A leakage protection method and system suitable for low-voltage AC-DC hybrid distribution network

Abstract

The application belongs to the technical field of electrical engineering, and discloses a leakage protection method and system suitable for a low-voltage AC-DC hybrid distribution network, which comprises the following steps: real-time detection of a total leakage current signal of the low-voltage AC-DC hybrid distribution network; determination of whether the total leakage current signal is abnormal according to a preset threshold; if the total leakage current signal is not abnormal, return to the next sampling point to extract the total leakage current signal for judgment; if the total leakage current signal is abnormal, extract the optimal feature group of the characteristic index of the total leakage current signal; realize detection of the leakage time; establish a leakage state recognition model based on a deep belief network using the index of the optimal feature group, and determine the leakage state type; and determine whether tripping is needed according to the corresponding safety protection criterion of the leakage state type, thereby avoiding the problem of misoperation of the leakage protection device of the low-voltage AC-DC hybrid distribution network caused by the determination of whether tripping is needed only according to the size of the total leakage current.

Description

Technical Field

[0001] This invention belongs to the field of electrical engineering technology, specifically relating to a leakage current protection method and system suitable for low-voltage AC / DC hybrid distribution networks. Background Technology

[0002] Currently, a large number of non-isolated distributed photovoltaic (PV) systems are connected to low-voltage distribution networks. In these non-isolated AC / DC hybrid distribution networks, a significant common-mode leakage current is generated, which then flows into the AC grid, increasing the difficulty of accurately detecting leakage faults. Currently, AC-type leakage current protection devices are commonly used in low-voltage distribution networks, operating on the principle of determining tripping based on the magnitude of the total leakage current. Operational experience shows that current leakage current protection devices often fail to operate correctly in low-voltage AC / DC hybrid distribution networks, or malfunction due to increased common-mode leakage current in the PV system, reducing their sensitivity and reliability. The main reason is that, in principle, leakage current protection devices cannot distinguish between leakage current from biological electric shock faults and common-mode leakage current from the PV system. Summary of the Invention

[0003] The purpose of this invention is to provide a leakage current protection method and system suitable for low-voltage AC / DC hybrid distribution networks, so as to solve the technical problem that the leakage current protection device in the prior art cannot distinguish between the leakage current of biological electric shock faults and the common mode leakage current of photovoltaic systems, resulting in low sensitivity and reliability of the leakage current protection device.

[0004] To achieve the above objectives, the present invention employs the following technical solution:

[0005] The first invention provides a leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks, comprising:

[0006] Real-time detection of total leakage current signal in low-voltage AC / DC hybrid distribution networks;

[0007] The total leakage current signal is used to determine whether a leakage current abnormality has occurred based on a preset threshold: if the total leakage current signal indicates no leakage current abnormality, the total leakage current signal of the next sampling point is extracted for judgment; if the total leakage current signal indicates a leakage current abnormality, the optimal feature group of the total leakage current signal feature index is extracted.

[0008] The optimal feature set index is used to establish a leakage current state identification model based on the deep belief network model, and the leakage current state type is determined according to the leakage current state identification model.

[0009] The corresponding safety protection criteria are executed based on the leakage current status type.

[0010] Preferably, the step of determining whether the total leakage current signal indicates an abnormal leakage current occurrence based on a preset threshold specifically involves:

[0011] If the total leakage current exceeds the set threshold and the detected value remains above the threshold for three consecutive cycles, it is determined that a leakage current abnormality has occurred at this moment; otherwise, it is determined that no leakage current abnormality has occurred at this moment.

[0012] Preferably, if the total leakage current signal indicates an abnormal leakage current, then the optimal feature set of the total leakage current signal characteristic indicators is extracted, specifically including:

[0013] The original features are obtained by extracting features from the total leakage current signal from the time domain and frequency domain perspectives, respectively.

[0014] The Relief-F algorithm is used to filter the original features and select the optimal feature group.

[0015] Preferably, in the step of extracting features from the total leakage current signal from the time domain and frequency domain respectively to obtain the original features, the features extracted from the time domain are time domain features, specifically including: maximum value, minimum value, average value, median, peak-to-peak value, rectified average value, variance, standard deviation, kurtosis, skewness, root mean square, mean square value, root square amplitude, waveform factor, peak factor, impulse factor, and margin factor.

