CN103257271B - A kind of micro-capacitance sensor harmonic wave based on STM32F107VCT6 and m-Acetyl chlorophosphonazo pick-up unit and detection method - Google Patents

Abstract

The invention discloses a microgrid harmonic and interharmonic detection device and detection method based on STM32F107VCT6, belonging to the technical field of power quality detection. The device includes a signal acquisition module, a signal processing module, and a signal display output module. In the signal acquisition module, a Hall voltage and current sensor is used as a collector for collecting three-phase voltage and current of the power grid; the signal processing module includes a voltage and current sensor. conversion circuit, anti-aliasing filter circuit, level adjustment circuit, core processor STM32F107VCT6 and processor peripheral circuit; the output terminal of the level adjustment circuit is connected to the A/D converter input terminal of the core processor STM32F107VCT6. The method includes signal collection and processing processes, and the device and method meet the requirements of calculation accuracy and calculation speed while satisfying flexibility and convenience in structure.

Description

A microgrid harmonic and interharmonic detection device and detection method based on STM32F107VCT6

technical field

The present invention belongs to the technical field of power quality detection, in particular, relates to a device and detection method capable of detecting harmonics and inter-harmonics of a micro-grid system, and more specifically, relates to a micro-grid harmonic and inter-harmonic based on STM32F107VCT6 Harmonic detection device and detection method.

Background technique

At present, most of the known harmonic detection devices are aimed at large power grids, and mostly detect harmonics, and there are not many detection devices for inter-harmonics. At present, the IEC6100-4-7 standard and most of the existing harmonic detection devices adopt the FFT (FastFourierTransform, Fast Fourier Transform) algorithm for the detection of interharmonics. However, the traditional FFT algorithm is relatively accurate for detecting integer harmonics, but when there are a large number of non-integer harmonics in the power grid, the algorithm will have large errors and even fail to detect, resulting in extremely unreliable calculation results. Therefore, a more accurate algorithm is urgently needed to improve the detection of harmonics and interharmonics.

With the vigorous promotion of smart grid construction and the rapid development of micro-grid technology, the power quality of micro-grid has attracted more and more attention. However, harmonics are a key indicator to measure the power quality of the power grid. In order to control the harmonics and avoid the injection of harmonics into the large power grid, it is very necessary to detect the harmonics of the micro-grid. The frequent switching of distributed power sources in the microgrid will generate harmonic currents, and the nonlinear characteristics of power electronic equipment will also generate a large number of harmonic voltages. The unstable output power of intermittent energy sources will cause voltage fluctuations and flicker, and the capacity of the microgrid Small loads are variable. In addition, since the microgrid has two stable operating states, namely, the grid-connected operating state and the islanded operating state, the fundamental frequency of the micro-grid will change with the change of the fundamental frequency of the external power grid during grid-connected operation; If the distributed power output does not match the load power consumption, it will also cause changes in the fundamental frequency of the microgrid. These characteristics of the microgrid make the voltage and current in the microgrid very easy to fluctuate, and the rated fundamental frequency is not a constant value, making the voltage and current a non-stationary state quantity, and there are a large number of interharmonic components in addition to the integer harmonic components , which is extremely detrimental to the stable operation of the power grid. The power quality problem in microgrid is a key technology that needs to be solved urgently, but there is no effective detection device designed for the characteristics of microgrid harmonics. Therefore, there is an urgent need for a harmonic detection device for microgrids, and with the continuous promotion of microgrids, the market also urgently needs the harmonic detection device.

China Patent No.: 200910043655.6, published on November 25, 2009, discloses a patent document titled a detection method for harmonic and inter-harmonic parameters, which aims to overcome the existing harmonic and inter-harmonic detection The deficiency of method, this invention provides harmonic and interharmonic parameter detection method, and its main realization step is: 1) according to the characteristic of sine, cosine function, the fundamental wave component in the voltage or current signal and each harmonic The wave components are converted into DC components respectively; 2) Extract the DC components by low-pass filtering, and calculate the parameters such as the amplitude and phase angle of the fundamental wave component and each harmonic component; 3) Subtract the fundamental wave and harmonics from the voltage signal Components to obtain a voltage or current signal containing only the inter-harmonics; 4) Obtain the inter-harmonic parameters by searching the maximum value of the amplitude spectrum. However, this method has problems such as low detection accuracy and difficulties in actual operation, and it is not suitable for the detection of harmonics between microgrids.

