Distributed wireless control system and method for transformer winding deformation detection

Abstract

The invention discloses a distributed wireless control system and method for transformer winding deformation detection. The distributed wireless control system comprises a plurality of transformers, a first deformation detection device, a second deformation detection device, an edge computing server and a cloud platform, wherein the first deformation detection device and the second deformation detection device are electrically connected with windings of the transformers; the first deformation detection device is used for generating a first fault signal when a winding deformation fault is detected; the edge computing server is used for analyzing according to the type of the fault winding and the short-circuit impedance variable quantity to obtain a first fault credibility level; and the cloud platform is used for obtaining the second fault credibility according to the historical data of the transformer winding, and determining a fault secondary judgment strategy according to the first fault credibility level and the second fault credibility. According to the invention, two different types of winding deformation detection devices are arranged to work cooperatively, and a secondary judgment strategy is executed based on judgment of the fault credibility level and the fault credibility, so that the problem of deformation fault false alarm in actual work is reduced, and the accuracy and reliability of deformation fault detection are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of transformer winding deformation detection, and in particular relates to a distributed wireless control system and method for transformer winding deformation detection. Background Art

[0002] Transformers are important power equipment in power transmission and distribution. Transformer failures often lead to power supply safety problems, so it is necessary to set up a variety of devices to detect transformer failures. Transformer winding deformation failure is a common type of transformer failure.

[0003] In the prior art, transformer winding deformation faults are detected by a variety of methods such as short-circuit impedance method, pulse signal method, frequency response method, vibration measurement method, and optical fiber sensing method. However, in practical applications, the problem of false alarm of deformation faults often occurs. The reason is that the working environment of the transformer is complex and changeable. It is often difficult to set an accurate fault threshold based on a deformation detection method. When the deformation amount of the transformer winding is near the fault threshold, it is often impossible to accurately detect it, resulting in a false alarm of deformation fault. Usually, the fault signal will cause the transformer to shut down and require staff to go to deal with it, which consumes a lot of manpower and material resources. Therefore, there is an urgent need for a technical solution that can accurately and reliably detect winding deformation faults to solve the above technical problems.

[0004] The invention patent application CN118408509A in the prior art discloses a transformer winding deformation detection system and method based on distributed sensing, including multiple transformers to be detected, multiple winding deformation detection devices, and a monitoring cloud platform; the multiple winding deformation detection devices include at least one standard detection device and multiple distributed wireless detection devices; the winding of the first transformer to be tested is connected to the standard detection device and the distributed wireless detection device; the winding of the second transformer to be tested is connected to the distributed wireless detection device. This invention patent application obtains more real and reliable detection data by comparing different deformation detection devices, thereby improving the accuracy of transformer winding deformation judgment. However, this invention patent application fails to solve technical problems such as false alarms of deformation faults. Summary of the invention

[0005] The present invention provides a distributed wireless control system, a system and a readable storage medium for transformer winding deformation detection, which are used to solve the technical problem of false alarm of deformation fault.

[0006] In a first aspect, the present invention provides a distributed wireless control system for transformer winding deformation detection, comprising a plurality of transformers, a first deformation detection device electrically connected to the winding of the transformer, a second deformation detection device, an edge computing server, and a cloud platform; the first deformation detection device and the second deformation detection device are both communicatively connected to the edge computing server through a wireless transmission module; the edge computing server is communicatively connected to the cloud platform; the first deformation detection device is used to generate a first fault signal when a winding deformation fault is detected; The edge computing server is used to receive the first fault signal, and obtain the first fault credibility level according to the type of the faulty winding and the short-circuit impedance change analysis; The cloud platform is used to obtain a second fault credibility based on historical data of the transformer winding, wherein the second fault credibility includes longitudinal fault credibility and transverse fault credibility; the cloud platform is also used to determine a secondary fault judgment strategy based on the first fault credibility level and the second fault credibility.

[0007] Furthermore, the edge computing server includes a fault credibility level calculation module, which is used to query the voltage level and power level of the faulty winding according to the working status information of the faulty winding; determine whether the faulty winding is a high-voltage winding or a low-voltage winding according to the voltage level and power level; obtain the short-circuit impedance change of the faulty winding in the first fault signal, if the faulty winding is a high-voltage winding, mark the short-circuit impedance change as a first change, if the faulty winding is a low-voltage winding, mark the short-circuit impedance change as a second change; determine the first fault credibility level according to the first change and / or the second change.

