Distributed distribution transformer high-voltage side intelligent terminal fault identification method and system

Abstract

The invention relates to the technical field of power distribution automation, and discloses a distributed distribution transformer high-voltage side intelligent terminal fault identification method and system. The method comprises the steps of collecting and preprocessing three-phase current signals on the high-voltage side of a distribution transformer, extracting effective values of each phase of current, calculating per unit values and three-phase imbalance characteristic parameters, collecting switch state codes as state identifiers, judging fault types in combination with the characteristic parameters, establishing a power distribution loop topological relation to verify a fault judgment result, and judging whether the fault is abnormal or not. And generating an isolation control instruction to realize fault isolation and power supply recovery. The problems of long response delay, low fault positioning precision and lack of refined fault type identification and topology verification mechanisms caused by centralized processing in an existing power distribution network fault detection system are solved, and the accuracy of fault identification of the high-voltage side of the distribution transformer and the fault isolation response speed are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution automation, and in particular to a distributed distribution transformer high-voltage side intelligent terminal fault identification method and system. BACKGROUND

[0002] As an important part of the power system, the distribution network undertakes the key task of distributing electric energy from the transmission network to the terminal user. The distribution transformer, as the core equipment of the distribution network, directly affects the power supply reliability and power quality. The traditional fault detection of the distribution network mainly relies on centralized fault indicators and manual line inspection. The centralized fault indicators determine the fault occurrence by monitoring the changes of line current and voltage, and the manual line inspection determines the fault location by on-site inspection. The power distribution automation system collects the operation data of the distribution equipment through the remote terminal unit and uploads them to the master station for centralized analysis and processing. The master station comprehensively determines the fault section according to the current and voltage data of multiple measuring points, and realizes fault isolation and power supply restoration by issuing control instructions. The distribution network has a complex topology and a large number of devices. The traditional centralized monitoring and control method has problems such as large communication delay, heavy processing load of the master station, and low fault positioning accuracy.

[0003] The existing technology has obvious deficiencies in fault identification of the high-voltage side of the distribution transformer. The centralized fault detection system relies on the centralized analysis and processing of the master station. When the scale of the distribution network expands, the master station needs to process a large amount of data, resulting in a sharp increase in calculation load. The delay and fault of the communication link will affect the timely transmission of fault information. The traditional fault indicator mainly detects the overcurrent and grounding signals of the line, and lacks fine identification ability for faults on the incoming line side, bus, and load side of the distribution transformer. The current information of a single measuring point cannot accurately distinguish whether the fault location is on the incoming line side, the outgoing bus, or the load side of the transformer. The fault determination relies on the setting of experience threshold values, and different capacity and load characteristics of the distribution transformer need to set criteria separately, increasing the configuration complexity. The fault isolation and power supply restoration decision of the centralized system relies on the control instructions issued by the master station, and the response time from fault detection to isolation control is long, affecting the power supply continuity of the non-fault area. SUMMARY

[0004] The present application provides a distributed distribution transformer high-voltage side intelligent terminal fault identification method and system, which solves the problems of long response delay, low fault positioning accuracy, lack of fine fault type identification, and topology verification mechanism in the existing distribution network fault detection system, and improves the accuracy of fault identification and the response speed of fault isolation of the high-voltage side of the distribution transformer.

[0005] In a first aspect, the present application provides a distributed distribution transformer high-voltage side intelligent terminal fault identification method, which comprises: The three-phase current signals on the high-voltage side of the distribution transformer are collected, the three-phase current signals are preprocessed to obtain preprocessed current signals; The effective values of the phases of the preprocessed current signals are extracted, the current effective values of the phases are compared with corresponding reference current values to obtain phase current unit values, and three-phase current imbalance characteristic parameters are calculated according to the phase current unit values; The state information of the switch on the high-voltage side of the distribution transformer is collected, the state information of the switch is encoded into state identifiers, and the fault type of the distribution transformer is determined in combination with the three-phase current imbalance characteristic parameters; The topological relationship data of a plurality of distribution transformers in a distribution loop are established, the determined fault type is verified according to the topological relationship data, and a fault determination result is confirmed; An isolation control instruction is generated according to the confirmed fault determination result, and the isolation control instruction is executed to realize fault area isolation and non-fault area power supply recovery.

[0006] In a second aspect, the application provides a distributed distribution transformer high-voltage side intelligent terminal fault identification system, which comprises: A processing module is configured to collect three-phase current signals on the high-voltage side of a distribution transformer, and pre-process the three-phase current signals to obtain preprocessed current signals; A calculation module is configured to extract the effective values of the phases of the preprocessed current signals, compare the current effective values of the phases with corresponding reference current values to obtain phase current unit values, and calculate three-phase current imbalance characteristic parameters according to the phase current unit values; An encoding module is configured to collect state information of a switch on the high-voltage side of the distribution transformer, encode the state information of the switch into state identifiers, and determine the fault type of the distribution transformer in combination with the three-phase current imbalance characteristic parameters; A verification module is configured to establish topological relationship data of a plurality of distribution transformers in a distribution loop, verify the determined fault type according to the topological relationship data, and confirm a fault determination result; An implementation module is configured to generate an isolation control instruction according to the confirmed fault determination result, and execute the isolation control instruction to realize fault area isolation and non-fault area power supply recovery.

[0007] In a third aspect, a distributed distribution transformer high-voltage side intelligent terminal fault identification device is provided, which comprises a memory and at least one processor, the memory stores instructions, and the at least one processor invokes the instructions in the memory to enable the distributed distribution transformer high-voltage side intelligent terminal fault identification device to perform the distributed distribution transformer high-voltage side intelligent terminal fault identification method described above.

[0008] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores instructions which, when executed on a computer, cause the computer to perform the distributed distribution transformer high-voltage side intelligent terminal fault identification method described above.

[0009] In the technical scheme provided in the present application, the three-phase current signals on the high-voltage side of the distribution transformer are collected and preprocessed to obtain preprocessed current signals, a multi-stage preprocessing method combining wavelet decomposition and Wiener filtering is used to remove random noise and harmonic interference, the signal is decomposed into approximation coefficients and detail coefficients according to the frequency scale by wavelet decomposition, the high-frequency detail coefficients are denoised by a threshold function to retain the main components of the power frequency, the Wiener filter realizes adaptive filtering by designing a transfer function according to the ratio of the signal power spectral density to the noise power spectral density in the frequency domain, and the multi-stage preprocessing significantly improves the signal-to-noise ratio of the current signal. The effective values of the current of each phase of the preprocessed current signal are extracted, and the current unit values of each phase are obtained by ratio calculation with the historical current reference values. The unit normalization method eliminates the influence of the absolute current value difference of the distribution transformer of different capacities, so that the distribution transformer of various capacities can use a unified fault criterion. According to the current unit values of each phase, the three-phase current imbalance characteristic parameters are calculated, the maximum current unit value is divided by the intermediate current unit value to quantitatively represent the imbalance degree of the three-phase current distribution. When the three-phase load is symmetrical, the parameter is close to 1, and when single-phase grounding or broken line fault occurs, the parameter significantly increases. The switching state information on the high-voltage side of the distribution transformer is collected and coded into a state identifier, and a mapping relationship between the switching state combination and the fault perceptible area is established. According to the state identifier, the fault perceptible position range is obtained by querying the fault perceptible area mapping table, and the fault type is determined by combining the numerical distribution relationship of the three-phase current imbalance characteristic parameters and the current unit values of each phase. The fine identification of the in-line side fault, the bus fault and the load side fault is realized. By establishing the topological relationship data of multiple distribution transformers in the distribution loop, the electrical propagation law of the fault in the distribution loop is used for cross verification. When the target distribution transformer is determined to be an in-line side fault or a load side fault, it is checked whether the previous distribution transformer is a bus fault and the subsequent distribution transformer is fault-free. If the single-point fault propagation law is met, the fault determination result is confirmed to be correct. The topological verification mechanism effectively reduces the misjudgment rate. According to the confirmed fault determination result, the isolation control instruction is generated and the fault area isolation and non-fault area power supply recovery are performed. The whole process is completed on the high-voltage side of the distribution transformer intelligent terminal without relying on the centralized processing of the master station, which significantly shortens the fault response time and improves the power supply reliability and self-healing ability of the distribution network.

