Power distribution network primary and secondary deep fusion FTU partial discharge test method and circuit

Abstract

The invention discloses a power distribution network primary and secondary deep fusion FTU partial discharge test method and circuit, and relates to the technical field of FTU partial discharge test, and the method comprises the steps: collecting power supply voltage and partial discharge signal data, and carrying out the data preprocessing; performing time sequence analysis on the power supply voltage fluctuation and the partial discharge data to obtain a dynamic time sequence characteristic evaluation result, and performing spatial distribution analysis on the partial discharge signal to generate a three-dimensional space PD distribution diagram; constructing a space-time correlation model according to the dynamic time sequence characteristic evaluation result and the three-dimensional space PD distribution map, and analyzing the correlation between the power supply voltage fluctuation and the partial discharge signal in the space-time dimension to obtain a space-time dimension analysis result; and optimizing partial discharge detection according to a space-time dimension analysis result, and detecting the FTU by adopting the optimized partial discharge detection to obtain a partial discharge detection result. The partial discharge detection method is optimized, the sensitivity and accuracy of detection are improved, and the accuracy and reliability of partial discharge detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of FTU partial discharge testing, and in particular to a FTU partial discharge testing method and circuit for deep integration of primary and secondary power distribution networks. Background Art

[0002] With the continuous advancement of primary and secondary deep integration technology, the partial discharge of equipment is required to be less than 20pC. In order to respond to the requirements of the State Grid, our company has put the solution of FTU (distribution switch monitoring terminal) partial discharge problem on the agenda. The purpose of partial discharge test is to find defects in equipment structure and manufacturing process. For example: the local electric field strength inside the insulation is too high; metal parts have sharp corners; insulation is mixed with impurities or has local defects. The electrical connection between the metal grounding parts and the conductors inside the product is poor, so as to eliminate these defects and prevent partial discharge from damaging the insulation.

[0003] Partial discharge refers to the non-penetrating discharge between electrodes of the insulating medium in high-voltage electrical appliances under the action of high electric field strength. Partial discharge occurs in the air gaps or bubbles inside one or several insulations. Since the discharge energy is very small, it does not affect the short-term insulation strength of electrical equipment. Some adverse effects caused by partial discharge, such as the generation of undesirable compounds, can slowly damage the insulation, and finally lead to the breakdown of the entire insulation and sudden failure.

[0004] When partial discharge occurs inside a dielectric, many phenomena occur. Some are electrical, such as the generation of electrical pulses, increased dielectric loss, and electromagnetic wave radiation; some are non-electrical, such as light, heat, noise, changes in gas pressure, and chemical changes. However, the FTU partial discharge is not a true partial discharge, but a partial discharge test failure caused by the fluctuation of the FTU power supply voltage. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed.

[0007] Therefore, the present invention provides a distribution network primary and secondary deep integration FTU partial discharge test method and circuit, which can solve the problems mentioned in the background technology.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] In a first aspect, the present invention provides a distribution network primary and secondary deep fusion FTU partial discharge test method, which includes collecting power supply voltage and partial discharge signal data, and performing data preprocessing;

[0010] Perform time series analysis on power supply voltage fluctuation and partial discharge data to obtain dynamic time series feature evaluation results, and perform spatial distribution analysis on partial discharge signals to generate a three-dimensional spatial PD distribution map;

[0011] A spatiotemporal correlation model is constructed based on the dynamic timing feature evaluation results and the three-dimensional PD distribution diagram to analyze the correlation between the power supply voltage fluctuation and the partial discharge signal in the spatiotemporal dimension to obtain the spatiotemporal dimension analysis results;

[0012] The partial discharge detection is optimized through the time-space dimension analysis result, and the FTU is tested using the optimized partial discharge detection to obtain the partial discharge detection result.

[0013] As an optimal solution of the FTU partial discharge test method for deep integration of primary and secondary distribution networks described in the present invention, the supply voltage includes instantaneous voltage value, voltage fluctuation frequency, voltage fluctuation amplitude and harmonic components; the partial discharge signal data includes discharge intensity, discharge frequency, discharge time interval, discharge waveform and discharge position.

[0014] As a preferred solution of the FTU partial discharge test method for deep integration of primary and secondary distribution networks described in the present invention, the time series analysis includes extracting the dynamic timing characteristics of the power supply voltage and the partial discharge signal, and using the extracted characteristics to construct a time series model for dynamic timing characteristic evaluation; the spatial distribution analysis includes using the Kriging interpolation method to generate a three-dimensional spatial PD distribution map.

[0015] As a preferred solution of the FTU partial discharge test method for the primary and secondary deep integration of the distribution network described in the present invention, wherein: the dynamic timing characteristics of the power supply voltage include the mean, variance, maximum value, minimum value and standard deviation of the instantaneous voltage value;

[0016] The dynamic time series characteristics of the partial discharge signal include the mean, variance, maximum, minimum and standard deviation of the discharge intensity;

[0017] Use the extracted features to build a time series model and evaluate the dynamic time series features. The calculation formula is as follows:

[0018]

[0019] Among them, T v (t) is the dynamic timing characteristic evaluation result of the supply voltage, T p (t) is the dynamic timing characteristic evaluation result of the partial discharge signal;

[0020] The spatial distribution analysis comprises:

[0021] The Kriging interpolation method is used to generate a three-dimensional PD distribution map. The distribution map calculation formula is as follows:

[0022]

[0023] Among them, μ is the global mean, λ i is the Kriging weight, P xyz (x, y, z) is the spatial distribution data of partial discharge intensity.

[0024] As an optimal solution for the primary and secondary deep fusion FTU partial discharge test method of the distribution network described in the present invention, the construction of the spatiotemporal correlation model includes combining the hybrid model of the long short-term memory network (LSTM) and the three-dimensional convolutional neural network (3D-CNN) to simultaneously process time series data and spatial data to capture the characteristics of power supply voltage fluctuations and partial discharge signals in time and space dimensions.

[0025] As an optimal solution of the primary and secondary deep fusion FTU partial discharge test method of the distribution network described in the present invention, the operation process of the spatiotemporal correlation model includes preparing time series data and spatial data, using 3D-CNN to process spatial data, and using LSTM layer to process time series data, fusing the outputs of 3D-CNN and LSTM, and obtaining the final output through the fully connected layer.