[0008] Further, the cloud platform includes a fault credibility analysis module, and the fault credibility analysis module includes a vertical credibility analysis module and a horizontal credibility analysis module; The longitudinal credibility analysis module is used to query the fault confirmation signal from the database, the fault confirmation signal includes a timestamp; obtain the fault time series signal according to the timestamp signal; obtain the first time period between each adjacent confirmed fault; obtain the N short-circuit impedance value sequences in the N first time periods closest to the current moment of the first fault signal; respectively compare the size relationship between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference with the fault-free time period among the N first time periods as the second time period; intercept data of the same time length as the fault-free time period in the second time period, and obtain the first trend curve and the second trend curve by curve fitting; perform correlation analysis on the first trend curve and the second trend curve, and calculate the correlation, which is the longitudinal fault credibility.

[0009] Furthermore, the lateral credibility analysis module is used to trace back the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; determine whether there is a fault confirmation signal in the comparison historical data; query other transformers of the same model to obtain short-circuit impedance historical data; fit the comparison historical data into a third trend curve by curve fitting, and fit the data of the winding corresponding to the first fault signal in the fault-free time period into a fourth trend curve by curve fitting; perform correlation analysis on the third trend curve and the fourth trend curve, and calculate the correlation, which is the lateral fault credibility.

[0010] In a second aspect, the present invention provides a distributed wireless control method for transformer winding deformation detection, comprising: Step S1, online monitoring of the deformation of several transformer windings based on the short-circuit impedance method, generating a first fault signal when an abnormality occurs; the first fault signal includes a fault identifier, working state information of the faulty winding, and a change in the short-circuit impedance of the faulty winding; the execution subject of the online monitoring is a first deformation detection device; the first deformation detection device is an online detection device for winding deformation based on the short-circuit impedance method; Step S2: The edge computing server receives the first fault signal, and obtains a first fault credibility level according to the type of the faulty winding and the change in short-circuit impedance; Step S3, analyzing historical data of the transformer winding based on the cloud platform to obtain a second fault credibility, including: Step S31, analyzing historical data of transformer windings to obtain longitudinal fault credibility; Step S32, analyzing the historical data of the transformer winding to obtain the lateral fault credibility; Step S4, determining a secondary fault judgment strategy according to the first fault credibility level and the second fault credibility; Step S5: The second deformation detection device performs deformation detection on the faulty winding corresponding to the first fault signal based on a frequency response method. If the detection is passed, the first fault signal is eliminated; otherwise, a fault confirmation signal is generated.

[0011] Furthermore, the step S2 comprises: Step S21: The edge computing server queries the voltage level and power level of the faulty winding according to the working status information of the faulty winding; Step S22: The edge computing server determines whether the faulty winding is a high-voltage winding or a low-voltage winding according to the voltage level and the power level; Step S23, obtaining the short-circuit impedance change of the faulty winding in the first fault signal, marking the short-circuit impedance change as the first change if the faulty winding is a high-voltage winding, and marking the short-circuit impedance change as the second change if the faulty winding is a low-voltage winding; Step S24: determining the first fault credibility level according to the first change amount and / or the second deformation amount.

[0012] Furthermore, the step S24 includes: Step S241: If the first change amount is greater than or equal to the first threshold, or the second change amount is greater than or equal to the second threshold, the first fault credibility level is high; Step S242: If the first change amount is less than the third threshold value and greater than or equal to the fifth threshold value, or the second change amount is less than the fourth threshold value and greater than or equal to the sixth threshold value, the first fault credibility level is low, and the process goes to step S3; Step S243: If the first change amount is greater than or equal to the third threshold and less than the first threshold, or the second change amount is greater than or equal to the fourth threshold and less than the second threshold, the first fault credibility level is medium, and the process proceeds to step S3; wherein the fifth threshold is the fault threshold of the high-voltage winding, the sixth threshold is the fault threshold of the low-voltage winding, and the fifth threshold < the sixth threshold < the third threshold < the fourth threshold < the first threshold < the second threshold; Step S244: If the first change is less than the fifth threshold, or the second change is less than the sixth threshold, the counter is incremented and timing starts, and the value of the counter within a preset time period is counted. If it exceeds the preset value, the first fault credibility level is low, and the process goes to step S3.