[0016] Preferably, in the step of extracting features from the total leakage current signal from the time domain and frequency domain respectively to obtain the original features, the features extracted from the frequency domain are frequency domain features, specifically including: centroid frequency, mean square frequency, root mean square frequency, frequency variance and frequency standard deviation.

[0017] Preferably, the leakage current state identification model based on deep belief network is a nonlinear mapping relationship between the total leakage current signal characteristic index and the electric shock of biological body and the leakage current state of photovoltaic system. The leakage current state identification model based on deep belief network includes a visible layer, an RBM network and an output layer.

[0018] Preferably, the number of nodes in the visible layer is the number of indicators in the optimal feature group;

[0019] The number of nodes in the output layer is equal to the number of leakage current state types;

[0020] The RBM network is a two-layer network consisting of a visible layer v and a hidden layer h. The random variable combination expressions for the visible layer and the hidden layer are v = [v1, v2, ..., v...]. n ]、h=[h1,h2L,h m Then, the energy function of the joint state (v,h) of all visible and hidden variables of the RBM is expressed as:

[0021]

[0022] In the formula, a i b j These are the biases for the visible layer and hidden layer variables, respectively; w ij The connection weights for visible and hidden variables;

[0023] RBM networks can obtain the joint probability distribution of training and label data. The joint probability distribution of the random variable P(v,h) is:

[0024]

[0025] Based on the above formula, we get

[0026]

[0027] In the formula, This is the activation function.

[0028] Preferably, the leakage current state type includes biological electric shock state and photovoltaic equipment leakage current state; if it is biological electric shock state, the biological safety protection criterion is executed; if it is photovoltaic equipment leakage current state, the equipment safety protection criterion is executed.

[0029] Preferably, the execution of the biological safety protection criterion specifically includes the following steps:

[0030] Using a biological method for calculating the branch current of electric shock, the branch current I is extracted from the total leakage current in real time. r And determine the current I in the electric shock branch. r The size of the preset value;

[0031] According to the current I in the electric shock branch r The result of comparing the current current with a preset value is used to calculate the duration of the current electric shock within a specified numerical range.

[0032] If the current in the electric shock branch is I r If the duration is less than the preset value, the timer will start recording the duration of the electric shock. If the duration of the electric shock is not less than 3 seconds, the leakage current protection device will trip and issue an alarm signal. If the duration is less than 3 seconds, the leakage current protection device will not trip and will only issue an alarm signal.

[0033] If the current in the electric shock branch is I r If the value is not less than the preset value, the leakage protection device will activate without delay and issue an alarm signal.

[0034] Preferably, the safety protection criterion for the execution device specifically includes the following steps:

[0035] Real-time acquisition of total leakage current I Σ And determine the total leakage current I. Σ size;

[0036] If the total leakage current I Σ When the current is greater than or equal to 10mA but less than 30mA, the leakage protection device will not activate and will only issue an alarm signal.

[0037] If the total leakage current I Σ When the current leakage current is greater than or equal to 30mA but less than 100mA, the duration of the current total leakage current level within the specified range is recorded.

[0038] 1) If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not operate and will only issue an alarm signal;

[0039] 2) If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal.

[0040] If the total leakage current I Σ When the current is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

[0041] The second invention provides a leakage current protection system suitable for low-voltage AC / DC hybrid distribution networks, comprising:

[0042] The detection unit is used to detect the total leakage current signal of the low-voltage AC / DC hybrid distribution network in real time.

[0043] The judgment unit is used to determine whether the total leakage current signal indicates a leakage abnormality based on a preset threshold. If the total leakage current signal indicates no leakage abnormality, it returns to extract the total leakage current signal of the next sampling point for judgment; if the total leakage current signal indicates a leakage abnormality, it extracts the optimal feature group of the total leakage current signal feature indicators.

[0044] The model building and judgment unit is used to build a leakage current state identification model based on a deep belief network model according to the indicators of the optimal feature group, and to determine the leakage current state type according to the leakage current state identification model.

[0045] The execution unit is used to execute the corresponding safety protection criteria according to the leakage current status type.

[0046] Preferably, in the execution unit, the leakage state type includes biological electric shock state and photovoltaic device leakage state; if it is biological electric shock state, the biological safety protection criterion is executed; if it is photovoltaic device leakage state, the device safety protection criterion is executed.

