Power distribution automation device supervisory collection system

Abstract

The application provides a power distribution automation equipment supervision acquisition system, which comprises the following steps: time synchronization and space registration are realized by adopting a space-time alignment algorithm on infrared signals, vibration signals and electrical signals; thermal imaging analysis is performed on the infrared signals after synchronization and registration, an overheated area is automatically framed and selected, and a temperature gradient is calculated; vibration signal analysis is performed on the vibration signals after synchronization and registration, vibration entropy is calculated; partial discharge analysis is performed on the electrical signals, and the number of discharge pulses is obtained; the temperature gradient, the vibration entropy and the number of discharge pulses are weighted and fused, the fused features are input into an LSTM model for time sequence correlation analysis, the fault type and the fault type probability are obtained, and the 750KV substation is warned according to the fault type probability. The application provides an automation equipment supervision acquisition system for the 750KV substation, and has the advantages of low missed detection rate and high accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power, and particularly relates to a power distribution automation equipment supervision and acquisition system. BACKGROUND

[0002] The SCADA (Supervisory Control And Data Acquisition) system is a computer-based DCS (Distributed Control System) and power automation monitoring system, and can be applied to data acquisition and monitoring control in the field of electric power.

[0003] In the electric power system, the SCADA system is most widely applied and its technology is most mature. The SCADA system plays an important role in the remote control system, can monitor and control the running equipment in the field, and realizes data acquisition, equipment control, measurement, parameter adjustment, and various signal alarms, and the RTU (Remote Terminal Unit) and the FTU (Feeder Terminal Unit) are important components of the SCADA system, and play a very important role in the current substation integrated automation construction.

[0004] The SCADA system is a computer-based production process control and dispatching automation system, can monitor and control the running equipment in the field, and realizes data acquisition, equipment control, measurement, parameter adjustment, and various signal alarms. SUMMARY

[0005] The present application provides a power distribution automation equipment supervision and acquisition system, which mainly aims to provide an automation equipment supervision and acquisition system for a 750KV substation, improves supervision accuracy, and reduces false alarm rate.

[0006] The present application provides a power distribution automation equipment supervision and acquisition system, which mainly aims to provide an automation equipment supervision and acquisition system for a 750KV substation, improves supervision accuracy, and reduces false alarm rate.

[0007] The infrared thermal imager obtains an infrared signal, the acoustic vibration sensor obtains a vibration signal, and the current sensor and the voltage sensor obtain an electrical signal;

[0008] The time-space alignment algorithm is used to realize time synchronization and space registration of the infrared signal, the vibration signal, and the electrical signal.

[0009] The infrared signals after synchronization registration are subjected to thermal imaging analysis, overheat regions are automatically framed and selected, and temperature gradients are calculated, the vibration signals after synchronization registration are subjected to vibration signal analysis, vibration entropy is calculated, the electrical signals are subjected to partial discharge analysis, and the number of discharge pulses is obtained;

[0010] The temperature gradient, the vibration entropy and the number of discharge pulses are input into an LSTM model for time sequence correlation analysis, a fault type and a fault type probability are obtained, and the 750KV substation is warned according to the fault type probability.

[0011] Further, the infrared signals, the vibration signals and the electrical signals are subjected to time and space alignment algorithm to realize time synchronization and space registration, and the steps include:

[0012] The sampling rate of the electrical signals is taken as a reference frequency, the infrared signals are subjected to cubic spline interpolation, the sampling rate of the infrared signals is improved to the reference frequency, the dynamic time warping algorithm is applied to the vibration signals and the electrical signals, mechanical vibration transmission delay is compensated, and time synchronization function is realized;

[0013] The temperature matrix output by the infrared thermal imager is converted into a three-dimensional point cloud, and the point cloud correspondence between the infrared signals and the real equipment in the 750KV substation is realized according to the three-dimensional point cloud and a transformation matrix, wherein the transformation matrix is obtained according to the installation position of the infrared thermal imager, the transformation matrix represents the correspondence between the three-dimensional point cloud and the real physical space coordinates, and space synchronization function is realized.

