Outlet short circuit state detection method based on photovoltaic outlet side circuit breaker

Abstract

The invention relates to an outlet short-circuit state detection method based on a photovoltaic outlet side circuit breaker, and the method comprises the steps: constructing short-circuit current and voltage time-varying vectors through a short-circuit time-varying vector theory, and constructing a time-varying vector modulus length calculation mathematical model; on the basis, the short-circuit time-varying vector modulus length is calculated, short-circuit state characteristics containing the short-circuit current and voltage time-varying vector modulus length are designed, the short-circuit state characteristics are extracted through the time-varying vector modulus length, and therefore a short-circuit state detection criterion is constructed; the short-circuit fault detection judgment of the photovoltaic outlet side is carried out by setting the short-circuit state detection threshold value of the photovoltaic outlet side. The method is beneficial to the efficiency, reliability and universality of short circuit state detection.

Description

Technical Field

[0001] The present invention relates to the technical field of power detection, and in particular to a method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker. Background Art

[0002] With the construction of new power systems, DC sources and loads represented by photovoltaics and energy storage are widely used in low-voltage AC systems, which poses new challenges to the operation and protection of traditional power systems. With the access of a large number of DC sources and loads, the short-circuit characteristics of low-voltage AC systems have undergone a series of fundamental changes.

[0003] In engineering practice, the access of distributed power sources usually requires the installation of grid-connected circuit breakers and the setting of lower protection settings to ensure that the short-circuit fault point can be promptly removed when a short circuit occurs in the low-voltage AC system. However, when a short-circuit fault occurs on the grid side of the low-voltage AC system, the surge in short-circuit current and voltage drop caused by the distributed power sources will have a significant impact on the system. Although the distributed power sources are equipped with active protection functions, which can trigger short-circuit protection when the circuit breaker protection fails and drive the semiconductor devices to quickly remove the short-circuit current, this operation often causes the semiconductor devices to operate at abnormal working currents, thereby significantly reducing their service life.

[0004] In addition, when the DC source load device is connected to the low-voltage AC system, the interference generated by the transient signal during its startup process also poses a challenge to the reliability of the short-circuit detection method. This interference manifests itself as the identification problem between the weak detection characteristics in the short-circuit loop of the distributed power source and the strong interference generated when the power conversion device is started, which further increases the complexity of the protection method. Therefore, in the process of building a new low-voltage distribution grid, the risk of false operation and refusal of the short-circuit protection method will be greatly increased, and the traditional short-circuit detection method is difficult to apply when facing the complex operating conditions brought by the DC source load.

[0005] In summary, with the rapid increase of DC sources and loads in low-voltage AC systems, new short-circuit state detection methods must be developed to adapt to the new operating environment and improve the safety and reliability of the system. Summary of the invention

[0006] The object of the present invention is to provide a method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker, which is beneficial to the efficiency, reliability and universality of short-circuit state detection.

[0007] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a method for detecting the short-circuit state of an outlet based on a circuit breaker on the photovoltaic outlet side, constructing short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, and constructing a mathematical model for calculating the modulus length of the time-varying vector; on this basis, calculating the modulus length of the short-circuit time-varying vector, designing a short-circuit state feature containing the modulus length of the short-circuit current and voltage time-varying vector, extracting the short-circuit state feature through the modulus length of the time-varying vector, thereby constructing a short-circuit state detection criterion, and performing short-circuit fault detection and judgment on the photovoltaic outlet side by setting the short-circuit state detection threshold on the photovoltaic outlet side.

[0008] Furthermore, when constructing the short-circuit current and voltage time-varying vectors through the short-circuit time-varying vector theory, since the full current and full voltage formulas of the low-voltage AC short circuit both have a steady-state periodic component and a DC attenuation component, the following formula is used to describe them:

[0009]

[0010] Where, if(t) and uf(t) represent the short-circuit current and short-circuit voltage at the detection point respectively; Z S Represents the equivalent impedance of the short-circuit loop; Z 1 is the equivalent impedance between the detection point and the short-circuit point; R 1 , L 1 Z 1 resistance and inductance; θ is the initial phase angle; t is the time; τ is the DC component attenuation coefficient; i(0) is the initial current value at the moment of short circuit; U m is the maximum value of system power; U m sin(ωt+θ) is the system power supply; ω is the angular velocity of rotation; is the short-circuit impedance angle; is the phase angle offset of the short-circuit voltage at the detection point;

[0011] Formulas (1) and (2) are mapped to vector space, and the time dimension is introduced to construct the time-varying vectors of short-circuit current and voltage.

