Alternating current and direct current power distribution network electric shock protection action judgment method and system

Abstract

The invention relates to the technical field of AC / DC power distribution network electric shock protection action, in particular to an AC / DC power distribution network electric shock protection action judgment method, system and device and a storage medium, and the method comprises the steps: obtaining a residual current signal of an AC / DC power distribution network electric shock state, decomposing the residual current signal of the electric shock state, and building an electric shock current regression extraction model; and extracting the electric shock branch current in real time by using the electric shock current regression extraction model, and making an electric shock protection action judgment. According to the method, the electric shock current component is accurately extracted from the complex residual current waveform, then the severity of a contact accident is judged according to the extracted contact current component, accurate and effective electric shock action protection is carried out, RCD protection misoperation caused by interference factors and refusal action in an electric shock state can be effectively prevented, and the reliability of the RCD protection is improved. And the safety of the AC-DC power grid is further improved. The problem that an existing traditional RCD based on current threshold detection has maloperation in a low-voltage power distribution network with distributed photovoltaic access is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of leakage protection operation of AC-DC distribution networks, and specifically provides a method, system, device and storage medium for judging the electric shock protection operation of AC-DC distribution networks, especially a criterion for judging the electric shock protection operation on the user side of a low-voltage distribution network with distributed photovoltaic access, as well as a method, system, device and storage medium for extracting electric shock current. Background Art

[0002] A Residual Current Device (RCD), also known as a leakage protection device or residual current protection device, is a safety protection device used in power systems and electrical equipment. Its main function is to detect and disconnect the residual current in the circuit to prevent electric shock injuries and fire accidents. As a technical means to prevent safety accidents such as personal electric shock and equipment leakage, it is widely used in low-voltage distribution networks. When a human electric shock accident occurs in a traditional low-voltage distribution network without distributed photovoltaic, the electric shock current is a power frequency current, and the traditional RCD operates according to the effective value of the residual current.

[0003] However, in recent years, with the large-scale access of distributed photovoltaic to low-voltage distribution networks, the photovoltaic leakage current flows into the AC-side power supply circuit. This change not only adds a large number of high-order harmonics to the power frequency residual current when an electric shock occurs, resulting in the residual current containing not only power frequency components but also a large number of high-order harmonic components; at the same time, due to the existence of high-order harmonics, the effective value of the residual current (such as the RMS value, Root Mean Square) will also increase to varying degrees. The effective value is a physical quantity that measures the thermal effect of the current, and it takes into account the energy contributions of all frequency components in the current waveform. When the effective value increases significantly, the harm degree of the electric shock accident will also increase accordingly. In addition, the unbalance of the three-phase lines and the existence of line-to-ground parasitic parameters in the low-voltage distribution network will also cause a significant increase in the residual current of the line. Therefore, when an electric shock accident occurs in an AC-DC distribution network, the traditional RCD that uses the size of the effective value of the residual current as the criterion for judging the electric shock protection operation can no longer meet the technical requirements such as the reliability and selectivity of the protection device.

[0004] Therefore, in a low-voltage AC-DC distribution network with large-scale access of distributed photovoltaic, studying the method for extracting the electric shock current when an electric shock accident occurs and proposing a new criterion for judging the residual current protection operation of the AC-DC distribution network are of great significance for ensuring the correct operation of the RCD and the safe and reliable operation of the low-voltage distribution network. Summary of the Invention

[0005] Aiming at the problem of misoperation of the traditional RCD based on current threshold detection in a low-voltage distribution network with distributed photovoltaic access in the prior art, the present invention provides a method, system, device and storage medium for judging the electric shock protection operation of AC-DC distribution networks.

[0006] To achieve the above object, the present invention is implemented by the following technical solutions: The present invention provides a method for judging the operation of electric shock protection in an AC-DC distribution network, including: Obtaining a real-time residual current signal of the electric shock state of the AC distribution network; Using a pre-established electric shock current regression extraction model to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time; Making a judgment on the operation of electric shock protection according to the real-time extracted electric shock branch current; The establishment process of the electric shock current regression extraction model is as follows: obtaining the residual current signal of the electric shock state of the AC-DC distribution network, and decomposing the residual current signal of the electric shock state to obtain modal components in different frequency bands; Using the modal components in different frequency bands to establish an electric shock current regression extraction model.