The reason why the traditional FFT algorithm will inevitably produce large errors in detecting the interharmonics of the microgrid is that the spectrum leakage of the FFT algorithm will generate pseudo interharmonics, thereby submerging the real interharmonics, and FFT is only suitable for the detection of fixed period quantities , but in fact the microgrid voltage and current fluctuations do not present a fixed cycle regularity. Therefore, there is an urgent need for an improved microgrid interharmonic detection method to achieve accurate detection of interharmonics.

The flaws in the above algorithms make it difficult to find an interharmonic detection device for microgrids, so research is urgently needed to solve them. For the detection of inter-harmonics in microgrids, the technical problems to be solved are: (1) how to realize the separation of harmonics and inter-harmonics; (2) how to suppress the mutual interference between Interference: (3) How to suppress harmonic components from submerging inter-harmonic components. In addition, most of the existing harmonic detection devices use 8-bit single-chip microcomputers, a few use 16-bit single-chip microcomputers, and there is no detection device that uses 32-bit single-chip microcomputers. The 8-bit single-chip microcomputer is characterized by low cost and flexible use, but the disadvantage is that the operation speed is slow, and generally it needs to expand RAM and ROM, and the hardware circuit is more complicated.

Contents of the invention

1. Problems to be solved

Aiming at the problems of slow computing speed, need to expand RAM and ROM, and complex hardware circuits in the detection of micro-grid harmonics in the prior art, the present invention provides a micro-grid harmonic and inter-harmonic detection device and detection method based on STM32F107VCT6 , it not only satisfies the calculation accuracy and calculation speed, but also meets the flexibility and convenience in structure.

2. Technical solution

In order to solve the above problems, the present invention adopts the following technical solutions.

A micro-grid harmonic and inter-harmonic detection device based on STM32F107VCT6, including a signal acquisition module, a signal processing module, and a signal display output module. The signal acquisition module uses a Hall voltage and current sensor as the acquisition grid three-phase voltage The collector of electric current; Described signal processing module comprises the conversion circuit of voltage and current, anti-aliasing filter circuit, level adjustment circuit, core processor STM32F107VCT6 and processor peripheral circuit; The Hall voltage current of described signal acquisition module The acquisition end of the sensor is connected to the power grid, and the output end is connected to the signal processing module. In the signal processing module, the current and voltage conversion circuit, the anti-aliasing filter circuit and the level adjustment circuit are sequentially passed through; the output end of the level adjustment circuit is connected to the core The A/D converter input terminal of the processor STM32F107VCT6, and the periphery of the STM32F107VCT6 is connected to the LCD display.

Preferably, the peripheral circuit of the processor includes a power supply circuit, a reset circuit and a crystal oscillator circuit; the power supply circuit provides power supplies of different voltages required by the entire system through voltage conversion; the reset chip is used in the reset circuit to manually reset mode; the crystal oscillator circuit provides a clock signal for the single-chip microcomputer.

Preferably, the output display module includes an LCD display screen, a keyboard, and a PC communication module.

Preferably, said output display module also includes a keyboard circuit, and the connection between the keyboard and the core processor STM32F107VCT6 is through the keyboard circuit.

In order to cooperate with the device of the present invention, the system software designed by the harmonic detection device of the present invention mainly includes data sampling, data processing, man-machine interface and communication, and mainly adopts the Keil development environment C language development tool. In terms of algorithm, on the basis of FFT algorithm, an interharmonic detection method based on quasi-synchronous sampling reconstruction signal is proposed. Accurate measurement of harmonic and interharmonic parameters. Due to the existence of inter-harmonics, the system voltage and current signals are no longer stable cycles, but non-stationary signals. The traditional FFT algorithm has a large error in the analysis of non-stationary signals, and the "elimination method" detection of harmonics and inter-harmonics will also cause large errors in the micro-grid, because the content of inter-harmonics in the micro-grid is more than that of the power grid. , the influence of inter-harmonics on harmonics cannot be ignored in the calculation. The invention proposes a detection algorithm based on quasi-synchronous sampling reconstruction signal, which is suitable for non-synchronous sampling, so the harmonic detection device can use a fixed sampling frequency to sample the power signal, thereby ensuring the validity of each sampled data.