[0026] As a preferred solution of the FTU partial discharge test method for the distribution network with deep integration of primary and secondary power distribution networks described in the present invention, the optimization of partial discharge detection includes:

[0027] Dynamically adjust the parameter settings of the detection equipment according to the comprehensive evaluation results of the time and space dimensions, including adjusting the sensitivity and sampling frequency of the sensor;

[0028] The dynamic adjustment includes:

[0029] Set the initial detection equipment parameters, including sensor sensitivity S 0 , sampling frequency f 0 , monitoring frequency ν 0 and coverage A 0 ;

[0030] Collect power supply voltage and partial discharge signal data in real time, extract dynamic time series features and analyze spatial distribution, and update the time and space dimension evaluation results M(t,x,y,z);

[0031] According to the latest spatiotemporal dimension assessment results M(t,x,y,z), the risk area is assessed and the assessment results are classified into low, medium and high risk areas;

[0032] When M(t,x,y,z)>first threshold θ 1 When the assessment area is determined to be a high-risk area;

[0033] When the first threshold θ 1 ≥M(t,x,y,z)>second threshold θ 2 When the risk is too high, the assessment area will be judged as a medium-risk area;

[0034] When the second threshold θ 2 When ≥M(t,x,y,z), the assessment area is judged as a low-risk area;

[0035] Dynamically adjust the parameter settings of the detection equipment according to the risk level, including:

[0036] In low-risk areas, the initial parameter settings are maintained, making the sensor sensitivity S = S 0 , sampling frequency f = f 0 ;

[0037] In medium-risk areas, appropriately increase monitoring frequency and coverage, and adjust sensor sensitivity to S 1 and the sampling frequency is f 1 ; The sensitivity of the medium-risk area is adjusted according to the discharge intensity and frequency characteristics. The calculation formula is as follows:

[0038]

[0039] Among them, k 1 is the first sensitivity adjustment coefficient;

[0040] The sampling frequency in the medium-risk area is adjusted according to the voltage fluctuation frequency characteristics.

[0041]

[0042] Among them, m 1 is the first fluctuation frequency adjustment coefficient;

[0043] Significantly increase monitoring frequency and coverage in high-risk areas and improve sensor sensitivity 2 and sampling frequency f 2 ;

[0044] The high-risk area significantly increases the sensitivity, and the calculation formula is as follows:

[0045]

[0046] Among them, k 2is the second sensitivity adjustment coefficient, k 2 >k 1 ;

[0047] The sampling frequency is significantly increased in high-risk areas, and the calculation formula is as follows:

[0048]

[0049] Among them, m 2 is the second fluctuation frequency adjustment coefficient, and m 2 >m 1 ;

[0050] Dynamically select and adjust the detection algorithm, and choose the detection method for the current situation based on real-time changes in spatiotemporal characteristics.

[0051] In a second aspect, the present invention provides a distribution network primary and secondary deep fusion FTU partial discharge test circuit, which includes: an AC-DC loop module and a DC-DC loop module;

[0052] The AC-DC loop module converts alternating current into direct current and is used to limit the voltage of the power supply input to the AC-DC loop module to be lower than a set value, thereby performing overvoltage protection;

[0053] The DC-DC loop module is electrically connected to the AC-DC loop module and is used to reduce the fluctuation of the voltage input to the DC-DC loop module power supply, so that the partial discharge reaches a minimum value.

[0054] As a preferred solution of the FTU partial discharge test method for the primary and secondary deep integration of the distribution network of the present invention, wherein: the AC-DC loop module includes an overvoltage protection unit, and the overvoltage protection unit includes a capacitor C2, a capacitor C3, a varistor and a rectifier unit;

[0055] The DC-DC loop module includes a bidirectional voltage regulator diode D7, a capacitor C14, a resistor R11, a resistor R14, a resistor R13, a capacitor C16, a capacitor C100, a resistor R15, a chip U2, a capacitor C12, a Schottky diode D6, an inductor L2, a capacitor CC1, a capacitor CC2, a capacitor C13, a resistor R12, a resistor R16, a thermistor Rt, a capacitor C15, a diode D9, a voltage regulator diode D8, a resistor R19, an integrated circuit chip QT2, a resistor R17, a resistor R20, a resistor R21 and a resistor R18.

[0056] In a third aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, the steps of the FTU partial discharge test method for deep integration of primary and secondary distribution networks are implemented.

[0057] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, the steps of a FTU partial discharge test method for deep integration of primary and secondary distribution networks are implemented.

[0058] Compared with the prior art, the beneficial effects of the present invention are that the power supply voltage and partial discharge signal data are collected and preprocessed, a comprehensive analysis of voltage fluctuations and discharge characteristics is achieved, the accuracy and reliability of the data are ensured, and thus the efficient processing of subsequent steps is facilitated; through time series analysis and three-dimensional spatial PD distribution map generation, the dynamic timing characteristics and spatial distribution characteristics of the power supply voltage and partial discharge signals are identified, thereby providing a solid foundation for spatiotemporal correlation analysis; by constructing a hybrid model combining LSTM and 3D-CNN, the correlation between the power supply voltage fluctuation and the partial discharge signal in the spatiotemporal dimension is analyzed, achieving the effect of accurately capturing and predicting the partial discharge characteristics; using the spatiotemporal dimension analysis results, the parameters and detection algorithms of the detection equipment are dynamically adjusted, the partial discharge detection method is optimized, the sensitivity and accuracy of the detection are improved, thereby effectively ensuring the safe operation of the power system; the organic combination of these steps not only improves the accuracy and reliability of partial discharge detection, but also provides important support for the intelligent management of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0060] Figure 1 A method flow chart of a distribution network primary and secondary deep integration FTU partial discharge test method and circuit provided by an embodiment of the present invention;

[0061] Figure 2 An internal structure diagram of a computer device for a distribution network primary and secondary deep integration FTU partial discharge test method and circuit provided by an embodiment of the present invention.

[0062] Figure 3 A circuit diagram of a distribution network primary and secondary deep integration FTU partial discharge test method and circuit provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned purposes, features and advantages of the present invention more understandable, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0064] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0065] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0066] Example 1

[0067] Reference Figure 1-Figure 3 , which is the first embodiment of the present invention, and provides a distribution network primary and secondary deep integration FTU partial discharge test method, including:

[0068] Before describing the embodiments of the present application in detail, some related concepts are first explained for the sake of clarity.

[0069] Partial discharge: Partial discharge (PD) refers to the local breakdown phenomenon that occurs in an electrical insulation system due to the local electric field strength exceeding the breakdown strength of the medium at that location. This discharge does not immediately lead to a complete breakdown of the insulation system, but its long-term existence will gradually deteriorate the insulating material. Partial discharge has the characteristics of short duration (usually at the nanosecond level), small amplitude (generally at the picosecond level), high frequency (up to the MHz level), and the location of occurrence is limited to a local area of ​​the insulator. It is an important phenomenon in the power system and has a significant impact on the safe operation of equipment. Detection methods include electrical, acoustic, chemical, and optical methods. In the preventive maintenance and fault diagnosis of power systems, the detection and analysis of partial discharge plays a key role, providing an important basis for the health status assessment and life prediction of equipment.