[0013] Furthermore, the step S31 includes: Step S311, querying a fault confirmation signal from a database, wherein the fault confirmation signal includes a timestamp; Step S312, obtaining a fault time series signal according to the timestamp signal; Step S313, obtaining a first time period between each adjacent confirmed fault; Step S314, obtaining a sequence of N short-circuit impedance values ​​in the N first time periods closest to the current moment of the first fault signal, where N is a preset positive integer; Step S315, respectively compare the magnitude relationships between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference with the fault-free time period among the N first time periods as the second time period; Step S316, intercepting data of the same time length as the fault-free time period at the maximum, and obtaining a first trend curve and a second trend curve by curve fitting; Step S317: perform correlation analysis on the first trend curve and the second trend curve, and calculate the correlation degree as the longitudinal fault credibility.

[0014] Furthermore, the step S32 includes: Step S321, tracing back the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; Step S322, judging whether there is a fault confirmation signal in the comparison historical data, if there is a fault confirmation signal in the other two phases, proceeding to step S41, otherwise proceeding to step S42; Step S323, query other transformers of the same model to obtain short-circuit impedance historical data, and return to step S31 until comparison historical data without a fault confirmation signal is found; Step S324, fitting the reference historical data into a third trend curve by curve fitting, and fitting the data of the winding corresponding to the first fault signal in the fault-free time period into a fourth trend curve by curve fitting; Step S325: performing a correlation analysis on the third trend curve and the fourth trend curve, and calculating the correlation degree as the lateral fault credibility.

[0015] Furthermore, the step S4 comprises: Step S41: If the first fault credibility level is low, and the longitudinal fault credibility is less than a preset value or the transverse fault credibility is less than a preset value, proceed to step S5; Step S42: if the first fault credibility level is medium, and the longitudinal fault credibility is less than a preset value and the transverse fault credibility is less than a preset value, proceed to step S5; Step S43: If the first fault credibility level is high, directly generate a fault confirmation signal.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides two different types of winding deformation detection devices, wherein the first deformation detection device is an online winding deformation detection device based on a short-circuit impedance method; the second deformation detection device is an offline winding deformation detection device and / or an online winding deformation detection device based on a frequency response method; the two types of detection devices complement each other and can more accurately detect winding deformation faults; The present invention judges the trust level and credibility of the fault signal generated by the first deformation detection device, and takes into account the different characteristics of the high-voltage winding and the low-voltage winding in the trust level judgment, which is more objective and accurate; in the credibility judgment, the vertical and horizontal dimensions are comprehensively considered, and the credibility level of the fault signal can be comprehensively judged, and based on this, it is decided whether to activate the second deformation detection device for secondary judgment of the deformation fault, which can greatly improve the detection efficiency.

[0017] The present invention can reduce the problem of false alarm of deformation faults in actual work and improve the accuracy and reliability of deformation fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of the structure of a distributed wireless control system for transformer winding deformation detection provided by an embodiment of the present invention; Figure 2 A structural diagram of an FIR filter system for transformer winding deformation detection provided by an embodiment of the present invention; Figure 3 A flow chart of a distributed wireless control method for transformer winding deformation detection provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The embodiment of the present invention provides a distributed wireless control system for transformer winding deformation detection. Figure 1 , Figure 1 A schematic diagram of the structure of a distributed wireless control system for transformer winding deformation detection provided by an embodiment of the present invention, the system includes: Several transformers, a first deformation detection device and a second deformation detection device electrically connected to the windings of the transformers, an edge computing server, and a cloud platform; The first deformation detection device and the second deformation detection device are both communicatively connected to the edge computing server via a wireless transmission module; The edge computing server is communicatively connected with the cloud platform; The first deformation detection device is used to generate a first fault signal when a winding deformation fault is detected; The edge computing server is used to receive the first fault signal, and obtain the first fault credibility level according to the type of the faulty winding and the short-circuit impedance change analysis, The cloud platform is used to obtain a second fault credibility according to historical data of the transformer winding, wherein the second fault credibility includes a longitudinal fault credibility and a transverse fault credibility; The cloud platform is also used to determine a secondary fault judgment strategy according to the first fault credibility level and the second fault credibility.