[0047] Preferably, the biological safety protection criteria include: an electric shock current threshold determination module, an electric shock duration timing module, and a leakage current protection device;

[0048] The electric shock current threshold determination module is used to extract the electric shock branch current I from the total leakage current using a biological electric shock branch current calculation method. r And determine the current I in the electric shock branch. r The size of the preset value;

[0049] The electric shock duration statistics module is used to count the duration of the current electric shock current within a specified numerical range;

[0050] The leakage current protection device is used to operate under specified conditions, specifically:

[0051] If the electric shock current threshold determination module determines the electric shock branch current I r If the duration is less than the preset value, the electric shock duration statistics module will work to time the duration of the electric shock. If the duration of the electric shock is greater than or equal to 3 seconds, the leakage current protection device will trip and issue an alarm signal; if the duration is less than 3 seconds, the leakage current protection device will not trip, but will only issue an alarm signal.

[0052] If the electric shock current threshold determination module determines that the electric shock branch current I r If the value is not less than the preset value, the leakage protection device will activate without delay and issue an alarm signal.

[0053] Preferably, the equipment safety protection criteria include: a total leakage current threshold determination module, a duration timing module, and a leakage protection device;

[0054] The total leakage current threshold determination module is used to collect the total leakage current I in real time. Σ And determine the total leakage current I. Σ size;

[0055] The duration statistics module is used to calculate the duration of the current total leakage current level within a specified range based on the judgment result of the total leakage current threshold determination module.

[0056] The leakage current protection device is used to operate under specified conditions, specifically:

[0057] When the total leakage current threshold determination module determines the total leakage current I Σ When the current is greater than or equal to 10mA but less than 30mA, the leakage protection device will not activate and will only issue an alarm signal.

[0058] When the total leakage current threshold determination module determines the total leakage current I Σ When the current leakage current is greater than or equal to 30mA but less than 100mA, the duration statistics module calculates the duration of the current total leakage current level within this range.

[0059] 1) If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not operate and will only issue an alarm signal;

[0060] 2) If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal.

[0061] When the total leakage current threshold determination module determines the total leakage current I Σ When the current is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

[0062] The third invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the leakage current protection method for low-voltage AC / DC hybrid distribution networks described in any of the preceding inventions.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] This invention discloses a leakage current protection method applicable to low-voltage AC / DC hybrid distribution networks. By real-time detection of the total leakage current signal of the low-voltage AC / DC hybrid distribution network, it can quickly respond to potential leakage situations, ensuring that problems are detected at the first moment. Simultaneously, based on the judgment of the total leakage current signal and a preset threshold, the leakage moment is detected. The leakage type is determined by combining multi-feature selection and deep belief networks (DBN). DBN, as a powerful deep learning model, can automatically learn and extract complex features from data, thereby achieving accurate judgment of the leakage type. Furthermore, this application executes corresponding action criteria based on the leakage type to determine whether tripping is necessary, avoiding the problem of false tripping of low-voltage AC / DC hybrid distribution network leakage current protection devices caused by judging tripping solely based on the magnitude of the total leakage current. The method disclosed in this application can not only detect the moment of leakage fault but also quickly identify leakage types in different states, and then execute corresponding protection criteria based on different leakage types, improving the sensitivity and reliability of the leakage current protection device. Attached Figure Description

[0065] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a flowchart of the method of the present invention;

[0067] Figure 2 This is a flowchart of an embodiment of the present invention;

[0068] Figure 3 This is a structural diagram of the DBN-based leakage current state identification model according to an embodiment of the present invention;

[0069] Figure 4 Biological safety protection criterion process of this invention embodiment

[0070] Figure 5 This is the device security protection criterion process in an embodiment of the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0072] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0073] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0074] In the description of the embodiments of the present invention, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0075] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0076] In the description of the embodiments of the present invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention according to the specific circumstances.

[0077] The present invention will now be described in further detail with reference to the accompanying drawings:

[0078] See Figure 1 This application discloses a leakage current protection method applicable to low-voltage AC / DC hybrid distribution networks, comprising:

[0079] S1: Real-time detection of total leakage current signal in low-voltage AC / DC hybrid distribution network;

[0080] S2: Determine whether the total leakage current signal indicates a leakage anomaly based on a preset threshold; the detection of leakage moment is achieved based on the threshold judgment: if the total leakage current signal indicates no leakage anomaly, return to extract the total leakage current signal of the next sampling point for judgment; if the total leakage current signal indicates a leakage anomaly, extract the optimal feature group of the total leakage current signal feature index.