[0007] Further, the method for obtaining the residual current signal of the electric shock state of the AC-DC distribution network and decomposing the residual current signal of the electric shock state to obtain modal components in different frequency bands is as follows: Obtaining the residual current signal of the electric shock state of the AC-DC low-voltage distribution network with distributed photovoltaic access; Using the variational mode decomposition method to decompose the residual current signal of the electric shock state to obtain modal components in different frequency bands.

[0008] Further, the method for using the variational mode decomposition method to decompose the residual current signal of the electric shock state to obtain modal components in different frequency bands is as follows: Decomposing the residual current signal into k IMF components; Performing Hilbert transform on each IMF component, modulating the spectrum of each IMF component to the corresponding base frequency band, and evaluating the bandwidth of each IMF component; Under the constraint condition that the sum of the IMF components is equal to the original residual current signal, constructing an extreme value problem to minimize the sum of the bandwidths of all IMF components, introducing a penalty parameter and a Lagrange multiplier, and transforming the constrained variational problem into an unconstrained variational problem; Using the alternating direction multiplier algorithm to perform iteration to solve the unconstrained variational problem until the convergence condition is satisfied, and outputting k IMF components.

[0009] Further, the method for using the modal components in different frequency bands to establish an electric shock current regression extraction model is as follows: Taking the modal components in different frequency bands as the training set and test set for model establishment, defining the loss function and penalty function of the XGBoost model, and initializing each hyperparameter of the XGBoost model; Construct grid nodes according to the parameter combination to be optimized, and set the search range and search step to be searched; Traverse the grid nodes from different directions, use the parameter combination to be optimized to train the XGBoost model, and obtain the predicted value and actual value of the samples corresponding to each group of parameter combinations to be optimized; Select a group of parameter combinations to be optimized with the smallest root mean square error between the predicted value and the actual value of the corresponding sample as the current optimal solution; Judge whether the current optimal solution reaches the threshold; if so, output the current parameter combination as the optimal parameter combination; if not, re-determine the search range and search step, and perform iteration until the current optimal solution reaches the threshold.

[0010] Further, the method for making an electric shock protection action judgment according to the electric shock branch current extracted in real time is as follows: Obtain the real-time electric shock branch current and calculate the effective value; Make an electric shock protection action judgment according to the effective value.

[0011] Further, the method for obtaining the real-time electric shock branch current and calculating the effective value I r is as follows:

[0012] wherein, I r is the effective value, k is the number of sampling points of the biological electric shock current signal; is the i th sampling value.

[0013] Further, the method for making an electric shock protection action judgment according to the effective value is as follows:

[0014] wherein, I r is the effective value, is the duration.

[0015] The present invention provides an AC / DC distribution network electric shock protection action judgment system, including: Residual current signal acquisition module: used to acquire the real-time residual current signal of the electric shock state of the AC distribution network; Real-time extraction module for electric shock branch current: used to use a pre-established electric shock current regression extraction model to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time; Electric shock protection action judgment module: used to make an electric shock protection action judgment according to the electric shock branch current extracted in real time; The establishment process of the electric shock current regression extraction model is as follows: Obtain the residual current signal of the electric shock state of the AC / DC distribution network, and decompose the residual current signal of the electric shock state to obtain modal components in different frequency bands; Use the modal components in different frequency bands to establish an electric shock current regression extraction model.

[0016] A terminal device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0017] A computer-readable storage medium stores a computer program, characterized in that when the computer program is executed by a processor, the steps of the above method are implemented.

[0018] Compared with the prior art, the present invention has the following beneficial effects: A method for judging the electric shock protection action of an AC / DC distribution network according to the present invention. This method obtains the residual current signal of the electric shock state of the AC / DC distribution network, and decomposes the residual current signal of the electric shock state to obtain modal components in different frequency bands; then, uses the modal components in different frequency bands to establish an electric shock current regression extraction model; finally, inputs the residual current signal of the real-time electric shock state of the AC / DC distribution network into the pre-established electric shock current regression extraction model to extract the electric shock branch current in real time and make a judgment on the electric shock protection action. This method accurately extracts the electric shock current component from the complex residual current waveform, and then judges the severity of the electric shock accident according to the extracted contact current component, and performs precise and effective electric shock action protection, which can effectively prevent the misoperation of the RCD protection caused by interference factors and the refusal to operate under the electric shock state, and further improve the safety of the AC / DC power grid.

