Multi-source coupling leakage current accurate traceability and cooperative suppression method and system of power distribution network

Abstract

The invention relates to the technical field of intelligent power distribution network safety detection and fault processing, in particular to a multi-source coupling leakage current accurate traceability and cooperative suppression method and system for a power distribution network, and the method comprises the steps: collecting leakage current, voltage phase and power fluctuation data through a synchronous phasor measurement device; based on the collected data, constructing a four-dimensional feature library, carrying out decoupling processing on leakage current signals, and then realizing accurate traceability positioning of the leakage current according to a matching result of the leakage current signals and the four-dimensional feature library and a time difference between sensors; data are uploaded to a distribution network dispatching center, leakage current is evaluated based on a dynamically adjusted safety threshold value, and then operation states of a photovoltaic inverter, an energy storage converter and a charging pile are cooperatively controlled to realize active suppression of the leakage current; and performing preliminary diagnosis on the leakage current at the edge side, uploading a result to a cloud end, and determining a final diagnosis conclusion. According to the invention, the accuracy of multi-source coupling leakage current tracing and the accuracy of cooperative suppression of the power distribution network are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent power distribution network safety detection and fault processing, in particular to a multi-source coupled leakage current accurate tracing and collaborative suppression method and system for a power distribution network. BACKGROUND

[0002] With the development of smart grids, the structure of the power distribution network has changed from the traditional "single power supply radial" to the complex form of "source-grid-load-storage multi-source interaction". The wide access of devices such as photovoltaic, energy storage and charging piles has improved flexibility while also brought new leakage current safety hazards. Existing technologies such as single power supply leakage current detection, local line monitoring and traditional leakage protection have obvious defects when facing multi-source coupling scenarios: first, the detection range is localized and cannot cover the entire domain, leading to mutual conduction of leakage currents between different devices causing misjudgment; second, signal cross interference is serious, and the leakage current signals generated by multiple source devices running simultaneously overlap each other, especially transient disturbances such as charging pile start-stop can mask the real fault signal, causing a tracing error of more than 30%; third, there is a lack of collaborative suppression mechanism, and the passive mode of "detection-cutting" is commonly used, which not only cannot realize multi-device linkage governance, but also easily leads to power supply interruption, affecting system reliability. Therefore, the existing technology cannot solve the core contradictions of single data, ignoring topology, poor real-time performance and weak model generalization ability, and an innovative solution that can integrate multi-modal data and achieve accurate tracing and active collaborative suppression is urgently needed. SUMMARY

[0003] The present application provides a multi-source coupled leakage current accurate tracing and collaborative suppression method and system for a power distribution network, which solves the three core problems of localized detection range, signal cross interference and lack of collaborative suppression of existing leakage current detection technology in the face of complex power distribution networks with "source-grid-load-storage" multi-source coupling.

[0004] The purpose of the present application can be achieved by the following technical solutions: The first aspect of the present application provides a multi-source coupled leakage current accurate tracing and collaborative suppression method for a power distribution network, comprising: S1: Synchronously collecting leakage current, voltage phase and power fluctuation data through synchronous phasor measurement devices deployed at each node of the source, grid, load and storage of the power distribution network; S2: Based on the collected leakage current, voltage phase and power fluctuation data, a four-dimensional feature library containing source-side, grid-side, load-side and storage-side features is constructed, and an improved variational mode decomposition algorithm is used to decouple the coupled leakage current signal. Through the matching results of the decoupled leakage current signal and the four-dimensional feature library, as well as the time difference between sensors, accurate tracing and positioning of the leakage current is realized. S3: Upload the source tracing results and real-time data to the distribution network dispatch center, evaluate the leakage current based on the dynamically adjusted safety threshold, and generate multi-device linkage suppression instructions; S4: Execute the multi-device linkage suppression command to coordinate and control the operating status of the photovoltaic inverter, energy storage converter and charging pile, so as to achieve active suppression of leakage current; S5: Perform a preliminary diagnosis on the edge side based on the decoupled leakage current to obtain preliminary diagnostic results. Then, upload the preliminary diagnostic results to the cloud. The cloud will verify and integrate the preliminary diagnostic results on the edge side with historical data and output the final diagnostic conclusion.

[0005] Furthermore, the specific process of decoupling the coupled leakage current signal using the improved variational mode decomposition algorithm includes: Step 1: First, determine the initial value of the number of modes K; perform VMD decomposition to determine the K modal components; Step 2: Based on the objective function of the improved variational mode decomposition algorithm, fine-tune and optimize the K modal components through iterative convergence to obtain the optimized K modal components corresponding to each mode number; Step 3: Perform a Fourier transform on each optimized modal component corresponding to each modal number to obtain the spectrum of each modal component; calculate the spectral correlation coefficient between adjacent components based on the spectra of two adjacent components. Step 4: If any spectral correlation coefficient is greater than the preset correlation coefficient threshold, the process jumps to step 5; if all spectral correlation coefficients are less than or equal to the preset correlation coefficient threshold, the process jumps to step 6. Step 5: Increment the modality number K by 1, then jump to step 1 to continue the iteration; Step 6: Output the final number of modes K and the corresponding optimized K modal components; The objective function of the improved variational mode decomposition algorithm is:

[0006] In the formula, Represents the first value obtained after decomposition. One eigenmode function Indicates the first Modal components The center frequency, Indicates the first One modal component, Indicates all A set of modal components, Indicates all A set of center frequencies L2 norm The square of, This represents the partial derivative with respect to time. For Hilbert transform pairs, Represents the unit impulse function. Represents the imaginary unit. Represents pi (π). This represents the convolution operation. Represents the natural constant. Indicates time, It is a complex exponential function used for frequency modulation; This represents a function that takes the minimum value. The spectral correlation coefficient is the Pearson correlation coefficient.

