Short-circuit fault detection and accurate positioning method and system for alternating-current micro-grid in island mode

Abstract

The invention relates to a short-circuit fault detection and accurate positioning method and system in an island mode of an alternating-current micro-grid, and the method comprises the following steps: collecting a current signal of normal operation of the alternating-current micro-grid in the island mode, and obtaining a maximum value E4 of a wavelet energy spectrum value during normal operation through a binary wavelet energy spectrum transformation algorithm; when a short-circuit fault occurs, a first peak value E4max1, a fault initial phase angle theta w, distributed power supply capacity SDG and transition resistance Rg of a wavelet energy spectrum waveform during the short-circuit fault are collected; judging the type of the short-circuit fault according to the wavelet energy spectrum value; and taking the ratio E4ratio, Rg, SDG and theta w of the size of the first peak value of the fault current wavelet energy spectrum waveform to the maximum value of the wavelet energy spectrum value during normal operation as characteristic values, inputting the characteristic values into a neural network algorithm based on an orthogonal least square method and a radial basis function, and positioning the occurrence position of the short-circuit fault.

Description

Technical Field

[0001] The present application relates to the field of power monitoring technology, and more specifically, to a method and system for detecting and accurately locating short-circuit faults in an AC microgrid island mode. Background Art

[0002] With the continuous growth of energy demand and the concern about the environmental impact of traditional energy, microgrid technology, as a new type of power supply mode, has been more and more widely used. Microgrids can integrate a variety of distributed energy resources, such as solar energy, wind energy, energy storage systems, etc., to achieve efficient power supply to local loads, improve energy utilization efficiency, and reduce dependence on centralized power grids. In the operation of microgrids, AC microgrid island mode is a special operating state. In island mode, the microgrid is disconnected from the main grid and only relies on local distributed power sources to power local loads. This mode has important application value in some remote areas, independent islands or emergency situations.

[0003] However, AC microgrids in island mode face many challenges, among which short-circuit faults are a serious problem. Short-circuit faults can cause a sharp increase in current, which may damage equipment, affect power supply reliability, and even endanger personnel safety. Therefore, timely and accurate detection of short-circuit faults and locating the fault point are crucial to ensure the safe and stable operation of AC microgrids in island mode.

[0004] Traditional short-circuit fault detection and location methods may have insufficient adaptability when facing AC microgrids in island mode. For example, some methods may not be able to accurately distinguish different types of short-circuit faults, or have limitations when considering factors such as the impact of distributed power sources, different power distribution methods (three-phase three-wire and three-phase four-wire), and transition resistance. In addition, the complex microgrid topology and variable operating conditions also put forward higher requirements for fault detection and location technology. Therefore, a more effective short-circuit fault detection and precise location method and system in AC microgrid island mode is needed to meet the needs of practical applications. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method and system for detecting and accurately locating short-circuit faults in an AC microgrid island mode.

[0006] The technical solution of the present invention is as follows:

[0007] The present invention proposes a short-circuit fault detection and precise positioning method in an AC microgrid island mode, comprising the following steps:

[0008] Step S1, collecting the current signal of the AC microgrid in normal operation in the island mode, and obtaining the maximum value E4 of the wavelet energy spectrum value during normal operation by using a binary wavelet energy spectrum transformation algorithm;

[0009] Step S2: when a short circuit fault occurs, collect the first peak value E4 of the wavelet energy spectrum waveform when the short circuit fault occurs max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ;

[0010] Step S3, judging the short-circuit fault type according to the wavelet energy spectrum value;

[0011] Step S4: calculate the ratio E4 of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation. ratio , R g , S DG , and θ w As the eigenvalue, it is input into the neural network algorithm based on the orthogonal least squares method combined with the radial basis function to locate the location of the short-circuit fault.

[0012] As a preferred implementation, the maximum value E4 of the wavelet energy spectrum value during normal operation is obtained by using a binary wavelet energy spectrum transformation algorithm. The specific calculation formula is as follows:

[0013]

[0014] Where: E(a) is the wavelet energy at scale a; W p (a, b) are the wavelet coefficients at scale a and translation b;

[0015] By integrating the wavelet coefficients at different scales, the maximum value E4 of the wavelet energy spectrum during normal operation is obtained.