A method for detecting harmonics and interharmonics of a microgrid, the steps of which are:

(1) In the case of satisfying the Nyquist sampling theorem, the power system signal is sampled at equal intervals at a fixed sampling frequency, that is, the equation

K·T _s ＝P·T ₀ (1-1)

Established, where T _s is the sampling period, T ₀ is the signal period, K is the number of sampling points, P is the number of signal periods, for asynchronous sampling, obviously P is not an integer;

(2) Use the adaptive tracking digital notch filter to filter out the fundamental component in the sampling sequence x(KT _s ) to obtain a new sampling sequence x'(KT _s ), so subtract x from x(KT _s ) '(KT _s ), the fundamental component x ₀ (KT _s ) can be obtained; compared with the traditional notch filter, the adaptive tracking digital notch filter makes the fundamental frequency in the It can also accurately filter out fluctuations around 50Hz. The frequency response of an ideal digital notch filter is

$\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; j\&omega;t</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 1</span>}& {\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Wherein e is a natural constant, j is an imaginary number unit, ω is a frequency, t is a time, and _ω0 is a notch frequency;

For a single notch frequency, its transfer function H(z ^-1 ) has the following symmetrical mirror form

$\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; pz</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

In the formula, z is a complex variable, p is a constant close to 1 but slightly smaller than 1, a ₁ is a parameter, which is determined by the notch frequency,

The notch frequency expression is

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; arccos</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;T</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

Where T is the sampling period, a ₁ (n) that changes with the sampling point is obtained by the adaptive least squares method, and the corresponding notch frequency also changes with time. In order to realize the adaptiveness, p changes with the sampling point as p(n); thus, the H(z ^-1 ) corresponding to each sampling point signal in x(KT _s ) will be different, and the adaptive adjustment is realized, and the fundamental wave is obtained by subtracting the original signal from the filtered signal Signal;

(3) Use the linear interpolation method to calculate the fundamental wave period of the signal, and set the function value at the node a=t ₀ <t ₁ <t ₂ ...<t _n =b to be y ₀ , y ₁ , y ₂ ,…,y _n , replace the curve with a straight line in each small interval [x _j ,x _j+1 ], then the time t _i when the signal crosses the threshold a for the ith time is

${\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

In the same way, the time t _i+20 when the signal crosses the threshold a for the i+20th time, the average value T* of the sampling time of 20 cycle lengths of the sampling sequence x(n) is:

$\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; *</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 20</span>}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 20</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

Then T* is the fundamental period of the signal, and 20 periods are used here to calculate the fundamental period more accurately;

(4) According to the calculated fundamental wave period, the harmonic components in the original sampling sequence are quasi-synchronized to obtain the reconstructed sequence x ^* (kλ _s ), P is not an integer when sampling asynchronously, and spectrum leakage will occur. In order Realize the quasi-synchronization of the non-synchronous sampling sequence, adjust the sampling period T _s , and use T* to calculate the quasi-synchronous sampling period λ _s

${\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 20</span>}{\mathrm{<span\; class="notranslate">\; T</span>}}^{<span\; class="notranslate">\; *</span>}}{{\mathrm{<span\; class="notranslate">\; L</span>}}^{<span\; class="notranslate">\; *</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Among them, L ^* is the number of sampling points in 20 periods of the new quasi-synchronous sampling sequence, but it can be approximated that L ^* = L to reduce the error, and L is the number of sampling points in the 20 signal periods of the original sampling sequence, Each sampling point in the quasi-synchronous sampling sequence is obtained by the cubic spline interpolation method, the sampling point in the original sampling sequence is used as the interpolation node, T _s is the interpolation node interval, and the starting point of a certain signal cycle is also the starting point of the sampling point , since the cubic spline difference function K(t) is a piecewise cubic polynomial on each interval [x _j , x _j+1 ], the second order derivative K''(t of the cubic spline function K(t) is known ) is a first-degree polynomial in each small interval, assuming K''(t _j )=M _j , K''(t _j+1 )=M _j+1 , then the expression of K''(t) is

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; \&Prime;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Wherein, h _j =t _j+1 -t _j .