[0070] LSTM: LSTM (Long Short-Term Memory) is a special recurrent neural network (RNN) proposed by Hochreiter and Schmidhuber in 1997. It is specially designed to solve the long-term dependency problem in long sequence learning and overcome the gradient vanishing problem of traditional RNN. The core of LSTM is a memory unit, which is controlled by the input gate, forget gate and output gate to control the flow of information. The input gate determines how much new information enters the memory unit, the forget gate determines how much old information is discarded, and the output gate determines how much information is output. This structure enables LSTM to learn long-term dependencies and process sequences of different lengths. LSTM performs well in natural language processing, speech recognition, time series prediction and other fields. In power systems, LSTM is often used to process time series data of power supply voltage fluctuations and partial discharge signals to capture the characteristics and changing laws of these data in the time dimension.

[0071] 3D-CNN: 3D-CNN (three-dimensional convolutional neural network) is an extension of convolutional neural network (CNN) specifically designed for processing three-dimensional data. Compared with traditional 2D-CNN, 3D-CNN uses three-dimensional convolution kernels and can perform convolution operations in both spatial and temporal dimensions. This enables it to capture the spatial-temporal characteristics of data and is particularly suitable for processing three-dimensional data such as videos and medical images. The core idea of ​​3D-CNN is to automatically learn high-level feature representations of data through a combination of three-dimensional convolutional layers, pooling layers, and fully connected layers. In the fields of computer vision and medical image analysis, 3D-CNN has shown strong performance. In power systems, 3D-CNN is often used to process three-dimensional spatial PD distribution maps, from which the spatial distribution characteristics of partial discharge are extracted, providing important information for risk assessment and fault diagnosis.

[0072] Kriging interpolation: The Kriging interpolation method is an advanced geostatistical interpolation technique, originally developed by D.G.Krige, a mining engineer in South Africa, and later theorized and formalized by G.Matheron, a French mathematician. This method is widely used in geology, hydrology, environmental science and other fields, and is particularly suitable for interpolation of spatially correlated data. The Kriging method is based on a basic assumption that points close in space have more similar values ​​than points far away. This spatial correlation is described by a variogram. Kriging is a BLUE-BestLinearUnbiasedEstimator. It determines the optimal weights by minimizing the estimated variance. For the points to be interpolated, the Kriging method assigns different weights to the surrounding known points. These weights take into account not only the distance but also the overall spatial structure. The Kriging method not only provides interpolation estimates, but also gives the error of the estimate, which is an important advantage.

[0073] In the prior art, different partial discharge detection standards and methods may often lead to differences in test results. If the FTU detection standards and methods are not strict enough or are not applicable to the current equipment, the test results may fail.

[0074] This application provides a method that can effectively solve the above-mentioned problems. Next, we will combine multiple embodiments to elaborate on how to implement the FTU partial discharge test method for the distribution network with deep integration of primary and secondary power distribution networks.

[0075] Figure 1 A method flow chart of a FTU partial discharge test method and circuit for deep integration of primary and secondary distribution networks is shown, including:

[0076] S1: Collect power supply voltage and partial discharge signal data, and perform data preprocessing;

[0077] Furthermore, the supply voltage includes instantaneous voltage value, voltage fluctuation frequency, voltage fluctuation amplitude and harmonic components;

[0078] The instantaneous voltage value is obtained through voltage sensors or power quality monitoring equipment. The instantaneous voltage value can help identify the frequency, amplitude and period of voltage fluctuations; the voltage fluctuation frequency is used to record the frequency changes of voltage fluctuations and identify the main frequency components of voltage fluctuations through spectrum analysis; the voltage fluctuation amplitude is used to measure the amplitude of voltage fluctuations, including the amplitude of voltage rise and fall, to help identify abnormal fluctuations; the harmonic component is collected by the harmonic component in the power supply voltage, and the influence of high-order harmonics in voltage fluctuations on partial discharge is understood through harmonic analysis;

[0079] The local discharge signal data includes discharge intensity, discharge frequency, discharge time interval, discharge waveform and discharge position;

[0080] The discharge intensity is measured by collecting the intensity of each discharge (pC) through a partial discharge test instrument, and recording the amplitude of the discharge signal; the discharge frequency is the number of partial discharges occurring per unit time, and the frequency distribution of the discharge is determined through statistical analysis; the discharge time interval is the time interval between two consecutive discharges, and the periodic characteristics of the discharge are identified through time series analysis; the discharge waveform is analyzed through waveform analysis to understand the specific characteristics of the discharge; the discharge position is determined through partial discharge positioning technology (such as ultrasonic positioning or electromagnetic wave positioning) to determine the location where the discharge occurs.

[0081] Furthermore, the data preprocessing includes data cleaning and data standardization;

[0082] Data cleaning includes data denoising and outlier processing;

[0083] Data denoising includes,

[0084] Use a low-pass filter or a median filter to remove high-frequency noise from the supply voltage and partial discharge signals;

[0085] Perform Fourier transform on the harmonic components to remove interference signals of non-fundamental components;

[0086] Outlier processing includes:

[0087] Use box plots or z-score methods to detect and remove outliers;

[0088] Missing or abnormal data points are filled using interpolation.

[0089] S2: Perform time series analysis on the power supply voltage fluctuation and partial discharge data to obtain dynamic time series feature evaluation results, and perform spatial distribution analysis on the partial discharge signal to generate a three-dimensional spatial PD distribution map;

[0090] The time series analysis comprises:

[0091] According to the needs of time series analysis, the dynamic time series characteristics of the power supply voltage and partial discharge signal are extracted;

[0092] The dynamic timing characteristics of the supply voltage V(t) include the mean value V of the instantaneous voltage value mean , variance Vvar, maximum value V max , minimum value V min and standard deviation V std ;

[0093] The dynamic time series characteristics of the partial discharge signal P(t) include the mean value of the discharge intensity P mean , Variance Pvar , maximum value P max , minimum value P min and standard deviation P std ;

[0094] Use the extracted features to build a time series model and evaluate the dynamic time series features. The calculation formula is as follows:

[0095]

[0096] Among them, T v (t) is the dynamic timing characteristic evaluation result of the supply voltage, T p (t) is the dynamic timing characteristic evaluation result of the partial discharge signal;

[0097] The spatial distribution analysis comprises:

[0098] The Kriging interpolation method is used to generate a three-dimensional PD distribution map. The distribution map calculation formula is as follows:

[0099]

[0100] Among them, μ is the global mean, λ i is the Kriging weight, P xyz (x, y, z) is the spatial distribution data of partial discharge intensity.