[0016] As a preferred implementation, the short circuit fault location is located by combining the orthogonal least squares method with a radial basis function neural network algorithm, wherein:

[0017] The radial basis function neural network algorithm consists of three layers of neurons, including: input layer, hidden layer and output layer; the input layer receives data and is responsible for processing input features; the hidden layer maps the nonlinear problem of the input space to a high-dimensional space through the radial basis function; the output layer is responsible for processing the output of the hidden layer to produce the final prediction or decision result; the radial basis function mapping process is shown in the following formula:

[0018]

[0019] Where: φ i is the basis function of the i-th hidden layer neuron; c i is the center point of the i-th hidden layer neuron; σ iis the center point width of the i-th hidden layer neuron; ‖xc i ‖ is the sample point x and c i The Euclidean distance of

[0020] The output layer and the hidden layer are connected using linear weights. The specific formula is as follows:

[0021]

[0022] Where: z is the actual value of the prediction point; is the predicted output value; i is the number of neurons in the hidden layer; r is the number of samples in the training set; λ i is the output weight of the i-th neuron in the hidden layer; ε is the error between the actual value and the predicted value;

[0023] Orthogonal least squares method is used to optimize the selection of neurons and parameter determination of the hidden layer of the neural network algorithm of radial basis function; the role of orthogonal least squares method is to improve the prediction accuracy of the model by optimizing the selection of appropriate neuron centers and widths; orthogonal least squares method selects the optimal hidden layer neurons and their parameters by minimizing the sum of squared errors to avoid overfitting and accelerate the training process. The objective function is as follows:

[0024]

[0025] Where: N is the number of samples; z k is the actual output value of the kth sample, is the predicted output value of the kth sample.

[0026] As a preferred embodiment, the E4 ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase three-wire system, the distributed power capacity S DG The change of E4 has no effect on the value of ratio ,θ w , R g To locate the short-circuit fault point, the specific positioning formula is as follows:

[0027]

[0028] Where, d is the distance from the short-circuit fault point to the detection point; θ w is the initial phase angle of the fault; E4 ratio R is the ratio of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation; g is the transition resistor.

[0029] As a preferred embodiment, the E4 ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase four-wire system in island mode, S DG The change of E4 will affect the value of ratio ,θ w , R g and S DG To locate the short-circuit fault point, the specific positioning formula is as follows:

[0030]

[0031] Where: S DG It is the distributed power supply capacity in the three-phase four-wire system.

[0032] On the other hand, the present invention also provides a short-circuit fault detection and precise positioning system in an AC microgrid island mode, comprising:

[0033] Run the data acquisition module to collect the distributed power capacity S in normal operation of the AC microgrid island mode DG and current signal, and obtain the maximum value E4 of the wavelet energy spectrum value during normal operation through the binary wavelet energy spectrum transformation algorithm;

[0034] Short-circuit fault data acquisition module, when a short-circuit fault occurs, collects the first peak value E4 of the wavelet energy spectrum waveform during the short-circuit fault max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ;

[0035] A short-circuit fault type judgment module judges the short-circuit fault type according to the wavelet energy spectrum value;

[0036] The short-circuit fault location module calculates the ratio of the first peak value of the fault current wavelet energy spectrum waveform to the maximum value of the wavelet energy spectrum during normal operation. ratio , R g , S DG , and θ w As the eigenvalue, it is input into the neural network algorithm based on the orthogonal least squares method combined with the radial basis function to locate the location of the short-circuit fault.

[0037] As a preferred implementation, the short-circuit fault location module, E4 ratio , R g , SDG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase three-wire system, the distributed power capacity S DG The change of E4 has no effect on the value of ratio ,θ w , R g To locate the short-circuit fault point, the specific positioning formula is as follows:

[0038]

[0039] Where, d is the distance from the short-circuit fault point to the detection point; θ w is the initial phase angle of the fault; E4 ratio R is the ratio of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation; g is the transition resistor.