[0022] Wherein, the first deformation detection device is an online detection device for winding deformation based on the short-circuit impedance method; The second deformation detection device is an offline detection device and / or an online detection device for winding deformation based on a frequency response method.

[0023] Preferably, the edge computing server comprises a fault trust level calculation module, and the fault trust level calculation module is used to query the voltage level and power level of the fault winding according to the working state information of the fault winding; Determining whether the faulty winding is a high voltage winding or a low voltage winding according to the voltage level and the power level; Obtaining a short-circuit impedance change of the faulty winding in the first fault signal, marking the short-circuit impedance change as a first change if the faulty winding is a high-voltage winding, and marking the short-circuit impedance change as a second change if the faulty winding is a low-voltage winding; The first fault credibility level is determined according to the first change amount and / or the second deformation amount.

[0024] Preferably, the cloud platform includes a fault credibility analysis module, and the fault credibility analysis module includes a vertical credibility analysis module and a horizontal credibility analysis module; The longitudinal credibility analysis module is used to query the fault confirmation signal from the database, the fault confirmation signal includes a timestamp; obtain the fault time series signal according to the timestamp signal; Obtain a first time period between each adjacent confirmed fault; obtain a sequence of N short-circuit impedance values ​​in the N most recent first time periods before the current moment of the first fault signal; Compare the magnitude relationships between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference between the N first time periods and the fault-free time period as the second time period; Intercepting data of the same time length as that of the fault-free time period at most, and obtaining a first trend curve and a second trend curve by curve fitting; A correlation analysis is performed on the first trend curve and the second trend curve, and the calculated correlation is the longitudinal fault credibility.

[0025] Preferably, the lateral credibility analysis module is used to trace back the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; Determining whether there is a fault confirmation signal in the comparison historical data; Query other transformers of the same model to obtain historical short-circuit impedance data; The comparison historical data is fitted into a third trend curve by curve fitting, and the data of the winding corresponding to the first fault signal in the fault-free time period is fitted into a fourth trend curve by curve fitting; A correlation analysis is performed on the third trend curve and the fourth trend curve, and the calculated correlation is the lateral fault credibility.

[0026] Preferably, the second deformation detection device includes a monitoring chip and adopts an embedded algorithm, including a digital filtering algorithm, a sinusoidal signal amplitude and phase solution algorithm, and a correlation coefficient algorithm; Preferably, according to the sampling law, the highest frequency of the input analog signal should be less than half of the sampling frequency. In order to prevent the high-frequency components in the analog input signal (for example, local discharge signals, space electromagnetic wave signals, etc. mixed therein) from affecting the sampling of low-frequency components, a low-pass filter should be used to filter out high-frequency interference. In the online transfer function detection, in order to eliminate the interference of the power frequency and its harmonic signals, a high-pass filter should be used to filter out low-frequency interference.

[0027] Therefore, the dedicated chip should have a digital filter function, which should be placed after the A / D conversion of each channel signal acquisition and before other processing. The digital filter can use a finite impulse response FIR filter, solidify its impulse response function h(n) in the dedicated chip, and process the data through the following convolution operation: , In the formula, , Input binary signal for convolution operation, , , , is the response function matrix element, , In response to the signal; The above process can be completed through shifting, multiplication and accumulation during data acquisition.

[0028] Preferably, the correlation coefficient solution algorithm includes: Suppose there are two transfer function amplitude sequences of length N , , k=0, l,…,N-1, the correlation coefficient can be calculated according to the following formula.