[0007] Furthermore, the precise source tracing and localization of leakage current is achieved by matching the decoupled leakage current signal with the four-dimensional feature library and using the time difference between sensors, including: Based on the final number of modes K corresponding to the optimized K modal components, the feature parameters in the four-dimensional feature library are compared and matched to determine the fault type of leakage current; the sensor that receives the signal first is selected as the reference sensor; the time difference between other sensors and the reference sensor is calculated, and the distance between other sensors and the reference sensor is calculated based on the time difference between other sensors and the reference sensor and the wave speed, so as to realize the fault location of leakage current. The specific feature parameters in the constructed four-dimensional feature library include: The common-mode current amplitude and inverter switching frequency on the source side, the ground capacitance and insulation impedance phase on the grid side, the high-frequency harmonic content and transient current slope on the load side, and the DC component and fluctuation frequency of the ion current on the storage side.

[0008] Furthermore, the specific process of evaluating leakage current based on a dynamically adjusted safety threshold includes: 1) Calculate the current dynamically adjusted safety threshold; 2) Obtain the current actual leakage current value; 3) Perform the assessment; the specific assessment process is as follows: if the current actual leakage current value is greater than the current dynamically adjusted safety threshold, the leakage current is determined to be excessive, there is a safety hazard, and the assessment result is abnormal; if the current actual leakage current value is less than or equal to the current dynamically adjusted safety threshold, it means that the current leakage current is within the allowable range, and the system continues to monitor. The current actual leakage current value is derived from the actual leakage current value after decoupling. The dynamically adjusted safety threshold model is specifically represented as follows:

[0009] In the formula, a dynamic safety threshold at time t, a reference threshold, a real-time humidity at time t, a reference humidity, a real-time temperature at time t, a reference temperature, a humidity weight coefficient, a temperature weight coefficient, both of which are preset values; According to the traceability result and the evaluation result of the leakage current, a multi-device linkage suppression instruction is generated through a distribution network dispatching center.

[0010] Further, the multi-device linkage suppression instruction is executed to cooperatively control the operating states of the photovoltaic inverter, the energy storage converter and the charging pile to actively suppress the leakage current, including: The cooperative control includes photovoltaic side suppression, energy storage side suppression and charging pile side suppression. The photovoltaic side suppression is to dynamically adjust the PWM modulation strategy of the inverter when the photovoltaic side leakage current exceeds the standard, wherein the common-mode voltage suppression formula in the modulation strategy is:

[0011] In the formula, represents the DC bus voltage, represents the duty ratio of the i-th bridge arm, represents the common-mode voltage. The energy storage side suppression is to activate the battery cluster equalization circuit when the energy storage side leakage current is abnormal, and the calculation formula of the equalization current

[0012] In the formula, represents the equalization current, represents the proportional coefficient, represents the maximum single cell voltage, represents the average voltage, represents the instantaneous voltage deviation, represents the integral coefficient, represents the integral of the voltage deviation, represents the proportional control, represents the integral control. The charging pile side suppression includes charging pile soft start control and harmonic suppression. ​​​​​The soft-start control process for charging piles is as follows: when the leakage current on the charging pile side exceeds the standard abnormally, the charging pile's own controller controls the rate of increase of the charging pile current, that is, extends the current time to form a smooth start-up process. The harmonic suppression process involves the built-in filter circuit of the charging pile actively filtering out high-frequency harmonic components in a specific frequency band.

[0013] Further, step S5 specifically includes: Step S51: At the edge nodes, extract the decoupled leakage current features, input the extracted leakage current features into a lightweight convolutional neural network model for forward propagation calculation, and obtain preliminary diagnostic results. Step S52: In the cloud, receive the preliminary diagnostic results of the edge node, and combine them with historical operation and maintenance data to calculate the probability of actual leakage using a Bayesian fusion algorithm; Step S53: Quantify the line aging index, and obtain the final leakage risk based on the actual probability of leakage, the line aging index, and humidity data; obtain the final diagnosis conclusion through the final leakage risk.

[0014] Furthermore, the forward propagation formula in step S51 is:

[0015] In the formula, Indicates the first Layered linear output, Indicates the first The activation value of the layer, Indicates the first Layer weights, Indicates the first Layer bias, Indicates the first The activation value of the layer, This represents the activation function. This indicates a convolution operation.

[0016] Furthermore, the model formula corresponding to the Bayesian fusion algorithm in step S52 is:

[0017] In the formula, This indicates the actual location of the leakage current event. This indicates the detection results for edge nodes. This represents historical data in the cloud. This represents the probability of a leakage actually occurring given the known edge alarm E and the current operating condition C. Represents prior probability; This represents the probability that the edge CNN model can correctly trigger an alarm under the condition that a leakage current L actually occurs; represents the probability of the current condition C occurring under the condition that the real leakage L occurs; represents the total probability of the edge alarm E and the current condition C occurring simultaneously.