[0101] S3: constructing a spatiotemporal correlation model according to the dynamic timing feature evaluation result and the three-dimensional spatial PD distribution diagram, analyzing the correlation between the power supply voltage fluctuation and the partial discharge signal in the spatiotemporal dimension, and obtaining the spatiotemporal dimension analysis result;

[0102] Furthermore, the construction of spatiotemporal correlation model includes:

[0103] Firstly, a hybrid model architecture combining long short-term memory network (LSTM) and three-dimensional convolutional neural network (3D-CNN) is designed. This architecture can process time series data and spatial data simultaneously, so as to capture the characteristics of power supply voltage fluctuation and partial discharge signal in time and space dimensions.

[0104] The basic structure of the model is as follows:

[0105] The 3D-CNN part processes spatial data (three-dimensional spatial PD distribution map);

[0106] The LSTM part processes time series data (dynamic time series features);

[0107] The outputs of the two networks are combined, and then the final output is obtained through the fully connected layer; the calculation formula of the spatiotemporal correlation model is as follows:

[0108] M(t,x,y,z)=LSTM(T v (t), T p (t))×3D-CNN(P xyz (x, y, z)

[0109] Among them, M(t, x, y, z) is the output result of the fully connected layer;

[0110] Furthermore, the operation process of the spatiotemporal correlation model includes:

[0111] Prepare time series and spatial data;

[0112] The time series data include the dynamic time series characteristic evaluation results of supply voltage fluctuation and partial discharge signal;

[0113] The spatial data is a three-dimensional PD distribution map;

[0114] Make sure the two types of data are aligned in the time dimension so that the model can correctly learn the spatiotemporal correlations;

[0115] 3D-CNN is used to process spatial data. It contains multiple 3D convolutional layers and pooling layers, which can effectively extract three-dimensional spatial features. The input of this part of the network is the three-dimensional spatial PD distribution map. The 3D convolution operation can capture the spatial distribution characteristics of partial discharge.

[0116] The LSTM layer is used to process time series data. It can capture long-term dependencies and is very suitable for processing the time series characteristics of power supply voltage fluctuations and partial discharge signals. The input of the LSTM layer is dynamic time series features, and the output is the time dimension encoding of these features.

[0117] The outputs of 3D-CNN and LSTM are fused by concatenating the outputs of the two networks. The fused features pass through two fully connected layers and finally output the prediction results.

[0118] Select the Adam optimizer and use the mean square error as the loss function to compile the model. Adam combines the advantages of the momentum method and the adaptive learning rate method, and can effectively handle sparse gradients and non-stationary targets. Considering that the partial discharge intensity is a continuous value, the mean square error (MSE) is a suitable choice;

[0119] Before training, the data set is divided into a training set, a validation set and a test set. For example, the present invention sets the training set to 70%, the validation set to 15%, and the test set to 15%, and then uses the prepared data set for training; during the training process, the model will simultaneously learn spatial features and temporal features, as well as the correlation between them;

[0120] Experiments were designed to verify whether the model truly captured the spatiotemporal correlation, including:

[0121] Fix spatial features, change temporal features, and observe output changes;

[0122] Fix the temporal features, change the spatial features, and observe the output changes;

[0123] Analyze the sensitivity of the model to different spatial and temporal combinations.

[0124] Specifically, the experimental design expansion includes time series impact experiment, spatial distribution impact experiment, power supply voltage fluctuation sensitivity experiment, time-space interaction experiment and extreme case test

[0125] The purpose of the time series impact experiment is to analyze the impact of different time modes on partial discharge. The method is to fix the spatial position, input different time series modes (such as normal, fluctuation, mutation, etc.), and observe the partial discharge prediction results;

[0126] The purpose of the spatial distribution effect experiment is to study the effect of different spatial distributions on partial discharge; the method is to fix the time characteristics, change the spatial distribution mode (such as concentrated, dispersed, random, etc.), and analyze the partial discharge prediction results;

[0127] The purpose of the supply voltage fluctuation sensitivity experiment is to evaluate the sensitivity of partial discharge to supply voltage fluctuations by gradually increasing the supply voltage fluctuation amplitude and recording the changes in the predicted intensity of partial discharge;

[0128] The purpose of the space-time interaction experiment is to explore the interaction between time and space characteristics.

[0129] The method is to change the time and space characteristics simultaneously and observe the nonlinear changes of the partial discharge prediction results;

[0130] The purpose of the extreme case test is to evaluate the performance of the model under extreme conditions. The method is to simulate extreme power supply voltage fluctuations and spatial distribution conditions and analyze the prediction results of the model. The specific data content is shown in the following table:

[0131] Table 1: Experimental results on time series impact

[0132]

[0133] Table 2: Experimental results on the effect of spatial distribution

[0134]

[0135]

[0136] Table 3: Test results of supply voltage fluctuation sensitivity

[0137]

[0138] These data tables provide examples of model prediction results and show the trend of partial discharge under different experimental conditions. By analyzing these data, we can draw the following preliminary conclusions:

[0139] The temporal pattern has a significant effect on partial discharge, especially in the case of sudden changes and large fluctuations;

[0140] Spatial distribution also affects the intensity of partial discharge, and concentrated distribution tends to lead to higher discharge intensity;

[0141] The fluctuation amplitude of the power supply voltage shows a nonlinear positive correlation with the intensity of partial discharge;

[0142] The sensitivity coefficient increases with the increase of the fluctuation amplitude of the supply voltage, indicating that the system is more vulnerable under larger fluctuations;

[0143] These experiments and data analysis help us better understand the impact of temporal and spatial characteristics on partial discharge, and provide an important basis for further optimizing power system operation and maintenance strategies.

[0144] S4: optimizing the partial discharge detection according to the time-space dimension analysis result, and detecting the FTU using the optimized partial discharge detection to obtain the partial discharge detection result.