[0029] Compute the standard deviation of two series: , , In the formula, , are two transfer function amplitude sequences, is the transfer function amplitude sequence The standard deviation of is the transfer function amplitude sequence The standard deviation of is the transfer function amplitude, amplitude, and length; Compute the covariance of two series: , In the formula, is the covariance of the two transfer function amplitude sequences; Compute the normalized covariance of two series: , In the formula, is the normalized covariance of the two transfer function amplitude sequences; Calculate the correlation coefficient: , In the formula, is the correlation coefficient; Preferably, in each link of generation, transmission and processing of monitoring signals, some interference signals will inevitably be introduced. To extract useful signals reflecting winding deformation, filtering must be performed. The present invention selects FIR digital filter and designs FIR filter based on FPGA to meet transformer winding deformation fault detection.

[0030] The digital filter order is 256. The specific implementation method is to divide the designed 256-order FIR filter into 64 4-order sub-filters, each filter is implemented by a 4-input LUT, and the filter coefficients have even symmetry, so the number of required LUTs can be halved, and the designed 256-order FIR filter only needs 32 4-input LUTs. The working process of designing a digital filter is: (1) Read 256 12-bit signed sampling data of the filter front-end ADC into the shift register one by one and perform shift buffering; (2) According to the symmetry of the filter coefficient, the data multiplied by the same coefficient in the input sample data are pre-added to form new 128 input data; (3) Perform serial-to-parallel conversion on the pre-added data, and combine the lowest bits of the input address into an address, with each four bits forming a group and sent to 32 lookup table modules respectively; (4) The second lowest bit of the input address is pieced together into an address, and each group of four bits is sent to 32 lookup tables for addressing. The outputs of each lookup table are added together, and the sum is the second partial product. The sum is added to the value stored in the register, and then shifted right by one bit and stored in the register. (5) Repeat steps (3) and (4) until the address consisting of the 11th bit of the input data is addressed. The eleventh partial product is obtained, which is added to the value in the register. The sum is shifted right one position and stored in the register. Finally, the value obtained by addressing the 12th bit (sign bit) of the sample value is subtracted to obtain the output result of the FIR filter.

[0031] According to the above working process, the FIR digital filter system can be divided into input delay module, pre-addition module, parallel-to-serial conversion module, lookup table module and shift accumulation module. For the structural block diagram, please refer to Figure 2 , Figure 2 A structural diagram of an FIR filter system for transformer winding deformation detection provided in an embodiment of the present invention.

[0032] The embodiment of the present invention also provides a distributed wireless control method for transformer winding deformation detection. Figure 3 , Figure 3 A flow chart of a distributed wireless control method for transformer winding deformation detection provided by an embodiment of the present invention, the method comprising the steps of: Step S1, online monitoring of deformation of a plurality of transformer windings based on a short-circuit impedance method, and generating a first fault signal when an abnormality occurs; The first fault signal includes a fault identifier, working state information of the fault winding, and a change in short-circuit impedance of the fault winding; The execution subject of online monitoring is the first deformation detection device; The first deformation detection device is an online detection device for winding deformation based on the short-circuit impedance method; Step S2: The edge computing server receives the first fault signal, and obtains a first fault credibility level according to the type of the faulty winding and the change in short-circuit impedance; Preferably, step S2 comprises: Step S21: The edge computing server queries the voltage level and power level of the faulty winding according to the working status information of the faulty winding; Step S22: The edge computing server determines whether the faulty winding is a high-voltage winding or a low-voltage winding according to the voltage level and the power level; Step S23, obtaining the short-circuit impedance change of the faulty winding in the first fault signal, marking the short-circuit impedance change as the first change if the faulty winding is a high-voltage winding, and marking the short-circuit impedance change as the second change if the faulty winding is a low-voltage winding; Step S24: determining the first fault credibility level according to the first change amount and / or the second deformation amount.