[0081] S3: Using the index of the optimal feature set based on the deep belief network model, a leakage current state identification model based on the deep belief network is established, and the leakage current state type is determined according to the leakage current state identification model; the leakage current type is determined by combining multi-feature selection and deep belief network.

[0082] S4: Execute the corresponding safety protection criteria based on the leakage current condition type. This determines whether tripping is necessary by executing the appropriate action criteria based on the leakage current type, avoiding the problem of malfunctioning leakage current protection devices in low-voltage AC / DC hybrid distribution networks that can occur if tripping is only determined based on the total leakage current magnitude.

[0083] In some embodiments, determining whether the total leakage current signal indicates a leakage anomaly based on a preset threshold specifically involves: if the detected total leakage current exceeds the preset threshold and the detected value remains above the threshold for three consecutive cycles, then a leakage anomaly is determined to have occurred at that moment; otherwise, it is determined that no leakage anomaly has occurred at that moment. The threshold value can be set according to specific circumstances, and determining the timing of a leakage fault based on the preset threshold can identify the exact moment.

[0084] In some embodiments, if the total leakage current signal indicates an abnormal leakage current, then extracting the characteristic indicators of the total leakage current signal specifically includes:

[0085] The original features are obtained by extracting features from the total leakage current signal from the time domain and frequency domain perspectives, respectively.

[0086] The Relief-F algorithm was used to filter the original features and select the optimal feature set. Although the above multi-angle features can completely express the essential characteristics of signals in different leakage current states, some redundant features increase the computational complexity. In order to improve the feature extraction efficiency and the leakage current state identification accuracy, the Relief-F algorithm was further used to filter the original feature space.

[0087] In some embodiments, in the step of extracting features from the total leakage current signal from the time domain and frequency domain to obtain the original features, the features extracted from the time domain are time domain features, as shown in Table 1, which are time domain feature variables and their calculation expressions, specifically including: maximum value, minimum value, average value, median, peak-to-peak value, rectified average value, variance, standard deviation, kurtosis, skewness, root mean square, mean square value, root square amplitude, waveform factor, peak factor, impulse factor, and margin factor.

[0088] Table 1. Time-domain characteristic variables and their calculation expressions

[0089]

[0090]

[0091] In some embodiments, in the step of extracting features from the total leakage current signal from the time domain and frequency domain respectively to obtain the original features, the features extracted from the frequency domain are frequency domain features, as shown in Table 2, specifically including: centroid frequency, mean square frequency, root mean square frequency, frequency variance and frequency standard deviation.

[0092] Table 2 Frequency Domain Characteristic Variables and Their Calculation Expressions

[0093]

[0094] In some embodiments, the leakage current state identification model based on deep belief networks is a nonlinear mapping relationship between the total leakage current signal characteristic index and the electric shock of biological organisms and the leakage current state of photovoltaic systems. The leakage current state identification model based on deep belief networks includes a visible layer, an RBM network, and an output layer.

[0095] More preferably, the number of nodes in the visible layer is the number of indicators in the optimal feature group;

[0096] The number of nodes in the output layer is equal to the number of leakage current state types;

[0097] The RBM network is a two-layer network consisting of a visible layer v and a hidden layer h. The random variable combination expressions for the visible layer and the hidden layer are v = [v1, v2, ..., v...]. n ]、h=[h1,h2L,h m Then, the energy function of the joint state (v,h) of all visible and hidden variables of the RBM is expressed as:

[0098]

[0099] In the formula, a i b j These are the biases for the visible layer and hidden layer variables, respectively; w ij The connection weights for visible and hidden variables;

[0100] RBM networks can obtain the joint probability distribution of training and label data. The joint probability distribution of the random variable P(v,h) is:

[0101]

[0102] Based on the above formula, we can obtain

[0103]

[0104] In the formula, This is the activation function.

[0105] In some embodiments, the leakage current state type includes biological electric shock state and photovoltaic device leakage current state; in the step of executing the corresponding safety protection criterion according to the leakage current state type, if the leakage current state type is biological electric shock state, then the biological safety protection criterion is executed; if the leakage current state type is photovoltaic device leakage current state, then the device safety protection criterion is executed.