[0019] The present invention also provides a system for judging the electric shock protection action of an AC / DC distribution network. Through the settings of the residual current signal acquisition module, the real-time extraction module of the electric shock branch current, and the electric shock protection action judgment module in the system, the acquisition of the residual current signal of the real-time electric shock state of the AC / DC distribution network, the real-time extraction of the electric shock branch current, and the electric shock protection action judgment process are realized. The system is sensitive in response, and can realize precise electric shock protection actions, with higher applicability and safety, and can better meet the safety requirements of the continuously developing power system and electrical equipment.

[0020] The present invention also provides a terminal device, including 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 computer program, the steps of the above method are implemented. The device has a simple structure, low transformation cost, and small resource occupancy.

[0021] A computer-readable storage medium stores a computer program, characterized in that when the computer program is executed by a processor, the steps of the above method are implemented. This storage medium has good portability and strong versatility. Description of the Drawings

[0022] Figure 1 It is a schematic flowchart of a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0023] Figure 2 It is a comparison diagram of the waveforms of the residual current and the current signal of the electric shock branch.

[0024] Figure 3 It is a flowchart of the VMD algorithm in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0025] Figure 4 shows the components of each frequency band of the residual current after VMD decomposition in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention, where a is the components of each frequency band of the residual current obtained after VMD decomposition; b is a comparison diagram of the IMF1 component and the current of the organism's electric shock.

[0026] Figure 5 It is a flowchart of the GSCV-XGBoost modeling method in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0027] Figure 6 It is the RMSE value of the optimal model after 10-fold cross-validation in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0028] Figure 7 It is a comparison diagram of the electric shock current extracted by the XGBoost model, the true electric shock current, and the low-frequency component of the residual current in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0029] Figure 8 It is a regression fitting line diagram of the electric shock current extracted by the XGBoost model and the true electric shock current in a method for judging the action of electric shock protection in an AC-DC distribution network according to the present invention.

[0030] Figure 9 It is a diagram for dividing the time / current region of the organism's current effect.

[0031] Figure 10 It is a detailed flowchart of the method for judging the action of electric shock protection in an AC-DC distribution network in an embodiment of the present invention.

[0032] Figure 11 It is a structural diagram of the system for judging the action of electric shock protection in an AC-DC distribution network according to the present invention. Detailed Embodiments

[0033] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0034] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0035] The present invention will be further described in detail below in conjunction with specific embodiments, which is an explanation rather than a limitation of the present invention.

[0036] Embodiment 1 See Figure 1 , the present invention discloses a method for judging the operation of electric shock protection in an AC / DC distribution network, including: S1: Obtain the real-time residual current signal of the electric shock state of the AC distribution network; S2: Use the pre-established electric shock current regression extraction model to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time; S3: Make a judgment on the operation of electric shock protection according to the electric shock branch current extracted in real time; The establishment process of the electric shock current regression extraction model is as follows: Obtain the residual current signal of the electric shock state of the AC / DC distribution network, and decompose the residual current signal of the electric shock state to obtain modal components of different frequency bands; Use the modal components of different frequency bands to establish an electric shock current regression extraction model.

[0037] This method accurately extracts the electric shock current component from the complex residual current waveform, then determines the severity of the electric shock accident based on the extracted contact current component, and performs precise and effective electric shock protection actions. It can effectively prevent the misoperation of the RCD protection caused by interference factors and the refusal to operate under the electric shock state, further improving the safety of AC and DC power grids.