[0012] Furthermore, the calculation formulas for the module lengths of the short-circuit current steady-state vector, the short-circuit voltage steady-state vector, the load current vector before the short circuit, and the load voltage vector before the short circuit are as follows:

[0013]

[0014] in, is the short-circuit current steady-state vector, and the initial phase angle is is the short-circuit voltage steady-state vector, and the initial phase angle is are the load current and voltage vectors before short circuit respectively; Z load is the load equivalent impedance.

[0015] In the time-varying vector space, the short-circuit current and voltage time-varying vectors are solved by geometry, and the expressions of the short-circuit current and voltage time-varying vector modulus lengths are obtained as follows:

[0016]

[0017] in, is the time-varying vector of short-circuit current; is the time-varying vector of short-circuit voltage; is the short-circuit current static decay vector, The direction of the short-circuit current is always kept stationary, and the modulus decays exponentially with time, with a decay coefficient of τ; ψ is the static decay vector of the short-circuit current Angle with the α axis.

[0018] Furthermore, the short-circuit time-varying vector modulus is calculated based on the minimum absolute deviation method. The specific method is as follows:

[0019] The original signal, i.e. the collected current and voltage values, is fitted using the minimum absolute deviation method to obtain the best estimated values ​​of the fitting line parameters k(n) and b(n). On this basis, the signal fitting value is obtained. Calculate the absolute error series between the fitted signal and the original signal:

[0020]

[0021] The data points with high interference degree are eliminated by absolute error to filter out the impulse noise in the original signal. To this end, the data in the absolute error series are clustered into two types of data clusters by K-means clustering algorithm. The clustering criterion function is defined as:

[0022]

[0023] Where N is the number of data in the sampling window, h j is the element in the cluster center matrix H;

[0024] By updating the cluster centers, the clustering criterion function value converges, that is, two types of data based on two cluster centers are obtained.

[0025] Furthermore, the implementation steps of the K-means clustering algorithm are as follows:

[0026] (a) Generate cluster centers according to the number of target categories and initialize the cluster centers. To reduce the iteration speed of cluster calculation, set the cluster initialization centers to the maximum and minimum error values;

[0027] (b) Calculate the Euclidean distance between the sample data and the cluster center hj, and assign the sample data to the cluster center point with the closest distance to form two cluster data clusters;

[0028] (c) Calculate the average value of the sample data in the cluster data cluster and use it as the cluster center hj of the cluster cluster;

[0029] (d) Based on the updated cluster centers, a new clustering criterion function value is calculated, and it is determined whether the errors of two successive clustering criterion function values ​​converge; if so, clustering ends; if not, steps (b) and (c) are repeated.

[0030] Furthermore, the clustering calculation amount can be reduced by adjusting the clustering criterion function value convergence allowable error or the maximum number of iterations.

[0031] Furthermore, the clustering results are distinguished: if the cluster centers of two data clusters are far apart, it means that there is impulse noise interference in the original signal. The original data corresponding to the cluster with a smaller cluster center is selected, and the minimum absolute error is performed again to obtain the corrected fitting line parameter estimate. Sum signal estimate If the values ​​of the two cluster centers are both small, it means that the noise interference is low and no data correction is required; the calculation formula for the time-varying vector modulus after the fitting line parameter correction is:

[0032]

[0033] in, is the modulus value of the nth sampling window of the fitting line; y 0° (n) is the value of the original signal in the nth sampling window under 0° deflection; y 90° (n) is the original signal value under the 90° deflection signal under the nth sampling window; is the best estimate of the slope of the fitted straight line under the nth sampling window; The best estimate of the intercept of the fitting line under the nth sampling window; f is the original signal frequency; T s is the signal sampling period; j is the position of the selected original data in the sampling window, and M is the number of selected original data.

[0034] Furthermore, by setting the short-circuit state detection threshold on the photovoltaic outlet side, short-circuit fault detection and judgment on the photovoltaic outlet side is performed. When the branch state quantity is greater than the set short-circuit state detection threshold, the photovoltaic outlet is considered to be in a short-circuit state, otherwise it is considered to be in a normal state.

