-9 KHz and 9-150 KHz to corresponding filter circuits; the first filter circuit consists of a first low-pass filter loop and a first high-pass filter loop which are connected in series and is used for extracting signals of 9KHz to 150KHz in an original input signal; the second filter circuit consists of a second low-pass filter circuit and a second high-pass filter circuit which are connected in series and is used for extracting signals of more than 2KHz and less than 9KHz from an original input signal; the first filter circuit and the second filter circuit are respectively connected with the input ends of the corresponding digital-to-analog converters through corresponding signal modulation circuits and are used for amplifying and lifting the filtered signals to preset sampling precision and resolution; the output end of the digital-to-analog converter is connected with a corresponding input port of the FPGA; the FPGA comprises the following steps: dividing the preset number of the frequency data into time slices with equal intervals; step two, taking a continuous preset number of sampling values in each time slice as a calculation time window, and performing DFT or FFT conversion to obtain the required frequency resolution; thirdly, taking a preset number of frequency components, taking the obtained frequency resolution as an interval, and respectively removing a preset number of lowest results and highest results; and fourthly, carrying out operations of taking the maximum value, taking the minimum value and taking the average value from the reserved result to obtain the maximum value, the minimum value and the average value of each higher harmonic.

2. The measurement method of the ultra-high harmonic measurement device according to claim 1, wherein: the amplitude-frequency characteristics of the first filter circuit and the second filter circuit of the band-pass filter at a preset frequency point are measured in advance to obtain a segmented filter coefficient; and obtaining amplitude-frequency characteristic filter coefficients of the 9-150KHz full frequency band by using the amplitude-frequency characteristics obtained through actual measurement and adopting an interpolation method to correct signals filtered by the first filter circuit and the second filter circuit.

3. The measurement method of an ultraharmonics measurement apparatus according to claim 1 or 2, wherein: in the second step, a preset number of sampling value calculation time windows are gradually increased according to the power of n of 2, DFT or FFT is respectively carried out, and a group of frequency resolutions are obtained; and determining the frequency resolution corresponding to the most suitable data window according to the hardware computing capacity as the required frequency resolution.