[0040] As a preferred implementation, the short-circuit fault location module, E4 ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase four-wire system in island mode, S DG The change of E4 will affect the value of ratio ,θ w , R g and S DG To locate the short-circuit fault point, the specific positioning formula is as follows:

[0041]

[0042] Where: S DG It is the distributed power supply capacity in the three-phase four-wire system.

[0043] On the other hand, the present invention also provides a computer-readable medium for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement a short-circuit fault detection and precise positioning method in an AC microgrid island mode as described in any embodiment of the present invention.

[0044] The present invention has the following beneficial effects: 1. Accurate positioning of single and mixed cable short-circuit faults in island mode is achieved through wavelet energy spectrum value and improved OLS-RBF neural network. 2. Transition resistance is introduced as a variable, and the adaptability of distributed source-load switching and different power distribution modes (three-phase three-wire and three-phase four-wire) is considered, which can be applied to all working scenarios in island mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 This is a schematic diagram of the method flow of Example 1;

[0047] Figure 2 This is the structure diagram of the microgrid simulation model. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0050] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0051] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0052] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0053] Embodiment 1:

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clear, the following will be combined with the specific embodiments of the present application and refer to the attached Figure 1 , clearly and completely describe the technical solution of the present invention.

[0055] To solve the problems in the prior art, the present invention provides a method for detecting and accurately locating short-circuit faults in an AC microgrid island mode, comprising the following steps:

[0056] Step S1, collecting the current signal of the AC microgrid in normal operation in the island mode, and obtaining the maximum value E4 of the wavelet energy spectrum value during normal operation by using a binary wavelet energy spectrum transformation algorithm;

[0057] The maximum value E4 of the wavelet energy spectrum value during normal operation is obtained by the binary wavelet energy spectrum transformation algorithm. The specific calculation formula is as follows:

[0058]

[0059] Where: E(a) is the wavelet energy at scale a; W p (a, b) are the wavelet coefficients at scale a and translation b;

[0060] By integrating the wavelet coefficients at different scales, the maximum value E4 of the wavelet energy spectrum during normal operation is obtained.

[0061] like Figure 2 Shown is a simulation model of a microgrid operating in island mode, with the fault points set at 100 m, 200 m, 300 m, 400 m, 500 m, and 600 m from the detection point.

[0062] When a short circuit fault occurs, collect the first peak value E4 of the wavelet energy spectrum waveform during the short circuit fault max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ;

[0063] like Figure 2 Shown is a simulation model of a microgrid operating in island mode, with the fault points set at 100 m, 200 m, 300 m, 400 m, 500 m, and 600 m from the detection point.

[0064] Step S3, judging whether a short-circuit fault occurs and the type of the short-circuit fault by using the wavelet energy spectrum value;

[0065] The wavelet energy spectrum value is used to determine whether a short circuit fault occurs, specifically by comparing whether the wavelet energy spectrum value is greater than the fault threshold. The fault threshold is set to 5 times the maximum value E4 of the wavelet energy spectrum value during normal operation.

[0066] The wavelet energy spectrum value is used to determine whether a short-circuit fault occurs and the type of the short-circuit fault, wherein:

[0067] For a three-phase three-wire system, ground faults and non-ground faults are distinguished by detecting the zero-sequence current. The zero-sequence current calculation formula is as follows:

[0068]

[0069] Where: I0 is the zero-sequence current of the three-phase three-wire system, I a ,I b ,I c They are the three-phase currents A, B and C of the three-phase three-wire system;

[0070] For a three-phase four-wire system, the ground fault and non-ground fault are distinguished by detecting the residual current. The residual current formula is as follows:

[0071]

[0072] Where: I d is the residual current; I a ,I b ,I c ,I n Represent the current vectors of phase a, phase b, phase c and phase n respectively.

[0073] In this embodiment:

[0074] In a three-phase three-wire system, the possibility of a ground fault is first determined by detecting the zero-sequence current. At the same time, the wavelet energy spectrum value is analyzed to assist in the judgment. For a ground fault, since the fault current has a grounding path, it will cause the zero-sequence current to change, and the distribution and size of the wavelet energy spectrum value of the fault phase current in certain frequency bands are different from those of a non-ground fault.