[0145] Further, optimization of partial discharge detection includes:

[0146] According to the comprehensive evaluation results M(t,x,y,z) in the time and space dimensions, the parameter settings of the detection equipment are dynamically adjusted, including adjusting the sensitivity and sampling frequency of the sensor to adapt to the current voltage fluctuations and discharge signal characteristics;

[0147] Dynamic adjustments include:

[0148] Set the initial detection equipment parameters, including sensor sensitivity S 0 , sampling frequency f 0 , monitoring frequency ν 0 and coverage A 0 ;

[0149] Collect power supply voltage and partial discharge signal data in real time, extract dynamic time series features and analyze spatial distribution, and update the time and space dimension evaluation results M(t,x,y,z);

[0150] According to the latest spatiotemporal dimension assessment results M(t,x,y,z), the risk area is assessed and the assessment results are classified into low, medium and high risk areas;

[0151] When M(t,x,y,z)>first threshold θ1 When the assessment area is determined to be a high-risk area;

[0152] When the first threshold θ 1 ≥M(t,x,y,z)>second threshold θ 2 When the risk is too high, the assessment area will be judged as a medium-risk area;

[0153] When the second threshold θ 2 When ≥M(t,x,y,z), the assessment area is judged as a low-risk area;

[0154] Dynamically adjust the parameter settings of the detection equipment according to the risk level, including:

[0155] In low-risk areas, the initial parameter settings are maintained, making the sensor sensitivity S = S 0 , sampling frequency f = f 0 ;

[0156] In medium-risk areas, appropriately increase monitoring frequency and coverage, and adjust sensor sensitivity to S 1 and the sampling frequency is f 1 ; The sensitivity of the medium-risk area is adjusted according to the discharge intensity and frequency characteristics. The calculation formula is as follows:

[0157]

[0158] Among them, k 1 is the first sensitivity adjustment coefficient;

[0159] The sampling frequency in the medium-risk area is adjusted according to the voltage fluctuation frequency characteristics.

[0160]

[0161] Among them, m 1 is the first fluctuation frequency adjustment coefficient;

[0162] Significantly increase monitoring frequency and coverage in high-risk areas, and significantly improve sensor sensitivity 2 and sampling frequency f 2 ;

[0163] The high-risk area significantly increases the sensitivity, and the calculation formula is as follows:

[0164]

[0165] Among them, k 2 is the second sensitivity adjustment coefficient, k 2 >k 1 ;

[0166] The sampling frequency is significantly increased in high-risk areas, and the calculation formula is as follows:

[0167]

[0168] Among them, m 2 is the second fluctuation frequency adjustment coefficient, and m 2 >m 1 ;

[0169] For abnormal fluctuations or high-risk areas, the monitoring frequency and coverage of detection equipment are increased to ensure the accuracy of test results;

[0170] Dynamically select and adjust the detection algorithm, and choose the most suitable detection method for the current situation according to the real-time changes in spatiotemporal characteristics;

[0171] Dynamically select and adjust detection algorithms including,

[0172] A comprehensive library of partial discharge detection algorithms has been established, covering traditional spectrum analysis methods (such as Fourier transform and wavelet transform), pulse current method, ultra-high frequency method, acoustic emission method, optical detection method, and modern machine learning methods (such as support vector machine, random forest and deep learning model). This diverse algorithm library provides the basis for coping with various detection scenarios.

[0173] Continuously extract and analyze spatiotemporal features in real time. The time domain focuses on the discharge amplitude, frequency, and duration. The frequency domain analyzes the main frequency components and harmonic content. The spatial domain tracks the discharge location and distribution pattern. These features provide important basis for subsequent algorithm selection.

[0174] Based on different spatiotemporal characteristics, clear algorithm selection criteria should be established. For example, when a high content of high-frequency harmonics is detected, the spectrum analysis method is preferred; when the discharge location is dispersed, the ultra-high frequency method or acoustic emission method may be more suitable; and when the discharge pattern is complex, the machine learning method may perform better;

[0175] A weighted scoring mechanism is used to evaluate the applicability of each algorithm in the current situation in real time. The evaluation indicators include accuracy, running speed, adaptability and resource consumption. Each indicator has a corresponding weight. The system selects the most suitable algorithm for the current situation by calculating the total score.

[0176] After selecting the algorithm, optimize the algorithm parameters according to the current spatiotemporal characteristics. For example, for frequency analysis methods, the window size and frequency range may need to be adjusted; for machine learning methods, the learning rate and batch size may need to be optimized. This parameter optimization ensures that the algorithm can perform best under the current circumstances.

[0177] In some complex situations, a single algorithm may not meet the needs. In this case, consider the combination and fusion of multiple algorithms, select the best performing algorithms, assign different weights according to their advantages, and then perform weighted fusion of their results to obtain more reliable and comprehensive detection results;

[0178] Continuously monitor the performance of the selected algorithm, including detection accuracy and processing speed. If the performance indicator falls below the preset threshold, trigger the re-evaluation and selection process to ensure that the most suitable algorithm is always used;

[0179] Adopt online learning technology to enable the algorithm to continuously learn and update from new data. For algorithms that support incremental learning, new data is directly used for updating; for other algorithms, customized online learning methods are implemented. This continuous learning ensures that the algorithm can adapt to the changing environment.

[0180] An exception handling mechanism is set up to quickly switch to a preset backup solution when the main algorithm fails. This mechanism is implemented by trying to execute the main algorithm, verifying the validity of the results, and switching to the backup algorithm when necessary, which ensures the continuity and reliability of the detection process;

[0181] In addition to real-time adjustments, comprehensive evaluations are performed regularly. This process uses historical data to comprehensively evaluate all algorithms in the algorithm library to ensure that the currently used algorithm is still the best choice. If a better choice is found, corresponding adjustments are made.

[0182] This embodiment also provides a computer device, which may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 2 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a comprehensive evaluation method suitable for orderly access of large-scale distributed power sources is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse.

[0183] This embodiment also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: collecting power supply voltage and partial discharge signal data, and performing data preprocessing;

[0184] Perform time series analysis on power supply voltage fluctuation and partial discharge data to obtain dynamic time series feature evaluation results, and perform spatial distribution analysis on partial discharge signals to generate a three-dimensional spatial PD distribution map;

[0185] A spatiotemporal correlation model is constructed based on the dynamic timing feature evaluation results and the three-dimensional PD distribution diagram to analyze the correlation between the power supply voltage fluctuation and the partial discharge signal in the spatiotemporal dimension to obtain the spatiotemporal dimension analysis results;

[0186] The partial discharge detection is optimized through the time-space dimension analysis result, and the FTU is tested using the optimized partial discharge detection to obtain the partial discharge detection result.

[0187] Example 2

[0188] Reference Figure 3 , which is the second embodiment of the present invention. In order to implement the test method of the present invention, the patent of the present invention provides a circuit for a distribution network primary and secondary deep integration FTU partial discharge test method.

[0189] The circuit includes an AC-DC loop module, which converts alternating current into direct current and is used to limit the voltage of the power supply input to the AC-DC loop module to be lower than a set value, thereby playing a role of overvoltage protection;

[0190] A DC-DC loop module is electrically connected to the AC-DC loop module and is used to reduce the voltage fluctuation of the power supply input to the DC-DC loop module so that the partial discharge reaches a minimum value.