Integrate K''(t) twice and bring it into the boundary condition to get the expression of K(t) as

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}$

(t∈[t _j ,t _j+1 ];j=0,1,…,n-1)(1-9)

Suppose the mth quasi-synchronous sampling point is located between the jth synchronous sampling point and the j+1th synchronous sampling point, then the quasi-synchronous sampling point can be calculated by the following formula

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}$

(t∈[t _j ,t _j+1 ];j=0,1,…,n-1)(1-10)

Repeating the calculation process of K(mλ _s ), you can get L ^* -1 calculated values. These L ^* -1 quasi-synchronous sampling points and the starting point together form a quasi-synchronous sampling sequence with a length of T ^* . So far, we have obtained Reconstruct all information of the sampling sequence x ^* (kλ _s ), that is, the position and sampling value of the reconstructed signal;

(5) Use the FIR notch filter to separate the harmonic and interharmonic components in x ^* (kλ _s ), and set multiple notch frequencies in the filter to form a comb; since the sampling frequency is the frequency of the inter-tooth interval of the filter Integer multiples, so the FIR notch filter has a good effect on synchronous sampling without energy loss, and is suitable for quasi-synchronous sampling sequences; since the notch frequency of the filter is an integer multiple of the fundamental, it can filter out the fundamental waves and all integer harmonics;

(6) For the inter-harmonic component x ^*I (kλ _s ) obtained after filtering with the FIR notch filter, use the double interpolation FFT algorithm with Hanning window to calculate the parameters of the inter-harmonic;

(7) Subtract the interharmonic component x ^*I (kλ _s ) from the quasi-synchronous sampling sequence x ^* (kλ _s ) to obtain the harmonic component x ^*H (kλ _s );

(8) Carry out DFT or FFT operation on x ^*H (kλ _s ), and calculate the harmonic parameters of each order from the result.

Further, the adaptive tracking digital notch filter is to increase adaptive control on the basis of the digital notch filter, which accepts an input signal with a notch frequency, and obtains a ₁ by a least squares adaptive algorithm , use a ₁ to form a digital notch filter H(z ^-1 ) to notch the signal with this notch frequency.

On the basis of the current detection technology, there are two key technical points for the superiority of this device:

1) Effectively limit the leakage of harmonics and avoid the loss of harmonic energy.

2) By separating harmonics and inter-harmonics, the interaction between harmonics and inter-harmonics is reduced.

In the analysis of the large power grid, the above technical key point 1) is more important than the technical key point 2), but in view of the increase in the intermediate harmonic content of the micro-grid, the technical key point 2) and the technical key point 1) are also the main points to be solved technical problems.

For the aforementioned technical key point 1), since the separated harmonics and inter-harmonics are all synchronous sampling, the fence effect and spectrum leakage effect caused by FFT asynchronous sampling are eliminated, thereby effectively limiting the leakage of harmonics. Minimize the technical requirements of harmonic energy loss.

For the aforementioned technical key point 2), a detection algorithm based on quasi-synchronous sampling reconstructed signals is proposed to quasi-synchronize non-synchronous sampling signals to achieve the purpose of suppressing spectrum leakage and fence effects. The algorithm first adopts adaptive tracking digital notch filter Preprocess the signal obtained under asynchronous sampling, filter out other frequency components other than the fundamental frequency, then use the method of comparing the filtered signal with the threshold value and use the linear interpolation method to obtain the fundamental period of the signal, according to the fundamental frequency period, the cubic spline interpolation algorithm is used to reconstruct the original sampling sequence, so that the reconstructed signal is similar to the synchronous sampling signal, and then the reconstructed signal is separated from the harmonics and inter-harmonics through the FIR notch filter. Since the spectrum leakage and the fence effect are significantly suppressed, the accurate parameters of each harmonic can be obtained according to the FFT result of the reconstructed signal. By separating the harmonic and interharmonic components in the signal, the method achieves the purpose of suppressing the mutual interference between the two, and can realize the accurate measurement of harmonic and interharmonic parameters at the same time.