[0075] Single-phase ground short circuit fault characteristics: In a single-phase ground short circuit fault, the wavelet energy spectrum value of the fault phase current may have a higher energy concentration in the low frequency band.

[0076] Characteristics of two-phase short-circuit fault: When a two-phase short-circuit fault occurs, the current changes of the two phases are similar, and their wavelet energy spectrum values ​​will show a certain symmetry in shape, peak value and distribution frequency band.

[0077] Characteristics of three-phase short circuit fault: When a three-phase short circuit fault occurs, the three-phase currents suddenly change at the same time, and their wavelet energy spectrum values ​​may have a higher amplitude and a more uniform distribution as a whole.

[0078] In the three-phase four-wire system, the residual current is detected to preliminarily judge the ground fault situation, and the wavelet energy spectrum value is combined to make a comprehensive judgment.

[0079] Characteristics of three-phase short-circuit fault: When a three-phase short-circuit fault occurs, the wavelet energy spectrum value of the three-phase current has a high amplitude as a whole and the change trend is consistent. The energy values ​​at multiple scales are large, and the difference in wavelet energy spectrum values ​​between different phases is relatively small, because the change of the three-phase current is similar due to the short circuit of the three phases at the same time. Unlike the three-phase three-wire system, due to the existence of the neutral line, a three-phase short-circuit fault under the three-phase four-wire system may have a certain impact on the neutral line current, which will also be reflected in the wavelet energy spectrum value. For example, the wavelet energy spectrum value of the neutral line current may fluctuate to a certain extent, but the overall amplitude is relatively small compared to the three-phase current.

[0080] Characteristics of two-phase short-circuit fault: In the case of a two-phase short-circuit fault, the wavelet energy spectrum values ​​of the faulty two phases are significantly different from those of the three-phase short-circuit fault under the three-phase four-wire system in terms of shape, peak value, and distribution frequency band. The wavelet energy spectrum values ​​of the faulty two phases will have higher peaks at certain scales, but the overall energy distribution is not as extensive as the three-phase short-circuit fault, and the wavelet energy spectrum values ​​of the non-fault phases are relatively low. Similar to the two-phase short-circuit fault under the three-phase three-wire system, there is a certain symmetry in the wavelet energy spectrum values ​​between the faulty two phases, but due to the influence of the neutral line, its specific characteristics may change. For example, in certain energy calculations or feature analyses involving neutral line current, different results will be shown from the three-phase three-wire system.

[0081] Characteristics of single-phase ground short circuit fault: In addition to the characteristics of low-frequency energy concentration of the fault phase similar to the three-phase three-wire system, the single-phase ground short circuit fault is also closely related to the wavelet energy spectrum value of the neutral line current under the three-phase four-wire system. The change of the wavelet energy spectrum value of the fault phase current in a specific frequency band will cause the corresponding change of the wavelet energy spectrum value of the neutral line current. The energy relationship between the two (such as ratio, difference or coordinated change characteristics, etc.) can be used as an important basis for judging the single-phase ground short circuit fault under the three-phase four-wire system. Moreover, due to the shunting effect of the neutral line, the size and change law of the wavelet energy spectrum value of the fault phase current during a single-phase ground short circuit fault may be slightly different from that under the three-phase three-wire system. For example, at the same fault distance, the wavelet energy spectrum value of the fault phase in the three-phase four-wire system may be relatively small, but its energy relationship characteristics with the neutral line current are more obvious.

[0082] Step S4, locating the short circuit fault location by using a neural network algorithm based on orthogonal least squares method combined with radial basis function.