[0018] Further, the step S53 specifically comprises: When the probability of the real leakage occurring is greater than or equal to the preset safety threshold, then the comprehensive decision link is entered to determine the final leakage risk; when the probability of the real leakage occurring is less than the preset safety threshold, then it is determined that the false alarm caused by the charging pile start-stop interference, and no subsequent operation is needed. When the comprehensive decision link is entered, the final leakage risk is obtained according to the probability of the real leakage occurring, the line aging index, and the humidity data; wherein the final leakage risk is specifically represented by the formula:

[0019] In the formula, represents the line aging index, represents the real-time relative humidity, represents the reference humidity, represents the extreme high humidity; represents the line aging weight, represents the humidity weight, represents the probability of the real leakage occurring under the condition that the edge alarm E and the current condition C are known; represents the final leakage risk; wherein the line aging index is specifically represented by the formula:

[0020] In the formula, represents the basic aging factor, represents the real-time electrical health factor, represents the environmental acceleration factor, represents the historical failure correction factor, is a first weight coefficient, represents a second weight coefficient, represents a third weight coefficient, represents a fourth weight coefficient; wherein the basic aging factor is specifically represented by the formula:

[0021] In the formula, represents the actual running life of the line, represents the design life of the line, represents the material correction coefficient; wherein the real-time electrical health factor is specifically represented by the formula:

[0022] In the formula, represents the current insulation impedance of the line, represents the initial insulation impedance of the line when the line is put into operation, represents the leakage current variation coefficient; wherein the leakage current variation coefficient is the ratio of the standard deviation to the average value of the leakage current data collected by the PMU device in the past 24 hours; wherein the environmental acceleration factor is specifically expressed by the formula:

[0023] In the formula, represents the real-time temperature, represents the reference temperature, represents the extreme high temperature; is a humidity weight coefficient, is a temperature weight coefficient, and both are preset values; wherein the historical fault correction factor is specifically expressed by the formula:

[0024] In the formula, represents the number of leakage fault times of the line in the past three years, represents the number of insulation damage fault times of the line in the past three years, represents the total operation and maintenance times of the line in the past three years; Finally, the leakage risk condition is determined, specifically, when the final leakage risk is greater than or equal to a preset determination threshold, it is determined that there is a real leakage risk, otherwise, there is no real leakage risk.

[0025] The second aspect of the application provides a multi-source coupled leakage current accurate tracing and collaborative suppression system of a power distribution network, comprising: A data acquisition module: for synchronously acquiring leakage current, voltage phase and power fluctuation data through synchronous phasor measurement devices deployed at each node of the source, network, load and storage of the power distribution network; A feature decoupling and tracing positioning module: for constructing a four-dimensional feature library containing source side, network side, load side and storage side features based on the acquired leakage current, voltage phase and power fluctuation data, and decoupling the coupled leakage current signal by using an improved variational mode decomposition algorithm, and realizing accurate tracing and positioning of the leakage current through the matching results of the decoupled leakage current signal and the four-dimensional feature library and the time difference between the sensors; A leakage current evaluation module: for uploading the tracing results and real-time data to the power distribution network dispatching center, evaluating the leakage current based on a dynamically adjusted safety threshold, and generating a multi-device linkage suppression instruction; Inhibition module: used for executing the multi-device linkage inhibition instruction, cooperatively controlling the operation states of the photovoltaic inverter, the energy storage converter and the charging pile, so as to realize active inhibition of the leakage current; Edge-cloud collaborative verification module: used for preliminarily diagnosing the decoupled leakage current on the edge side to obtain a preliminary diagnosis result, then uploading the preliminary diagnosis result to the cloud, and verifying and fusing the preliminary diagnosis result on the edge side by the cloud in combination with historical data to output a final diagnosis conclusion.

[0026] Compared with the prior art, the beneficial effects of the present application are: through the synchronous phasor measurement devices deployed at the nodes of the source, network, load and storage of the power distribution network, the leakage current, voltage phase and power fluctuation data are synchronously collected; the synchronous and accurate collection of global electrical data is realized, providing a high-reliability and high-consistency data basis for subsequent analysis; based on the collected data, a four-dimensional feature library containing the characteristics of the source side, network side, load side and storage side is constructed, and an improved variational mode decomposition algorithm is used to decouple the coupled leakage current signal; through the matching result of the decoupled leakage current signal and the four-dimensional feature library and the time difference between the sensors, the accurate tracing and positioning of the leakage current are realized; through signal decoupling and feature matching, the leakage current fault source is accurately identified and positioned from the complex coupled signal; the tracing result and real-time data are uploaded to the distribution network dispatching center, the leakage current is evaluated based on the dynamically adjusted safety threshold, and a multi-device linkage inhibition instruction is generated; intelligent decision-making based on real-time risk assessment is realized, and an accurate cross-device collaborative inhibition strategy is generated; the multi-device linkage inhibition instruction is executed, the operation states of the photovoltaic inverter, the energy storage converter and the charging pile are cooperatively controlled, so as to realize active inhibition of the leakage current; through multi-device linkage control, the leakage current is actively inhibited without interrupting power supply, and the system reliability is improved; the leakage current is preliminarily diagnosed on the edge side, and the result is uploaded to the cloud, and the cloud verifies and fuses the edge diagnosis result in combination with historical data to output a final diagnosis conclusion; through edge and cloud collaborative verification, the real-time and accuracy of fault diagnosis are taken into account, and the false alarm rate is greatly reduced; the three core problems of localized detection range, signal cross interference and lack of collaborative inhibition of the existing leakage current detection technology in the face of complex power distribution network with multi-source coupling of "source-network-load-storage" are solved. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0028] Figure 1A step flow schematic diagram of a multi-source coupled leakage current accurate tracing and collaborative suppression method of a power distribution network is provided for the present application. Figure 2 A module flow schematic diagram of a multi-source coupled leakage current accurate tracing and collaborative suppression system of a power distribution network is provided for the present application. DETAILED DESCRIPTION

[0029] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0030] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] In view of the problems existing in the background art, a multi-source coupled leakage current accurate tracing and collaborative suppression method and system of a power distribution network are designed and researched, which has important practical significance.