[0014] Further, the infrared signals after synchronization registration are subjected to thermal imaging analysis, overheat regions are automatically framed and selected, and temperature gradients are calculated, and the steps include:

[0015] For the transformer, the GIS device and the overhead line, time difference is performed on 10 consecutive thermal images of each device respectively, and the temperature change matrix, the binary mask and the time energy graph of each device are obtained;

[0016] The temperature change matrix of each device is superimposed with the infrared signals of each device, the superimposed data are input into a YOLOv5 model, the overheat regions of each device are automatically framed and selected, and the temperature gradient of each device is output;

[0017] According to the overheat regions of each device and the transformation matrix and the rotation and translation matrix, the standby infrared fault points of each device are matched.

[0018] Further, the vibration signals after synchronization registration are subjected to vibration signal analysis, and the vibration entropy is calculated, and the steps include:

[0019] The vibration signal after synchronous registration is decomposed by wavelet packet to extract a power frequency component and a high frequency residual signal;

[0020] The high frequency residual signal is decomposed by EMD to obtain an intrinsic mode function component representing mechanical looseness;

[0021] The nonlinear kinetic entropy of the screened intrinsic mode function component is calculated to obtain the vibration entropy and the sideband energy ratio;

[0022] The instantaneous frequency of the screened intrinsic mode function component is obtained by Hilbert transform, and the modulation phenomenon is detected according to the instantaneous frequency and the sideband energy ratio, thereby obtaining the standby vibration fault point.

[0023] Further, the screened intrinsic mode function component is obtained by the following steps:

[0024] The false components of the intrinsic mode function are removed by the correlation coefficient method, and the first N intrinsic mode function components with cumulative energy > 85% are retained, N being a positive integer.

[0025] Further, the electrical signal is analyzed for partial discharge to obtain the number of discharge pulses, and the step comprises:

[0026] The electrical signal is pulse detected to generate the number of discharge pulses and a PRPD spectrum;

[0027] The PRPD spectrum is input into an improved VGG16 model to obtain a discharge type probability and a pulse density prediction value of each phase interval, and the full connection layer in the improved VGG16 model is replaced by an attention mechanism module;

[0028] According to the discharge type probability and the pulse density prediction value of each phase interval, a standby electrical fault point is generated.

[0029] Further, the temperature gradient, the vibration entropy and the number of discharge pulses are input into an LSTM model for time series correlation analysis to obtain a fault type and a fault type probability, and the step comprises:

[0030] The temperature gradient, the vibration entropy and the number of discharge pulses are input into an LSTM model, if the temperature gradient exceeds a preset gradient threshold and the vibration entropy is within a normal range, the fault type corresponding to the temperature gradient and the fault type probability are output, and a secondary warning is performed;

[0031] If the vibration entropy exceeds a preset baseline and the temperature gradient is within a normal range, the fault type corresponding to the vibration entropy and the fault type probability are output, and a secondary warning is performed.

[0032] Further, the temperature gradient, the vibration entropy and the number of discharge pulses are input into the LSTM model for time sequence correlation analysis to obtain a fault type and a fault type probability.

[0033] If the temperature gradient exceeds the preset gradient threshold and the vibration entropy exceeds the preset baseline, the fault type is mechanical looseness leading to increased contact resistance, and a first-level warning is performed.

[0034] If the temperature gradient exceeds the preset gradient threshold and the number of discharge pulses exceeds the preset pulse threshold, the fault type is insulation deterioration causing partial discharge, and a first-level warning is performed.

[0035] If the temperature gradient exceeds the preset gradient threshold, the vibration entropy exceeds the preset baseline, and the number of discharge pulses exceeds the preset pulse threshold, the fault type is an arc fault precursor, and a first-level warning is performed.

[0036] Further, the infrared thermal imager is installed at the transformer, circuit breaker, disconnecting switch and bus joint of the 750KV substation and transmission line, the acoustic vibration sensor is installed on the large equipment shell of the 750KV substation, the current sensor is installed at the current transformer, and the voltage sensor is installed at the voltage transformer of the bus, incoming line and outgoing line interval of the 750KV substation.

[0037] The power distribution automation equipment supervision and acquisition system provided by the application can realize early warning of the 750KV substation by installing infrared thermal imagers, acoustic vibration sensors, current sensors and voltage sensors at various places of the 750KV substation and processing the collected data. The acquired infrared signals, vibration signals and electrical signals are first time-synchronized and space-registered by using a time-space registration algorithm; then the registered infrared signals are analyzed for thermal imaging to automatically frame the overheated area and calculate the temperature gradient, the registered vibration signals are analyzed to obtain the vibration entropy, and the electrical signals are analyzed for partial discharge to obtain the number of discharge pulses; finally, the temperature gradient, the vibration entropy and the number of discharge pulses are weighted and fused, the fused features are input into an LSTM model for time sequence correlation analysis to obtain the fault type and the fault probability, and early warning is performed.