[0106] Specifically, the implementation of biological safety protection criteria includes the following steps:

[0107] Using a biological method for calculating the branch current of electric shock, the branch current I is extracted from the total leakage current in real time. r And determine the current I in the electric shock branch. r The size of the preset value;

[0108] According to the current I in the electric shock branch r The result of comparing the current current with a preset value is used to calculate the duration of the current electric shock within a specified numerical range.

[0109] If the current in the electric shock branch is I rIf the duration is less than the preset value, the timer will start recording the duration of the electric shock. If the duration is greater than or equal to 3 seconds, the leakage current protection device will trip and issue an alarm signal. If the duration is less than 3 seconds, the leakage current protection device will not trip but will only issue an alarm signal.

[0110] If the current in the electric shock branch is I r If the value is not less than the preset value, the leakage protection device will activate without delay and issue an alarm signal.

[0111] Specifically, the execution of equipment safety protection criteria includes the following steps:

[0112] Real-time acquisition of total leakage current I Σ And determine the total leakage current I. Σ size;

[0113] If the total leakage current I Σ When the current is greater than or equal to 10mA but less than 30mA, the leakage protection device will not activate and will only issue an alarm signal.

[0114] If the total leakage current I Σ When the current leakage current is greater than or equal to 30mA but less than 100mA, the duration of the current total leakage current level within the specified range is recorded.

[0115] 1) If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not operate and will only issue an alarm signal;

[0116] 2) If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal.

[0117] If the total leakage current I Σ When the current is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

[0118] In some embodiments, a leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks is provided, with... Figure 2 As shown, the specific steps include:

[0119] S1: Real-time detection of total leakage current signal in low-voltage AC / DC hybrid distribution network.

[0120] S2: Determine abnormal changes in leakage current. When the total leakage current exceeds the set threshold (the threshold value can be set according to the specific situation; in this invention, it is set to 10mA), and the discrimination element determines that the detected value is continuously higher than the threshold within 3 cycles, then it is determined that an abnormal leakage current has occurred at this moment; otherwise, it is determined that no abnormal leakage current has occurred at this moment.

[0121] S3: If the determination result of S2 is that no leakage current abnormality has occurred, then return to extract the total leakage current signal of the next sampling point, and repeat S1~S3 to determine the leakage current abnormality based on the latest collected total leakage current sampling value; if it is determined that a leakage current abnormality has occurred, then extract the characteristic indicators of the total leakage current signal. The specific steps are as follows:

[0122] First, features of the total leakage current signal are extracted from both the time and frequency domains. The time domain features and expressions are shown in Table 1, including 17 features: maximum value, minimum value, average value, median, peak-to-peak value, rectified average value, variance, standard deviation, kurtosis, skewness, root mean square (RMS), mean square value, root mean square amplitude, waveform factor, peak factor, impulse factor, and margin factor. The frequency domain features and expressions are shown in Table 2, including 5 features: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0123] While the above multi-angle features can fully express the essential characteristics of signals under different leakage current states, some redundant features increase computational complexity. To improve feature extraction efficiency and leakage current state identification accuracy, the Relief-F algorithm is further used to filter the original feature space. Relief-F is a common filtering feature selection method that calculates weights for each feature based on relevant statistics to select the optimal feature set. The principle of the Relief-F algorithm can be described as follows:

[0124] (1) Randomly select a sample R in the target sample set, and simultaneously search for k nearest neighbor samples in the samples with the same attribute of R to form a set H;

[0125] (2) Search for k nearest neighbor samples in different attribute samples of R to form a set M;

[0126] (3) Calculate the distances between samples in R and H and M respectively, and obtain the weights W of all features based on the distances. A The calculation formula is as follows:

[0127]

[0128] In the formula, W 0(A) Here, n represents the initial weights for feature A; n is the number of random samplings of the sample; Diff(A,R) i H j ) is R i and H j The distance relative to feature A; p(c) is the proportion of class c in the total number of samples in the original dataset; Diff(A,R) i M j (c) is the sample R i and M j(c) Distance relative to feature A; class(Ri) is the class of Ri; M j (c) is a category The j-th nearest neighbor sample.