[0032] As shown in Figure 1 The first aspect of the present application is to provide a multi-source coupled leakage current accurate tracing and collaborative suppression method of a power distribution network, comprising the following steps: Step S1: Synchronously collecting leakage current, voltage phase and power fluctuation data by deploying synchronous phasor measurement devices at each node of the source, network, load and storage of the power distribution network.

[0033] Specifically, high-precision synchronous phasor measurement devices (PMU) are deployed at key electrical nodes of the power distribution network, such as source (photovoltaic inverter outlet), network (line branch point, switch station), load (charging pile access point), and storage (energy storage system DC side and AC interface). A uniform high-precision clock source is provided for all PMUs in the network, usually provided by a Beidou / GPS satellite timing system, to ensure that the sampling time synchronization error of all PMUs in the network is within microseconds.

[0034] Then, the leakage current, voltage phase, and power fluctuation data are synchronously collected by the PMU.

[0035] At this point, the data collection is completed.

[0036] Step S2: Based on the collected leakage current, voltage phase, and power fluctuation data, a four-dimensional feature library containing source-side, network-side, load-side, and storage-side features is constructed, and an improved variational mode decomposition algorithm is used to decouple the coupled leakage current signal. Through the matching results of the decoupled leakage current signal and the four-dimensional feature library, as well as the time difference between sensors, the precise tracing and positioning of the leakage current are realized.

[0037] Specifically, the process of decoupling the leakage current signal is as follows: Step 1: First, determine the initial value of the mode number K; perform VMD (Variational Mode Decomposition) decomposition to determine K modal components; Step 2: According to the objective function of the improved variational mode decomposition algorithm, fine-tune and optimize the K modal components through iterative convergence to obtain the optimized K modal components corresponding to each modal number; Step 3: Perform Fourier transform on each optimized modal component corresponding to each modal number to obtain the frequency spectrum of each modal component; calculate the frequency spectrum correlation coefficient between adjacent components according to the frequency spectrum of the two adjacent components; Step 4: If any frequency spectrum correlation coefficient is greater than a preset correlation coefficient threshold, it indicates that modal aliasing has occurred, and the current decomposition effect is not good, then the process jumps to step 5; if all frequency spectrum correlation coefficients are less than or equal to the preset correlation coefficient threshold, it indicates that the modal separation is good, and the iteration is successful, then the process jumps to step 6; Step 5: Add 1 to the modal number K, then jump to step 1 to continue iteration; Step 6: Output the final modal number K and the corresponding optimized K modal components.

[0038] The objective function of the improved VMD algorithm is:

[0039] In the formula, indicates the first eigenmode function obtained after decomposition, indicates the center frequency of the first modal component, indicates the first modal component, indicates the set of all modal components, indicates the set of all center frequencies, indicates the square of the L2 norm , indicates the partial derivative with respect to time, is the Hilbert transform pair, indicates the unit impulse function (Dirac delta function), indicates the imaginary unit, indicates the ratio, indicates the circular constant, indicates the convolution operation, indicates the natural constant, indicates time, is a complex exponential function for frequency modulation (mixing); indicates the minimum value function.

[0040] The spectral similarity of adjacent modes after VMD decomposition is evaluated by the spectral correlation coefficient, which is used as a feedback signal to guide and improve the VMD algorithm (such as adaptively adjusting the mode number K or the decomposition parameter) to solve the mode aliasing problem. The spectral correlation coefficient is the Pearson correlation coefficient, and the calculation of the Pearson correlation coefficient is a known technology, which will not be described in detail here.

[0041] The Fourier transform is a known technology, which will not be described in detail here. In this embodiment, the preset correlation coefficient threshold is 0.8, and in this embodiment, the preset correlation coefficient threshold is not specifically limited, and the implementer can determine it according to the specific situation.

[0042] According to the final mode number K corresponding to the optimized K modal components, the characteristic parameters in the four-dimensional feature library are compared and matched to determine the fault type of the leakage current. The sensor that receives the signal first is selected as the reference sensor; the time difference of other sensors relative to the reference sensor is calculated, and the distance between other sensors and the reference sensor is calculated according to the time difference of other sensors relative to the reference sensor and the wave speed, so as to realize the fault positioning of the leakage current.

[0043] The specific feature parameters in the constructed four-dimensional feature library include: Source side (photovoltaic): common-mode current amplitude, inverter switching frequency, where the characteristic range of the switching frequency is 2-150kHz; Grid side (line): capacitance to ground, insulation impedance phase, wherein the characteristic range of capacitance to ground is 0.1-10uF; Load side (charging pile): High-frequency harmonic content, transient current slope, wherein the characteristic range of high-frequency harmonic content is 150kHz-1MHz; Storage side (energy storage): DC component of ion current and fluctuation frequency, wherein the characteristic range of fluctuation frequency is 0-100Hz.

[0044] This completes the tracing of the leakage current.

[0045] Step S3: Upload the source tracing results and real-time data to the distribution network dispatch center, evaluate the leakage current based on the dynamically adjusted safety threshold, and generate multi-device linkage suppression instructions.

[0046] Specifically, the process for conducting a safety assessment of leakage current is as follows: 1) Calculate the current dynamically adjusted safety threshold; 2) Obtain the current actual leakage current value (this value comes from the actual leakage current value after decoupling); 3) Perform the assessment; the specific assessment process is as follows: if the current actual leakage current value is greater than the current dynamically adjusted safety threshold, the leakage current is determined to be excessive, there is a safety hazard, and the assessment result is abnormal; if the current actual leakage current value is less than or equal to the current dynamically adjusted safety threshold, it means that the current leakage current is within the allowable range, and the system continues to monitor.