[0038] The advantages of the embodiments of the application are as follows:

[0039] (1) In the embodiments of the application, the detection sensitivity of weak temperature rise signals can be improved by 4 times by time difference, and the accuracy of early warning is improved.

[0040] (2), the temperature change matrix in the embodiment of the application can inhibit reflection interference, solves the problem that surface discharge heat is easily misjudged as sunlight reflection in the traditional method, thereby reducing the missed detection rate.

[0041] (3), the embodiment of the application enhances the saliency of key alarm through weight fusion, and uses LSTM to mine the time sequence causal relationship between multiple features, realizes the upgrade from "single-point alarm" to "state evolution prediction", and improves the accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 An execution process flowchart of a power distribution automation equipment supervision and acquisition system provided in the embodiment of the application is provided.

[0043] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0044] The embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar notations represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0045] In order to enable persons skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor fall within the scope of protection of the present application.

[0046] In the embodiments of the present application, at least one means one or more, and multiple means two or more than two. In the description of the present application, the terms "first", "second", "third" and the like are only used for distinguishing purposes of description, and cannot be understood as indicating or implying relative importance, nor can be understood as indicating or implying order. In addition, the terms "first", "second" are only used for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more features. In the description of the present application, the meaning of "multiple" is two or more than two, unless otherwise specifically limited.

[0047] Reference to "one embodiment" or "some embodiments" or "one implementation" or "some implementations" etc. in the present description means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearance of the phrases in various places in the specification are not therefore meant to be taken as a sign that all of the features, structures, or characteristics etc. described in connection with the embodiment are included in one of every embodiment. The terms "including", "containing", "having" and variations thereof as used in the specification are meant to encompass the case where nothing else is included, contained or having.

[0048] It should be noted that the "connection" in the embodiments of the application can be understood as an electrical connection, and the connection between two electrical elements can be direct or indirect connection between the two electrical elements. For example, A and B are connected, which can be direct connection between A and B, or indirect connection between A and B through one or more other electrical elements.

[0049] Figure 1 The execution process flow chart of the power distribution automation equipment supervision acquisition system provided by the embodiments of the application is shown in FIG. 7, in which an infrared thermal imager, an acoustic vibration sensor, a current sensor and a voltage sensor are installed at a 750KV substation, and the supervision acquisition system realizes the early warning function of the 750KV substation through the following steps: Figure 1

[0050] Specifically, the infrared thermal imager is installed at the transformer, circuit breaker, disconnector and bus joint of the 750KV substation and power transmission line, the acoustic vibration sensor is installed on the shell of the large equipment of the 750KV substation, the current sensor is installed at the current transformer, and the voltage sensor is installed at the voltage transformer between the bus, incoming line and outgoing line interval of the 750KV substation.

[0051] S10, acquiring infrared signals through the infrared thermal imager, acquiring vibration signals through the acoustic vibration sensor, and acquiring electrical signals through the current sensor and the voltage sensor;

[0052] The infrared signals of the infrared thermal imager, the vibration signals of the acoustic vibration sensor, the current signals of the current sensor and the voltage signals of the voltage sensor are collected according to the collection frequency, and the current signals and the voltage signals are collectively referred to as electrical signals.

[0053] In order to solve the data time synchronization and spatial registration problems of sensors with different sampling rates (thermal imaging 1Hz, vibration 10kHz, electrical quantity 4kHz), the data needs to be time-synchronized and spatially registered first.

[0054] S20, time synchronization and spatial registration are realized by using a time-space alignment algorithm on the infrared signals, the vibration signals and the electrical signals;

[0055] ​S21, using a sampling rate of the electrical signal as a reference frequency, performing cubic spline interpolation on the infrared signal to increase the sampling rate of the infrared signal to the reference frequency, applying a dynamic time warping algorithm to the vibration signal and the electrical signal to compensate for mechanical vibration transmission delay and realize time synchronization function;

[0056] In the embodiment of the application, the sampling rate of the electrical signal is used as the reference frequency, the infrared signal and other low-frequency signals are interpolated by cubic spline interpolation, and the infrared signal is increased from 1HZ to 4KHZ to maintain the same frequency as the electrical quantity; then, dynamic warping matching is performed on the vibration signal and the electrical signal, a dynamic time warping (DTW) algorithm is applied, and mechanical vibration transmission delay is compensated.