[0010] The application applies wavelet decomposition and Wiener filtering algorithm to signal pretreatment in the field of power distribution network fault diagnosis and self-healing control, and the contribution of the algorithm features to the scheme is reflected in that the power frequency component and noise harmonic component of the current signal are effectively separated through multi-stage pretreatment, the accuracy of subsequent fault feature extraction is significantly improved, the per unit normalization algorithm eliminates the influence of the capacity difference of distribution transformers on the fault criterion, so that the distributed intelligent terminal can adopt a unified criterion threshold, the calculation method of three-phase current unbalance feature parameter quantitatively characterizes the degree of asymmetry of three-phase current distribution through the ratio of the maximum current per unit value to the intermediate current per unit value, and provides reliable feature parameters for fault type determination, the switch state coding and fault perceptible area mapping algorithm establishes the association between switch state combination and electrical position range, combined with the current unbalance feature parameter, the fine positioning of the fault position is realized, the power distribution loop topology verification algorithm uses the propagation law of the fault in the loop, and the correctness of the single-point fault determination result is cross-verified by calculating the bus fault matching degree of the fault type of the previous power distribution transformer and the fault-free matching degree of the fault type of the subsequent power distribution transformer, the application of the algorithm enables the distributed intelligent terminal to have the ability of local fault identification, topology verification and automatic isolation control, and the fast fault positioning and power supply recovery can be realized without relying on the centralized processing of the master station, and the intelligent level and fault response speed of the power distribution network are improved as a whole. BRIEF DESCRIPTION OF DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings. Figure 1 An embodiment schematic diagram of the fault identification method of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the present application; Figure 2 An embodiment schematic diagram of the fault identification system of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the present application; Figure 3 An embodiment schematic diagram of the fault identification system of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the present application; DETAILED DESCRIPTION

[0012] The embodiment of the present application provides a distributed distribution transformer high-voltage side intelligent terminal fault identification method and system. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Please refer to Figure 1 One embodiment of the distributed distribution transformer high-voltage side intelligent terminal fault identification method in the embodiment of the present application includes: Step S101, collecting three-phase current signals of the high-voltage side of the distribution transformer, performing signal preprocessing on the three-phase current signals, and obtaining preprocessed current signals; Step S102, extracting the effective value of each phase current of the preprocessed current signal, performing ratio calculation on the current effective value of each phase and the corresponding reference current value, obtaining the unit value of each phase current, and calculating the three-phase current imbalance characteristic parameter according to the unit value of each phase current; Step S103, collecting the state information of the switch of the high-voltage side of the distribution transformer, encoding the state information of the switch into a state identifier, and determining the fault type of the distribution transformer in combination with the three-phase current imbalance characteristic parameter; Step S104, establishing the topological relationship data of a plurality of distribution transformers in a distribution loop, verifying the determined fault type according to the topological relationship data, and confirming the fault determination result; Step S105, generating an isolation control instruction according to the confirmed fault determination result, and executing the isolation control instruction to realize fault area isolation and non-fault area power supply recovery.

[0014] It can be understood that the execution subject of the present application can be a distributed distribution transformer high-voltage side intelligent terminal fault identification system, and can also be a terminal or a server, and the specific place is not limited. The embodiment of the present application takes the server as the execution subject for example.