[0035] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides an outlet short-circuit state detection method based on the photovoltaic outlet side circuit breaker, which constructs short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, and analyzes the vector change process under short circuit; on this basis, the real-time calculation method of short-circuit time-varying vector modulus is studied, and the branch short-circuit state characteristics containing short-circuit current and voltage time-varying vector modulus are designed, so as to construct the short-circuit state detection criterion, thereby improving the rapidity, reliability and universality of short-circuit state detection. Compared with the traditional detection method, the outlet short-circuit state detection method proposed by the present invention has better detection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 1 is a schematic diagram of an implementation principle of a method for detecting an outlet short-circuit state according to an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of a simulation experiment topology in an embodiment of the present invention;

[0038] Figure 3 It is a time-varying vector diagram of short-circuit current and voltage constructed based on the introduction of the time dimension in an embodiment of the present invention;

[0039] Figure 4 is a short-circuit time-varying vector diagram of a new energy source in an embodiment of the present invention;

[0040] Figure 5 is a comparison of the detection time of the method in the embodiment of the present invention and the conventional detection method;

[0041] Figure 6 is the time-varying vector modulus based on the photovoltaic outlet side in the embodiment of the present invention;

[0042] Figure 7 It is a schematic diagram of the mechanism of action of the branch state quantity in the embodiment of the present invention. DETAILED DESCRIPTION

[0043] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0044] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0045] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0046] The present invention provides an outlet short-circuit state detection method based on a photovoltaic outlet side circuit breaker. On the basis of summarizing the transient and steady-state characteristics of short-circuit faults in low-voltage AC systems containing DC sources and loads, the need for constructing short-circuit state characteristics is proposed. This method constructs short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, and analyzes the vector change process under short circuit; on this basis, the real-time calculation method of the short-circuit time-varying vector modulus is studied, and the branch short-circuit state characteristics containing short-circuit current and voltage time-varying vector modulus are designed, so as to construct a short-circuit state detection criterion to improve the rapidity, reliability and universality of short-circuit state detection. Compared with traditional detection methods, this dynamic vector modulus short-circuit state detection method has better detection effect, and the specific effects are as follows: Figure 5 In order to achieve the above purpose, the present invention adopts the following technical solution:

[0047] like Figure 1 As shown, this embodiment provides an outlet short-circuit state detection method based on the photovoltaic outlet side circuit breaker, constructs short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, and constructs a mathematical model for calculating the time-varying vector modulus; on this basis, calculates the short-circuit time-varying vector modulus, designs short-circuit state features including short-circuit current and voltage time-varying vector modulus, extracts short-circuit state features through the time-varying vector modulus, and constructs a short-circuit state detection criterion, and sets the short-circuit state detection threshold on the photovoltaic outlet side to perform short-circuit fault detection and judgment on the photovoltaic outlet side, thereby improving the rapidity, reliability, and universality of short-circuit state detection. Compared with traditional detection methods, this method has better detection effects, and the specific effects are as follows: Figure 5 shown.

[0048] When constructing short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, since the full current and full voltage formulas of low-voltage AC short circuits have steady-state periodic components and DC attenuation components, the following formulas are used to describe them:

[0049]

[0050] Where, if(t) and uf(t) represent the short-circuit current and short-circuit voltage at the detection point respectively; Z S Represents the equivalent impedance of the short-circuit loop; Z 1 is the equivalent impedance between the detection point and the short-circuit point; R 1 , L1 Z 1 resistance and inductance; θ is the initial phase angle; t is the time; τ is the DC component attenuation coefficient; i(0) is the initial current value at the moment of short circuit; U m is the maximum value of system power; U m sin(ωt+θ) is the system power supply; ω is the angular velocity of rotation; is the short-circuit impedance angle; is the phase angle offset of the short-circuit voltage at the detection point.

[0051] Map equations (1) and (2) to vector space, and introduce the time dimension to construct the time-varying vectors of short-circuit current and voltage, as follows: Figure 2 shown.

[0052] In this embodiment, the time-varying vector diagram of the short-circuit current and voltage constructed based on the introduction of the time dimension is as follows: Figure 3 shown. Figure 3 The vectors in are defined as follows: is the short-circuit current steady-state vector, and the initial phase angle is is the short-circuit voltage steady-state vector, and the initial phase angle is are the load current and voltage vectors before short circuit; is the short-circuit current static attenuation vector, ψ is the angle between the vector and the α axis; is the time-varying vector of short-circuit current; is the time-varying vector of short-circuit voltage. Among them, Rotate counterclockwise with the pole as the center, and the angular velocity is ω; The direction of is always kept stationary, and the modulus decays exponentially with time, with a decay coefficient of τ. The α and β axes form a rectangular coordinate system, and the projection of each vector on the α axis is the original signal. Figure 3 In (b), the red line segment indicates Figure 3 The voltage drop of the short-circuit current static decay vector on the short-circuit impedance in (a) remains stationary in the direction, and the modulus decays exponentially. load is the load equivalent impedance.