[0038] Embodiment 2 See Figure 1 , the present invention discloses a method for judging the electric shock protection action of an AC-DC distribution network, including: S1: Obtain the real-time residual current signal of the electric shock state of the AC distribution network: S2: Use the pre-established electric shock current regression extraction model to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time; S3: Make a judgment on the electric shock protection action according to the electric shock branch current extracted in real time; The establishment process of the electric shock current regression extraction model is as follows: Obtain the residual current signal of the electric shock state of the AC-DC distribution network, and decompose the residual current signal of the electric shock state to obtain modal components of different frequency bands; Use the modal components of different frequency bands to establish an electric shock current regression extraction model.

[0039] The specific method for obtaining the residual current signal of the electric shock state of the AC-DC distribution network and decomposing the residual current signal of the electric shock state to obtain modal components of different frequency bands is as follows: Obtain the residual current signal in the electric shock state of the AC-DC low-voltage distribution network with distributed photovoltaic access: Collect the waveform of the residual current data in the electric shock accident state of the AC-DC low-voltage distribution network with distributed photovoltaic access, and compare the waveform of the residual current with the waveform of the electric shock branch current signal. See Figure 2 , it can be seen that the residual current signal is quite different from the electric shock current waveform, and the collected residual current signal cannot reflect the degree of electric shock, which may cause misoperation and refusal to operate of the protection device. To accurately reflect the electric shock situation, it is necessary to obtain the electric shock current from the residual current; Use the Variational Mode Decomposition (VMD) method to decompose the residual current signal in the electric shock state to obtain modal components of different frequency bands. The VMD decomposition process is as follows: Decompose the residual current signal into k IMF components. The method is:

[0040] Among them, is the residual current signal, is the modal function of the IMF component,i is the number of IMF components.

[0041] Perform Hilbert transform on each IMF component, modulate the spectrum of each IMF component to the corresponding base frequency band, and evaluate the bandwidth of each IMF component. The method is as follows: To obtain the unilateral spectrum of the mode, by performing Hilbert transform on each, the analytical signal of each mode is obtained:

[0042] where is the impulse function, t is the time, and j is the imaginary unit; Multiply equation (2) by , and modulate the spectrum of each to the corresponding base frequency band, where is the corresponding center frequency:

[0043] Calculate the square of the envelope of the above corresponding demodulated signal and evaluate the width of the mode; Under the constraint condition that the sum of IMF components is equal to the original residual current signal, construct an extremum problem to minimize the sum of the bandwidths of all IMF components, and introduce a penalty parameter and Lagrange multipliers to transform the constrained variational problem into an unconstrained variational problem. Specifically: Under the constraint condition, with the minimum sum of the bandwidths of each mode as the optimal objective, construct an extremum problem:

[0044] where is the original residual current signal, is the original signal decomposed into the K th intrinsic mode function.

[0045] To solve the above optimal problem, introduce the augmented Lagrangian function, introduce the penalty parameter and the Lagrange multiplier , and finally transform the constrained variational problem into the form of an unconstrained variational problem:

[0046] Use the alternating direction multiplier algorithm to perform iteration to solve the unconstrained variational problem until the convergence condition is satisfied, and output k IMF components, that is: The VMD solution uses the alternating direction multiplier algorithm to solve the above variational problem. It converts each modal component in the time domain into a component in the frequency domain, and continuously updates the center frequency and bandwidth of each IMF component in turn to obtain the center point of the above formula. Modal component and the center frequency The iterative update formulas are as follows:

[0047] Among them, represents the center frequency corresponding to the power spectrum of the modal function at this time; is the inverse Fourier transform; See Figure 3 , the iterative operation steps of VMD are: Initialize , , and n , n is the number of iterations, , and respectively represent the initial values of the modal functions of each IMF component, the initial values of the center frequencies, and the initial values of the Lagrange multipliers Initial values; n = n + 1, update and , and execute the entire loop; Define the fidelity coefficient , and update the Lagrange multiplier according to the following formula

[0048]

[0049] Set the precision , and the given discrimination formula is:

[0050] If the above formula is satisfied, the iteration terminates and K IMF components are output; otherwise, continue to loop until the maximum number of iterations is reached.