[0015] Specifically, the intelligent terminal synchronously collects A-phase current signals, B-phase current signals and C-phase current signals of the incoming line switch position, the outgoing line switch position and the load switch position through a current transformer installed at the high-voltage side of the distribution transformer. The current transformer is an electromagnetic device that transforms large current into small current in proportion. Usually, it converts kilo-ampere-level current at the high-voltage side into a standard signal of 5 amperes or 1 ampere for measurement according to a specific transformation ratio. In the collection process, three-phase current signals are required to be sampled at the same time to ensure phase synchronization. The original three-phase current signals collected contain noise interference and harmonic components, so preprocessing is needed. First, the original three-phase current signals are decomposed by a wavelet basis to obtain approximation coefficients and multi-layer detail coefficients. Wavelet decomposition is a time-frequency analysis method that realizes multi-resolution analysis of signals by decomposing signals into wavelet coefficients of different frequency scales. The approximation coefficients represent the low-frequency main components of the signal, and the detail coefficients represent the high-frequency noise components of the signal. The multi-layer detail coefficients are denoised by a threshold function. The threshold function sets a threshold value to zero or performs shrinkage processing on the detail coefficients with an amplitude less than the threshold value to remove high-frequency noise. After denoising, the processed approximation coefficients and detail coefficients are reconstructed by inverse wavelet transform to obtain the denoised current signal. Then, the denoised current signal is converted to the frequency domain by fast Fourier transform. Fast Fourier transform is an algorithm for efficiently calculating discrete Fourier transform. It converts time-domain signals into frequency-domain signals represented as the superposition of different frequency components. In the frequency domain, a frequency-domain transfer function of a Wiener filter is applied for filtering processing. The Wiener filter is an optimal linear filter based on the least mean square error criterion. Its frequency-domain transfer function is designed according to the ratio of signal power spectral density to noise power spectral density. It attenuates different frequency components to different degrees. After filtering, the signal is converted back to the time domain by inverse fast Fourier transform to obtain the final preprocessed current signal. This preprocessing process not only removes random noise but also filters harmonic interference. The effective value of the preprocessed current signal is extracted. The current effective value refers to the value equivalent to the heat effect of the alternating current in one cycle. It is obtained by averaging the square of the current instantaneous value in one cycle and then taking the square root. A-phase current effective value, B-phase current effective value and C-phase current effective value are obtained. Then, A-phase reference current value, B-phase reference current value and C-phase reference current value before the fault occur are read from a historical current reference value storage queue. The historical current reference value storage queue is a data structure that stores the effective values of each phase current during normal operation in chronological order. The queue saves the steady-state current values in a period before the fault occurs as a reference. The A-phase current effective value is divided by the A-phase reference current value to obtain the A-phase current unit value. The B-phase current effective value is divided by the B-phase reference current value to obtain the B-phase current unit value. The C-phase current effective value is divided by the C-phase reference current value to obtain the C-phase current unit value. The unit value is a normalized relative value representation method.The influence of absolute value size is eliminated by ratio with reference value, which facilitates unified judgment of different capacity distribution transformers. The maximum current per unit and the intermediate current per unit are selected from the A-phase current per unit, the B-phase current per unit and the C-phase current per unit. The maximum current per unit is divided by the intermediate current per unit to obtain a three-phase current imbalance characteristic parameter, which reflects the unbalance degree of three-phase current distribution. When three-phase load is symmetrical, the parameter is close to 1. When single-phase grounding or broken wire fault occurs, the parameter will significantly increase. The incoming line switch state, outgoing line switch state and load switch state of the distribution transformer high-voltage side are collected. The open state is defined as 0, indicating that the switch is open. The closed state is defined as 1, indicating that the switch is closed. The switch state triplets are encoded according to the combination of the three switch states to obtain a state identifier. The state identifier is a code composed of three binary numbers. From high to low, it represents the state of the incoming line switch, the outgoing line switch and the load switch. The power transmission direction identifier of the distribution transformer in the distribution loop is obtained. The power transmission direction identifier indicates whether the power flows from the incoming line side to the load side or flows from the other end. The state identifier is used to query the fault perceptible area mapping table to obtain the perceptible fault position range. The fault perceptible area mapping table is a pre-established table. The table records the electrical position range that the intelligent terminal can detect the fault corresponding to different switch state combinations. The three-phase current imbalance characteristic parameter is compared with the preset threshold value. The preset threshold value is a criterion value obtained by statistical analysis of a large amount of historical fault data. When the three-phase current imbalance characteristic parameter is greater than the preset threshold value, it indicates that the three-phase current distribution is abnormal. Combined with the numerical distribution relationship of the per unit values of the currents of the three phases, the fault type of the distribution transformer in the perceptible fault position range is determined. The numerical distribution relationship includes that if a per unit value of a phase current significantly decreases, it is determined as a broken wire fault of the phase; if a per unit value of a phase current or two-phase current significantly increases, it is determined as a short-circuit fault. The loop topology state matrix is constructed according to the connection order of the distribution transformers in the distribution loop. The loop topology state matrix is a one-dimensional or two-dimensional data structure. Each element in the matrix corresponds to the position of a distribution transformer in the distribution loop. The element value records the fault type identifier of the distribution transformer at the position. The power transmission direction matrix of the same dimension is constructed to record the power transmission direction identifier of each position distribution transformer. The target distribution transformer with fault type of incoming line side fault or load side fault is extracted from the loop topology state matrix. The incoming line side fault refers to the fault point located between the incoming line switch and the power supply of the distribution transformer. The load side fault refers to the fault point located between the load switch and the load of the distribution transformer. The pre-sequence distribution transformer fault type set and the post-sequence distribution transformer fault type set of the target distribution transformer are obtained. The pre-sequence distribution transformer refers to the distribution transformer upstream of the target distribution transformer in the distribution loop. The post-sequence distribution transformer refers to the distribution transformer downstream of the target distribution transformer. The power transmission direction identifier value of the target distribution transformer is read from the power transmission direction matrix.According to the power transmission direction identifier value, the position range of the pre-sequence power distribution transformer fault type set and the post-sequence power distribution transformer fault type set is determined, busbar fault type matching degree calculation is performed on the pre-sequence power distribution transformer fault type set, the busbar fault refers to the fault on the outgoing busbar of the power distribution transformer, the matching degree calculation is to obtain the pre-sequence matching degree value by ratio operation of the number of busbar fault types in the set and the total number of elements in the set, the post-sequence power distribution transformer fault type set is subjected to no-fault type matching degree calculation, the no-fault type matching degree value is obtained by ratio operation of the number of no-fault types in the set and the total number of elements in the set, when the pre-sequence matching degree value and the post-sequence matching degree value are both equal to 1, it indicates that all power distribution transformers upstream of the target power distribution transformer detect busbar faults and all power distribution transformers downstream do not detect faults, which conforms to the electrical propagation rule of single-point fault, the loop state identifier value corresponding to the position of the loop topology state matrix is set to the single-point fault locatable state, when the pre-sequence matching degree value or the post-sequence matching degree value is less than 1, it indicates that there is a situation that does not conform to the single-point fault propagation rule, the loop state identifier value is set to the multi-point fault to be verified state, the loop state identifier value is read from the loop topology state matrix, when the loop state identifier value is the single-point fault locatable state, the position index and the fault type of the fault power distribution transformer are extracted, the position index is the position number of the power distribution transformer in the loop topology state matrix, the target isolation switch combination is obtained according to the fault type by querying the fault isolation strategy mapping table, the fault isolation strategy mapping table is a pre-planned table, which records the switch combination that needs to be opened corresponding to different fault types, the switch opening instruction sequence is generated according to the target isolation switch combination, the switch opening instruction sequence is a series of switch control instructions arranged in operation order, the switch opening instruction sequence is sent to the high-voltage side switch actuator of the corresponding power distribution transformer through the communication link, the communication link is a data transmission channel between the intelligent terminal and the switch actuator, the switches in the target isolation switch combination are controlled to execute the opening operation to complete the fault area isolation, the set of healthy power distribution transformers which are in the same power distribution loop with the fault power distribution transformer and are located in the non-fault area is identified from the loop topology state matrix, the set of healthy power distribution transformers contains all power distribution transformers with no fault type, the power supply paths of each power distribution transformer in the set of healthy power distribution transformers are subjected to topology analysis, the topology analysis is to find the available power supply path from the standby power source to the healthy power distribution transformer by traversing the connection relationship of the power distribution loop, the switch closing instruction sequence is generated to control the switches on the standby power supply path to execute the closing operation, and the power supply recovery of the set of healthy power distribution transformers is completed, in a power distribution loop with five power distribution transformers connected in sequence, when the state identifier of the incoming switch closing, the outgoing switch opening, and the load switch closing of the No. 3 power distribution transformer is detected, the A-phase current signal waveform data is synchronously collected through the current transformer, db4 wavelet basis is used for 5-layer wavelet decomposition, the detail coefficients of the 3rd to 5th layers after decomposition are processed by a soft threshold function, and the threshold is set to 3 times the standard deviation of the detail coefficients of each layer,The processed detail coefficients and the approximation coefficients are inverse wavelet transformed to reconstruct a denoised current signal. The reconstructed denoised current signal is converted to a frequency domain through a 2048-point fast Fourier transform. A transfer function of a Wiener filter is applied in the frequency domain. The transfer function is designed according to a ratio of a signal power spectral density to a noise power spectral density. The transfer function retains 50 Hz fundamental wave and integer harmonic components and attenuates other frequency components. After filtering, the filtered signal is converted back to a time domain through an inverse fast Fourier transform to obtain a preprocessed current signal. A root mean square value of the preprocessed current signal is extracted. The root mean square value of the A-phase current is obtained by summing squares of sampling point data in 20 milliseconds in one power frequency cycle, dividing the sum by the number of sampling points, and then taking a square root. An A-phase reference current value in 10 seconds before a fault occurs is read from a historical current reference value storage queue. The reference value is obtained by averaging A-phase current root mean square values in 10 seconds. The A-phase current root mean square value is divided by the A-phase reference current value to obtain an A-phase current unit value. B-phase and C-phase current unit values are obtained in the same way. A maximum current unit value and an intermediate current unit value are selected from the three-phase current unit values. The maximum current unit value is divided by the intermediate current unit value to obtain a three-phase current imbalance characteristic parameter. The parameter is compared with a preset threshold value 1.5. It is found that the three-phase current imbalance characteristic parameter is greater than the preset threshold value. In combination with a significant decrease of the A-phase current unit value to 0.1 and a distribution relationship of the B-phase and C-phase current unit values remaining above 0.95, a fault can be sensed according to a state identifier. A mapping table of a fault sensing area is queried. A fault sensing area corresponding to the switch state combination includes an incoming line side and a bus area. In combination with a distribution relationship of the current unit values, it is determined that the fault type is an incoming line side A-phase open line fault. A loop topology state matrix is constructed to record fault types of the five distribution transformers. The No. 3 distribution transformer is recorded as an incoming line side fault. The No. 1 and No. 2 distribution transformers are recorded as bus faults. The No. 4 and No. 5 distribution transformers are recorded as no faults. A power transmission direction identifier value of the No. 3 distribution transformer is read from a power transmission direction matrix as a positive power transmission direction. It is determined that the No. 1 and No. 2 distribution transformers are previous sequence distribution transformers, and the No. 4 and No. 5 distribution transformers are subsequent sequence distribution transformers. A matching degree of a previous sequence distribution transformer fault type set is calculated. The set contains two bus fault type elements. A total number of elements in the set is 2. A ratio operation obtains a previous sequence matching degree value as 1. A matching degree of a subsequent sequence distribution transformer fault type set is calculated. The set contains two no fault type elements. A total number of elements in the set is 2. A ratio operation obtains a subsequent sequence matching degree value as 1. Since the previous sequence matching degree value and the subsequent sequence matching degree value are both equal to 1, a loop state identifier value is set as a single point fault locatable state. The loop state identifier value is read from the loop topology state matrix as the single point fault locatable state. A position index of the No. 3 distribution transformer is extracted as 3 and a fault type is extracted as an incoming line side A-phase open line fault. A fault isolation strategy mapping table is queried according to the fault type. A target isolation switch combination includes an incoming line switch of the No. 3 distribution transformer and an outgoing line switch of the No. 2 distribution transformer. A switch opening instruction sequence is generated. The incoming line switch of the No. 3 distribution transformer is controlled to open first, and then the outgoing line switch of the No. 2 distribution transformer is controlled to open.After the fault area isolation is completed, a set of healthy distribution transformers including the No. 4 and No. 5 distribution transformers is identified, a power supply path is analyzed in topology, it is found that a standby power source is available at the other end of a distribution loop, the standby power source can be accessed by closing a tie switch of the No. 5 distribution transformer, a switch closing instruction sequence is generated, the tie switch is controlled to perform a closing operation, and power supply recovery of the No. 4 and No. 5 distribution transformers is completed.

[0016] In an embodiment, three-phase current signals at a high-voltage side of a distribution transformer are collected, and signal preprocessing is performed on the three-phase current signals to obtain preprocessed current signals, including: A-phase, B-phase and C-phase current signals of an incoming switch, an outgoing switch and a load switch at the high-voltage side of the distribution transformer are synchronously collected by a current transformer to obtain original three-phase current signals; The original three-phase current signals are decomposed by a wavelet basis in multiple layers to obtain approximation coefficients and multiple layers of detail coefficients, the multiple layers of detail coefficients are denoised by a threshold function, and denoised current signals are obtained by inverse wavelet transform reconstruction; The denoised current signals are converted to a frequency domain by fast Fourier transform, are filtered by a frequency domain transfer function of a Wiener filter, are converted to a time domain by inverse fast Fourier transform, and preprocessed current signals are obtained.