3. Beneficial effects

Compared with the prior art, the present invention has the following beneficial effects:

(1) The present invention is a micro-grid harmonic and inter-harmonic detection device based on STM32F107VCT6, fully considering the distributed power structure of the micro-grid itself and the influence of the main network on the micro-grid, especially the separation of the micro-grid and the main grid Or when the microgrid maintains the unbalanced voltage of the main grid and operates with a large amount of harmonics in the microgrid, it can solve the above problems well. It can detect harmonics and interharmonics within the 51st order of the system, and can calculate and display single harmonics Wave distortion rate, total harmonic distortion rate, crest factor, active power, reactive power and other values, in addition, the size of the detection device is very suitable for hand-held, easy to carry;

(2) The present invention is a micro-grid harmonic and inter-harmonic detection device based on STM32F107VCT6. The use of the hardware is based on the Cortex-M3 core 32-bit single-chip STM32F107VCT6, which has fast processing speed and low cost. In terms of harmonic detection have good performance;

(3) The present invention is a method for detecting harmonics and interharmonics in a microgrid. It proposes a new detection algorithm for quasi-synchronous sampling and reconstruction signals. It designs an adaptive tracking digital notch filter to calculate the fundamental wave period. The design It is very targeted for the characteristics of the periodic fluctuation of the microgrid signal. It uses the method of reconstructing the signal with the cubic spline interpolation function to realize the quasi-synchronization of the non-synchronous signal. It avoids the spectrum leakage and Fence effect, the design of the FIR notch filter can realize the separation of harmonics and inter-harmonics, avoid the mutual interference between harmonics and inter-harmonics, and improve the calculation accuracy.

Description of drawings

1 is a schematic diagram of the overall structure of the interharmonic detection device of the present invention;

Fig. 2 is a schematic flow chart of the harmonic and inter-harmonic algorithm of the present invention.

detailed description

The present invention will be described in detail below in conjunction with the accompanying drawings.

The device workflow as shown in Figure 1 is: the present invention adopts the following technical solutions.

A micro-grid harmonic and inter-harmonic detection device based on STM32F107VCT6, including a signal acquisition module, a signal processing module, and a signal display output module. The signal acquisition module uses a Hall voltage and current sensor as the acquisition of the three-phase voltage and current of the power grid The signal processing module includes a voltage and current conversion circuit, an anti-aliasing filter circuit, a level adjustment circuit, a core processor STM32F107VCT6 and a peripheral circuit of the processor; the acquisition end of the Hall voltage and current sensor of the signal acquisition module is connected to the power grid, and the output The terminal is connected to the signal processing module, and in the signal processing module, it passes through the current-voltage conversion circuit, anti-aliasing filter circuit and level adjustment circuit in sequence; the output terminal of the level adjustment circuit is connected to the A/D converter input terminal of the core processor STM32F107VCT6, The peripheral of STM32F107VCT6 is connected to the LCD display. The core processor STM32F107VCT6 is produced by STMicroelectronics (ST) Group.

The peripheral circuit of the processor includes a power supply circuit, a reset circuit and a crystal oscillator circuit; the power supply circuit provides power supplies of different voltages required by the entire system through voltage conversion; the reset circuit adopts a reset chip manual reset method; the crystal oscillator circuit provides a clock signal for the single-chip microcomputer. The output display module includes LCD display screen, keyboard and PC communication module. The output display module also includes a keyboard circuit, and the connection between the keyboard and the core processor STM32F107VCT6 is through the keyboard circuit.

The acquisition terminal of the Hall voltage and current sensor of the signal acquisition module is connected to the power grid, and the output terminal is connected to the signal processing module. First, the current signal is converted into a voltage signal through the current-voltage conversion circuit. Voltage conversion and voltage clamping. After the signal comes out of the current-voltage conversion circuit, it enters the anti-aliasing filter circuit. Considering the filtering effect and delay time, the circuit uses a second-order Butterworth filter for analog filtering. The filter cut-off frequency is 2500Hz and can filter out more than 51 times. The purpose of filtering here is to satisfy the sampling theorem during subsequent data acquisition and avoid frequency aliasing during sampling. After the signal passes through the anti-aliasing filter circuit, it enters the level adjustment circuit. This circuit is composed of two operational amplifiers and resistors. The circuit is supplied with a 1.6V reference voltage from the outside. After the signal passes through the circuit, due to the proportional circuit formed by the operational amplifier and the reference voltage The adjustment makes the signal amplitude become half of the original and the waveform is raised as a whole. In addition, due to the action of the two operational amplifiers, a differential signal is formed, and then the signal is sent to the ADC pin of the STM32F107VCT6, and the A/D converter inside the STM32F107VCT6 is used for analog-to-digital conversion. STM32F107VCT6 has two 12-bit A/D converters with a total of 18 channels, and two channels can work at the same time. In addition to meeting the requirements of the IEC standard, that is, the width of the analysis window is 200ms, and the frequency resolution of DFT is 5Hz. In order to improve the calculation accuracy of the fundamental period and the harmonic frequency resolution, the sampling frequency of the device is set to 10240Hz. The length of the processed sample sequence is 4096 points. After the signal is sampled and converted into a digital signal, it is analyzed and calculated in STM32F107VCT6.