[0047] The dynamically adjusted safety threshold model is specifically represented as follows:

[0048] In the formula, Indicates time Dynamic security threshold, Indicates the baseline threshold. Indicates time Real-time humidity, Indicates the reference humidity. Indicates time Real-time temperature, Indicates the reference temperature. Humidity weighting coefficient These are temperature weighting coefficients, and both are preset values. In this embodiment, they are taken as 0.15 and 0.1 respectively. and The value of is not specifically limited; the implementer can determine it according to the specific circumstances.

[0049] Based on the source tracing results and leakage current assessment results, the distribution network dispatch center generates multi-device linkage suppression commands.

[0050] At this point, a multi-device linkage suppression command was obtained.

[0051] Step S4: Execute the multi-device linkage suppression command to coordinate the operation status of the photovoltaic inverter, energy storage converter and charging pile, so as to achieve active suppression of leakage current.

[0052] The coordinated control includes photovoltaic-side suppression, energy storage-side suppression, and charging pile-side suppression. Photovoltaic-side suppression involves dynamically adjusting the inverter's PWM (Pulse Width Modulation) modulation strategy when the photovoltaic leakage current exceeds the limit. The common-mode voltage suppression formula is as follows:

[0053] In the formula, Indicates the DC bus voltage. Indicates the first Duty cycle of each bridge arm, This refers to the common-mode voltage. The detailed process is as follows: the system dynamically adjusts the common-mode voltage using a pre-set optimized modulation strategy library. When an excess is detected, the scheduling center selects and issues a specific PWM modulation mode (such as changing the modulation wave or using a specific space vector sequence) from the strategy library, instructing the inverter to switch and execute it.

[0054] Among them, energy storage side suppression is as follows: when the leakage current on the energy storage side is abnormal, the battery cluster equalization circuit is activated, and its equalization current... The calculation formula is:

[0055] In the formula, Indicates the equalization current. Represents the proportionality coefficient. Indicates the maximum single-cell voltage. Indicates average voltage. Indicates instantaneous voltage deviation. Represents the integral coefficient. The integral representing the voltage deviation. Indicates proportional control. This indicates integral control.

[0056] Among them, charging pile side suppression includes charging pile soft start control and harmonic suppression; The soft-start control process for charging piles is as follows: when the leakage current on the charging pile side exceeds the standard abnormally, the charging pile's own controller controls the rate of increase of the charging pile current, that is, extends the current time to form a smooth start-up process, avoiding the impact of sudden current changes on the detection system.

[0057] The harmonic suppression process involves the built-in filter circuit of the charging pile actively filtering out high-frequency harmonic components in a specific frequency band to reduce interference with the leakage current detection signal.

[0058] Thus, active suppression of leakage current is achieved.

[0059] Step S5: Perform a preliminary diagnosis on the edge side based on the decoupled leakage current to obtain preliminary diagnosis results. Then, upload the preliminary diagnosis results to the cloud. The cloud will verify and integrate the preliminary diagnosis results on the edge side with historical data and output the final diagnosis conclusion.

[0060] Step S51: At the edge nodes, extract the decoupled leakage current features, input the extracted leakage current features into a lightweight convolutional neural network model for forward propagation calculation, and obtain preliminary diagnostic results; the forward propagation formula is:

[0061] In the formula, Indicates the first Layered linear output, Indicates the first The activation value of the layer, Indicates the first Layer weights Indicates the first Layer bias, Indicates the first The activation value of the layer, This represents the activation function. This indicates a convolution operation.

[0062] Step S52: In the cloud, receive the preliminary diagnostic results of the edge node, and combine them with historical operation and maintenance data to calculate the probability of actual leakage current occurrence using a Bayesian fusion algorithm; wherein, the model formula corresponding to the Bayesian fusion algorithm is:

[0063] In the formula, This indicates the actual location of the leakage current event. This indicates the detection results for edge nodes. This represents historical data in the cloud. This represents the probability (i.e., posterior probability) of a leakage actually occurring given the known edge alarm E and the current operating condition C. represents the prior probability (the inherent possibility of the line itself leaking electricity without any new evidence, obtained according to historical statistical data); represents the probability that the edge CNN model can correctly alarm under the condition that the leakage L actually occurs; represents the probability of the current working condition C (such as high humidity, high aging index) under the condition that the leakage L actually occurs; represents the total probability of the occurrence of the edge alarm E and the current working condition C.

[0064] Step S53: Quantifying the line aging index, obtaining the final leakage risk according to the probability of the actual occurrence of leakage, the line aging index, and the humidity data; obtaining the final diagnosis conclusion through the final leakage risk.

[0065] When the probability of the actual occurrence of leakage is greater than or equal to the preset safety threshold, the comprehensive decision link is entered to determine the final leakage risk; when the probability of the actual occurrence of leakage is less than the preset safety threshold, it is determined that the false alarm is caused by the charging pile start-stop interference, and no subsequent operation is required. In this embodiment, the preset safety threshold is 0.9, and in this embodiment, the preset safety threshold is not specifically limited, and the implementer can determine it according to the specific situation.

[0066] When entering the comprehensive decision link, the final leakage risk is obtained according to the probability of the actual occurrence of leakage, the line aging index, and the humidity data; wherein the final leakage risk is specifically represented by the formula:

[0067] In the formula, represents the line aging index, represents the real-time relative humidity, represents the reference humidity (taking the normal operation humidity of the power distribution network as 40%), represents the extreme high humidity (taking 95%, corresponding to the plum rain / moist environment); represents the line aging weight, represents the humidity weight, represents the probability of the actual occurrence of leakage under the condition that the edge alarm E and the current working condition C are known; represents the final leakage risk.