[0017] Specifically, the current transformer is installed at the incoming line switch position, outgoing line switch position and load switch position of the high voltage side of the distribution transformer, the current transformer converts the large current of the high voltage side into small current signal according to the fixed transformation ratio, for example, the primary side current of 400 amperes is converted into the secondary side current of 5 amperes according to the transformation ratio of 80 to 1, the intelligent terminal synchronously collects the A-phase current signal, B-phase current signal and C-phase current signal output by the current transformer at the three positions through the analog-digital converter, the synchronous collection refers to sampling the three-phase current signal at the same time, the sampling frequency is set to 10000 sampling points per second, 200 sampling points are collected within 20 milliseconds corresponding to 50 hertz of each power frequency cycle, the original three-phase current signal collected contains random noise and harmonic interference, the original three-phase current signal is subjected to multi-layer wavelet decomposition using the wavelet base, the wavelet base is a kind of base function with time-frequency localization characteristics, the db4 wavelet base is suitable for power signal analysis due to its symmetry, the multi-layer wavelet decomposition decomposes the signal into approximation coefficients and detail coefficients according to different frequency scales, the approximation coefficients represent the low frequency part of the signal, i.e. the main power frequency component, the detail coefficients represent the high frequency part of the signal, i.e. the noise and harmonic components, taking the A-phase current signal as an example for 5-layer wavelet decomposition, the 1st layer decomposition divides the signal into approximation coefficients with a frequency range of 0 to 2500 hertz and detail coefficients with a frequency range of 2500 to 5000 hertz, and the decomposition is sequentially performed until the 5th layer, finally obtaining the 5th layer approximation coefficients and 5 groups of detail coefficients from the 1st to 5th layers, the multi-layer detail coefficients are subjected to denoising processing using the threshold function, the threshold function sets the coefficients with an amplitude less than the threshold value to zero or shrinks them, the soft threshold function retains the sign of the coefficients greater than the threshold value after subtracting the threshold value, the threshold value is set to 3 times the standard deviation of the detail coefficients of each layer, the denoised detail coefficients and the 5th layer approximation coefficients are reconstructed through the inverse wavelet transform, the inverse wavelet transform is the inverse process of the wavelet decomposition, the approximation coefficients and detail coefficients of each layer are weighted and summed to reconstruct the time domain signal according to the wavelet base function, finally obtaining the denoised current signal, the denoised current signal is converted to the frequency domain through the fast Fourier transform, the fast Fourier transform decomposes the time domain signal into the superposition of sine and cosine components of different frequencies, the frequency domain signal is obtained after the transformation, each frequency point in the frequency domain signal corresponds to a complex value, the modulus of the complex number represents the amplitude of the frequency component, the frequency domain transfer function of the Wiener filter is applied for filtering in the frequency domain, the frequency domain transfer function of the Wiener filter is designed according to the ratio of the signal power spectral density to the noise power spectral density, the signal power spectral density is obtained by squaring the frequency domain amplitude of the denoised current signal, the transfer function of the Wiener filter is close to 1 at the frequency points with large signal power spectral density to retain the frequency components, and is close to 0 at the frequency points with large noise power spectral density to suppress the frequency components, the complex value of each frequency point of the frequency domain signal is multiplied by the transfer function value of the corresponding frequency point to obtain the filtered frequency domain signal, the filtered frequency domain signal is converted back to the time domain through the inverse fast Fourier transform, the inverse fast Fourier transform is the inverse process of the fast Fourier transform,The frequency components of the frequency domain signal are combined according to the corresponding amplitude and phase to form a time domain signal to obtain a preprocessed current signal. The high-voltage side incoming line switch position detection of the distribution transformer outputs 5 amperes of A-phase current, corresponding to 400 amperes of primary side current. The original A-phase current signal is obtained by an analog-to-digital converter within 20 milliseconds of a power frequency cycle, which has 200 sampling points. The original A-phase current signal is decomposed by a db4 wavelet base to 5 layers. The first layer decomposition reduces the number of sampling points from 200 to 100, which is divided into 100 first layer approximation coefficients and 100 first layer detail coefficients. The second layer decomposition reduces the first layer approximation coefficients from 100 to 50, and the decomposition is sequentially performed to the fifth layer to obtain 6 fifth layer approximation coefficients and layer detail coefficients. The standard deviation of the first layer detail coefficients is calculated. The sum of the squares of the 100 detail coefficients is divided by 100 and then the square root is taken. The threshold is set to 3 times the standard deviation. The coefficients with an amplitude less than the threshold in the first layer detail coefficients are set to zero. The coefficients with an amplitude greater than the threshold in the first layer detail coefficients are subtracted from the threshold and the sign is retained. The threshold of the second layer to the fifth layer detail coefficients is calculated by the same method for denoising processing. The fifth layer approximation coefficients and the layer detail coefficients are reconstructed by inverse wavelet transform. The fifth layer approximation coefficients are upsampled and interpolated. The fifth layer detail coefficients are added after convolution operation to reconstruct the fourth layer approximation coefficients. The reconstruction is sequentially performed to the first layer until the final reconstruction of the denoised current signal with 200 sampling points. The 200 sampling points of the denoised current signal are subjected to 256-point fast Fourier transform, and the remaining 56 sampling points are zero-filled. The frequency domain data of 128 frequency points are obtained after transformation. The power spectral density of the denoised current signal is calculated. The amplitudes of the 128 frequency points are squared. The transfer function of the Wiener filter at each frequency point is equal to the signal power spectral density divided by the sum of the signal power spectral density and the noise power spectral density. The transfer function value near 50 Hz is close to 1 to retain the power frequency component, and the transfer function value far from 50 Hz in the high frequency region is close to 0 to suppress high frequency noise. The complex values of the 128 frequency points of the frequency domain signal are multiplied by the corresponding transfer function values to obtain the filtered frequency domain signal. The 128 frequency points after filtering are subjected to inverse fast Fourier transform to obtain a time domain signal with 256 sampling points. The first 200 sampling points are taken as the preprocessed current signal.

[0018] In a specific embodiment, the effective values of the preprocessed current signal are extracted. The current effective value of each phase is calculated by the ratio of the current effective value of each phase to the corresponding reference current value to obtain the current unit value of each phase. The three-phase current imbalance characteristic parameters are calculated according to the current unit value of each phase, including: The effective value extraction processing is performed on the preprocessed current signal to obtain the A-phase current effective value, the B-phase current effective value and the C-phase current effective value, respectively. read the A-phase reference current value, the B-phase reference current value and the C-phase reference current value before the fault occurs from the historical current reference value storage queue, divide the A-phase current effective value by the A-phase reference current value to obtain the A-phase current unit value, divide the B-phase current effective value by the B-phase reference current value to obtain the B-phase current unit value, and divide the C-phase current effective value by the C-phase reference current value to obtain the C-phase current unit value; select the maximum current unit value and the intermediate current unit value from the A-phase current unit value, the B-phase current unit value and the C-phase current unit value, divide the maximum current unit value by the intermediate current unit value to obtain the three-phase current imbalance characteristic parameter.

[0019] In particular, the preprocessed current signal is subjected to a root-mean-square extraction process, which is a value equivalent to the heat effect of an alternating current in one cycle, and the calculation method is to square the sum of the current instantaneous value in one power frequency cycle and then divide by the number of sampling points and take the square root. The intelligent terminal segments the preprocessed current signal according to the power frequency cycle. Each power frequency cycle corresponds to 20 milliseconds of 50 Hz, and under the sampling frequency, it corresponds to 200 sampling points. The current instantaneous values of the 200 sampling points are squared and summed, and the sum is divided by 200 to obtain the average value. The average value is taken to the power of the A-phase current effective value of the power frequency cycle. Similarly, the B-phase current effective value and the C-phase current effective value are calculated. The historical current reference value storage queue is a first-in-first-out data structure that stores the effective values of the current of each phase when the distribution transformer is operating normally in chronological order. The queue length is set to store the current data in the recent period. When a fault trigger signal is detected, the intelligent terminal reads the A-phase reference current value, the B-phase reference current value and the C-phase reference current value before the fault from the historical current reference value storage queue. The reference current value selects the steady-state current value in a certain time window before the fault. The A-phase current effective value detected at present is divided by the A-phase reference current value read from the queue to obtain the A-phase current unit value. The unit value is a kind of normalization method, which eliminates the influence of absolute value size by comparing with the reference value, so that distribution transformers of different capacities can use unified criteria. The B-phase current effective value is divided by the B-phase reference current value to obtain the B-phase current unit value. The C-phase current effective value is divided by the C-phase reference current value to obtain the C-phase current unit value. The three-phase current unit values reflect the change multiples of the current of each phase relative to the normal operating state. The A-phase current unit value, the B-phase current unit value and the C-phase current unit value are compared and sorted to select the maximum current unit value and the intermediate current unit value. The maximum current unit value is divided by the intermediate current unit value to obtain the three-phase current imbalance characteristic parameter, which reflects the imbalance degree of the three-phase current distribution. When the three-phase load is completely symmetrical, the three-phase current unit values are equal, and the ratio of the maximum current unit value to the intermediate current unit value is close to 1. When a single-phase ground fault or a broken line fault occurs, the current of a certain phase significantly increases compared to the currents of the other two phases, and the ratio of the maximum current unit value to the intermediate current unit value deviates significantly from 1. The intelligent terminal on the high-voltage side of the distribution transformer extracts the root-mean-square value of the preprocessed A-phase current signal of the incoming line switch position. The preprocessed signal contains 200 sampling points in one power frequency cycle. The instantaneous values of the 200 sampling points are squared, and the instantaneous value of the first sampling point is squared and added to the instantaneous value of the second sampling point, and so on, to the instantaneous value of the 200th sampling point. The cumulative sum is divided by 200 to obtain the average value. The average value is taken to the power of the A-phase current effective value. Similarly, the B-phase current effective value and the C-phase current effective value are obtained.The historical current reference value storage queue stores the effective values of each phase current in the last 10 seconds before the fault occurs, and the queue contains 500 power frequency cycle data. When a switch state change is detected to trigger fault judgment, the steady-state period before the last power frequency cycle is read from the queue as the reference value. The A-phase current effective value at the fault occurrence time is divided by the A-phase reference current value read from the queue to obtain the A-phase current unit value. Assuming that the A-phase current effective value decreases significantly, the A-phase current unit value is less than 1, the B-phase current effective value is divided by the B-phase reference current value to obtain the B-phase current unit value close to 1, and the C-phase current effective value is divided by the C-phase reference current value to obtain the C-phase current unit value close to 1. The three-phase current unit values are compared in size. The B-phase current unit value and the C-phase current unit value are close in value and greater than the A-phase current unit value. The maximum current unit value is selected as the larger one of the B-phase and C-phase, and the intermediate current unit value is selected as the smaller one of the B-phase and C-phase. The maximum current unit value is divided by the intermediate current unit value. Since the two values are close, the ratio is close to 1, indicating that the B-phase and C-phase current distribution is balanced, while the A-phase current unit value deviates significantly from the maximum current unit value and the intermediate current unit value. The ratio obtained by dividing the maximum current unit value by the A-phase current unit value is significantly greater than 1. This ratio is used as a three-phase current imbalance characteristic parameter to reflect the degree of three-phase imbalance caused by the abnormal decrease of the A-phase current.