[0057] S22, converting the temperature matrix output by the infrared thermal imager into a three-dimensional point cloud, and realizing the point cloud correspondence relationship between the infrared signal and the real equipment in the 750KV substation according to the three-dimensional point cloud and the transformation matrix, wherein the transformation matrix is obtained according to the installation position of the infrared thermal imager, the transformation matrix represents the correspondence relationship between the three-dimensional point cloud and the real physical space coordinates, and the space synchronization function is realized.

[0058] The temperature matrix (640*512 pixels) output by the infrared thermal imager is essentially two-dimensional image data, which only contains pixel coordinates (x, y) and temperature values (T); after being converted into a three-dimensional point cloud (x, y, T), a mapping relationship can be established with the real physical space coordinates of the equipment. For example, if a high-temperature point is found at the position (120, 300) of the point cloud, the corresponding flange bolt position on the actual equipment can be determined through the transformation matrix.

[0059] The point cloud data retains the temperature distribution of the geometric topological structure of the equipment, such as the temperature distribution of the protruding parts such as the sleeve and the flange, which is the basis for realizing the spatial registration with the vibration sensor. For example, when iterative optimization is performed, the Euclidean distance between the high-temperature area in the point cloud and the monitoring point of the vibration sensor needs to be calculated, and the real spatial distance cannot be obtained by using only two-dimensional images. The monitoring point of the vibration sensor is usually marked with three-dimensional coordinates (such as 1.5m from the ground on the side of the transformer oil tank), and the infrared data must be raised to three-dimensional space to establish the correlation.

[0060] The transformation relationship from the thermal imager coordinate system to the global coordinate system of the substation is determined through joint calibration of the thermal imager and the total station; the rotation and translation matrix is optimized, and during the operation of the equipment, the rotation matrix R and the translation matrix t are iteratively optimized through the ICP (Iterated Closest Points) algorithm, so that the distance between the point cloud and the feature points of the equipment model is minimized.

[0061] The pixel coordinates in a two-dimensional thermal image cannot directly reflect the actual spatial distance. After point cloud conversion, the temperature gradient in the real space can be calculated, which is a key indicator for assessing equipment overheating. The ICP algorithm requires the input to be a three-dimensional spatial point set, as two-dimensional data cannot be used for rotation and translation calculations. The acceleration direction of the vibration sensor needs to be aligned with the spatial direction of the temperature distribution, which requires three-dimensional coordinates.

[0062] S30, perform thermal imaging analysis on the synchronously registered infrared signal, automatically select the overheated area and calculate the temperature gradient, perform vibration signal analysis on the synchronously registered vibration signal, calculate the vibration entropy, and perform partial discharge analysis on the electrical signal to obtain the number of discharge pulses;

[0063] Specifically, the steps of performing thermal imaging analysis on the synchronously registered infrared signal, automatically selecting overheated areas, and calculating temperature gradients include:

[0064] S31A: For transformers, GIS equipment and overhead lines, time difference is performed on 10 consecutive frames of thermal images of each device to obtain the temperature change matrix, binarized mask and time-domain energy map of each device.

[0065] In power equipment monitoring scenarios, "equipment" refers to an electrical unit with independent functions that is continuously monitored, and the same spatial reference point must be maintained when photographing the equipment. For each device, a temporal difference algorithm is executed for 10 consecutive frames (typically a 30-second time span, 1Hz sampling rate) to obtain the device's temperature change matrix, binarized mask, and temporal energy map. The temperature change matrix represents the rate of temperature rise or fall for each pixel and is often used to capture rapidly developing hotspots. The binarized mask highlights areas of significant change and can be represented by temperature changes exceeding a temperature threshold, often used to reduce background interference. The temporal energy map represents the cumulative temperature change and is frequently used to identify persistent degradation, such as contact oxidation.

[0066] For example, in a typical differential result example, under normal conditions, the temperature change matrix is ​​randomly distributed within ±0.3℃; under fault conditions, there is a continuous positive growth area with ΔT>1℃ / s in a local area (loose bolts), or a ring-shaped diffusion ΔT distribution (insufficient oil in the casing).