[0083] The short circuit fault location is located by combining the orthogonal least squares method with the radial basis function neural network algorithm, wherein:

[0084] The radial basis function neural network algorithm consists of three layers of neurons, including: input layer, hidden layer and output layer; the input layer receives data and is responsible for processing input features; the hidden layer maps the nonlinear problem of the input space to a high-dimensional space through the radial basis function; the output layer is responsible for processing the output of the hidden layer to produce the final prediction or decision result; the radial basis function mapping process is shown in the following formula:

[0085]

[0086] Where: φ i is the basis function of the i-th hidden layer neuron; c i is the center point of the i-th hidden layer neuron; σ i is the center point width of the i-th hidden layer neuron; ‖xc i ‖ is the sample point x and c i The Euclidean distance of

[0087] The output layer and the hidden layer are connected using linear weights. The specific formula is as follows:

[0088]

[0089] Where: z is the actual value of the prediction point; is the predicted output value; i is the number of neurons in the hidden layer; r is the number of samples in the training set; λ i is the output weight of the i-th neuron in the hidden layer; ε is the error between the actual value and the predicted value;

[0090] Orthogonal least squares method is used to optimize the selection of neurons and parameter determination of the hidden layer of the neural network algorithm of radial basis function; the role of orthogonal least squares method is to improve the prediction accuracy of the model by optimizing the selection of appropriate neuron centers and widths; orthogonal least squares method selects the optimal hidden layer neurons and their parameters by minimizing the sum of squared errors to avoid overfitting and accelerate the training process. The objective function is as follows:

[0091]

[0092] Where: N is the number of samples; z k is the actual output value of the kth sample, is the predicted output value of the kth sample.

[0093] In a three-phase three-wire system, for different initial fault phase angles θ w , the first peak value E4 of the fault current wavelet energy spectrum waveform max1It decreases as the distance from the short-circuit point to the detection point increases. At the same time, the line resistance changes caused by transition resistance, cable material differences, and mixed structure cables must also be considered. Therefore, E4 is defined max The ratio of 1 to the maximum value of the wavelet energy spectrum E4 during normal operation is defined as E4 ratio During fault location, use E4 ratio As a characteristic value, the influence of cable type and material differences can be eliminated, thereby improving the accuracy of fault location. DG The change of has no effect on the E4 value. ratio , fault phase angle θ w , transition resistance R g The neural network can learn the complex relationship between transition resistance and fault location, and accurately locate short-circuit faults. This method takes into account transition resistance, different cable types, and cable structure differences, ensuring accurate fault location results even in complex line environments.

[0094] During operation, the transition resistance R g is the fault site measurement quantity. The fault accurate location method of three-phase three-wire system in island mode is based on θ w 、E4 ratio , R g is a variable, and the distance d from the fault point to the detection point is the objective function. Figure 2 For example, d∈{100,200,300,400,500,600}. To ensure the accuracy of the model, we take θ w ∈{0,30,60,90,120,150}, R g ∈{0.1,0.5,1}. Set up single-phase ground short circuit faults at 100m, 200m, 300m, 400m, 500m and 600m away from the detection point, and record the E detected at the detection point. 4ratio The E values ​​of 6×6×3=108 fault points with single-phase ground short circuit faults are obtained. 4ratio The total number of sample points is 108. 90 sample points are randomly selected as the training set of the OLS-RBF neural network algorithm, and the remaining 18 sample points are the test set. The radial basis function center point c i , output weight λ i And the set center point width is substituted The distance d from the short-circuit point to the detection point and θ are obtained w 、E4 ratio , R g The functional relationship between them is That is, the precise positioning formula for the three-phase three-wire system in island mode.

[0095] The orthogonal least squares method is combined with the radial basis function neural network algorithm to locate the short-circuit fault location. The positioning formula for the short-circuit fault point of the three-phase four-wire system in the island mode is as follows:

[0096]

[0097] Where: S DG It is the distributed power supply capacity in the three-phase four-wire system.

[0098] In a three-phase four-wire system, the distributed power supply capacity S DG The change of E4 will affect the value of E4. 4ratio , R g ,θ w , S DG Realize accurate positioning of short-circuit faults in three-phase four-wire systems in island mode.