[0053] Short-circuit current steady-state vector Short circuit voltage steady state vector Load current vector before short circuit Load voltage vector before short circuit The formula for calculating the modulus length is:

[0054]

[0055] In the time-varying vector space, the short-circuit current and voltage time-varying vectors are solved by geometry, and the expressions of the short-circuit current and voltage time-varying vector modulus lengths are obtained as follows:

[0056]

[0057] In this embodiment, the short-circuit time-varying vector modulus is calculated based on the minimum absolute deviation method, and the specific method is:

[0058] Due to the complex electromagnetic environment of the low-voltage AC system, the sampled signal mainly contains two types of white noise and pulse noise. In order to further improve the confidence of the modulus length result, it is necessary to perform noise reduction processing on the calculation process. The original signal, that is, the collected current and voltage values, is fitted using the minimum absolute deviation method to obtain the best estimated values ​​of the fitting line parameters k(n) and b(n). On this basis, the signal fitting value is obtained. Calculate the absolute error series between the fitted signal and the original signal:

[0059]

[0060] The data points with high interference degree are eliminated by absolute error to filter out the impulse noise in the original signal. To this end, the data in the absolute error series are clustered into two types of data clusters through the K-means clustering algorithm; the clustering criterion function is defined as:

[0061]

[0062] Where N is the number of data in the sampling window, h j is the element in the cluster center matrix H.

[0063] By updating the cluster centers, the clustering criterion function value converges, that is, two types of data based on two cluster centers are obtained.

[0064] The implementation steps of the K-means clustering algorithm are as follows:

[0065] (a) Generate cluster centers according to the number of target categories and initialize the cluster centers. To reduce the iteration speed of cluster calculation, set the cluster initialization centers to the maximum and minimum error values;

[0066] (b) Calculate the Euclidean distance between the sample data and the cluster center hj, and assign the sample data to the cluster center point with the closest distance to form two cluster data clusters;

[0067] (c) Calculate the average value of the sample data in the cluster data cluster and use it as the cluster center hj of the cluster cluster;

[0068] (d) Based on the updated cluster centers, a new clustering criterion function value is calculated, and it is determined whether the errors of two successive clustering criterion function values ​​converge; if so, clustering ends; if not, steps (b) and (c) are repeated.

[0069] In practical applications, the amount of clustering calculation can be reduced by adjusting the convergence allowable error or the maximum number of iterations of the clustering criterion function value.

[0070] Distinguish the clustering results: If the cluster centers of two data clusters are far apart, it means that there is impulse noise interference in the original signal. Select the original data corresponding to the cluster with a smaller cluster center, and perform the minimum absolute error again to obtain the corrected fitting line parameter estimate. Sum signal estimate If the values ​​of the two cluster centers are both small, it means that the noise interference is low and no data correction is required. The calculation formula for the time-varying vector modulus after the fitting line parameter correction is:

[0071]

[0072] in, is the modulus value of the nth sampling window of the fitting line; y 0° (n) is the value of the original signal in the nth sampling window under 0° deflection; y 90° (n) is the original signal value under the 90° deflection signal under the nth sampling window; is the best estimate of the slope of the fitted straight line under the nth sampling window; The best estimate of the intercept of the fitting line under the nth sampling window; f is the original signal frequency; T s is the signal sampling period; j is the position of the selected original data in the sampling window, and M is the number of selected original data.

[0073] This method proposes a short-circuit state detection threshold setting method, specifically:

[0074] According to the distribution of the maximum branch state quantity of the circuit breaker on the photovoltaic outlet side, for the short-circuit protection of the direct-connected load and the DC source load device, a judgment criterion is set to improve the rapidity of short-circuit detection. The short-circuit fault detection and judgment on the photovoltaic outlet side are performed by setting the short-circuit state detection threshold on the photovoltaic outlet side. When the branch state quantity is greater than the short-circuit state early detection threshold we set, we believe that the photovoltaic outlet is in a short-circuit state, otherwise it is considered to be in a normal state.