[0033] The present invention has found in actual experiments that, for the same degree of deformation, the short-circuit reactance change caused by the deformation of the low-voltage winding is greater than the short-circuit reactance change caused by the deformation of the high-voltage winding; that is, the low voltage is more sensitive and the detection is more accurate, and in practice, deformation often occurs in the low-voltage winding. Therefore, the present application considers the fault credibility level of the high-voltage winding and the low-voltage winding respectively; Preferably, the step S24 comprises: Step S241: If the first change amount is greater than or equal to the first threshold, or the second change amount is greater than or equal to the second threshold, the first fault credibility level is high; Step S242: If the first change amount is less than the third threshold value and greater than or equal to the fifth threshold value, or the second change amount is less than the fourth threshold value and greater than or equal to the sixth threshold value, the first fault credibility level is low, and the process goes to step S3; Step S243: If the first change amount is greater than or equal to the third threshold and less than the first threshold, or the second change amount is greater than or equal to the fourth threshold and less than the second threshold, the first fault credibility level is medium, and the process proceeds to step S3; wherein the fifth threshold is the fault threshold of the high-voltage winding, the sixth threshold is the fault threshold of the low-voltage winding, and the fifth threshold < the sixth threshold < the third threshold < the fourth threshold < the first threshold < the second threshold; Step S244: If the first change is less than the fifth threshold, or the second change is less than the sixth threshold, the counter is incremented and timing starts, and the value of the counter within a preset time period is counted. If it exceeds the preset value, the first fault credibility level is low, and the process goes to step S3.

[0034] In step S244, if a fault signal is received, but the change in short-circuit impedance is very small and less than the fault threshold, it means that it must be a false alarm. This may be the influence of accidental errors, resulting in a certain lag between the fault signal and the measured value of short-circuit impedance. If this situation occurs repeatedly in a short period of time, it means that the change in the measured value of short-circuit impedance is very small, but it fluctuates in the vicinity of the fault threshold. Therefore, it is necessary to further judge the fault credibility.

[0035] Step S3: Analyze historical data of the transformer winding based on the cloud platform to obtain a second fault credibility, where the second fault credibility includes a longitudinal fault credibility and a transverse fault credibility; Step S31, analyzing historical data of transformer windings to obtain longitudinal fault credibility; Preferably, the step S31 includes: Step S311, querying a fault confirmation signal from a database, wherein the fault confirmation signal includes a timestamp; Step S312, obtaining a fault time series signal according to the timestamp signal; Step S313, obtaining a first time period between each adjacent confirmed fault; Step S314, obtaining a sequence of N short-circuit impedance values ​​in the N first time periods closest to the current moment of the first fault signal, where N is a preset positive integer; Step S315, respectively compare the magnitude relationships between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference with the fault-free time period among the N first time periods as the second time period; Step S316, intercepting data of the same time length as the fault-free time period at the maximum, and obtaining a first trend curve and a second trend curve by curve fitting; Step S317: perform correlation analysis on the first trend curve and the second trend curve, and calculate the correlation degree as the longitudinal fault credibility.

[0036] Step S32, analyzing the historical data of the transformer winding to obtain the lateral fault credibility; Preferably, the step S32 includes: Step S321, tracing back the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; Step S322, judging whether there is a fault confirmation signal in the comparison historical data, if there is a fault confirmation signal in the other two phases, proceeding to step S41, otherwise proceeding to step S42; Step S323, query other transformers of the same model to obtain short-circuit impedance historical data, and return to step S31 until comparison historical data without a fault confirmation signal is found; Step S324, fitting the reference historical data into a third trend curve by curve fitting, and fitting the data of the winding corresponding to the first fault signal in the fault-free time period into a fourth trend curve by curve fitting; Step S325: performing a correlation analysis on the third trend curve and the fourth trend curve, and calculating the correlation degree as the lateral fault credibility.

[0037] Step S4, determining a secondary fault judgment strategy according to the first fault credibility level and the second fault credibility; Preferably, the step S4 comprises: Step S41: If the first fault credibility level is low, and the longitudinal fault credibility is less than a preset value or the transverse fault credibility is less than a preset value, proceed to step S5; Step S42: if the first fault credibility level is medium, and the longitudinal fault credibility is less than a preset value and the transverse fault credibility is less than a preset value, proceed to step S5; Step S43: If the first fault credibility level is high, directly generate a fault confirmation signal.

[0038] Step S5: The second deformation detection device performs deformation detection on the faulty winding corresponding to the first fault signal based on a frequency response method. If the detection is passed, the first fault signal is eliminated; otherwise, a fault confirmation signal is generated.