[0067] In this embodiment of the invention, the detection sensitivity of weak temperature rise signals can be improved by 4 times through time difference.

[0068] S32A superimposes the temperature change matrix of each device with the infrared signal of each device, inputs the superimposed data into the YOLOv5 model, automatically selects the overheated area of ​​each device and outputs the temperature gradient of each device.

[0069] The infrared signal of each device is the original thermal image collected by the infrared thermal imager, the temperature change matrix of each device is superimposed with the infrared signal to form 4-channel input, that is, [R, G, B, Delta T], wherein RGB is a pseudo-color thermal image, and Delta T is a normalized value of a difference matrix.

[0070] The superimposed data is input into a YOLOv5 model, the YOLOv5 model includes two detection heads, one of which outputs a predicted bounding box and a category, and the other directly outputs a temperature rise rate. That is, the YOLOv5 model outputs an overheating area, a defect type and a temperature rise rate, the data format of the overheating area is [x_center, y_center, w, h], and the defect type is a classification probability vector, wherein x_center represents the horizontal coordinate of the center point of the bounding box, y_center represents the vertical coordinate of the center point of the bounding box, w represents the width of the bounding box, and h represents the height of the bounding box.

[0071] The YOLOv5 model is a target detection model based on deep learning, which can realize rapid detection and positioning of multiple targets in an image by using modules such as a backbone network, a detection head and a loss function. The YOLOv5 model includes a backbone network, a feature fusion network and a detection head.

[0072] In the embodiment of the application, the temperature change matrix can suppress reflection interference, and solves the problem that surface discharge heating is easily misjudged as sunlight reflection in the traditional method, thereby reducing the missed detection rate.

[0073] S33A, according to the overheating area of each device and the transformation matrix and the rotation and translation matrix, a standby infrared fault point of each device is matched.

[0074] For example, in the time difference stage: the contact position Delta T of the 5th frame is +1.2 DEG C (the average of adjacent pixels is +0.3 DEG C), and the Delta T of the 8th frame increases to +3.8 DEG C, forming a significant hot spot area.

[0075] YOLOv5 model analysis: output bounding box (0.52, 0.71, 0.05, 0.04);

[0076] classified as "poor contact" (confidence 92%);

[0077] Temperature rise rate 5.4 DEG C / min (exceeds threshold 2 DEG C / min).

[0078] Therefore, it can be determined that the outgoing head of the disconnecting switch is overheated, and the standby infrared fault point is the disconnecting switch.

[0079] Therefore, the time difference processing upgrades the thermal image analysis from "static temperature distribution observation" to "dynamic thermal process capture", combined with the improved YOLOv5 model, to realize the closed-loop analysis from pixel-level change to equipment defect diagnosis. Although the three-dimensional point cloud does not directly participate in this process, it provides a spatial reference for subsequent multi-sensor data fusion.

[0080] Specifically, the vibration signal analysis on the vibration signal after synchronization registration is performed, and vibration entropy is calculated, and the step includes:

[0081] S31B, the vibration signal after synchronization registration is decomposed by wavelet packet, and the power frequency component and the high frequency residual signal are extracted;

[0082] The vibration signal after registration is decomposed by wavelet packet, and the power frequency component and the high frequency residual signal are extracted, wherein the frequency of the power frequency component can be 50HZ and 100HZ, and the high frequency residual signal is called the high frequency residual signal.

[0083] S32B, the high frequency residual signal is decomposed by EMD to obtain an intrinsic mode function component representing mechanical looseness;

[0084] The high frequency residual signal is decomposed by EMD (Empirical Mode Decomposition) to obtain an intrinsic mode function component representing mechanical looseness. The intrinsic mode function component can be referred to as IMF component. Each IMF component represents a vibration mode of a specific time scale in the original signal, and the frequency is arranged from high to low, and the IMF1 component has the highest frequency. For example, when the transformer winding is loose, the IMF3 component usually corresponds to the local resonance frequency of the mechanical structure.

[0085] S33B, the nonlinear dynamics entropy value of the screened intrinsic mode function component is calculated to obtain the vibration entropy and the sideband energy ratio;

[0086] The screened IMF component is usually IMF2, IMF3 and IMF4. The nonlinear dynamics entropy value of the screened IMF component is calculated to quantify the signal complexity and obtain the vibration entropy. The normal value range of the vibration entropy is 0.6-0.8. If the vibration entropy is greater than 1.2, it indicates that the fault state is entered, and the complexity is increased due to the increase of impact component.