[0099] A method for accurately locating faults in a three-phase four-wire system under island mode based on E4 in microgrid ratio , R g ,θ w , S DG is a variable, and the distance d from the fault point to the detection point is the objective function. d∈{100,200,300,400,500,600}. To ensure the accuracy of the model, S DG ∈{80,85,90,95,100,105,110,115,120}, θ w ∈{0,30,60,90,120,150}, R g ∈{0.1,0.5,1}. Set single-phase ground short circuit faults at 100m, 200m, 300m, 400m, 500m and 600m from the detection point, and record the E4 detected at the detection point. ratio The value is 6×9×6×3=972, and each fault point has a single-phase ground short circuit fault E4 ratio The total number of sample points is 972. 900 sample points are randomly selected as the training set of the OLS-RBF neural network algorithm, and the remaining 72 sample points are the test set. The radial basis function center point c i , output weight λ i And the set center point width is substituted The distance d from the short-circuit point to the detection point and E4 are obtained ratio , R g ,θ w , S DG The functional relationship between That is the precise positioning formula.

[0100] Embodiment 2:

[0101] This embodiment provides a short-circuit fault detection and precise positioning system in an AC microgrid island mode, comprising:

[0102] Run the data acquisition module to collect the distributed power capacity S in normal operation of the AC microgrid island mode DG and current signal, and obtain the maximum value E4 of the wavelet energy spectrum value during normal operation through the binary wavelet energy spectrum transformation algorithm;

[0103] Short-circuit fault data acquisition module, when a short-circuit fault occurs, collects the first peak value E4 of the wavelet energy spectrum waveform during the short-circuit fault max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ;

[0104] A short-circuit fault type judgment module judges the short-circuit fault type according to the wavelet energy spectrum value;

[0105] The short-circuit fault location module calculates the ratio of the first peak value of the fault current wavelet energy spectrum waveform to the maximum value of the wavelet energy spectrum during normal operation. ratio , R g , S DG , and θ w As the eigenvalue, it is input into the neural network algorithm based on the orthogonal least squares method combined with the radial basis function to locate the location of the short-circuit fault.

[0106] Embodiment three:

[0107] This embodiment provides an electronic device having a computer program stored thereon. When the computer program is executed by a processor, a method for detecting and accurately locating short-circuit faults in an AC microgrid island mode as described in any embodiment of the present invention is implemented.

[0108] Embodiment 4:

[0109] This embodiment provides a computer-readable medium for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement a short-circuit fault detection and precise positioning method in an AC microgrid island mode as described in any embodiment of the present invention.

[0110] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0111] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0114] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for detecting and accurately locating short-circuit faults in an AC microgrid island mode, characterized in that: The following steps are involved: Step S1, collecting the current signal of the AC microgrid in normal operation in the island mode, and obtaining the maximum value E4 of the wavelet energy spectrum value during normal operation by using a binary wavelet energy spectrum transformation algorithm; Step S2: when a short circuit fault occurs, collect the first peak value E4 of the wavelet energy spectrum waveform when the short circuit fault occurs max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ; Step S3, judging the short-circuit fault type according to the wavelet energy spectrum value; Step S4: calculate the ratio E4 of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation. ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares method combined with radial basis function is input to locate the location of the short circuit fault.

2. The method for detecting and accurately locating short-circuit faults in an AC microgrid island mode according to claim 1, characterized in that: The maximum value E4 of the wavelet energy spectrum value during normal operation is obtained by using the binary wavelet energy spectrum transformation algorithm. The specific calculation formula is as follows: Where: E(a) is the wavelet energy at scale a; W p (a, b) are the wavelet coefficients at scale a and translation b; By integrating the wavelet coefficients at different scales, the maximum value E4 of the wavelet energy spectrum during normal operation is obtained.