[0129] (4) Arrange all features in descending order of weight, and extract the first i features whose weights sum to 0.8 to obtain the optimal feature subset.

[0130] S4: Combine deep learning models to obtain leakage current status identification results.

[0131] Based on the Deep Belief Network (DBN) model in deep learning, a nonlinear mapping relationship is constructed between the features selected by the Relief-F above and the electric shock state of biological organisms and the leakage state of photovoltaic systems. A leakage state identification model based on DBN is established. By inputting the leakage current feature data to be measured into the leakage state identification model, the leakage state type can be obtained.

[0132] The structure of the DBN-based leakage current state identification model constructed in this invention is shown in the appendix. Figure 3 As shown, it consists of a visible layer, an RBM network, and an output layer.

[0133] The number of visible layer nodes is the number of features after Relief-F optimization; the number of output layer nodes is the number of two types of leakage states. According to the classification criteria of leakage states, a certain state is represented by 1, and a state that does not meet the criteria is represented by 0. The expected output matrix corresponding to each leakage state is (1,0) for biological electric shock. T The leakage current of the photovoltaic system is (0,1). T .

[0134] An RBM network is a two-layer network consisting of a visible layer v and a hidden layer h. The random variable combination expressions for the visible and hidden layers are v = [v1, v2, ..., v...]. n ]、h=[h1,h2L,h m Then, the energy function of the joint state (v,h) of all visible and hidden variables in the RBM network can be expressed as:

[0135]

[0136] In the formula, a i b j These are the biases for the visible layer and hidden layer variables, respectively; w ij The connection weights for visible and hidden variables.

[0137] RBM networks can obtain the joint probability distribution of training and label data. The joint probability distribution of the random variable P(v,h) is:

[0138]

[0139] Based on the above formula, we can obtain

[0140]

[0141] In the formula, This is the activation function.

[0142] S5: For identification results indicating an electric shock state in a living organism, execute the biological safety protection criterion; for identification results indicating a leakage state in a photovoltaic device, execute the device safety protection criterion.

[0143] The implementation process of the biological safety protection criteria is as follows: Figure 4 As shown, it includes two modules: an electric shock current threshold determination module and an electric shock duration timing module. The function of the electric shock current threshold determination module is to extract the electric shock branch current I from the total leakage current using a biological electric shock branch current calculation method. r The function of the electric shock duration statistics module is to count the duration of the current electric shock current within a specified numerical range.

[0144] Combination Figure 4 The action process for determining the safety of living organisms can be summarized as follows:

[0145] Its action criterion is: real-time calculation of the amplitude of the electric shock current; if the amplitude of the electric shock current I... r If the current is less than 10mA, the electric shock duration statistics module will start working. If the electric shock duration is greater than or equal to 3 seconds, the leakage current protection device will trip and issue an alarm signal; if the duration is less than 3 seconds, the leakage current protection device will not trip, but will only issue an alarm signal. If the electric shock current amplitude I... r If the current is ≥10mA, the leakage protection device will activate without delay and issue an alarm signal.

[0146] The implementation process of the equipment safety protection criteria is as follows: Figure 5 As shown, it includes two main sub-modules: the total leakage current threshold determination module and the duration timing module. The function of the total leakage current threshold determination module is to acquire the total leakage current I. Σ The data is processed and used as the basis for determining the operation of the leakage current protection device; the function of the duration statistics module is to count the duration of the current total leakage current level within the specified value range.

[0147] Combination Figure 5 The action flow of the equipment safety protection criteria can be summarized as follows:

[0148] 1) When the total leakage current I is greater than or equal to 10mA and less than 30mA, the leakage protection device will not operate and will only issue an alarm signal.

[0149] 2) When the total leakage current I is greater than or equal to 30mA but less than 100mA, two situations may occur:

[0150] If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not activate and will only issue an alarm signal.

[0151] If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal.

[0152] 3) When the total leakage current I is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

[0153] This application also discloses a leakage current protection system suitable for low-voltage AC / DC hybrid distribution networks, comprising:

[0154] The detection unit is used to detect the total leakage current signal of the low-voltage AC / DC hybrid distribution network in real time.

[0155] The judgment unit is used to determine whether the total leakage current signal indicates a leakage abnormality based on a preset threshold: if the total leakage current signal indicates no leakage abnormality, it returns to extract the total leakage current signal of the next sampling point for judgment; if the total leakage current signal indicates a leakage abnormality, it extracts the optimal feature group of the feature indicators of the total leakage current signal.