[0191] Preferably, the AC-DC loop module includes an overvoltage protection unit, which includes a capacitor C2, a capacitor C3 and a varistor; the first end of the capacitor C3 is electrically connected to the neutral line of the input power supply, the second end of the capacitor C3 is electrically connected to the live line of the input power supply, the first end of the capacitor C2 is electrically connected to the first end of the capacitor C3, the second end of the capacitor C2 is electrically connected to the second end of the capacitor C3, the first end of the varistor is electrically connected to the first end of the capacitor C2, and the second end of the varistor is electrically connected to the second end of the capacitor C2.

[0192] Preferably, the AC-DC loop module further includes a rectifier unit, a first end of the rectifier unit is electrically connected to a first end of the varistor, and a second end of the rectifier unit is electrically connected to a second end of the varistor.

[0193] Preferably, the AC-DC loop module further includes an anti-interference unit, which includes capacitor C5, capacitor C7, capacitor C8, capacitor C6, inductor L1, capacitor C1, capacitor C9, resistor R7, MOS tube Q1, diode D2, resistor R4, capacitor C10, resistor R9, Schottky diode D5, photocoupler U1, resistor R6, Schottky diode D4, thyristor QT1, resistor R5, resistor R8, resistor R10 and capacitor C4;

[0194] The first end of the capacitor C5 is electrically connected to the third end of the rectifier unit, the second end of the capacitor C5 is electrically connected to the fourth end of the rectifier unit, the first end of the capacitor C7 is electrically connected to the first end of the capacitor C5, the second end of the capacitor C7 is electrically connected to the second end of the capacitor C5, the first end of the capacitor C8 is electrically connected to the second end of the capacitor C5, and the first end of the capacitor C6 is electrically connected to the second end of the capacitor C5;

[0195] The first end of the inductor L1 is electrically connected to the first end of the resistor R6, the second end of the inductor L1 is electrically connected to the first end of the capacitor C1, the second end of the capacitor C1 is electrically connected to the first end of the capacitor C9, the second end of the capacitor C9 is electrically connected to the second end of the capacitor C6 and the source of the MOS tube Q1, the first end of the resistor R7 is electrically connected to the first end of the capacitor C1, the second end of the resistor R7 is electrically connected to the second end of the capacitor C1 and the drain of the MOS tube Q1, the gate of the MOS tube Q1 is electrically connected to the cathode of the Schottky diode D5, and the anode of the Schottky diode D5 is electrically connected to the source of the MOS tube and grounded;

[0196] The anode of the diode D2 is electrically connected to the first end of the inductor L1, the cathode of the diode D2 is electrically connected to the first end of the resistor R4, the second end of the resistor R4 is electrically connected to the first end of the capacitor C10, the second end of the capacitor C10 is electrically connected to the anode of the Schottky diode D5, the first end of the resistor R9 is electrically connected to the cathode of the Schottky diode D5, and the second end of the resistor R9 is electrically connected to the anode of the Schottky diode D5;

[0197] The second end of the resistor R4 is electrically connected to the photosensitive device of the photocoupler U1, the photosensitive device of the photocoupler U1 is electrically connected to the first end of the resistor R9, the first end of the resistor R6 is electrically connected to the first end of the resistor R4, the second end of the resistor R6 is electrically connected to the light-emitting device of the photocoupler U1 and the cathode of the Schottky diode D4, the light-emitting device of the photocoupler U1 is electrically connected to the cathode of the thyristor QT1, and the anode of the thyristor QT1 is electrically connected to the anode of the Schottky diode D4 and the second end of the capacitor C10;

[0198] The first end of the resistor R5 is electrically connected to the first end of the resistor R6, the second end of the resistor R5 is electrically connected to the first end of the resistor R8, the second end of the resistor R8 is electrically connected to the control electrode of the thyristor QT1 and the first end of the resistor R10, the second end of the resistor R10 is electrically connected to the anode of the thyristor QT1, the first end of the capacitor C4 is electrically connected to the first end of the resistor R5, and the second end of the capacitor C4 is electrically connected to the second end of the resistor R10.

[0199] Preferably, the DC-DC loop module includes a bidirectional voltage regulator diode D7, a capacitor C14, a resistor R11, a resistor R14, a resistor R13, a capacitor C16, a capacitor C100, a resistor R15, a chip U2, a capacitor C12, a Schottky diode D6, an inductor L2, a capacitor CC1, a capacitor CC2, a capacitor C13, a resistor R12, a resistor R16, a thermistor Rt, a capacitor C15, a diode D9, a voltage regulator diode D8, a resistor R19, an integrated circuit chip QT2, a resistor R17, a resistor R20, a resistor R21 and a resistor R18;

[0200] One anode of the bidirectional voltage regulator diode D7 is grounded, the other anode of the bidirectional voltage regulator diode D7 is electrically connected to the first end of the capacitor C14 and the VIN pin of the chip U2, the first end of the resistor R17 is electrically connected to the other anode of the bidirectional voltage regulator diode D7, the second end of the resistor R17 is electrically connected to the first end of the resistor R20, and the second end of the resistor R20 is electrically connected to the first end of the resistor R21;

[0201] The first end of the capacitor C14 is electrically connected to the first end of the resistor R11 and the first end of the resistor R17, the second end of the capacitor C14 is grounded, the second end of the resistor R11 is electrically connected to the first end of the resistor R14 and the EN pin of the chip U2, the second end of the resistor R14 is electrically connected to the ETPAD pin of the chip U2, the first end of the resistor R13 is electrically connected to the first end of the capacitor C100 and the COMP pin of the chip U2, the second end of the resistor R13 is electrically connected to the first end of the capacitor C16, the second end of the capacitor C16 is electrically connected to the second end of the resistor R14, the second end of the capacitor C100 is electrically connected to the second end of the capacitor C16 and grounded, the first end of the resistor R15 is electrically connected to the RT / CLK pin of the chip U2, and the second end of the resistor R15 is electrically connected to the second end of the capacitor C100;

[0202] The BOOT pin of the chip U2 is electrically connected to the first end of the capacitor C12, the second end of the capacitor C12 is electrically connected to the first end of the inductor L2 and the SW pin of the chip U2, the second end of the capacitor C12 is electrically connected to the cathode of the Schottky diode D6, the anode of the Schottky diode D6 is grounded, the second end of the inductor L2 is electrically connected to the first end of the resistor R12, the second end of the resistor R12 is electrically connected to the first end of the resistor R16, the first end of the capacitor C15 and the FB pin of the chip U2, the second end of the capacitor C15 is grounded, the second end of the resistor R16 is electrically connected to the first end of the thermistor Rt, and the second end of the thermistor Rt is grounded;

[0203] The first end of the resistor R18 is electrically connected to the first end of the resistor R17, the second end of the resistor R18 is electrically connected to the cathode of the integrated circuit chip QT2 and the first end of the resistor R19, the anode of the integrated circuit chip QT2 is electrically connected to the second end of the resistor R21 and is grounded, the reference electrode of the integrated circuit chip QT2 is electrically connected to the first end of the resistor R21, the second end of the resistor R19 is electrically connected to the cathode of the Zener diode D8, the anode of the Zener diode D8 is electrically connected to the anode of the diode D9, and the cathode of the diode D9 is electrically connected to the first end of the capacitor C15;

[0204] The first end of the capacitor CC1 is electrically connected to the second end of the inductor L2, the first end of the capacitor CC2 and the first end of the capacitor C13. The second end of the capacitor CC1 is electrically connected to the second end of the capacitor CC2 and the second end of the capacitor C13 and is grounded.