[0075] The invention proposes a method for detecting the short-circuit state of a photovoltaic outlet circuit breaker. The specific implementation principle is as follows: Figure 1As shown in the figure. Firstly, the short-circuit time-varying vector analysis theory is proposed, the short-circuit current and voltage time-varying vectors are constructed, and the time-varying vector modulus calculation mathematical model is designed; then, the short-circuit state characteristics are extracted through the time-varying vector modulus, the short-circuit state detection criterion is designed, and the short-circuit state detection threshold setting method is proposed; finally, the response characteristics of the proposed short-circuit state detection algorithm are analyzed through the short-circuit fault simulation model of the low-voltage AC system with DC source and load, and compared with the detection algorithm based on the transient and steady-state characteristics of the short-circuit current, the technical advantages of the proposed short-circuit state detection method are further explained.

[0076] According to the short-circuit time-varying vector analysis theory, the line short-circuit current and voltage time-varying vectors are constructed, and a mathematical model for calculating the modulus of the time-varying vector is designed. Since the full current and full voltage formulas of low-voltage AC short circuits have steady-state periodic components and DC attenuation components, the time dimension is introduced to construct the short-circuit current and voltage time-varying vectors. In the time-varying vector space, the short-circuit current and voltage time-varying vectors can be solved by geometry, and the short-circuit current and voltage time-varying vectors can be easily obtained.

[0077] Analyzing formula (4), the time-varying vector modulus of the short-circuit current is determined by the steady-state vector modulus, the static attenuation vector modulus and the angle between the two. Considering that the steady-state vector modulus is constant at the time of short circuit, the initial value of the static attenuation component modulus is also determined, and the angle between the two fluctuates and decays with the static attenuation component in a short time; since the k coefficient is approximately 0, it can be approximately considered that the time-varying vector modulus of the short-circuit voltage jumps to the steady-state vector modulus value. It can be seen that the time-varying vector modulus of the short-circuit current and voltage maintains a stable growth trend, and the phase effect of the process vector is limited, which can more intuitively describe the short-circuit state of the system.

[0078] Since the short-circuit output current of new energy sources such as photovoltaics and energy storage is controlled by the grid-connected point voltage, the converter adopts a unity power factor control method, and the controller responds quickly. It can be approximately considered that the inverter output short-circuit current time-varying vector always tracks the grid-connected point voltage vector after the short circuit, and the phases of the two time-varying vectors are the same. Therefore, according to the short-circuit voltage time-varying vector, the distributed power supply output short-circuit current time-varying vector under the grid-connected point short circuit can be constructed, such as Figure 4 shown.

[0079] The time-varying vector modulus is calculated using the minimum absolute deviation method, and the K-means clustering algorithm is used to improve the noise resistance performance.

[0080] (a) Generate cluster centers according to the number of target categories and initialize the cluster centers. To reduce the iteration speed of cluster calculation, the cluster initialization centers can be set to the maximum and minimum error values;

[0081] (b) Calculate the Euclidean distance between the sample data and the cluster center hj, and assign the sample data to the cluster center point with the closest distance to form two cluster data clusters;

[0082] (c) Calculate the average value of the sample data in the cluster data cluster and use it as the cluster center hj of the cluster cluster;

[0083] (d) Based on the updated cluster center, calculate the new clustering criterion function value and determine whether the error between the two clustering criterion function values ​​converges. If so, clustering ends. If not, repeat steps (b) and (c).

[0084] For the clustering results, it is necessary to distinguish: if the cluster centers of two data clusters are far apart, it means that there is impulse noise interference in the original signal. The original data corresponding to the cluster with a smaller cluster center can be selected, and the minimum absolute error can be performed again to obtain the corrected fitting line parameter estimate. Sum signal estimate If the values ​​of the two cluster centers are both small, it means that the noise interference is low and no data correction is required.

[0085] Construct short-circuit state characteristics. On the basis of short-circuit current and voltage time-varying vector modulus calculation, construct a composite feature of branch state quantity through current and voltage time-varying vector modulus. Use the change of current and voltage phase time-varying modulus during short-circuit fault to judge short-circuit fault. The specific effect is as follows: Figure 6 This simulation is set to have a three-phase short circuit at 34.7ms and remove the short circuit point at 144.8ms.