[0156] The model building and judgment unit is used to establish a leakage current state identification model based on the index of the optimal feature group based on the deep belief network model, and to determine the leakage current state type based on the leakage current state identification model.

[0157] The execution unit is used to execute the corresponding safety protection criteria according to the leakage current status type.

[0158] In some embodiments, the leakage state type in the execution unit includes biological electric shock state and photovoltaic device leakage state; if it is biological electric shock state, the biological safety protection criterion is executed; if it is photovoltaic device leakage state, the device safety protection criterion is executed.

[0159] More preferably, the biological safety protection criteria include: an electric shock current threshold determination module, an electric shock duration timing module, and a leakage current protection device;

[0160] The electric shock current threshold determination module is used to extract the electric shock branch current I from the total leakage current using a biological electric shock branch current calculation method. r And determine the current I in the electric shock branch.r The size of the preset value;

[0161] The electric shock duration statistics module is used to count the duration of the current electric shock current within a specified numerical range;

[0162] The leakage current protection device is used to operate under specified conditions, specifically:

[0163] If the electric shock current threshold determination module determines the electric shock branch current I r If the duration is less than the preset value, the electric shock duration statistics module will work to time the duration of the electric shock. If the duration of the electric shock is greater than or equal to 3 seconds, the leakage current protection device will trip and issue an alarm signal; if the duration is less than 3 seconds, the leakage current protection device will not trip, but will only issue an alarm signal.

[0164] If the electric shock current threshold determination module determines that the electric shock branch current I r If the value is not less than the preset value, the leakage protection device will activate without delay and issue an alarm signal.

[0165] More preferably, the equipment safety protection criteria include: a total leakage current threshold determination module, a duration timing module, and a leakage protection device;

[0166] The total leakage current threshold determination module is used to collect the total leakage current I in real time. Σ And determine the total leakage current I. Σ size;

[0167] The duration statistics module is used to calculate the duration of the current total leakage current level within a specified range based on the judgment result of the total leakage current threshold determination module.

[0168] The leakage current protection device is used to operate under specified conditions, specifically:

[0169] When the total leakage current threshold determination module determines the total leakage current I Σ When the current is greater than or equal to 10mA but less than 30mA, the leakage protection device will not activate and will only issue an alarm signal.

[0170] When the total leakage current threshold determination module determines the total leakage current I Σ When the current leakage current is greater than or equal to 30mA but less than 100mA, the duration statistics module calculates the duration of the current total leakage current level within this range.

[0171] 1) If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not operate and will only issue an alarm signal;

[0172] 2) If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal.

[0173] When the total leakage current threshold determination module determines the total leakage current I Σ When the current is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

[0174] This application also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the leakage current protection method for low-voltage AC / DC hybrid distribution networks described in any of the preceding claims.

[0175] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0176] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0177] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0178] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks, characterized in that, include: Real-time detection of total leakage current signal in low-voltage AC / DC hybrid distribution networks; The total leakage current signal is used to determine whether a leakage current abnormality has occurred based on a preset threshold: if the total leakage current signal indicates no leakage current abnormality, the total leakage current signal of the next sampling point is extracted for judgment; if the total leakage current signal indicates a leakage current abnormality, the optimal feature group of the total leakage current signal feature index is extracted. The optimal feature set index is used to establish a leakage current state identification model based on the deep belief network model, and the leakage current state type is determined according to the leakage current state identification model. Execute the corresponding safety protection criteria based on the leakage current status type; The leakage current state identification model based on deep belief network is a nonlinear mapping relationship between the total leakage current signal characteristic index and the electric shock of biological body and the leakage current state of photovoltaic system. The leakage current state identification model based on deep belief network includes a visible layer, an RBM network and an output layer. The number of nodes in the visible layer is the number of indicators in the optimal feature group; The number of nodes in the output layer is equal to the number of leakage current state types; The RBM network is a network consisting of a view layer. v and hidden layers h The combined random variable expressions for the visible and hidden layers of the two-layer network are as follows: , The joint state of all visible and hidden variables in an RBM ( v , h The energy function of is expressed as: In the formula, These are the biases for the visible layer and hidden layer variables, respectively; The connection weights for visible and hidden variables; RBM networks can obtain the joint probability distribution of training data and label data, random variables. The joint probability distribution is: Based on the energy function and the joint probability distribution, we get In the formula, This is the activation function.