[0020] In a specific embodiment, the state information of the switch on the high-voltage side of the distribution transformer is collected, and the state information of the switch is encoded as a state identifier. The three-phase current imbalance characteristic parameter is used to determine the fault type of the distribution transformer, including: The state information of the switch on the high-voltage side of the distribution transformer is collected, and the state information of the switch is encoded as a state identifier. The three-phase current imbalance characteristic parameter is used to determine the fault type of the distribution transformer, including: The power transmission direction identifier of the distribution loop where the distribution transformer is located is obtained, and the state identifier is used to query the fault perceptible area mapping table to obtain the perceptible fault position range. The three-phase current imbalance characteristic parameter is compared with a preset threshold value. When the three-phase current imbalance characteristic parameter is greater than the preset threshold value, the numerical distribution relationship of the phase current unit value is determined within the perceptible fault position range to determine the fault type of the distribution transformer.

[0021] Specifically, the intelligent terminal collects the incoming line switch state, outgoing line switch state and load switch state of the high-voltage side of the distribution transformer through a digital input interface, and the switch state signals are output through auxiliary contacts. When the switch is open, the low-level signal output by the auxiliary contact corresponds to the open state, and is defined as 0; when the switch is closed, the high-level signal output by the auxiliary contact corresponds to the closed state, and is defined as 1. The states of the three switches are combined in the order of incoming line switch, outgoing line switch and load switch to form a switch state triple, and the switch state triple is binary coded to obtain a state identifier. The state identifier is a three-bit binary number, the highest bit represents the incoming line switch state, the middle bit represents the outgoing line switch state, and the lowest bit represents the load switch state. When the incoming line switch is closed, the outgoing line switch is open, and the load switch is closed, the state triple is 101, which is converted to a decimal value of 5 as the state identifier. The power transmission direction identifier records the direction of power flow, and forward power transmission indicates that power flows from the incoming line side of the distribution transformer to the load side, and reverse power transmission indicates that power flows from the other end of the distribution transformer. The power transmission direction identifier is stored in the configuration parameters of the distribution transformer. The intelligent terminal reads the power transmission direction identifier, queries the fault perceptible area mapping table according to the state identifier, and the fault perceptible area mapping table is a pre-established table. Each row in the table records a perceptible fault location range corresponding to a state identifier. The perceptible fault location range corresponding to the state identifier 5 includes the incoming line side and the bus area, and the perceptible fault location range corresponding to the state identifier 7 includes the incoming line side, the bus area and the load side. The intelligent terminal looks up the corresponding row in the mapping table according to the state identifier, reads the perceptible fault location range field of the row to obtain the electrical location range that the intelligent terminal can detect under the current switch state combination, compares the calculated three-phase current imbalance characteristic parameter with the preset threshold value, and determines the preset threshold value according to a large amount of historical fault data statistical analysis. When the three-phase load is completely symmetrical, the three-phase current imbalance characteristic parameter is close to 1, and when a fault occurs, the parameter significantly increases, and the preset threshold value is set to 1.5, when the three-phase current imbalance characteristic parameter is greater than the preset threshold value, it is judged that the three-phase current distribution is abnormal, the fault type is determined according to the numerical distribution relationship of the current per unit value of each phase, the numerical distribution relationship includes that the current per unit value of a phase is significantly smaller than the current per unit value of the other two phases, the current per unit value of a phase is significantly larger than the current per unit value of the other two phases, the current per unit value of two phases is significantly smaller than the current per unit value of the other phase, etc. Different modes, when the A-phase current per unit value is significantly smaller than the B-phase and C-phase current per unit value, it is determined that the A-phase wire breaking fault occurs, when the A-phase current per unit value is significantly larger than the B-phase and C-phase current per unit value, it is determined that the A-phase ground fault occurs, the fault type of the distribution transformer is determined within the range of the perceivable fault position, when the range of the perceivable fault position includes the incoming line side and the bus area, the fault position is determined to be located in the incoming line side or the bus area according to the current per unit value distribution relationship, the high voltage side intelligent terminal of the distribution transformer collects the high level signal output by the auxiliary contact of the incoming line switch through the digital input interface, records the state of the incoming line switch as 1, collects the low level signal output by the auxiliary contact of the outgoing line switch, records the state of the outgoing line switch as 0, collects the high level signal output by the auxiliary contact of the load switch, records the state of the load switch as 1, combines the states of the three switches in order as the switch state triple 101, binary encodes the triple, the highest bit 1 represents the closing of the incoming line switch, the middle bit 0 represents the opening of the outgoing line switch, and the lowest bit 1 represents the closing of the load switch. The binary number 101 is converted into the decimal number 5 as the state identifier, the forward power transmission direction identifier is read from the configuration parameter storage area of the distribution transformer, the fault perceivable area mapping table is queried according to the state identifier 5, the perceivable fault position range corresponding to the state identifier 5 in the mapping table is the incoming line side and the bus area, the perceivable fault position range field of the row is read, the calculated three-phase current imbalance characteristic parameter is compared with the preset threshold value 1.5, the three-phase current imbalance characteristic parameter is greater than the preset threshold value, it is judged that the three-phase current distribution is abnormal, the A-phase current per unit value, the B-phase current per unit value and the C-phase current per unit value are read, the A-phase current per unit value is significantly smaller than the B-phase and C-phase current per unit value, the B-phase current per unit value and the C-phase current per unit value are close in value, and it is determined that the A-phase wire breaking fault occurs. Within the range of the perceivable fault position, combined with the state of the outgoing line switch, the fault type is determined to be the incoming line side A-phase wire breaking fault, when the outgoing line switch is closed, the fault type is determined to be the bus A-phase wire breaking fault.

[0022] In a specific embodiment, the topological relationship data of a plurality of distribution transformers in a distribution loop is established, and the determined fault type is verified according to the topological relationship data to confirm the fault determination result, including: A loop topological state matrix is constructed according to the connection order of the distribution transformers in the distribution loop, the loop topological state matrix records the fault type of each position distribution transformer, and a same dimension power transmission direction matrix is constructed to record the power transmission direction identifier of each position distribution transformer; Extracting the target distribution transformer with the fault type of feeder side fault or load side fault from the loop topology state matrix, obtaining the fault type set of the precedent distribution transformer and the fault type set of the subsequent distribution transformer of the target distribution transformer; Checking whether the fault type set of the precedent distribution transformer is all busbar faults, checking whether the fault type set of the subsequent distribution transformer is all no faults, and confirming that the fault determination result is correct when the checking passes.