[0086] A branch state quantity is defined to extract short-circuit state characteristics to achieve rapid detection of low-voltage AC short circuits with DC sources and loads. The constructed branch state quantity is as follows:

[0087]

[0088] In formula (8), BSV k (n) is the real-time branch state quantity of branch k; is the time-varying vector modulus of the current and voltage of branch k; UN is the rated voltage; I fmin_k is the reference value of branch k, and I fmin_k It can be set to 1.5 times the rated output current of the converter; I fmin_k Take the minimum short-circuit current effective value of the short circuit at the end of this branch. By differentially processing the current and voltage moduli, the current and voltage disturbances caused by the normal operation of the system are suppressed, the characteristic difference between the short-circuit fault and the system operation disturbance is increased, and the robustness of the short-circuit state detection is improved. In addition, by normalizing the modulus of the current and voltage time-varying vectors, the difference in the short-circuit branch state caused by different short-circuit power sources can be reduced, and the short-circuit characteristics of the branch state can be improved. Figure 7As shown in the schematic diagram of the mechanism of branch state quantity, when the branch is unloaded, since the current vector modulus is 0, BSV is approximately -1; at t1, the branch load starts, the current vector modulus increases, and the voltage vector modulus remains unchanged. Since the current reference value is the minimum short-circuit current of the branch, the per-unit value of the current time-varying vector modulus is small, and BSV is slightly greater than -1, but still negative; at t2, the branch is short-circuited, and BSV increases rapidly to a positive value. The closer the short-circuit point is to the head end of the branch, the greater the BSV gradually increases; at t3, the short-circuit fault is removed, and BSV returns to a negative value.

[0089] In order to improve the stability and rapidity of short-circuit fault detection, a short-circuit state detection criterion based on the photovoltaic outlet side circuit breaker is constructed according to the branch state quantity, that is, the short-circuit state early detection criterion, as shown in formula (9):

[0090] BSV k (n)≥F act.1 (9)

[0091] Among them, F act.1 The short-circuit state early detection threshold is set by the photovoltaic outlet side to detect and judge the short-circuit fault. When the branch state quantity is greater than the short-circuit state early detection threshold we set, we believe that the photovoltaic outlet is in a short-circuit state, otherwise it is considered to be in a normal state.

[0092] The simulation model of the short-circuit fault of the output including the photovoltaic side circuit breaker is analyzed and verified. The short-circuit state detection simulation experiment topology is as follows: Figure 2 As shown in the figure, when a short circuit occurs in the photovoltaic branch, the branch state quantity changes significantly, which is more conducive to our judgment of the short circuit state detection on the photovoltaic outlet side. Compared with existing short circuit detection methods such as the short circuit current detection algorithm of Fourier series or the short circuit early detection algorithm of wavelet decomposition, the proposed short circuit state detection method does not cause false operation under typical source and load condition disturbances, does not refuse to operate for various short circuit faults, and greatly shortens the short circuit detection time.

[0093] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0094] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0095] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0097] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker, characterized in that: The short-circuit current and voltage time-varying vectors are constructed through the short-circuit time-varying vector theory, and a mathematical model for calculating the time-varying vector modulus is constructed; on this basis, the short-circuit time-varying vector modulus is calculated, and the short-circuit state characteristics including the short-circuit current and voltage time-varying vector modulus are designed. The short-circuit state characteristics are extracted through the time-varying vector modulus, and the short-circuit state detection criterion is constructed. The short-circuit fault detection and judgment on the photovoltaic export side is performed by setting the short-circuit state detection threshold on the photovoltaic export side.

2. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker according to claim 1, characterized in that: When constructing short-circuit current and voltage time-varying vectors through short-circuit time-varying vector theory, since the full current and full voltage formulas of low-voltage AC short circuits have steady-state periodic components and DC attenuation components, the following formulas are used to describe them: Where, if(t) and uf(t) represent the short-circuit current and short-circuit voltage at the detection point respectively; Z S represents the equivalent impedance of the short-circuit loop; Z1 is the equivalent impedance between the detection point and the short-circuit point; R1 and L1 are the resistance and inductance of Z1 respectively; θ is the initial phase angle; t is the time; τ is the DC component attenuation coefficient; i(0) is the initial current value at the moment of short circuit; U m is the maximum value of system power; U m sin(ωt+θ) is the system power supply; ω is the angular velocity of rotation; is the short-circuit impedance angle; is the phase angle offset of the short-circuit voltage at the detection point; Formulas (1) and (2) are mapped to vector space, and the time dimension is introduced to construct the time-varying vectors of short-circuit current and voltage.

3. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker according to claim 2, characterized in that: The calculation formulas for the module lengths of the short-circuit current steady-state vector, the short-circuit voltage steady-state vector, the load current vector before the short circuit, and the load voltage vector before the short circuit are as follows: in, is the short-circuit current steady-state vector, and the initial phase angle is is the short-circuit voltage steady-state vector, and the initial phase angle is are the load current and voltage vectors before short circuit respectively; Z load is the load equivalent impedance; In the time-varying vector space, the short-circuit current and voltage time-varying vectors are solved by geometry, and the expressions of the short-circuit current and voltage time-varying vector modulus lengths are obtained as follows: in, is the time-varying vector of short-circuit current; is the time-varying vector of short-circuit voltage; is the short-circuit current static decay vector, The direction of the short-circuit current is always kept stationary, and the modulus decays exponentially with time, with a decay coefficient of τ; ψ is the static decay vector of the short-circuit current Angle with the α axis.

4. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker according to claim 3, characterized in that: The short-circuit time-varying vector modulus is calculated based on the minimum absolute deviation method. The specific method is: The original signal, i.e. the collected current and voltage values, is fitted using the minimum absolute deviation method to obtain the best estimated values ​​of the fitting line parameters k(n) and b(n). On this basis, the signal fitting value is obtained. Calculate the absolute error series between the fitted signal and the original signal: The data points with high interference degree are eliminated by absolute error to filter out the impulse noise in the original signal. To this end, the data in the absolute error series are clustered into two types of data clusters by K-means clustering algorithm. The clustering criterion function is defined as: Where N is the number of data in the sampling window, h j is the element in the cluster center matrix H; By updating the cluster centers, the clustering criterion function value converges, that is, two types of data based on two cluster centers are obtained.

5. A method for detecting an outlet short-circuit state based on a photovoltaic outlet circuit breaker according to claim 4, characterized in that: The implementation steps of the K-means clustering algorithm are as follows: (a) Generate cluster centers according to the number of target categories and initialize the cluster centers. To reduce the iteration speed of cluster calculation, set the cluster initialization centers to the maximum and minimum error values; (b) Calculate the Euclidean distance between the sample data and the cluster center hj, and assign the sample data to the cluster center point with the closest distance to form two cluster data clusters; (c) Calculate the average value of the sample data in the cluster data cluster and use it as the cluster center hj of the cluster cluster; (d) Based on the updated cluster centers, a new clustering criterion function value is calculated, and it is determined whether the errors of two successive clustering criterion function values ​​converge; if so, clustering ends; if not, steps (b) and (c) are repeated.

6. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker according to claim 5, characterized in that: The amount of clustering calculation can be reduced by adjusting the convergence allowable error or the maximum number of iterations of the clustering criterion function value.

7. A method for detecting an outlet short-circuit state based on a photovoltaic outlet side circuit breaker according to claim 4, characterized in that: Distinguish the clustering results: If the cluster centers of two data clusters are far apart, it means that there is impulse noise interference in the original signal. Select the original data corresponding to the cluster with a smaller cluster center, and perform the minimum absolute error again to obtain the corrected fitting line parameter estimate. Sum signal estimate If the values ​​of the two cluster centers are both small, it means that the noise interference is low and no data correction is required; the calculation formula for the time-varying vector modulus after the fitting line parameter correction is: in, is the modulus value of the nth sampling window of the fitting line; y 0° (n) is the value of the original signal in the nth sampling window under 0° deflection; y 90° (n) is the original signal value under the 90° deflection signal under the nth sampling window; is the best estimate of the slope of the fitted straight line under the nth sampling window; The best estimate of the intercept of the fitting line under the nth sampling window; f is the original signal frequency; T s is the signal sampling period; j is the position of the selected original data in the sampling window, and M is the number of selected original data.

8. The method for detecting an outlet short-circuit state based on a photovoltaic outlet circuit breaker according to claim 4, characterized in that: The short-circuit fault detection and judgment on the photovoltaic outlet side is performed by setting the short-circuit state detection threshold on the photovoltaic outlet side. When the branch state quantity is greater than the set short-circuit state detection threshold, the photovoltaic outlet is considered to be in a short-circuit state, otherwise it is considered to be in a normal state.

Images