[0051] In the electric shock state, the components of each frequency band of the residual current obtained after VMD decomposition are shown in Figure 4a . Among them, the IMF1 component includes the electric shock current component of the organism, the line-to-ground leakage current component, and the low-frequency component in the PV leakage current. The comparison between it and the electric shock current of the organism is shown in Figure 4b . As can be seen from Figure 4, the IMF1 component in the low-frequency band of the residual current has a higher coupling degree with the original residual current signal and the actual electric shock branch signal, but both the residual current and its low-frequency component have a large difference from the actual electric shock current and cannot reflect the actual electric shock degree.

[0052] The specific method for establishing the regression extraction model of electric shock current by using modal components of different frequency bands is as follows: Refer to Figure 5 , establish a regression extraction model of electric shock current by optimizing Extreme Gradient Boosting (XGBoost) based on Grid Search & Cross Validation (GSCV). Use the modal components of different frequency bands as the training set and test set for model establishment, define the loss function and penalty function of the XGBoost model, and initialize each hyperparameter of the XGBoost model, that is, the acquisition and preprocessing of the sample data set: To reduce the influence of data non-stationarity on model establishment and improve the extraction accuracy, obtain the low-frequency component of the residual current with a higher coupling degree with the electric shock current through S1, and use this low-frequency component as the training set and test set for model establishment. The training set and test set are 100 groups and 20 groups respectively; define the loss function and penalty function of the XGBoost model, and initialize each hyperparameter of the XGBoost model; When establishing a regression extraction model of electric shock current based on XGBoost, it is necessary to optimize the initial hyperparameters of the model to find the optimal parameter combination. The optimized hyperparameters selected in the present invention include the maximum depth of the tree, the learning rate, the minimum weight of the leaf node, and the penalty coefficient. The names of each hyperparameter are, max_depth, learning_rate, min_child_weigh, and gamma respectively. Use GridSearchCV (GSCV) to optimize each hyperparameter, and its specific steps are as follows: Construct grid nodes according to the parameter combination to be optimized, and set the search range and search step size to be searched; Traverse the grid nodes in different directions, use the parameter combination to be optimized to train the XGBoost model, and obtain the predicted value and actual value of the sample corresponding to each group of parameter combinations to be optimized: Divide the training set into 10 subsets equally, use any one subset as the test set, and the remaining 9 as the training set to establish a model. Loop 10 times and calculate the results, and obtain the predicted value and actual value of the sample corresponding to each group of parameters.

[0053] Set the root mean squared error (RMSE) of the sample predicted value and actual value as the evaluation index, and select the group with the smallest RMSE of the XGBoost model under each parameter combination as the current optimal solution.

[0054] Judge whether the RMSE reaches the threshold. If it does, output the current parameter combination as the optimal solution; otherwise, adjust the step size and range, and traverse the grid nodes again in different directions.

[0055] After obtaining the optimal parameter combination of the model, use the test set to test the electric shock current regression analysis model. Take the low-frequency component of the residual current as the input to obtain the regression analysis value of the electric shock current. Compare the extracted electric shock current with the actual electric shock branch current, and use the RMSE and the coefficient of determination (R Squared, R2) as the indicators to evaluate the performance of the model for the regression extraction of the electric shock current. The RMSE value of the optimal model after 10-fold cross-validation is shown in Figure 6 。

[0056] The specific method for using the electric shock current regression extraction model in S2 to extract the electric shock branch current in real time is as follows: After obtaining the low-frequency component of the residual current in the real-time electric shock state through S1, input its time series into the pre-established electric shock current regression analysis model to extract the electric shock current component in the residual current. See Figure 7 For the comparison of the electric shock current extracted by the XGBoost model with the real electric shock current and the low-frequency component of the residual current, see Figure 8 For the regression fitting line of the electric shock current extracted by the XGBoost model and the real electric shock current, from Figure 7 and Figure 8 It can be seen that the biological electric shock current extracted by the XGBoost electric shock current regression model is in good fit with the real value signal, and the waveform is basically the same as the real waveform.