[0205] Preferably, the DC-DC loop module includes a capacitor C11, a first end of the capacitor C11 is electrically connected to a first end of the resistor R11, and a second end of the resistor R11 is grounded.

[0206] The beneficial effects of the circuit are as follows: when the input DC voltage is higher than the set value, the cathode and anode of the thyristor QT1 are turned on to reduce the input DC voltage, the photocoupler and the MOS tube work, and the inductor L1 and the resistor R7 are connected in series in the power supply circuit of the MOS tube to further play an overvoltage protection role, reduce voltage fluctuations, and prevent current mutations when the MOS tube is turned on, so that the circuit is in a stable working state; when the input DC voltage rises rapidly to the set value, the cathode and anode of the integrated circuit chip QT2 are turned on, the diode is in a cut-off state, the DC-DC loop module enters a normal working state, and the input voltage fluctuates only in a small range, thereby reducing the fluctuation of the input DC voltage and minimizing the partial discharge, that is, meeting the detection standard.

[0207] Example 3

[0208] Reference Figure 1 - Figure 3 , which is the third embodiment of the present invention, provides a distribution network primary and secondary deep integration FTU partial discharge test method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0209] The specific implementation steps are as follows:

[0210] Install voltage sensors and partial discharge test instruments on the target 10kV distribution network lines.

[0211] Power quality monitoring equipment is configured to obtain the instantaneous value of the supply voltage, voltage fluctuation frequency, voltage fluctuation amplitude, and harmonic components.

[0212] The partial discharge test instrument is used to collect the discharge intensity, discharge frequency, discharge time interval, discharge waveform and discharge position.

[0213] Data cleaning is performed, including using a low-pass filter to remove high-frequency noise in the power supply voltage and partial discharge signal, and performing Fourier transform on the harmonic components to remove interference signals of non-fundamental components.

[0214] Handle outliers, use the box plot method to detect and remove outliers, and use interpolation to fill in missing or abnormal data points.

[0215] The dynamic timing characteristics of the power supply voltage and partial discharge signals are extracted, the time series model is constructed, and the dynamic timing characteristics are evaluated.

[0216] The Kriging interpolation method is used to generate a three-dimensional PD distribution map, and the spatial distribution data of the partial discharge signal is used for interpolation calculation to obtain the spatial distribution of the discharge intensity.

[0217] A hybrid model architecture combining long short-term memory network (LSTM) and three-dimensional convolutional neural network (3D-CNN) is designed to process time series data and spatial data.

[0218] The extracted dynamic time series features and three-dimensional spatial PD distribution map are used to build a spatiotemporal correlation model to perform feature learning and correlation analysis in the spatiotemporal dimension.

[0219] According to the comprehensive evaluation results in the time and space dimensions, the parameter settings of the detection equipment, including the sensitivity and sampling frequency of the sensor, are dynamically adjusted to adapt to the current voltage fluctuations and discharge signal characteristics.

[0220] Dynamically select and adjust the detection algorithm, and choose the detection method that best suits the current situation based on changes in temporal and spatial characteristics.

[0221] Table 4 Test data record table

[0222]

[0223]

[0224] Through the above table data, we can see the key parameters of the power supply voltage and partial discharge signal at each test point. The following conclusions can be drawn from the analysis of these data:

[0225] The instantaneous voltage value of each test point is basically stable at about 230V, with a small fluctuation, indicating that the present invention has significant advantages in the stability control of the power supply voltage.

[0226] The voltage fluctuation frequency is basically maintained at 50Hz, with slight fluctuations, which is within the normal voltage fluctuation range.

[0227] There is an obvious correlation between the voltage fluctuation amplitude and the parameters such as discharge intensity, frequency, and time interval. For example, the voltage fluctuation amplitude at test point 4 is large (5.2V), and the corresponding discharge intensity and frequency are also relatively high, indicating that voltage fluctuation has a significant impact on partial discharge.

[0228] The harmonic component of test point 5 is relatively high (3.2%), and its discharge intensity and frequency are also relatively high, indicating that the harmonic component has a significant impact on partial discharge. The present invention can effectively identify and control the impact of harmonics on partial discharge through harmonic analysis.

[0229] The discharge intensity, frequency and time interval of each test point show the differences in the spatial distribution of partial discharge. The present invention can accurately capture these differences and make targeted optimization adjustments through the construction of a three-dimensional PD distribution map and a spatiotemporal correlation model.

[0230] By comparing with the prior art, the present invention has creativity and novelty in the following aspects:

[0231] Dynamically adjust detection equipment parameters: According to the comprehensive evaluation results in the time and space dimensions, dynamically adjust the sensitivity and sampling frequency of the sensor to improve the sensitivity and accuracy of detection.

[0232] Multi-algorithm comprehensive detection: By building a comprehensive library containing multiple partial discharge detection algorithms, the optimal algorithm can be selected for different detection scenarios, improving the adaptability and efficiency of detection.

[0233] Spatiotemporal correlation analysis: A hybrid model combining LSTM and 3D-CNN can simultaneously process temporal and spatial data, accurately capture the correlation between voltage fluctuations and partial discharge signals in the spatiotemporal dimensions, and provide more comprehensive and in-depth analysis results.

[0234] In summary, the present invention significantly improves the accuracy and reliability of partial discharge detection in 10kV distribution network through an innovative dynamic adjustment mechanism and a multi-algorithm comprehensive detection method, thus providing a strong guarantee for the safe operation of the distribution network.

[0235] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0236] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0237] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0238] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0239] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0240] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0241] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A distribution network primary and secondary deep integration FTU partial discharge test method, characterized by: include, Collect power supply voltage and partial discharge signal data, and perform data preprocessing; Perform time series analysis on power supply voltage fluctuations and partial discharge signals to obtain dynamic time series feature evaluation results, and perform spatial distribution analysis on partial discharge signals to generate a three-dimensional spatial PD distribution map; A spatiotemporal correlation model is constructed based on the dynamic timing feature evaluation results and the three-dimensional PD distribution diagram to analyze the correlation between the power supply voltage fluctuation and the partial discharge signal in the spatiotemporal dimension to obtain the spatiotemporal dimension analysis results; The partial discharge detection is optimized through the time-space dimension analysis result, and the FTU is tested using the optimized partial discharge detection to obtain the partial discharge detection result.