The specific implementation steps are shown in Figure 2:

A method for detecting harmonics and interharmonics of a microgrid, the steps of which are:

(1) In the case of satisfying the Nyquist sampling theorem, the power system signal is sampled at equal intervals at a fixed sampling frequency, that is, the equation

K·T _s ＝P·T ₀ (1-1)

Established, where T _s is the sampling period, T ₀ is the signal period, K is the number of sampling points, P is the number of signal periods, for asynchronous sampling, obviously P is not an integer;

(2) Use the adaptive tracking digital notch filter to filter out the fundamental component in the sampling sequence x(KT _s ) to obtain a new sampling sequence x'(KT _s ), so subtract x from x(KT _s ) '(KT _s ), the fundamental component x ₀ (KT _s ) can be obtained; compared with the traditional notch filter, the adaptive tracking digital notch filter makes the fundamental frequency in the It can also accurately filter out fluctuations near 50Hz. The adaptive tracking digital notch filter here is to add adaptive control on the basis of the digital notch filter, it accepts the input signal with the notch frequency, obtains a ₁ through the least squares adaptive algorithm, and uses a ₁ constitutes a digital notch filter H(z ^-1 ) to trap signals with this notch frequency. The frequency response of an ideal digital notch filter is

$\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; j\&omega;t</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 1</span>}& {\mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; \&omega;</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&omega;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

Wherein e is a natural constant, j is an imaginary number unit, ω is a frequency, t is a time, and _ω0 is a notch frequency;

For a single notch frequency, its transfer function H(z ^-1 ) has the following symmetrical mirror form

$\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; pz</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

In the formula, z is a complex variable, p is a constant close to 1 but slightly smaller than 1, a ₁ is a parameter, which is determined by the notch frequency, and the notch frequency expression is

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; arccos</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>}{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;T</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

Where T is the sampling period, a ₁ (n) that changes with the sampling point is obtained by the adaptive least squares method, and the corresponding notch frequency also changes with time. In order to achieve self-adaptability, p changes with the sampling point as is p(n), thus, the H(z ^-1 ) corresponding to each sampling point signal in x(KT _s ) will be different, and the adaptive adjustment is realized, and the base signal can be obtained by subtracting the original signal from the filtered signal wave signal;

(3) Use the linear interpolation method to calculate the fundamental wave period of the signal, and set the function value at the node a=t ₀ <t ₁ <t ₂ ...<t _n =b to be y ₀ , y ₁ , y ₂ ,…,y _n , replace the curve with a straight line in each small interval [x _j ,x _j+1 ], then the time t _i when the signal crosses the threshold a for the ith time is

${\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>\frac{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

In the same way, the time t _i+20 when the signal crosses the threshold a for the i+20th time, the average value T* of the sampling time of 20 cycle lengths of the sampling sequence x(n) is:

$\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; *</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 20</span>}}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 20</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

Then T* is the fundamental period of the signal, and 20 periods are used here to calculate the fundamental period more accurately;

(4) According to the calculated fundamental wave period, the harmonic components in the original sampling sequence are quasi-synchronized to obtain the reconstructed sequence x ^* (kλ _s ), P is not an integer when sampling asynchronously, and spectrum leakage will occur. In order Realize the quasi-synchronization of the non-synchronous sampling sequence, adjust the sampling period T _s , and use T* to calculate the quasi-synchronous sampling period λ _s

${\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 20</span>}{\mathrm{<span\; class="notranslate">\; T</span>}}^{<span\; class="notranslate">\; *</span>}}{{\mathrm{<span\; class="notranslate">\; L</span>}}^{<span\; class="notranslate">\; *</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Among them, L ^* is the number of sampling points in 20 periods of the new quasi-synchronous sampling sequence, but it can be approximated that L ^* = L to reduce the error, and L is the number of sampling points in the 20 signal periods of the original sampling sequence, Each sampling point in the quasi-synchronous sampling sequence is obtained by the cubic spline interpolation method, the sampling point in the original sampling sequence is used as the interpolation node, T _s is the interpolation node interval, and the starting point of a certain signal cycle is also the starting point of the sampling point , since the cubic spline difference function K(t) is a piecewise cubic polynomial on each interval [x _j , x _j+1 ], the second order derivative K''(t of the cubic spline function K(t) is known ) is a first-degree polynomial in each small interval, assuming K''(t _j )=M _j , K''(t _j+1 )=M _j+1 , then the expression of K''(t) is

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; \&Prime;</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Wherein, h _j =t _j+1 -t _j .