[0068] In this embodiment, the line aging weight is 0.3, and the humidity weight is 0.25; wherein the two weights are not specifically limited.

[0069] In the formula, the line aging index is specifically represented by the formula:

[0070] wherein, represents a line aging index, represents a basic aging factor, represents a real-time electrical health factor, represents an environmental acceleration factor, represents a historical fault correction factor, is a first weight coefficient, represents a second weight coefficient, represents a third weight coefficient, represents a fourth weight coefficient. The four weight coefficients are all preset values, in the embodiment , , , ; wherein in the embodiment , , , are not specifically limited, and the implementer can determine according to the specific situation.

[0071] wherein, the basic aging factor is specifically expressed by a formula:

[0072] wherein, represents a line actual operation life, represents a line design life, represents a material correction coefficient. Wherein, the copper core cable, the aluminum core cable and the overhead line are selected as the common materials of the distribution network; wherein in the embodiment, the material correction coefficient of the copper core cable is 0.8, the material correction coefficient of the aluminum core cable is 1.0, and the material correction coefficient of the overhead line is 1.1; wherein it is not specifically limited.

[0073] wherein, the real-time electrical health factor is specifically expressed by a formula:

[0074] wherein, represents a line current insulation impedance, represents a line initial insulation impedance when put into operation, represents a leakage current variation coefficient, represents a real-time electrical health factor.

[0075] wherein, the leakage current variation coefficient is a ratio of the standard deviation to the average value of the leakage current data collected by the PMU device in the last 24 hours; used to reflect the stability of the leakage current, the larger the variation coefficient, the worse the insulation.

[0076] wherein, the environmental acceleration factor is specifically expressed by a formula:

[0077] in the formula, represents an environmental acceleration factor, represents real-time relative humidity, represents a reference humidity (taking the normal operation humidity 40% of the power distribution network), represents an extreme high humidity (taking 95%, corresponding to the plum rain / moist environment), represents real-time temperature, represents a reference temperature (the normal operation temperature of the power distribution network, taking 25℃), represents an extreme high temperature (summer high temperature environment, taking 45℃); is a humidity weight coefficient, is a temperature weight coefficient, and both are preset values, which are respectively taken as 0.15 and 0.1 in the embodiment, wherein the values of and are not specifically limited and can be determined according to specific conditions.

[0078] wherein the historical fault correction factor is specifically expressed by a formula as:

[0079] in the formula, represents a historical fault correction factor, represents the number of line leakage fault times in the past three years, represents the number of line insulation damage fault times in the past three years, represents the total number of line operation and maintenance times in the past three years.

[0080] Finally, the leakage risk condition is determined, specifically, when the final leakage risk is greater than or equal to a preset determination threshold, it is determined that there is a real leakage risk, otherwise, there is no real leakage risk. Wherein the preset determination threshold is 0.95 in the embodiment, wherein the preset determination threshold is not specifically limited and can be determined according to specific conditions in the embodiment.

[0081] Thus, the embodiment is completed.

[0082] As shown in Figure 2 , the second aspect of the present application provides a multi-source coupled leakage current accurate tracing and collaborative suppression system of a power distribution network, comprising: a data acquisition module 101: for synchronously acquiring leakage current, voltage phase and power fluctuation data through synchronous phasor measurement devices deployed at each node of the source, network, load and storage of the power distribution network; The feature decoupling and traceable positioning module 102 is configured to construct a four-dimensional feature library containing source side, network side, load side and storage side features based on the collected leakage current, voltage phase and power fluctuation data, decouple the coupled leakage current signal by using an improved variational mode decomposition algorithm, and realize accurate traceable positioning of the leakage current by matching the decoupled leakage current signal with the four-dimensional feature library and the time difference between the sensors. The leakage current evaluation module 103 is configured to upload the traceable result to a distribution network dispatching center, evaluate the leakage current based on a dynamically adjusted safety threshold, and generate a multi-device linkage suppression instruction. The suppression module 104 is configured to execute the multi-device linkage suppression instruction, cooperatively control the operating states of the photovoltaic inverter, the energy storage converter and the charging pile, and realize active suppression of the leakage current. The edge-cloud collaborative verification module 105 is configured to preliminarily diagnose according to the decoupled leakage current on the edge side, obtain a preliminary diagnosis result, upload the preliminary diagnosis result to the cloud, verify and fuse the preliminary diagnosis result on the edge side by the cloud in combination with historical data, and output a final diagnosis conclusion.

[0083] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0084] The present application is described with reference to flowcharts and / or block diagrams of methods, systems and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts 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, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0085] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0086] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, so that the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flow Figure 1 one or more processes and / or blocks Figure 1 the function specified in the one or more blocks.

[0087] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit it, although the above embodiments of the present application have been described in detail, those skilled in the art should understand: the specific embodiments of the present application can be modified or replaced by the same, without departing from the spirit and scope of the present application, any modification or equivalent replacement, which should be covered within the scope of protection of the present application.