[0087] The IMF component and the corresponding fault type are as follows: the IMF1 component, the typical frequency range is 1KHZ-5KHZ, and the associated fault type is the transient vibration caused by partial discharge; the IMF3 component, the typical frequency range is 200HZ-800HZ, and the associated fault type is mechanical looseness of the structure; the IMF5 component, the typical frequency range is 50HZ-150HZ, and the associated fault type is the magnetic strain vibration of the iron core.

[0088] The correlation coefficient method can eliminate false components by screening IMF components, for example, discarding IMFs with a correlation coefficient less than 0.3 with the original signal, and retaining the first N IMFs with cumulative energy greater than 85%, N is usually 4-6, that is, the screened IMF components can be obtained.

[0089] In S34B, the instantaneous frequency of the screened intrinsic mode function component is obtained by Hilbert transform, and whether there is modulation phenomenon and the sideband energy ratio are detected according to the instantaneous frequency, so as to obtain the standby vibration fault point.

[0090] The instantaneous frequency of the screened IMF component is obtained by Hilbert transform, and whether there is modulation phenomenon is detected, for example, 50Hz±fn sideband appears when the bolt is loose.

[0091] For example, the original signal: vibration acceleration 0.8m / s 2 (normal baseline 0.3m / s 2 ).

[0092] IMF3 analysis: center frequency: 573Hz; sample entropy: 1.15 (normal 0.72); sideband energy ratio: 0.28 (normal <0.1)

[0093] Diagnosis conclusion: the insufficient bolt pretightening force leads to mechanical resonance. The bolt is the standby vibration fault point.

[0094] Specifically, the partial discharge analysis on the electrical signal to obtain the discharge pulse number, the step includes:

[0095] S31C, pulse detection is performed on the electrical signal to generate the discharge pulse number and PRPD spectrum;

[0096] The short-time energy threshold method is used to extract pulses from the electrical signal to generate the discharge pulse number and PRPD (Phase Resolved Partial Discharge) spectrum, the horizontal axis of the PRPD spectrum is the power frequency phase, the vertical axis is the pulse amplitude, and the chroma is the normalized count of pulse density.

[0097] S32C, the PRPD spectrum is input into the improved VGG16 model to obtain the discharge type probability and the pulse density prediction value of each phase interval, and the full connection layer in the improved VGG16 model is replaced by an attention mechanism module;

[0098] The PRPD pattern is input into an improved VGG16 model, which is a traditional VGG16 model with full-linkage layers replaced by an attention module + regression head, which outputs discharge type probability and pulse density distribution, where the discharge type probability includes surface discharge, internal discharge and corona.

[0099] S33C, according to the discharge type probability and the pulse density prediction value of each phase interval, a standby electrical fault point is generated.

[0100] Then according to the discharge type probability and the pulse density prediction value of each phase interval, the standby electrical fault point which may exist fault is screened out.

[0101] S40, input the temperature gradient, vibration entropy and discharge pulse number into the LSTM model for time sequence correlation analysis, obtain the fault type and fault type probability, and according to the fault type probability, the 750KV substation is warned.

[0102] Specifically, the temperature gradient, vibration entropy and discharge pulse number are weighted and fused, and the step includes:

[0103] S41, input the temperature gradient, vibration entropy and discharge pulse number into the LSTM model, if the temperature gradient exceeds the preset gradient threshold and the vibration entropy is in the normal range, output the fault type corresponding to the temperature gradient and the fault type probability, and perform secondary warning;

[0104] S42, if the vibration entropy exceeds the preset baseline and the temperature gradient is in the normal range, output the fault type corresponding to the vibration entropy and the fault type probability, and perform secondary warning;

[0105] S43, if the temperature gradient exceeds the preset gradient threshold and the vibration entropy exceeds the preset baseline, the fault type is mechanical looseness leading to increased contact resistance, and a primary warning is performed;

[0106] S44, if the temperature gradient exceeds the preset gradient threshold and the discharge pulse number exceeds the preset pulse threshold, the fault type is insulation deterioration causing partial discharge, and a primary warning is performed;

[0107] S45, if the temperature gradient exceeds the preset gradient threshold, the vibration entropy exceeds the preset baseline and the discharge pulse number exceeds the preset pulse threshold, the fault type is arc fault precursor, and a primary warning is performed.