[0023] In detail, the power distribution loop contains multiple power distribution transformers, a loop topology state matrix is constructed according to the physical connection order of the power distribution transformers in the loop, the loop topology state matrix is a one-dimensional array structure, each element of the array corresponds to the position of a power distribution transformer in the power distribution loop, the index number of the element represents the position number of the power distribution transformer in the loop, and the value of the element records the fault type identification of the power distribution transformer at the position; a power transmission direction matrix of the same dimension is constructed to record the power transmission direction identification of each position power distribution transformer, the power transmission direction matrix has the same array length as the loop topology state matrix, and each element records the power transmission direction of the corresponding position power distribution transformer; the power transmission direction identification includes forward power transmission and reverse power transmission; elements in the loop topology state matrix are traversed one by one, and the power distribution transformers with fault types of incoming line side fault or load side fault are taken as target power distribution transformers; the position index of the target power distribution transformer in the matrix is recorded, the fault type set of the preceding power distribution transformer and the fault type set of the subsequent power distribution transformer of the target power distribution transformer are obtained, the preceding power distribution transformer refers to the power distribution transformer upstream of the target power distribution transformer in the power distribution loop, and the subsequent power distribution transformer refers to the power distribution transformer downstream of the target power distribution transformer; the power transmission direction identification of the target power distribution transformer is read from the power transmission direction matrix; when the power transmission direction identification is forward power transmission, the preceding power distribution transformer is all the power distribution transformers with position indexes smaller than that of the target power distribution transformer, and the subsequent power distribution transformer is all the power distribution transformers with position indexes greater than that of the target power distribution transformer; the fault type identification of the corresponding position of the preceding power distribution transformer is read from the loop topology state matrix, and the fault type identification is combined to form the fault type set of the preceding power distribution transformer; the fault type identification of the corresponding position of the subsequent power distribution transformer is read from the loop topology state matrix, and the fault type identification is combined to form the fault type set of the subsequent power distribution transformer; each element in the fault type set of the preceding power distribution transformer is checked, and it is judged whether all the elements are bus fault type identifications; all the elements in the set are compared with the bus fault type identification, and when all the elements in the set are equal to the bus fault type identification, the fault type set of the preceding power distribution transformer passes the check; each element in the fault type set of the subsequent power distribution transformer is checked, and it is judged whether all the elements are no fault type identifications; all the elements in the set are compared with the no fault type identification, and when all the elements in the set are equal to the no fault type identification, the fault type set of the subsequent power distribution transformer passes the check; when the fault type set of the preceding power distribution transformer and the fault type set of the subsequent power distribution transformer both pass the check, it is confirmed that the fault determination result is correct; the power distribution loop contains five power distribution transformers, a loop topology state matrix is constructed according to the connection order, the matrix contains five elements, index 0 to index 4 correspond to the first to the fifth power distribution transformers respectively, and after the intelligent terminal of each power distribution transformer determines the fault type,Write the fault type identification into the corresponding position of the loop topology state matrix, record no fault at index 0 position, busbar fault at index 1 position, busbar fault at index 2 position, incoming line side fault at index 3 position, and no fault at index 4 position, construct a power transmission direction matrix, the power transmission direction identification of index 0 to index 4 is forward power transmission, traverse the elements in the loop topology state matrix, the fault type at index 3 position is incoming line side fault, extract the distribution transformer at index 3 position as the target distribution transformer, read the power transmission direction identification at index 3 position from the power transmission direction matrix as forward power transmission, the preceding distribution transformer is the distribution transformer with index less than 3, including the distribution transformers at index 0, index 1 and index 2 positions, the subsequent distribution transformer is the distribution transformer with index greater than 3, including the distribution transformer at index 4 position, read the fault type identification at index 0, index 1 and index 2 positions from the loop topology state matrix, the fault type identification at index 0 position is no fault, the fault type identification at index 1 position is busbar fault, and the fault type identification at index 2 position is busbar fault, combine the three fault type identifications to form a preceding distribution transformer fault type set, traverse the preceding distribution transformer fault type set, the fault type at index 0 position is not equal to busbar fault, the preceding distribution transformer fault type set fails to pass the check, read the fault type identification at index 4 position from the loop topology state matrix as no fault, combine the fault type identification to form a subsequent distribution transformer fault type set, traverse the subsequent distribution transformer fault type set, the fault type at index 4 position is equal to no fault, the subsequent distribution transformer fault type set passes the check, since the preceding distribution transformer fault type set fails to pass the check, the fault determination result is not confirmed to be correct.

[0024] In an embodiment, the preceding distribution transformer fault type set is checked whether all are busbar faults, and the subsequent distribution transformer fault type set is checked whether all are no faults, when the checks all pass, the fault determination result is confirmed to be correct, comprising: Read the power transmission direction identification value of the target distribution transformer from the power transmission direction matrix, and determine the position range of the preceding distribution transformer fault type set and the subsequent distribution transformer fault type set according to the power transmission direction identification value; Calculate the busbar fault type matching degree of the preceding distribution transformer fault type set, and obtain a preceding matching degree value by ratio operation of the number of busbar fault types in the set to the total number of elements in the set; calculate the no fault type matching degree of the subsequent distribution transformer fault type set, and obtain a subsequent matching degree value by ratio operation of the number of no fault types in the set to the total number of elements in the set; When the preceding matching degree value and the subsequent matching degree value are both equal to 1, set the loop state identification value at the corresponding position of the loop topology state matrix as single-point fault locatable state; when the preceding matching degree value or the subsequent matching degree value is less than 1, set the loop state identification value as multi-point fault to be checked state.

[0025] Specifically, the power transmission direction identifier value of the target distribution transformer is read from the power transmission direction matrix, the power transmission direction identifier value includes two types of forward power transmission and reverse power transmission, the position ranges of the pre-sequence distribution transformer fault type set and the post-sequence distribution transformer fault type set are determined according to the power transmission direction identifier value, when the power transmission direction identifier value is forward power transmission, the pre-sequence distribution transformer is all the distribution transformers before the position index of the target distribution transformer in the loop topology state matrix, and the post-sequence distribution transformer is all the distribution transformers after the position index, when the power transmission direction identifier value is reverse power transmission, the pre-sequence distribution transformer is all the distribution transformers after the position index, and the post-sequence distribution transformer is all the distribution transformers before the position index, busbar fault type matching degree calculation is performed on the pre-sequence distribution transformer fault type set, each element in the pre-sequence distribution transformer fault type set is traversed, the number of elements with fault type identifier equal to busbar fault is counted, and ratio operation is performed on the number of busbar fault types and the total number of set elements, the number of busbar fault types is taken as the numerator, and the total number of set elements is taken as the denominator for division calculation to obtain a pre-sequence matching degree value, the pre-sequence matching degree value ranges from 0 to 1, the pre-sequence matching degree value is equal to 1 when all the elements in the set are busbar faults, and the pre-sequence matching degree value is equal to 0 when there is no busbar fault in the set, no fault type matching degree calculation is performed on the post-sequence distribution transformer fault type set, each element in the post-sequence distribution transformer fault type set is traversed, the number of elements with fault type identifier equal to no fault is counted, and ratio operation is performed on the number of no fault types and the total number of set elements, the number of no fault types is taken as the numerator, and the total number of set elements is taken as the denominator for division calculation to obtain a post-sequence matching degree value, the post-sequence matching degree value ranges from 0 to 1, the post-sequence matching degree value is equal to 1 when all the elements in the set are no faults, and the post-sequence matching degree value is equal to 0 when there is no no fault in the set, whether the pre-sequence matching degree value and the post-sequence matching degree value are equal to 1 is judged, when the pre-sequence matching degree value is equal to 1 and the post-sequence matching degree value is equal to 1, it indicates that all the distribution transformers upstream of the target distribution transformer detect busbar faults, and all the distribution transformers downstream of the target distribution transformer do not detect faults, which accords with the electrical propagation rule of single-point faults, the loop state identifier value corresponding to the position of the target distribution transformer in the loop topology state matrix is set to a single-point fault locatable state, the single-point fault locatable state indicates that the fault point position has been accurately located and the fault type judgment is correct, when the pre-sequence matching degree value is less than 1 or the post-sequence matching degree value is less than 1, it indicates that there is a non-busbar fault type in the pre-sequence distribution transformer or a non-no fault type in the post-sequence distribution transformer, which does not accord with the electrical propagation rule of single-point faults, the loop state identifier value is set to a multiple-point fault to be verified state, the multiple-point fault to be verified state indicates that there are multiple fault points in the distribution loop or the fault judgment result needs to be further verified, and the distribution loop includes five distribution transformers,The distribution transformer at the index 3 position of the loop topology state matrix is determined as the target distribution transformer for the incoming line side fault, the power supply direction identification value at the index 3 position of the power supply direction matrix is read as the forward power supply, the position range of the preceding distribution transformer is determined as index 0 to index 2 and the position range of the subsequent distribution transformer is index 4 according to the forward power supply, the fault type identification at the index 0 to index 2 positions of the loop topology state matrix is read, the index 0 position is no fault, the index 1 position is busbar fault, and the index 2 position is busbar fault, the three fault type identifications are combined to form a fault type set of the preceding distribution transformer, the total number of elements in the set is 3, the fault type identification at the index 1 position is equal to busbar fault, the fault type identification at the index 2 position is equal to busbar fault, the number of busbar fault types is 2, and the number 2 of busbar fault types is divided by the total number 3 of elements in the set to obtain a preceding matching degree value of about 0.67, the fault type identification at the index 4 position of the loop topology state matrix is read as no fault, and the fault type identification is combined to form a fault type set of the subsequent distribution transformer, the total number of elements in the set is 1, the fault type identification at the index 4 position is equal to no fault, the number of no fault types is 1, and the number 1 of no fault types is divided by the total number 1 of elements in the set to obtain a subsequent matching degree value equal to 1, it is judged that the preceding matching degree value 0.67 is not equal to 1, the loop state identification value at the index 3 position of the loop topology state matrix is set as a multi-point fault to be verified state, when another condition in the loop is that the fault type identifications at the index 0 to index 2 positions are all busbar fault, the number of busbar fault types is 3, the total number of elements in the set is 3, and 3 divided by 3 obtains a preceding matching degree value equal to 1, the subsequent matching degree value is still 1, the preceding matching degree value is equal to 1 and the subsequent matching degree value is equal to 1, and the loop state identification value is set as a single-point fault locatable state.