Integrate K''(t) twice and bring it into the boundary condition to get the expression of K(t) as

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; t</span>}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}$

(t∈[t _j ,t _j+1 ];j=0,1,…,n-1)(1-9)

Assuming that the mth quasi-synchronous sampling point is located between the jth synchronous sampling point and the j+1th synchronous sampling point, then the quasi-synchronous sampling point can be calculated by the following formula

$\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\frac{{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 3</span>}}}{{\mathrm{<span\; class="notranslate">\; 6</span>}\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 6</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>\frac{{\mathrm{<span\; class="notranslate">\; m\&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}{{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}}$

(t∈[t _j ,t _j+1 ];j=0,1,…,n-1)(1-10)

Repeating the calculation process of K(mλ _s ), you can get L ^* -1 calculated values. These L ^* -1 quasi-synchronous sampling points and the starting point together form a quasi-synchronous sampling sequence with a length of T ^* . So far, we have obtained Reconstruct all information of the sampling sequence x ^* (kλ _s ), that is, the position and sampling value of the reconstructed signal;

(5) Use the FIR notch filter to separate the harmonic and interharmonic components in x ^* (kλ _s ), and set multiple notch frequencies in the filter to form a comb; since the sampling frequency is the frequency of the inter-tooth interval of the filter Integer multiples, so the FIR notch filter has a good effect on synchronous sampling without energy loss, and is suitable for quasi-synchronous sampling sequences; since the notch frequency of the filter is an integer multiple of the fundamental, it can filter out the fundamental waves and all integer harmonics;

(6) For the inter-harmonic component x ^*I (kλ _s ) obtained after filtering with the FIR notch filter, use the double interpolation FFT algorithm with Hanning window to calculate the parameters of the inter-harmonic;

(7) Subtract the interharmonic component x ^*I (kλ _s ) from the quasi-synchronous sampling sequence x ^* (kλ _s ) to obtain the harmonic component x ^*H (kλ _s );

(8) Carry out DFT or FFT operation on x ^*H (kλ _s ), and calculate the harmonic parameters of each order from the result.

Can find out that the present invention has the following advantages by embodiment:

1) The design of the interharmonic detection device fully considers the characteristics of the harmonics of the microgrid.

2) The structure is simple, the overall cost is low, and it is easy to implement. Since the signal reconstruction technology is used in the algorithm, the phase-locked circuit on the hardware is saved, which not only makes the detection results of inter-harmonics more accurate, but also saves costs.

3) A 32-bit interconnected single-chip microcomputer STM32F107 based on the Cortex-M3 core is selected as the core processor. The high computing speed of this processor can give full play to the superiority of the signal reconstruction algorithm and realize the effective cooperation between the algorithm and the hardware.

4) In the detection method, the method of signal reconstruction is adopted, which overcomes the mutual interference of harmonics and interharmonics and the spectrum leakage of traditional FFT algorithms.

5) By designing the FIR filter to separate the harmonics and inter-harmonics, the harmonics and inter-harmonics can be calculated separately, avoiding mutual interference and improving the calculation accuracy.

Claims (2)