Claims

1. A method for precise tracing and collaborative suppression of multi-source coupled leakage current of a power distribution network, characterized in that, The method comprises the following steps: S1: Synchronously collecting leakage current, voltage phase and power fluctuation data through the synchronous phasor measurement device deployed at the source, network, load and storage nodes of the power distribution network; S2: Based on the collected leakage current, voltage phase and power fluctuation data, a four-dimensional feature library containing source-side, network-side, load-side and storage-side features is constructed, and an improved variational mode decomposition algorithm is used to decouple the coupled leakage current signal, and the precise source positioning of the leakage current is realized through the matching result of the decoupled leakage current signal and the four-dimensional feature library and the time difference between the sensors; S3: The source positioning result and real-time data are uploaded to the distribution network dispatching center, the leakage current is evaluated based on the dynamically adjusted safety threshold, and a multi-device linkage suppression instruction is generated; S4: The multi-device linkage suppression instruction is executed to cooperatively control the operating state of the photovoltaic inverter, energy storage converter and charging pile to achieve active suppression of the leakage current; S5: Preliminary diagnosis is performed on the decoupled leakage current on the edge side to obtain a preliminary diagnosis result, and then the preliminary diagnosis result is uploaded to the cloud side, and the cloud side combines historical data to verify and fuse the preliminary diagnosis result on the edge side to output a final diagnosis conclusion.

2. The method of claim 1, wherein, The decoupled processing of the coupled leakage current signal by the improved variational mode decomposition algorithm comprises the following specific processes: Step 1: First, determine the initial value of the mode number K; perform VMD decomposition to determine K modal components; Step 2: According to the objective function of the improved variational mode decomposition algorithm, the K modal components are fine-tuned and optimized through iterative convergence to obtain the optimized K modal components corresponding to each modal number; Step 3: Fourier transform is performed on each modal component corresponding to the optimized modal number to obtain the frequency spectrum of each modal component; the frequency spectrum correlation coefficient between adjacent components is calculated according to the frequency spectrum of the two adjacent components; Step 4: If there is any one frequency spectrum correlation coefficient greater than the preset correlation coefficient threshold, then the process jumps to step 5; if all frequency spectrum correlation coefficients are less than or equal to the preset correlation coefficient threshold, then the process jumps to step 6; Step 5: Add 1 to the mode number K, and then jump to step 1 to continue iteration; Step 6: Output the final mode number K and the corresponding optimized K modal components; Wherein, the objective function of the improved variational mode decomposition algorithm is: In the formula, represents the first eigenmode function obtained after decomposition, represents the first modal component center frequency, represents the first modal component, represents the set of all modal components, represents the set of all center frequencies, represents the square of the L2 norm , represents the partial derivative with respect to time, is the Hilbert transform pair, represents the unit impulse function, represents the imaginary unit, represents the circular constant, represents the convolution operation, represents the natural constant, represents time, is a complex exponential function used for frequency modulation; represents the minimum value function; Wherein, the frequency spectrum correlation coefficient is the Pearson correlation coefficient.

3. The method of claim 2, wherein, The precise source positioning of the leakage current through the matching result of the decoupled leakage current signal and the four-dimensional feature library and the time difference between the sensors comprises: According to the K optimized modal components corresponding to the final mode number K, the feature parameters in the four-dimensional feature library are compared and matched to determine the fault type of the leakage current; the sensor that receives the signal first is selected as the reference sensor; the time difference of other sensors relative to the reference sensor is calculated, and the distance between other sensors and the reference sensor is calculated according to the time difference of other sensors relative to the reference sensor and the wave speed, so as to realize the fault positioning of the leakage current; Wherein, the specific feature parameters in the constructed four-dimensional feature library include: The source side common mode current amplitude and the inverter switching frequency, the net side ground capacitance value and insulation impedance phase, the load side high frequency harmonic content and transient current slope, the storage side ion current DC component and fluctuation frequency.

4. The method of claim 1, wherein, The safety threshold based on dynamic adjustment evaluates the leakage current, including the following specific processes: 1) calculating the current dynamic adjustment safety threshold; 2) obtaining the current actual leakage current value; 3) performing evaluation; the specific evaluation process is: if the current actual leakage current value is greater than the current dynamic adjustment safety threshold, it is determined that the leakage current is out of standard, there is a safety hazard, and the evaluation result is abnormal; If the current actual leakage current value is less than or equal to the current dynamic adjustment safety threshold, it means that the current leakage current is within the allowable range, and the system continues to monitor; Wherein, the current actual leakage current value is derived from the actual leakage current value after decoupling; Wherein, the dynamic adjustment safety threshold model is specifically represented as: In the formula, a dynamic safety threshold at time point , a reference threshold, a real-time humidity at time point , a reference humidity, a real-time temperature at time point , a reference temperature, is a humidity weight coefficient, is a temperature weight coefficient, and the humidity weight coefficient and the temperature weight coefficient are preset values; According to the traceability result and the evaluation result of the leakage current, a multi-device linkage suppression instruction is generated through the distribution network dispatching center.

5. The method of claim 1, wherein, The execution of the multi-device linkage suppression instruction cooperatively controls the operating state of the photovoltaic inverter, the energy storage converter and the charging pile to actively suppress the leakage current, including: The cooperative control includes photovoltaic side suppression, energy storage side suppression and charging pile side suppression; Wherein, the photovoltaic side suppression is: when the photovoltaic side leakage current is out of standard, the PWM modulation strategy of the inverter is dynamically adjusted, wherein the common mode voltage suppression formula in the modulation strategy is: In the formula, denotes the direct current bus voltage, denotes the duty cycle of the bridge arm, denotes the common mode voltage; The energy storage side inhibition is: when the energy storage side leakage current is abnormal, the battery cluster equalization circuit is activated, and the equalization current The calculation formula is: wherein represents the equalization current, represents a proportional factor, represents the maximum cell voltage, represents the average voltage, represents the instantaneous voltage deviation, represents an integration factor, represents the integral of the voltage deviation, represents the proportional control, represents the integral control; Wherein, the charging pile side suppression includes charging pile soft start control and harmonic suppression; Wherein, the process of charging pile soft start control is: when the charging pile side leakage current is out of standard, the charging pile current rising rate is controlled by the charging pile controller itself, that is, the current time is prolonged to form a smooth starting process; Wherein, the process of harmonic suppression is: the filter circuit built-in the charging pile actively filters out the high frequency harmonic components of specific frequency bands.