[0026] In an embodiment, isolation control instructions are generated according to the confirmed fault determination result, and the isolation control instructions are executed to realize fault area isolation and non-fault area power supply recovery, including: The loop state identification value is read from the loop topology state matrix, when the loop state identification value is a single-point fault locatable state, the position index and the fault type of the fault distribution transformer are extracted, and the target isolation switch combination is obtained according to the fault type query fault isolation strategy mapping table; Switch opening instructions are generated according to the target isolation switch combination, and the switch opening instructions are sent to the high-voltage side switch actuator of the corresponding distribution transformer through the communication link to control the switches in the target isolation switch combination to perform opening operation and complete fault area isolation; A set of healthy distribution transformers in the same distribution loop as the faulted distribution transformer and located in the non-faulted area is identified from the loop topology state matrix, and a topology analysis is performed on the power supply paths of each distribution transformer in the set of healthy distribution transformers to generate a switch closing instruction sequence to control the switches on the standby power supply path to perform closing operations, thereby completing power supply restoration of the set of healthy distribution transformers.

[0027] In particular, the loop state identification value is read from the loop topology state matrix, the loop state identification value is recorded in the matrix position corresponding to the target distribution transformer, when the loop state identification value is a single-point fault locatable state, the position index and fault type of the fault distribution transformer are extracted, the position index is the position number of the distribution transformer in the loop topology state matrix, and the fault type is the fault type identification determined by the intelligent terminal of the distribution transformer, the target switch combination is obtained by querying the fault isolation strategy mapping table according to the fault type, the fault isolation strategy mapping table is a pre-planned table, each row of the table records the switch combination that needs to be opened corresponding to a fault type, when the fault type is an incoming line fault, the target switch combination includes the incoming line switch of the fault distribution transformer and the outgoing line switch of the last distribution transformer, when the fault type is a load side fault, the target switch combination includes the load switch of the fault distribution transformer, the switch opening instruction sequence is generated according to the target switch combination, the switch opening instruction sequence is a series of switch control instructions arranged in operation order, each instruction contains the position index, switch type and opening operation code of the target distribution transformer, the switch opening instruction sequence is sent to the high-voltage side switch actuator of the corresponding distribution transformer through the communication link, the communication link includes wireless communication or wired communication, the intelligent terminal sends each instruction in the instruction sequence in order, the switch actuator receives the instruction and drives the switch operating mechanism to execute the opening operation, controls the switches in the target switch combination to execute the opening operation and completes the fault area isolation, the fault area isolation refers to electrically disconnecting the fault point from the power supply and the healthy area through the opening operation, a set of healthy distribution transformers in the same distribution loop as the fault distribution transformer and located in the non-fault area is identified from the loop topology state matrix, the set of healthy distribution transformers includes all distribution transformers with no fault type identification, all elements in the loop topology state matrix are traversed, and distribution transformers with no fault type identification and position index greater than that of the fault distribution transformer are screened out, the distribution transformers are combined to form the set of healthy distribution transformers, the power supply paths of the distribution transformers in the set of healthy distribution transformers are topologically analyzed, the topological analysis is to traverse the connection relationship of the distribution loop to find the available power supply path from the standby power supply to the healthy distribution transformer, the distribution loop usually has a ring network structure, in addition to the main power supply path, there is also a standby power supply path, the standby power supply path is connected with the standby power supply through a tie switch, the topological analysis first determines the positions of the distribution transformers in the set of healthy distribution transformers, then finds the standby power supply at the other end of the distribution loop, traces all switch states between the standby power supply and the healthy distribution transformer, generates a switch closing instruction sequence to control the switches on the standby power supply path to execute the closing operation, the switch closing instruction sequence includes control instructions of the tie switch and sectionalizing switch that need to be closed, the switch closing instruction sequence is sent to the corresponding switch actuator through the communication link, the switch actuator drives the switch operating mechanism to execute the closing operation,In the power supply recovery of the complete set of distribution transformers, the distribution transformer at index 3 in the distribution loop is determined to be an in-line fault, and the loop state identifier value is a single-point fault locatable state, the loop state identifier value at index 3 in the loop topology state matrix is read as a single-point fault locatable state, the position index of the fault distribution transformer is extracted as 3, and the fault type is an in-line fault. According to the in-line fault type, the fault isolation strategy mapping table is queried, the target isolation switch combination corresponding to the in-line fault type in the mapping table includes the in-line switch of the distribution transformer at index 3 and the out-line switch of the distribution transformer at index 2, a switch opening instruction sequence is generated, the first instruction includes position index 3, switch type is in-line switch, and operation code is opening, the second instruction includes position index 2, switch type is out-line switch, and operation code is opening, the first instruction is sent to the in-line switch actuator of the distribution transformer at index 3 through the communication link, the switch actuator receives the instruction and drives the in-line switch to open, the second instruction is sent to the out-line switch actuator of the distribution transformer at index 2 through the communication link, the switch actuator drives the out-line switch to open, and the fault area isolation is completed. All elements in the loop topology state matrix are traversed, the fault type at index 4 is identified as no fault, and the position index is greater than the position index of the fault distribution transformer 3. The distribution transformer at index 4 forms a complete set of distribution transformers, and the power supply path of the distribution transformer at index 4 is analyzed. There is a standby power supply at the other end of the distribution loop, the standby power supply is connected with the distribution transformer at index 4 through a tie switch, the tie switch is currently in an open state, a switch closing instruction sequence is generated, the instruction includes the position identifier of the tie switch and the closing operation code, the instruction is sent to the tie switch actuator through the communication link, the switch actuator drives the tie switch to close, the distribution transformer at index 4 obtains power supply from the standby power supply, and the power supply recovery is completed.

[0028] The above describes the fault identification method of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the application. The fault identification system of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the application is described below. Please refer to Figure 2 An embodiment of the fault identification system of the distributed distribution transformer high-voltage side intelligent terminal in the embodiment of the application includes: A processing module is configured to collect three-phase current signals at the high-voltage side of a distribution transformer, perform signal preprocessing on the three-phase current signals, and obtain preprocessed current signals. A calculation module is configured to extract effective values of the phases of the preprocessed current signals, perform ratio calculation on the current effective values of the phases and corresponding reference current values, obtain current unit values of the phases, and calculate three-phase current imbalance characteristic parameters according to the current unit values of the phases. The coding module is configured to collect state information of the switch on the high-voltage side of the distribution transformer, encode the state information of the switch as a state identifier, and determine the fault type of the distribution transformer in combination with the three-phase current imbalance characteristic parameter; The checking module is configured to establish topological relationship data of a plurality of distribution transformers in a distribution loop, check the determined fault type according to the topological relationship data, and confirm a fault determination result. The implementation module is configured to generate an isolation control instruction according to the confirmed fault determination result, and implement fault area isolation and non-fault area power supply recovery by executing the isolation control instruction.

[0029] The above Figure 2 The distributed distribution transformer high-voltage side intelligent terminal fault identification system in the embodiment of the application is described in detail from the perspective of a modular functional entity, and the distributed distribution transformer high-voltage side intelligent terminal fault identification device in the embodiment of the application is described in detail from the perspective of hardware processing.

[0030] Referring to Figure 3 The embodiment of the application also provides a distributed distribution transformer high-voltage side intelligent terminal fault identification device, which can be a server, and the internal structure thereof can be as shown in Figure 3 The distributed distribution transformer high-voltage side intelligent terminal fault identification device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. The processor of the computer is used to provide computing and control capabilities. The memory of the distributed distribution transformer high-voltage side intelligent terminal fault identification device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the distributed distribution transformer high-voltage side intelligent terminal fault identification device is used to store the corresponding data in the embodiment. The network interface of the distributed distribution transformer high-voltage side intelligent terminal fault identification device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement the above method.

[0031] Those skilled in the art can understand Figure 3 The structure shown in the embodiment of the application is only a block diagram of part of the structure related to the scheme of the application, and does not constitute a limitation on the distributed distribution transformer high-voltage side intelligent terminal fault identification device to which the scheme of the application is applied.

[0032] The application further provides a computer readable storage medium, which can be a nonvolatile computer readable storage medium or a volatile computer readable storage medium, and the computer readable storage medium stores instructions, and the instructions make a computer execute the steps of the distributed distribution transformer high-voltage side intelligent terminal fault identification method when the instructions are run on the computer.