1. A microgrid harmonic and interharmonic detection method, the steps of which are: (1) In the case of satisfying the Nyquist sampling theorem, the power system signal is sampled at equal intervals at a fixed sampling frequency, that is, the equation K·T _s ＝P·T ₀ (1-1) Established, where T _s is the sampling period, T ₀ is the signal period, K is the number of sampling points, P is the number of signal periods, for asynchronous sampling, obviously P is not an integer; (2) Use an adaptive tracking digital notch filter to filter out the fundamental component in the sampling sequence x(KT _s ), and obtain a new sampling sequence x'(KT _s ), so subtract x from x(KT _s ) '(KT _s ), the fundamental component x ₀ (KT _s ) can be obtained; compared with the traditional notch filter, the adaptive tracking digital notch filter makes the fundamental frequency in the It can also accurately filter out fluctuations around 50Hz. The frequency response of an ideal digital notch filter is Wherein e is a natural constant, j is an imaginary number unit, ω is a frequency, t is a time, and _ω0 is a notch frequency; For a single notch frequency, its transfer function H(z ^-1 ) has the following symmetrical mirror form In the formula, z is a complex variable, p is a constant close to 1 but slightly smaller than 1, a ₁ is a parameter, which is determined by the notch frequency, and the notch frequency expression is Where T is the sampling period, a ₁ (n) that changes with the sampling point is obtained by the adaptive least squares method, and the corresponding notch frequency also changes with time. In order to achieve self-adaptability, p changes with the sampling point as is p(n); (3) Use the linear interpolation method to calculate the fundamental wave period of the signal, and set the function value at the node a=t ₀ <t ₁ <t ₂ ...<t _n =b to be y ₀ , y ₁ , y ₂ ,...,y _n , replace the curve with a straight line in each small interval [x _j ,x _j+1 ], then the time t _i when the signal crosses the threshold a for the ith time is In the same way, the time t _i+20 when the signal crosses the threshold a for the i+20th time, the average value T ^* of the sampling time of the 20 cycle length of the sampling sequence x(n) is: Then T ^* is the fundamental period of the signal; (4) According to the calculated fundamental period, the harmonic components in the original sampling sequence are quasi-synchronized to obtain the reconstructed sequence x ^* (kλ _s ), P is not an integer when sampling asynchronously, and spectrum leakage will occur. In order Realize the quasi-synchronization of the non-synchronous sampling sequence, adjust the sampling period T _s , and use T ^* to calculate the quasi-synchronous sampling period λ _s Wherein L ^* is the number of sampling points in 20 periods of the new quasi-synchronous sampling sequence, but it can be approximated to take L ^* = L to reduce the error, and L is the number of sampling points in the original sampling sequence of 20 signal periods, Each sampling point in the quasi-synchronous sampling sequence is obtained by the cubic spline interpolation method, the sampling point in the original sampling sequence is used as the interpolation node, T _s is the interpolation node interval, and the starting point of a certain signal cycle is also the starting point of the sampling point , since the cubic spline difference function K(t) is a piecewise cubic polynomial on each interval [x _j , x _j+1 ], the second order derivative K”(t) of the cubic spline function K(t) is known It is a first-degree polynomial in each small interval, assuming K”(t _j )=M _j , K”(t _j+1 )=M _j+1 , then the expression of K”(t) is Among them, h _j =t _j+1 -t _j ; Integrate K"(t) twice and bring it into the boundary condition to get the expression of K(t) as Assuming that the mth quasi-synchronous sampling point is located between the jth synchronous sampling point and the j+1th synchronous sampling point, then the quasi-synchronous sampling point can be calculated by the following formula Repeating the calculation process of K(mλ _s ), you can get L ^* -1 calculated values. These L ^* -1 quasi-synchronous sampling points and the starting point together form a quasi-synchronous sampling sequence with a length of T ^* . So far, we have obtained Reconstruct all information of the sampling sequence x ^* (kλ _s ), that is, the position and sampling value of the reconstructed signal; (5) Utilize the FIR notch filter to separate the harmonic and interharmonic components in x ^* (kλ _s ), and the filter sets multiple notch frequencies to form a comb; (6) Interharmonic component x ^{* I} (kλ _s ) obtained after FIR notch filter filtering, use the double interpolation FFT algorithm that adds Hanning window to calculate the parameter of interharmonic; (7) Subtract the inter-harmonic component x ^*I (kλ _s ) from the quasi-synchronous sampling sequence x ^* (kλ _s ) to obtain the harmonic component x ^*H (kλ _s ); (8) Carry out DFT or FFT operation on x ^*H (kλ _s ), and calculate the harmonic parameters of each order from the result. 2. A kind of microgrid harmonic and interharmonic detection method according to claim 1, is characterized in that: described self-adaptive tracking digital notch filter is to increase self-adaptive on the basis of digital notch filter Control, which accepts the input signal with the notch frequency, calculates a ₁ through the least squares adaptive algorithm, and uses a ₁ to form a digital notch filter H(z ^-1 ) to trap and remove the signal with the notch frequency.