3. The method for detecting and accurately locating short-circuit faults in an AC microgrid island mode according to claim 1, characterized in that: The short circuit fault location is located by combining the orthogonal least squares method with the radial basis function neural network algorithm, wherein: The radial basis function neural network algorithm consists of three layers of neurons, including: input layer, hidden layer and output layer; the input layer receives data and is responsible for processing input features; the hidden layer maps the nonlinear problem of the input space to a high-dimensional space through the radial basis function; the output layer is responsible for processing the output of the hidden layer to produce the final prediction or decision result; the radial basis function mapping process is shown in the following formula: Where: φ i is the basis function of the i-th hidden layer neuron; c i is the center point of the i-th hidden layer neuron; σ i is the center point width of the i-th hidden layer neuron; ‖xc i ‖ is the sample point x and c i The Euclidean distance of The output layer and the hidden layer are connected using linear weights. The specific formula is as follows: Where: z is the actual value of the prediction point; is the predicted output value; i is the number of neurons in the hidden layer; r is the number of samples in the training set; λ i is the output weight of the i-th neuron in the hidden layer; ε is the error between the actual value and the predicted value; Orthogonal least squares method is used to optimize the selection of neurons and parameter determination of the hidden layer of the neural network algorithm of radial basis function; the role of orthogonal least squares method is to improve the prediction accuracy of the model by optimizing the selection of appropriate neuron centers and widths; orthogonal least squares method selects the optimal hidden layer neurons and their parameters by minimizing the sum of squared errors to avoid overfitting and accelerate the training process. The objective function is as follows: Where: N is the number of samples; z k is the actual output value of the kth sample, is the predicted output value of the kth sample.

4. The method for detecting and accurately locating short-circuit faults in an AC microgrid island mode according to claim 1, characterized in that: The E4 ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase three-wire system, the distributed power capacity S DG The change of E4 has no effect on the value of ratio ,θ w , R g To locate the short-circuit fault point, the specific positioning formula is as follows: Where, d is the distance from the short-circuit fault point to the detection point; θ w is the initial phase angle of the fault; E4 ratio R is the ratio of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation; g is the transition resistor.

5. The method for detecting and accurately locating short-circuit faults in an AC microgrid island mode according to claim 1, characterized in that: The E4 ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase four-wire system in island mode, S DG The change of E4 will affect the value of ratio ,θ w , R g and S DG To locate the short-circuit fault point, the specific positioning formula is as follows: Where: S DG It is the distributed power supply capacity in the three-phase four-wire system.

6. A short-circuit fault detection and precise positioning system in an AC microgrid island mode, characterized in that: include: Run the data acquisition module to collect the distributed power capacity S in normal operation of the AC microgrid island mode DG and current signal, and obtain the maximum value E4 of the wavelet energy spectrum value during normal operation through the binary wavelet energy spectrum transformation algorithm; Short-circuit fault data acquisition module, when a short-circuit fault occurs, collects the first peak value E4 of the wavelet energy spectrum waveform during the short-circuit fault max1 , initial fault phase angle θ w , Distributed power capacity S DG and transition resistance R g ; A short-circuit fault type judgment module determines the short-circuit fault type according to the wavelet energy spectrum value; The short-circuit fault location module calculates the ratio of the first peak value of the fault current wavelet energy spectrum waveform to the maximum value of the wavelet energy spectrum during normal operation. ratio , R g , S DG , and θ w As the eigenvalue, it is input into the neural network algorithm based on the orthogonal least squares method combined with the radial basis function to locate the location of the short-circuit fault.

7. The short-circuit fault detection and precise positioning system in the AC microgrid island mode according to claim 6 is characterized by: The short-circuit fault location module will ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase three-wire system, the distributed power capacity S DG The change of E4 has no effect on the value of ratio ,θ w , R g To locate the short-circuit fault point, the specific positioning formula is as follows: Where, d is the distance from the short-circuit fault point to the detection point; θ w is the initial phase angle of the fault; E4 ratio R is the ratio of the first peak value of the wavelet energy spectrum waveform of the fault current to the maximum value of the wavelet energy spectrum during normal operation; g is the transition resistor.

8. The short-circuit fault detection and precise positioning system in the AC microgrid island mode according to claim 6 is characterized by: The short-circuit fault location module will ratio , R g , S DG , and θ w As the eigenvalue, the neural network algorithm based on orthogonal least squares combined with radial basis function is input to locate the location of the short-circuit fault. For the three-phase four-wire system in island mode, S DG The change of E4 will affect the value of ratio ,θ w , R g and S DG To locate the short-circuit fault point, the specific positioning formula is as follows: Where: S DG It is the distributed power supply capacity in the three-phase four-wire system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, a short-circuit fault detection and precise positioning method in an AC microgrid island mode is implemented as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a short-circuit fault detection and precise positioning method in an AC microgrid island mode as described in any one of claims 1 to 5 is implemented.

Images