2. The leakage current protection method applicable to low-voltage AC / DC hybrid distribution networks according to claim 1, characterized in that, The specific steps for determining whether the total leakage current signal indicates an abnormal leakage current occurrence based on a preset threshold are as follows: If the total leakage current exceeds the set threshold and the detected value remains above the threshold for three consecutive cycles, it is determined that a leakage current abnormality has occurred at this moment; otherwise, it is determined that no leakage current abnormality has occurred at this moment.

3. The leakage current protection method applicable to low-voltage AC / DC hybrid distribution networks according to claim 1, characterized in that, If the total leakage current signal indicates an abnormal leakage current, then the optimal feature set of the total leakage current signal characteristic indicators is extracted, specifically including: The original features are obtained by extracting features from the total leakage current signal from the time domain and frequency domain perspectives, respectively. The Relief-F algorithm is used to filter the original features and select the optimal feature group.

4. A leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks according to claim 1, characterized in that, The leakage current status types include biological electric shock status and photovoltaic equipment leakage current status; in the step of executing the corresponding safety protection criteria according to the leakage current status type, if the leakage current status type is biological electric shock status, then the biological safety protection criteria are executed; if the leakage current status type is photovoltaic equipment leakage current status, then the equipment safety protection criteria are executed.

5. A leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks according to claim 4, characterized in that, The implementation of the biological safety protection criteria specifically includes the following steps: Using a biological method for calculating the branch current of electric shock, the branch current of electric shock is extracted from the total leakage current in real time. And determine the current in the branch circuit affected by the electric shock. The size of the preset value; Based on the current of the electric shock branch The result of comparing the current current with a preset value is used to calculate the duration of the current electric shock within a specified numerical range. If the current in the branch of electric shock If the duration is less than the preset value, the timer will start recording the duration of the electric shock. If the duration of the electric shock is not less than 3 seconds, the leakage current protection device will trip and issue an alarm signal. If the duration is less than 3 seconds, the leakage current protection device will not trip and will only issue an alarm signal. If the current in the branch of electric shock If the value is not less than the preset value, the leakage protection device will activate without delay and issue an alarm signal.

6. A leakage current protection method suitable for low-voltage AC / DC hybrid distribution networks according to claim 4, characterized in that, The security protection criteria for the execution device specifically include the following steps: Real-time acquisition of total leakage current And determine the total leakage current. size; If the total leakage current When the current is greater than or equal to 10mA but less than 30mA, the leakage protection device will not activate and will only issue an alarm signal. If the total leakage current When the current leakage current is greater than or equal to 30mA but less than 100mA, the duration of the current total leakage current level within the specified range is recorded. 1) If a temporary increase in leakage current occurs and the duration is less than 2 hours, the leakage protection device will not operate and will only issue an alarm signal; 2) If the leakage current increases continuously and lasts for more than 2 hours, the leakage protection device will simultaneously issue a trip command and an alarm signal; If the total leakage current When the current is greater than or equal to 100mA, the leakage protection device will immediately trip and issue an alarm signal.

7. A leakage current protection system suitable for low-voltage AC / DC hybrid distribution networks, characterized in that, The steps for implementing the leakage current protection method for low-voltage AC / DC hybrid distribution networks according to any one of claims 1-6 include: The detection unit is used to detect the total leakage current signal of the low-voltage AC / DC hybrid distribution network in real time. The judgment unit is used to determine whether the total leakage current signal indicates a leakage abnormality based on a preset threshold. If the total leakage current signal indicates no leakage abnormality, it returns to extract the total leakage current signal of the next sampling point for judgment; if the total leakage current signal indicates a leakage abnormality, it extracts the optimal feature group of the total leakage current signal feature indicators. The model building and judgment unit is used to build a leakage current state identification model based on a deep belief network model according to the indicators of the optimal feature group, and to determine the leakage current state type according to the leakage current state identification model. The execution unit is used to execute the corresponding safety protection criteria according to the leakage current status type.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the leakage current protection method for a low-voltage AC / DC hybrid distribution network as described in any one of claims 1-6.

Images