6. The method of claim 1, wherein, The step S5 specifically includes: Step S51: extracting the decoupled leakage current features at the edge node, inputting the extracted leakage current features into a lightweight convolutional neural network model for forward propagation calculation to obtain a preliminary diagnosis result; Step S52: receiving the preliminary diagnosis result of the edge node in the cloud, combining historical operation and maintenance data, and calculating the probability of real leakage occurrence by using a Bayesian fusion algorithm; Step S53: quantifying the line aging index, obtaining the final leakage risk according to the probability of real leakage occurrence, the line aging index and the humidity data; and obtaining the final diagnosis conclusion through the final leakage risk.

7. The method of claim 6, wherein, The forward propagation formula in step S51 is: wherein, represents the activation value of the layer, represents the linear output of the layer, represents the weight of the layer, represents the bias of the layer, represents the activation value of the layer, represents an activation function, represents a convolution operation.

8. The method of claim 6, wherein, The model formula corresponding to the Bayesian fusion algorithm in step S52 is: In the formula, represents a real leakage current position event, represents the detection result of the edge node, represents cloud historical data, represents the probability of a real leakage occurring under the condition of a known edge alarm E and a current working condition C; represents a prior probability; represents the probability of the edge CNN model being able to correctly alarm under the condition of a real leakage L occurring; represents the probability of the current working condition C occurring under the condition of a real leakage L occurring; represents the total probability of the edge alarm E and the current working condition C occurring simultaneously.

9. The method of claim 6, wherein, The step S53 specifically includes: When the probability of real leakage occurrence is greater than or equal to a preset safety threshold, the final leakage risk is determined in the comprehensive decision-making link; when the probability of real leakage occurrence is less than the preset safety threshold, it is determined that the false alarm is caused by charging pile start-stop interference, and no subsequent operation is required. When entering the comprehensive decision link, the final electric leakage risk is obtained according to the probability of real electric leakage, the line aging index and the humidity data; wherein the final electric leakage risk is specifically expressed by the formula: wherein, represents a line aging index, represents a real-time relative humidity, represents a reference humidity, represents an extreme high humidity; represents a line aging weight, represents a humidity weight, represents the probability of a real occurrence of an electric leakage, in the condition of a known edge alarm E and a current operating condition C; represents the final electric leakage risk; Wherein the line aging index is specifically expressed by the formula: wherein represents a base aging factor, represents a real-time electrical health factor, represents an environmental acceleration factor, represents a historical failure correction factor, is a first weight coefficient, represents a second weight coefficient, represents a third weight coefficient, represents a fourth weight coefficient; Wherein the basic aging factor is specifically expressed by the formula: In the formula, represents the actual operation life of the line, represents the design life of the line, represents the material correction coefficient; Wherein the real-time electrical health factor is specifically expressed by the formula: In the formula, represents the current insulation impedance of the line, represents the initial insulation impedance of the line when the line is put into operation, represents the leakage current variation coefficient; wherein the leakage current variation coefficient is the ratio of the standard deviation to the average value of the leakage current data collected by the PMU device in the past 24 hours. Wherein the environmental acceleration factor is specifically expressed by the formula: In the formula, represents a real-time temperature, represents a reference temperature, represents an extreme high temperature; is a humidity weight coefficient, is a temperature weight coefficient, and both are preset values; Wherein the historical fault correction factor is specifically expressed by the formula: In the formula, represents the number of line leakage fault times in the past three years, represents the number of line insulation damage fault times in the past three years, represents the total number of line operation and maintenance times in the past three years; Finally, the electric leakage risk condition is judged, specifically, when the final electric leakage risk is greater than or equal to the preset judgment threshold, it is judged that there is a real electric leakage risk, otherwise, there is no real electric leakage risk.

10. A multi-source coupled leakage current accurate tracing and collaborative suppression system of a power distribution network, characterized in that, Comprise: A data acquisition module: for synchronously collecting electric leakage current, voltage phase and power fluctuation data through the synchronous phasor measurement device deployed at each node of the power distribution network source, network, load and storage; A feature decoupling and traceability positioning module: for constructing a four-dimensional feature library containing source side, network side, load side and storage side features based on the collected electric leakage current, voltage phase and power fluctuation data, and decoupling the coupled electric leakage current signal by using an improved variational mode decomposition algorithm, and realizing accurate traceability positioning of the electric leakage current through the matching result of the decoupled electric leakage current signal and the four-dimensional feature library and the time difference between sensors; An electric leakage current evaluation module: for uploading the traceability result and real-time data to the distribution network dispatching center, evaluating the electric leakage current based on a dynamically adjusted safety threshold, and generating a multi-device linkage suppression instruction; A suppression module: for executing the multi-device linkage suppression instruction to cooperatively control the operating state of the photovoltaic inverter, the energy storage converter and the charging pile to realize active suppression of the electric leakage current; An edge-cloud collaborative verification module: for preliminarily diagnosing according to the decoupled electric leakage current on the edge side to obtain a preliminary diagnosis result, and then uploading the preliminary diagnosis result to the cloud side to verify and fuse the preliminary diagnosis result of the edge side in combination with historical data to output a final diagnosis conclusion.

Images