2. The distribution network primary and secondary deep integration FTU partial discharge test method as claimed in claim 1, characterized in that: The power supply voltage and the partial discharge signal data are obtained through an AC-DC loop module and a DC-DC loop module; the power supply voltage includes an instantaneous voltage value, a voltage fluctuation frequency, a voltage fluctuation amplitude, and harmonic components; the partial discharge signal data includes a discharge intensity, a discharge frequency, a discharge time interval, a discharge waveform, and a discharge position.

3. The distribution network primary and secondary deep integration FTU partial discharge test method as claimed in claim 2 is characterized by: The time series analysis includes extracting dynamic time series features of the power supply voltage and the partial discharge signal, and using the extracted dynamic time series features to construct a time series model, and performing dynamic time series feature evaluation through the time series model; the spatial distribution analysis includes generating a three-dimensional spatial PD distribution map using a Kriging interpolation method.

4. The distribution network primary and secondary deep integration FTU partial discharge test method as claimed in claim 3 is characterized by: The dynamic timing characteristics of the power supply voltage include the mean, variance, maximum value, minimum value and standard deviation of the instantaneous voltage value; The dynamic time series characteristics of the partial discharge signal include the mean, variance, maximum, minimum and standard deviation of the discharge intensity; Use the extracted features to build a time series model and evaluate the dynamic time series features. The calculation formula is as follows: Among them, T v (t) is the dynamic timing characteristic evaluation result of the supply voltage, T p (t) is the dynamic timing characteristic evaluation result of the partial discharge signal; The spatial distribution analysis comprises: The Kriging interpolation method is used to generate a three-dimensional PD distribution map. The distribution map calculation formula is as follows: Among them, μ is the global mean, λ i is the Kriging weight, P xyz (x, y, z) is the spatial distribution data of partial discharge intensity.

5. The distribution network primary and secondary deep integration FTU partial discharge test method as claimed in claim 4 is characterized by: The construction of the spatiotemporal correlation model includes combining a hybrid model of a long short-term memory network LSTM and a three-dimensional convolutional neural network 3D-CNN to simultaneously process time series data and spatial data, and capture the characteristics of power supply voltage fluctuations and partial discharge signals in time and space dimensions; The operation process of the spatiotemporal correlation model includes preparing time series data and spatial data, processing spatial data using 3D-CNN, processing time series data using LSTM layer, fusing the outputs of 3D-CNN and LSTM, and obtaining the final output through the fully connected layer.

6. The distribution network primary and secondary deep integration FTU partial discharge test method as claimed in claim 5, characterized in that: The optimization of partial discharge detection comprises: Dynamically adjust the parameter settings of the detection equipment according to the comprehensive evaluation results of the time and space dimensions, including adjusting the sensitivity and sampling frequency of the sensor; The dynamic adjustment includes: Set the initial detection equipment parameters, including sensor sensitivity S0, sampling frequency f0, monitoring frequency v0 and coverage range A0; Collect power supply voltage and partial discharge signal data in real time, extract dynamic time series features and analyze spatial distribution, and update the time and space dimension evaluation results M(t,x,y,z); According to the latest spatiotemporal dimension assessment results M(t,x,y,z), the risk area is assessed and the assessment results are classified into low, medium and high risk areas; When M(t,x,y,z)>the first threshold θ1, the assessment area is determined as a high-risk area; When the first threshold θ1≥M(t,x,y,z)>the second threshold θ2, the assessment area is determined as a medium-risk area; When the second threshold θ2 ≥ M(t, x, y, z), the evaluation area is determined as a low-risk area; Dynamically adjust the parameter settings of the detection equipment according to the risk level, including: The initial parameter settings are maintained in the low-risk area, so that the sensor sensitivity S = S0 and the sampling frequency f = f0; In medium-risk areas, the monitoring frequency and coverage should be appropriately increased, and the sensor sensitivity should be adjusted to S1 and the sampling frequency should be adjusted to f1. In medium-risk areas, the sensitivity should be adjusted according to the discharge intensity and frequency characteristics. The calculation formula is as follows: Wherein, k1 is the first sensitivity adjustment coefficient; The sampling frequency in the medium-risk area is adjusted according to the voltage fluctuation frequency characteristics. Wherein, m1 is the first fluctuation frequency adjustment coefficient; Significantly increase monitoring frequency and coverage in high-risk areas, and improve sensor sensitivity S2 and sampling frequency f2; The high-risk area significantly increases the sensitivity, and the calculation formula is as follows: Wherein, k2 is the second sensitivity adjustment coefficient, k2>k1; The sampling frequency is significantly increased in high-risk areas, and the calculation formula is as follows: Wherein, m2 is the second fluctuation frequency adjustment coefficient, and m2>m1; Dynamically select and adjust the detection algorithm, and choose the detection method for the current situation based on real-time changes in spatiotemporal characteristics.

7. A distribution network primary and secondary deep integration FTU partial discharge test circuit, based on the distribution network primary and secondary deep integration FTU partial discharge test method according to any one of claims 1 to 6, characterized in that: Including AC-DC loop module and DC-DC loop module; The AC-DC loop module converts alternating current into direct current and is used to limit the voltage of the power supply input to the AC-DC loop module to be lower than a set value, thereby performing overvoltage protection; The DC-DC loop module is electrically connected to the AC-DC loop module and is used to reduce the fluctuation of the voltage input to the DC-DC loop module power supply, so that the partial discharge reaches a minimum value.

8. The distribution network primary and secondary deep integration FTU partial discharge test circuit as claimed in claim 7, characterized in that: The AC-DC loop module includes an overvoltage protection unit, and the overvoltage protection unit includes a capacitor C2, a capacitor C3, a varistor and a rectifier unit; The DC-DC loop module includes a bidirectional voltage regulator diode D7, a capacitor C14, a resistor R11, a resistor R14, a resistor R13, a capacitor C16, a capacitor C100, a resistor R15, a chip U2, a capacitor C12, a Schottky diode D6, an inductor L2, a capacitor CC1, a capacitor CC2, a capacitor C13, a resistor R12, a resistor R16, a thermistor Rt, a capacitor C15, a diode D9, a voltage regulator diode D8, a resistor R19, an integrated circuit chip QT2, a resistor R17, a resistor R20, a resistor R21 and a resistor R18.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distribution network primary and secondary deep integration FTU partial discharge testing method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distribution network primary and secondary deep integration FTU partial discharge testing method described in any one of claims 1 to 7 are implemented.

Images