[0039] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0040] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0041] Finally, it should be noted that, in this article, relationships such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

Claims

1. A distributed wireless control system for transformer winding deformation detection, comprising a plurality of transformers, a first deformation detection device electrically connected to the transformer winding, a second deformation detection device, an edge computing server and a cloud platform; The first deformation detection device and the second deformation detection device are both connected to the edge computing server through a wireless transmission module; the edge computing server is connected to the cloud platform; characterized in that: The first deformation detection device is used to generate a first fault signal when a winding deformation fault is detected; The edge computing server is used to receive the first fault signal, and obtain the first fault credibility level according to the type of the faulty winding and the short-circuit impedance change analysis; The cloud platform is used to obtain a second fault credibility based on historical data of the transformer winding, wherein the second fault credibility includes longitudinal fault credibility and transverse fault credibility; the cloud platform is also used to determine a secondary fault judgment strategy based on the first fault credibility level and the second fault credibility.

2. A distributed wireless control system for transformer winding deformation detection according to claim 1, characterized in that: The edge computing server includes a fault trust level calculation module, and the fault trust level calculation module is used to query the voltage level and power level of the fault winding according to the working state information of the fault winding; Determine whether the faulty winding is a high-voltage winding or a low-voltage winding according to the voltage level and the power level; obtain the short-circuit impedance change of the faulty winding in the first fault signal, if the faulty winding is a high-voltage winding, mark the short-circuit impedance change as the first change, if the faulty winding is a low-voltage winding, mark the short-circuit impedance change as the second change; determine the first fault credibility level according to the first change and / or the second change.

3. A distributed wireless control system for transformer winding deformation detection according to claim 1, characterized in that: The cloud platform includes a fault credibility analysis module, and the fault credibility analysis module includes a vertical credibility analysis module and a horizontal credibility analysis module; The longitudinal credibility analysis module is used to query the fault confirmation signal from the database, the fault confirmation signal includes a timestamp; obtain the fault time series signal according to the timestamp signal; obtain the first time period between each adjacent confirmed fault; obtain the N short-circuit impedance value sequences in the N first time periods closest to the current moment of the first fault signal; respectively compare the size relationship between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference with the fault-free time period among the N first time periods as the second time period; intercept data of the same time length as the fault-free time period in the second time period, and obtain the first trend curve and the second trend curve by curve fitting; perform correlation analysis on the first trend curve and the second trend curve, and calculate the correlation, which is the longitudinal fault credibility.

4. A distributed wireless control system for transformer winding deformation detection according to claim 3, characterized in that: The horizontal credibility analysis module is used to trace the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; determine whether there is a fault confirmation signal in the comparison historical data; query other transformers of the same model to obtain short-circuit impedance historical data; fit the comparison historical data into a third trend curve by curve fitting, and fit the data of the winding corresponding to the first fault signal in the fault-free time period into a fourth trend curve by curve fitting; A correlation analysis is performed on the third trend curve and the fourth trend curve, and the calculated correlation is the lateral fault credibility.

5. A distributed wireless control method for transformer winding deformation detection, applied to the distributed wireless control system for transformer winding deformation detection as claimed in any one of claims 1 to 4, characterized in that: The method comprises: Step S1, online monitoring of the deformation of several transformer windings based on the short-circuit impedance method, generating a first fault signal when an abnormality occurs; the first fault signal includes a fault identifier, working state information of the faulty winding, and a change in the short-circuit impedance of the faulty winding; the execution subject of the online monitoring is a first deformation detection device; the first deformation detection device is an online detection device for winding deformation based on the short-circuit impedance method; Step S2: The edge computing server receives the first fault signal, and obtains a first fault credibility level according to the type of the faulty winding and the change in short-circuit impedance; Step S3, analyzing historical data of the transformer winding based on the cloud platform to obtain a second fault credibility, including: Step S31, analyzing historical data of transformer windings to obtain longitudinal fault credibility; Step S32, analyzing the historical data of the transformer winding to obtain the lateral fault credibility; Step S4, determining a secondary fault judgment strategy according to the first fault credibility level and the second fault credibility; Step S5: The second deformation detection device performs deformation detection on the faulty winding corresponding to the first fault signal based on a frequency response method. If the detection is passed, the first fault signal is eliminated; otherwise, a fault confirmation signal is generated.