[0108] If only the temperature gradient exceeds the threshold value, and the vibration entropy and the number of discharge pulses are in the normal range, such as the temperature gradient > 2℃ / min, specifically, the local continuous positive growth area of ΔT > 1℃ / s appears, the fault type is bolt loosening, or the annular diffusion ΔT distribution appears, the fault type is casing oil shortage, and the secondary independent alarm can be given.

[0109] If only the vibration entropy exceeds the baseline of 30% and the characteristic frequency of 573HZ side frequency exists, and the other temperature gradient and the number of discharge pulses are in the normal range, it can be determined that the fault type is the overheat of the outgoing head of the disconnector, and the secondary independent alarm can be given.

[0110] When the two types of alarms are triggered in the same device and within the time window (±10 seconds), the weight promotion mechanism is started. If the temperature gradient exceeds the preset gradient threshold value and the vibration entropy exceeds the preset baseline, and the number of discharge pulses is in the normal range, the fault type is mechanical loosening leading to increased contact resistance, and a first-level warning is given; if the temperature gradient exceeds the preset gradient threshold value and the number of discharge pulses exceeds the preset pulse threshold value, and the vibration entropy is in the normal range, the fault type is insulation deterioration causing partial discharge, and a first-level warning is given; if the temperature gradient exceeds the preset gradient threshold value, the vibration entropy exceeds the preset baseline, and the number of discharge pulses exceeds the preset pulse threshold value, the fault type is arc fault precursor.

[0111] Among them, the input layer of the LSTM model inputs the time step and the feature dimension, and the feature dimension includes three dimensions of temperature gradient, vibration entropy and discharge pulse number; the hidden layer of the LSTM model includes 2 layers of LSTM, each layer has 64 neurons, and the Dropout=0.2 prevents overfitting; the output layer of the LSTM model outputs the fault probability and predicts the feature change trend in the next 3 steps (15 seconds).

[0112] When the fault probability output by the LSTM is >0.7, the following actions are triggered: automatically generating a work order containing fault location (such as "transformer A phase casing") and fault type, and adjusting the protection setting value (such as shortening the over-current protection action time by 20%).

[0113] The embodiment of the application enhances the significance of key alarms by weight fusion, and uses LSTM to mine the time sequence causal relationship between multiple features, realizes the upgrade from "single-point alarm" to "state evolution prediction", and improves the accuracy.

[0114] The above-mentioned various modules in the power distribution automation device supervision and acquisition system can be realized by software, hardware and combinations thereof, wholly or partially. The above-mentioned various modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned various modules.

[0115] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of functional units and modules is exemplified. In actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the above-described functions.

[0117] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, but not to limit it. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Such modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A power distribution automation device supervisory collection system, comprising: Infrared thermal imager, acoustic vibration sensor, current sensor and voltage sensor are installed at 750KV substation, and the supervision acquisition system realizes early warning function of the 750KV substation through the following steps: Infrared signals are acquired by the infrared thermal imager, vibration signals are acquired by the acoustic vibration sensor, and electrical signals are acquired by the current sensor and the voltage sensor; Time synchronization and space registration are realized for the infrared signals, the vibration signals and the electrical signals by using a space-time alignment algorithm; Thermal imaging analysis is performed on the infrared signals after synchronization and registration, overheat areas are automatically framed and temperature gradient is calculated, vibration signal analysis is performed on the vibration signals after synchronization and registration, vibration entropy is calculated, and partial discharge analysis is performed on the electrical signals to obtain discharge pulse number; The temperature gradient, the vibration entropy and the discharge pulse number are input into an LSTM model for time sequence correlation analysis to obtain fault type and fault type probability, and early warning is performed on the 750KV substation according to the fault type probability; The step of realizing time synchronization and space registration for the infrared signals, the vibration signals and the electrical signals by using a space-time alignment algorithm comprises: Taking the sampling rate of the electrical signals as a reference frequency, the sampling rate of the infrared signals is improved to the reference frequency by using cubic spline interpolation, and the dynamic time warping algorithm is applied to the vibration signals and the electrical signals to compensate for mechanical vibration transmission delay, thereby realizing time synchronization function; The temperature matrix output by the infrared thermal imager is converted into three-dimensional point cloud, and the point cloud correspondence between the infrared signals and real equipment in the 750KV substation is realized according to the three-dimensional point cloud and a transformation matrix, wherein the transformation matrix is obtained according to the installation position of the infrared thermal imager, and the transformation matrix represents the correspondence between the three-dimensional point cloud and real physical space coordinates, thereby realizing space synchronization function; The step of inputting the temperature gradient, the vibration entropy and the discharge pulse number into an LSTM model for time sequence correlation analysis to obtain fault type and fault type probability comprises: The temperature gradient, the vibration entropy and the discharge pulse number are input into an LSTM model, if the temperature gradient exceeds a preset gradient threshold value and the vibration entropy is within a normal range, the fault type corresponding to the temperature gradient and the fault type probability are output, and secondary early warning is performed; If the vibration entropy exceeds a preset baseline and the temperature gradient is within a normal range, the fault type corresponding to the vibration entropy and the fault type probability are output, and secondary early warning is performed.