[0057] The specific method for making an electric shock protection action judgment according to the real-time extracted electric shock branch current in S3 is as follows: First is the design of the protection action boundary. When a biological electric shock accident occurs on the user side of the low-voltage distribution network, the degree of harm suffered by the organism depends on the magnitude of the current flowing through the organism and the duration of the electric shock. The longer the electric shock time and the larger the electric shock current, the physiological reaction of the organism will gradually progress from fright, involuntary muscle contraction, difficulty breathing, rising blood pressure to ventricular fibrillation, and the harm caused by the electric shock to the organism will become greater and greater. See Figure 9 For the current-time effect of the biological current from the left hand to both feet given by the GB / T 13870.0-2022 / IEC 60479-1:2018 standard, which is applicable to alternating current with a frequency of 15 - 100 Hz. Regarding Figure 9 In

[0058] The present invention divides the electric shock current into several ranges according to the harm of the current to the organism. First, it is 0 - 5 mA. In this range, the organism can freely break away from the electric shock current and it will not cause harm to the organism, and the residual current protection does not need to act. Secondly, it is 5 - 30 mA. When the electric shock current is in this range, the organism will not have irreversible organ damage or ventricular fibrillation. However, if the electric shock time is relatively long, there may be a phenomenon of being unable to break away, and the residual current protection can trip with a time delay. When the electric shock current is above 30 mA, the organism begins to show pathological physiological phenomena, and the residual current protection should act immediately.

[0059] Therefore, the present invention takes 5 mA and 30 mA of the effective value of the organism's electric shock current as the threshold values of the criterion for organism electric shock safety protection. Therefore, the specific setting of making an electric shock protection action judgment according to the electric shock branch current extracted in real time is as follows: Obtain the real-time electric shock branch current and calculate the effective value:

[0060] Wherein, I r is the effective value, k is the number of sampling points of the organism's electric shock current signal; is the i th sampling value; According to the effective value, make an electric shock protection action judgment:

[0061] Wherein, I r is the effective value, is the duration.

[0062] See Figure 10 , the specific steps of the implementation strategy for organism electric shock safety protection are as follows: 1) After the leakage type identification module determines that it is an organism electric shock fault, start to execute the organism safety protection criterion, and extract the organism electric shock branch current from the low-frequency component of the residual current I r ; 2) When the electric shock current I r is less than 5 mA, the protection device does not alarm and does not act, and maintains the normal detection state.

[0063] 3) When the electric shock current I r reaches 5 mA and is less than 30 mA, the protection starts timing and issues an alarm signal. If the timing duration does not exceed 1 s, the protection does not act; if the timing time reaches 1 s, the protection trips instantaneously.

[0064] 4) When the electric shock current I r reaches 30 mA, the protection immediately operates and issues an alarm signal.

[0065] See Figure 11 , the present invention provides a system for judging the operation of electric shock protection in an AC / DC distribution network, including: Residual current signal acquisition module: used to acquire the real-time residual current signal of the electric shock state in the AC distribution network; Real-time extraction module for the electric shock branch current: used to use a pre-established regression extraction model for the electric shock current to extract the electric shock branch current of the real-time residual current signal of the electric shock state in the AC distribution network in real time; Electric shock protection operation judgment module: used to make a judgment on the electric shock protection operation according to the real-time extracted electric shock branch current; The establishment process of the electric shock current regression extraction model is as follows: acquire the residual current signal of the electric shock state in the AC / DC distribution network, and decompose the residual current signal of the electric shock state to obtain modal components in different frequency bands; Use the modal components in different frequency bands to establish an electric shock current regression extraction model.

[0066] The present invention provides a terminal device including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0067] The computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention.

[0068] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0069] The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0070] The memory can be used to store the computer program and / or module(s). By running or executing the computer program and / or module(s) stored in the memory, and by invoking the data stored in the memory, the processor implements various functions of the terminal device.

[0071] If the modules / units integrated in the terminal device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0072] In summary, the present invention provides a method, system, device, and storage medium for judging the operation of electric shock protection in an AC / DC distribution network. By accurately extracting the electric shock current component from a complex residual current waveform and then judging the severity of the electric shock accident based on the extracted contact current component, precise and effective electric shock operation protection can be achieved, effectively preventing misoperation of the RCD protection caused by interference factors and refusal to operate under the electric shock state, and further improving the safety of the AC / DC power grid.

[0073] The above are only the preferred embodiments of the present invention and are not intended to limit the technical solutions of the present invention. Those skilled in the art should understand that, without departing from the spirit and principle of the present invention, the technical solutions can be subject to several simple modifications and substitutions, and these modifications and substitutions also fall within the protection scope covered by the claims.