6. The distributed wireless control method for transformer winding deformation detection according to claim 5, characterized in that: The step S2 comprises: Step S21: The edge computing server queries the voltage level and power level of the faulty winding according to the working status information of the faulty winding; Step S22: The edge computing server determines whether the faulty winding is a high-voltage winding or a low-voltage winding according to the voltage level and the power level; Step S23, obtaining the short-circuit impedance change of the faulty winding in the first fault signal, marking the short-circuit impedance change as the first change if the faulty winding is a high-voltage winding, and marking the short-circuit impedance change as the second change if the faulty winding is a low-voltage winding; Step S24: determining the first fault credibility level according to the first change amount and / or the second deformation amount.

7. The distributed wireless control method for transformer winding deformation detection according to claim 6, characterized in that: The step S24 comprises: Step S241: If the first change amount is greater than or equal to the first threshold, or the second change amount is greater than or equal to the second threshold, the first fault credibility level is high; Step S242: If the first change amount is less than the third threshold value and greater than or equal to the fifth threshold value, or the second change amount is less than the fourth threshold value and greater than or equal to the sixth threshold value, the first fault credibility level is low, and the process goes to step S3; Step S243: If the first change amount is greater than or equal to the third threshold and less than the first threshold, or the second change amount is greater than or equal to the fourth threshold and less than the second threshold, the first fault credibility level is medium, and the process proceeds to step S3; wherein the fifth threshold is the fault threshold of the high-voltage winding, the sixth threshold is the fault threshold of the low-voltage winding, and the fifth threshold < the sixth threshold < the third threshold < the fourth threshold < the first threshold < the second threshold; Step S244: If the first change is less than the fifth threshold, or the second change is less than the sixth threshold, the counter is incremented and timing starts, and the value of the counter within a preset time period is counted. If it exceeds the preset value, the first fault credibility level is low, and the process goes to step S3.

8. The distributed wireless control method for transformer winding deformation detection according to claim 5, characterized in that: The step S31 comprises: Step S311, querying a fault confirmation signal from a database, wherein the fault confirmation signal includes a timestamp; Step S312, obtaining a fault time series signal according to the timestamp signal; Step S313, obtaining a first time period between each adjacent confirmed fault; Step S314, obtaining a sequence of N short-circuit impedance values ​​in the N first time periods closest to the current moment of the first fault signal, where N is a preset positive integer; Step S315, respectively compare the magnitude relationships between the N first time periods and the fault-free time period before the current moment of the first fault signal, and take the first time period with the smallest absolute value difference with the fault-free time period among the N first time periods as the second time period; Step S316, intercepting data of the same time length as the fault-free time period at the maximum, and obtaining a first trend curve and a second trend curve by curve fitting; Step S317: perform correlation analysis on the first trend curve and the second trend curve, and calculate the correlation degree as the longitudinal fault credibility.

9. The distributed wireless control method for transformer winding deformation detection according to claim 5, characterized in that: The step S32 comprises: Step S321, tracing back the comparison historical data of the other two-phase windings of the transformer corresponding to the first fault signal and the fault-free time period; Step S322, judging whether there is a fault confirmation signal in the comparison historical data, if there is a fault confirmation signal in the other two phases, proceeding to step S41, otherwise proceeding to step S42; Step S323, query other transformers of the same model to obtain short-circuit impedance historical data, and return to step S31 until comparison historical data without a fault confirmation signal is found; Step S324, fitting the reference historical data into a third trend curve by curve fitting, and fitting the data of the winding corresponding to the first fault signal in the fault-free time period into a fourth trend curve by curve fitting; Step S325: performing a correlation analysis on the third trend curve and the fourth trend curve, and calculating the correlation degree as the lateral fault credibility.

10. The distributed wireless control method for transformer winding deformation detection according to claim 5, characterized in that: The step S4 comprises: Step S41: If the first fault credibility level is low, and the longitudinal fault credibility is less than a preset value or the transverse fault credibility is less than a preset value, proceed to step S5; Step S42: if the first fault credibility level is medium, and the longitudinal fault credibility is less than a preset value and the transverse fault credibility is less than a preset value, proceed to step S5; Step S43: If the first fault credibility level is high, directly generate a fault confirmation signal.

Images