2. The power distribution automation equipment supervisory collection system of claim 1, wherein, The step of performing thermal imaging analysis on the infrared signals after synchronization and registration, automatically framing overheat areas and calculating temperature gradient comprises: For transformers, GIS devices and overhead lines, time difference is performed on 10 consecutive thermal images of each device to obtain temperature change matrix, binary mask and time energy graph of each device; Superimpose the temperature change matrix of each device with the infrared signal of each device, input the superimposed data into the YOLOv5 model, automatically frame the overheating area of each device and output the temperature gradient of each device; According to the overheating area of each device and the transformation matrix, the rotation and translation matrix, match to the standby infrared fault point of each device.

3. The power distribution automation equipment supervisory collection system of claim 1, wherein, The step of performing vibration signal analysis on the vibration signals after synchronization registration and calculating the vibration entropy includes: Wavelet packet decomposition is performed on the vibration signals after synchronization registration to extract the power frequency component and high frequency residual signal; EMD decomposition is performed on the high frequency residual signal to obtain the intrinsic mode function component representing mechanical looseness; The nonlinear kinetic entropy value of the screened intrinsic mode function component is calculated to obtain the vibration entropy and the sideband energy ratio; The instantaneous frequency of the screened intrinsic mode function component is obtained by Hilbert transform, and the existence of modulation phenomenon and the sideband energy ratio are detected according to the instantaneous frequency, so as to obtain the standby vibration fault point.

4. The power distribution automation equipment supervisory collection system of claim 3, wherein, The screened intrinsic mode function component is obtained by the following steps: False components of the intrinsic mode function are removed by the correlation coefficient method, and the first N intrinsic mode function components with cumulative energy > 85% are retained, N is a positive integer.

5. The power distribution automation equipment supervision and collection system of claim 1, wherein, The step of performing partial discharge analysis on the electrical signal to obtain the discharge pulse number includes: Pulse detection is performed on the electrical signal to generate the discharge pulse number and the PRPD spectrum; The PRPD spectrum is input into the improved VGG16 model to obtain the discharge type probability and the pulse density prediction value of each phase interval, and the full connection layer in the improved VGG16 model is replaced by an attention mechanism module; According to the discharge type probability and the pulse density prediction value of each phase interval, the standby electrical fault point is generated.

6. The power distribution automation equipment supervision and collection system of claim 1, wherein, The step of inputting the temperature gradient, the vibration entropy and the discharge pulse number into the LSTM model for time sequence correlation analysis to obtain the fault type and the fault type probability further includes: If the temperature gradient exceeds the preset gradient threshold and the vibration entropy exceeds the preset baseline, the fault type is mechanical looseness leading to increased contact resistance, and a first level warning is performed; If the temperature gradient exceeds the preset gradient threshold and the discharge pulse number exceeds the preset pulse threshold, the fault type is insulation deterioration causing partial discharge, and a first level warning is performed; If the temperature gradient exceeds the preset gradient threshold, the vibration entropy exceeds the preset baseline and the discharge pulse number exceeds the preset pulse threshold, the fault type is arc fault precursor, and a first level warning is performed.

7. The power automation equipment supervisory collection system according to any one of claims 1 to 6, characterized by, The infrared thermal imager is installed at the transformer, circuit breaker, disconnecting switch and bus joint of the 750KV substation and transmission line, the acoustic vibration sensor is installed on the shell of the large equipment of the 750KV substation, the current sensor is installed at the current transformer, and the voltage sensor is installed at the voltage transformer of the bus, incoming line and outgoing line interval of the 750KV substation.

Images