Claims

1. A method for determining the action of electric shock protection in an AC / DC distribution network, characterized in that: include: Obtain real-time residual current signal of electric shock status in AC distribution network; The pre-established electric shock current regression extraction model is used to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time; Make an electric shock protection action judgment based on the electric shock branch current extracted in real time; The establishment process of the electric shock current regression extraction model is as follows: obtaining the residual current signal of the electric shock state of the AC / DC power distribution network, and decomposing the residual current signal of the electric shock state to obtain modal components of different frequency bands; A regression extraction model for electric shock current is established using modal components in different frequency bands.

2. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 1, characterized in that: The method of obtaining the residual current signal of the electric shock state of the AC / DC power distribution network and decomposing the residual current signal of the electric shock state to obtain modal components of different frequency bands is: Obtain the residual current signal of the AC / DC low-voltage distribution network with distributed photovoltaic access in the electric shock state; The variational mode decomposition method is used to decompose the residual current signal under electric shock state to obtain the modal components of different frequency bands.

3. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 2, characterized in that: The method of using the variational modal decomposition method to decompose the residual current signal in the electric shock state to obtain modal components of different frequency bands is: Decompose the residual current signal into k IMF components; Perform Hilbert transform on each IMF component, modulate the spectrum of each IMF component to the corresponding baseband, and evaluate the bandwidth of each IMF component; Under the constraint that the sum of the IMF components is equal to the original residual current signal, an extreme value problem is constructed to minimize the sum of the bandwidths of all IMF components, and a penalty parameter and Lagrange multiplier are introduced to transform the constrained variational problem into an unconstrained variational problem. The alternating direction multiplier algorithm is used to iteratively solve the unconstrained variational problem until the convergence condition is met and k IMF components are output.

4. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 1, characterized in that: The method of establishing the electric shock current regression extraction model by using modal components of different frequency bands is as follows: Use the modal components of different frequency bands as the training set and test set for model establishment, define the loss function and penalty function of the XGBoost model, and initialize the various hyperparameters of the XGBoost model; Construct grid nodes according to the parameter combination to be optimized, and set the search range and search step size; Traverse the grid nodes from different directions, train the XGBoost model using the parameter combination to be optimized, and obtain the predicted value and actual value of the sample corresponding to each set of parameter combination to be optimized; Select a set of parameter combinations to be optimized with the smallest root mean square error between the predicted value and the actual value of the corresponding sample as the current optimal solution; Determine whether the current optimal solution reaches the threshold; if yes, output the current parameter combination as the optimal parameter combination; If not, redetermine the search range and search step size and iterate until the current optimal solution reaches the threshold.

5. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 1, characterized in that: The method for determining the electric shock protection action based on the electric shock branch current extracted in real time is: Obtain the real-time electric shock branch current and calculate the effective value; According to the effective value, the electric shock protection action is judged.

6. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 5, characterized in that: The real-time electric shock branch current is obtained and the effective value is calculated. I r The method is: in, I r is a valid value, k is the number of sampling points of the biological body electric shock current signal; For the i The sampling value of each point.

7. The method for determining the action of electric shock protection in AC / DC power distribution network according to claim 5, characterized in that: The method for judging the action of electric shock protection according to the effective value is: in, I r is a valid value, For duration.

8. An AC / DC distribution network electric shock protection action judgment system, characterized in that: include: Residual current signal acquisition module: used to obtain the real-time residual current signal of the electric shock state of the AC distribution network; Real-time extraction module for electric shock branch current: used to extract the electric shock branch current of the real-time residual current signal of the electric shock state of the AC distribution network in real time by using the pre-established electric shock current regression extraction model; Electric shock protection action judgment module: used to make electric shock protection action judgment based on the electric shock branch current extracted in real time; The establishment process of the electric shock current regression extraction model is as follows: obtaining the residual current signal of the electric shock state of the AC / DC power distribution network, and decomposing the residual current signal of the electric shock state to obtain modal components of different frequency bands; A regression extraction model for electric shock current is established using modal components in different frequency bands.

9. A terminal 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 computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images