[0033] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0034] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the application or the whole or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for making a distributed distribution transformer high-voltage side intelligent terminal fault identification device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0035] The above embodiments are only used to illustrate the technical solutions of the application, rather than limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these 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 application.

Claims

1. A method for fault identification in a distributed distribution transformer high-voltage side intelligent terminal, characterized in that, The method includes: The three-phase current signal on the high-voltage side of the distribution transformer is collected, and the three-phase current signal is preprocessed to obtain the preprocessed current signal. Extract the effective value of each phase current of the preprocessed current signal, calculate the ratio of the current effective value of each phase current to the corresponding reference current value, obtain the per-unit value of each phase current, and calculate the three-phase current imbalance characteristic parameters based on the per-unit value of each phase current. The status information of the high-voltage side switch of the distribution transformer is collected, the status information of the switch is encoded into a status identifier, and the fault type of the distribution transformer is determined by combining the three-phase current imbalance characteristic parameters. Establish topological relationship data for multiple distribution transformers within the distribution loop, verify the determined fault type based on the topological relationship data, and confirm the fault determination result; Based on the confirmed fault determination result, an isolation control command is generated, and the isolation control command is executed to achieve isolation of the faulty area and restoration of power supply to the non-faulty area.

2. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 1, characterized in that, The process of acquiring the three-phase current signal from the high-voltage side of the distribution transformer, and performing signal preprocessing on the three-phase current signal to obtain a preprocessed current signal includes: The original three-phase current signals are obtained by synchronously acquiring the A-phase current signal, B-phase current signal, and C-phase current signal at the positions of the incoming switch, outgoing switch, and load switch on the high-voltage side of the distribution transformer using a current transformer. The original three-phase current signal is decomposed into multiple wavelet bases to obtain approximation coefficients and multiple detail coefficients. The multiple detail coefficients are then denoised using a threshold function. The denoised current signal is then reconstructed by inverse wavelet transform. The denoised current signal is converted to the frequency domain by Fast Fourier Transform, filtered by the frequency domain transfer function of the Wiener filter, and then converted to the time domain by Inverse Fast Fourier Transform to obtain the preprocessed current signal.

3. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 1, characterized in that, The process involves extracting the effective values ​​of each phase current from the preprocessed current signal, calculating the ratio between the current effective values ​​of each phase current and the corresponding reference current values ​​to obtain the per-unit value of each phase current, and calculating the three-phase current imbalance characteristic parameters based on the per-unit values ​​of each phase current, including: The preprocessed current signal is subjected to effective value extraction processing to obtain the effective values ​​of phase A current, phase B current and phase C current respectively; Read the A-phase reference current value, B-phase reference current value, and C-phase reference current value before the fault occurred from the historical current reference value storage queue. Divide the effective value of the A-phase current by the A-phase reference current value to obtain the per-unit value of the A-phase current. Divide the effective value of the B-phase current by the B-phase reference current value to obtain the per-unit value of the B-phase current. Divide the effective value of the C-phase current by the C-phase reference current value to obtain the per-unit value of the C-phase current. The maximum current per-unit value and the intermediate current per-unit value are selected from the current per-unit values ​​of phase A, phase B, and phase C. The maximum current per-unit value is divided by the intermediate current per-unit value to obtain the three-phase current imbalance characteristic parameters.

4. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 1, characterized in that, The process of collecting the status information of the high-voltage side switch of the distribution transformer, encoding the status information of the switch as a status identifier, and determining the fault type of the distribution transformer by combining the three-phase current imbalance characteristic parameters includes: The status of the incoming switch, outgoing switch and load switch on the high-voltage side of the distribution transformer is collected. The open state is defined as 0 and the closed state is defined as 1. The status identifier is obtained by encoding the switch status triplet according to the combination of the three switch states. Obtain the power supply direction identifier of the power distribution loop where the power distribution transformer is located, and query the fault-sensible area mapping table according to the status identifier to obtain the range of the perceptible fault location. The three-phase current imbalance characteristic parameters are compared with a preset threshold. When the three-phase current imbalance characteristic parameters are greater than the preset threshold, the fault type of the distribution transformer is determined by combining the numerical distribution relationship of the per-unit values ​​of the phase currents within the range of the perceptible fault location.

5. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 4, characterized in that, The process of establishing topological relationship data for multiple distribution transformers within a distribution loop, verifying the determined fault type based on the topological relationship data, and confirming the fault determination result includes: A loop topology state matrix is ​​constructed according to the connection sequence of the distribution transformers in the distribution loop. The loop topology state matrix records the fault type of the distribution transformer at each location. A power transmission direction matrix of the same dimension is constructed to record the power transmission direction identifier of the distribution transformer at each location. Extract the target distribution transformer with fault type of incoming side fault or load side fault from the loop topology state matrix, and obtain the set of fault types of the preceding distribution transformer and the set of fault types of the following distribution transformer for the target distribution transformer; Check whether the set of fault types of the preceding distribution transformers are all bus faults, and check whether the set of fault types of the following distribution transformers are all fault-free. When all checks pass, the fault determination result is confirmed to be correct.

6. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 5, characterized in that, The process of checking whether the preceding set of distribution transformer fault types all represent bus faults and checking whether the subsequent set of distribution transformer fault types all represent no faults, confirming the fault determination result is correct when all checks pass, includes: Read the power transmission direction identifier value of the target distribution transformer from the power transmission direction matrix, and determine the location range of the fault type set of the preceding distribution transformer and the fault type set of the following distribution transformer based on the power transmission direction identifier value; The bus fault type matching degree is calculated for the set of preceding distribution transformer fault types. The ratio of the number of bus fault types in the set to the total number of elements in the set is calculated to obtain the preceding matching degree value. The fault-free type matching degree is calculated for the set of subsequent distribution transformer fault types. The ratio of the number of fault-free types in the set to the total number of elements in the set is calculated to obtain the subsequent matching degree value. When both the preceding and following matching degree values ​​are equal to 1, the loop status identifier value at the corresponding position in the loop topology status matrix is ​​set to a single-point fault locationable state; when either the preceding or following matching degree value is less than 1, the loop status identifier value is set to a multi-point fault pending verification state.

7. The method for fault identification of a distributed distribution transformer high-voltage side intelligent terminal according to claim 6, characterized in that, The step of generating isolation control commands based on the confirmed fault determination results and executing the isolation control commands to achieve fault area isolation and power restoration to non-fault areas includes: Read the loop status identifier value from the loop topology state matrix. When the loop status identifier value is a single-point fault locationable state, extract the location index of the faulty distribution transformer and the fault type. Query the fault isolation strategy mapping table according to the fault type to obtain the target disconnect switch combination. A switch-opening instruction sequence is generated based on the target disconnector switch combination. The switch-opening instruction sequence is then sent to the high-voltage side switch actuator of the corresponding distribution transformer via a communication link, thereby controlling the switches in the target disconnector switch combination to perform the opening operation and complete the isolation of the fault area. Identify a set of healthy distribution transformers that are in the same distribution loop as the faulty distribution transformer and located in a non-faulty area from the loop topology state matrix. Perform topology analysis on the power supply path of each distribution transformer in the healthy distribution transformer set, generate a switch closing command sequence to control the switch on the backup power supply path to perform the closing operation, and complete the power supply restoration of the healthy distribution transformer set.

8. A fault identification system for a distributed distribution transformer high-voltage side intelligent terminal, characterized in that, For implementing the distributed distribution transformer high-voltage side intelligent terminal fault identification method as described in any one of claims 1-7, the distributed distribution transformer high-voltage side intelligent terminal fault identification system comprises: The processing module is used to acquire the three-phase current signal on the high-voltage side of the distribution transformer, perform signal preprocessing on the three-phase current signal, and obtain the preprocessed current signal. The calculation module is used to extract the effective values ​​of each phase current of the preprocessed current signal, calculate the ratio of the current effective value of each phase current to the corresponding reference current value, obtain the per-unit value of each phase current, and calculate the three-phase current imbalance characteristic parameters based on the per-unit value of each phase current. The encoding module is used to collect the status information of the high-voltage side switch of the distribution transformer, encode the status information of the switch into a status identifier, and determine the fault type of the distribution transformer by combining the three-phase current imbalance characteristic parameters. The verification module is used to establish topological relationship data of multiple distribution transformers within the distribution loop, verify the determined fault type based on the topological relationship data, and confirm the fault determination result. The implementation module is used to generate isolation control instructions based on the confirmed fault determination results, and execute the isolation control instructions to realize the isolation of the fault area and the restoration of power supply to the non-fault area.

9. A fault identification device for a distributed distribution transformer high-voltage side intelligent terminal, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the distributed distribution transformer high-voltage side intelligent terminal fault identification method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to execute the distributed distribution transformer high-voltage side intelligent terminal fault identification method as described in any one of claims 1 to 7.

Images