Power transformer winding turn-to-turn discharge early warning system, method, medium and equipment

Abstract

The invention discloses a power transformer winding turn-to-turn discharge early warning system and method, a medium and equipment, and in the system, a signal acquisition module is used for collecting multi-physical discharge signals generated between turns of a power transformer winding in real time; the data cleaning module is used for carrying out preprocessing operation on the collected multi-physical discharge signals; the map construction module is used for extracting characteristic parameters of the cleaned multi-physical discharge signals and constructing a PRPD distribution map of the corresponding multi-physical discharge signals; the discharge severity identification module is used for generating an identification result of whether discharge is generated between winding turns according to the characteristic parameters, and analyzing a PRPD atlas to identify the severity of the winding turn-to-turn discharge; and the grading alarm module is used for automatically triggering a corresponding alarm mechanism according to the turn-to-turn discharge identification result, and triggering different grades of alarm responses according to different discharge severity degrees.

Description

Technical Field

[0001] The present invention relates to the technical field of power transformers, and in particular to a power transformer winding turn-to-turn discharge early warning system, method, medium and equipment. Background Art

[0002] As the core equipment in the power system, the power transformer undertakes the important tasks of voltage conversion and electric energy distribution. Its safe and stable operation is crucial to ensure the reliability and economy of the power grid. Interturn discharge of power transformer windings is a common failure mode, which will lead to the degradation of the insulation performance of power equipment and threaten the safe operation of the equipment. If maintenance is not carried out in time, long-term discharge may even cause the equipment to explode and cause irreversible damage to the insulation. Therefore, it is of great practical significance to study the real-time monitoring and timely warning technology of interturn discharge of power transformer windings.

[0003] Existing technologies for identifying the severity of discharge in power transformers mostly focus on using a single physical quantity, such as an electrical signal or an acoustic signal, while ignoring the correlation and complementarity between multiple physical quantities, resulting in incomplete feature extraction and high false alarm and missed alarm rates. In addition, existing early warning systems are mostly based on fixed threshold judgments, and perform online or offline analysis of collected real-time signals, which makes it difficult to adapt to early warning needs under different operating conditions and environmental conditions. Moreover, with the continuous expansion of the scale of the power system and the volume of equipment, the electromagnetic interference between equipment and the sound waves generated during operation cause many noise signals to be mixed in the collected real discharge signals. Traditional methods have made it difficult to accurately identify the severity of the discharge, and data cleaning of the signals is required to remove noise interference. The present invention proposes an intelligent identification and graded early warning system for the severity of discharge between turns of power transformer windings to address the deficiencies in the prior art and improve the safety and reliability of power transformer operation.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0005] The present invention provides a power transformer winding interturn discharge early warning system, method, medium and equipment, which use a variety of monitoring technologies and data analysis methods to achieve real-time monitoring of transformer winding interturn discharge phenomena, signal processing, feature extraction, and intelligent identification of discharge severity and graded early warning. It is of great significance for ensuring the safe operation of the power system, reducing the probability of failure, and improving the efficiency of equipment maintenance and repair, and is suitable for the field of online monitoring of power transformer winding interturn discharge.

[0006] A power transformer winding turn-to-turn discharge early warning system comprises:

[0007] A signal acquisition module, which is used to collect multi-physical discharge signals generated between turns of power transformer windings in real time;

[0008] A data cleaning module, which is used to perform preprocessing operations on the collected multi-physical discharge signals;

[0009] A spectrum construction module, which is used to extract characteristic parameters of the multi-physical discharge signal after cleaning and construct a PRPD distribution spectrum of the corresponding multi-physical discharge signal;

[0010] A discharge severity identification module, which is used to generate an identification result of whether discharge occurs between winding turns according to the characteristic parameters, and analyze the PRPD spectrum to identify the severity of the discharge between winding turns;

[0011] The graded alarm module is used to automatically trigger the corresponding alarm mechanism according to the inter-turn discharge identification results, and trigger different levels of alarm responses according to the severity of the discharge.

[0012] In the described power transformer winding inter-turn discharge warning system, the signal acquisition module includes a sensor array unit and a data acquisition unit, the sensor array unit includes a high-frequency current sensor subunit for real-time acquisition of current signals generated between turns of the power transformer winding, a voltage sensor subunit for real-time acquisition of voltage signals generated between turns of the power transformer winding, an ultrasonic sensor subunit for real-time acquisition of ultrasonic signals generated between turns of the power transformer winding, and an ultra-high frequency sensor subunit for real-time acquisition of ultra-high frequency signals generated between turns of the power transformer winding, and the data acquisition unit includes a high-precision data acquisition card subunit for ensuring synchronous acquisition and calibration of multiple physical signals.

[0013] In the described inter-turn discharge warning system for power transformer windings, the data cleaning module includes an outlier detection unit and a signal denoising unit, the outlier detection unit includes an outlier identification subunit for performing outlier identification on the multi-physical discharge signal based on a support vector machine algorithm and an outlier removal subunit for removing the identified outliers, and the signal denoising unit includes a Kalman filter subunit for performing noise removal on the signal after the outliers have been removed.

[0014] In the described inter-turn discharge warning system for power transformer windings, the spectrum construction module includes an extractor unit for performing time domain waveform and spectrum analysis on multi-physical discharge to extract time-frequency domain characteristic parameter features of multi-physical discharge signals and a spectrum constructor unit for generating a PRPD distribution spectrum of multi-physical discharge signals.

[0015] In the described early warning system for interturn discharge of power transformer windings, the discharge severity identification module includes a model training unit and a real-time identification unit for storing a learning model that has been trained in advance. The model training unit uses the abnormal values ​​of current, voltage, ultrasonic signal, and ultra-high frequency signal in the historical data set to complete the training of the long short-term memory neural network model in advance, and trains the long short-term memory neural network model through at least 20 rounds, 1000 iterations per round, to ensure that the accuracy of the long short-term memory neural network model in identifying whether interturn discharge occurs and the severity during operation is above 95%. The real-time identification unit includes an interturn discharge real-time identification subunit for inputting the characteristic parameters into the model training unit to realize the identification of the occurrence of interturn discharge and a discharge severity real-time identification subunit for inputting the PRPD spectrum into the model training unit to realize the identification of the severity of interturn discharge.

[0016] In the power transformer winding interturn discharge early warning system, the historical data set includes normal operation data and known fault data, and the discharge severity is divided into two levels: slight and severe.

[0017] In the described early warning system for inter-turn discharge of power transformer windings, the hierarchical alarm module includes a hierarchical alarm communication unit that sets a three-level early warning mechanism and automatically triggers it according to the identification result, and an early warning tracking and recording unit that is used to track the progress of alarm event processing and record alarm logs.

[0018] A method for early warning of interturn discharge of power transformer windings comprises the following steps:

[0019] Real-time collection of multi-physics discharge signals generated between turns of power transformer windings;

[0020] Performing preprocessing operations on the multi-physical discharge signal to clean and remove noise;

[0021] Extract characteristic parameters of the multi-physical discharge signal after cleaning and construct the PRPD distribution map of the corresponding signal;

[0022] Identify the generation of winding inter-turn discharge according to the characteristic parameters, and analyze the generated multi-physical discharge signal PRPD spectrum to identify the severity of the winding inter-turn discharge;

[0023] According to the inter-turn discharge identification results, the corresponding alarm mechanism is automatically triggered, and different levels of alarm responses are triggered according to the severity of the discharge.

[0024] A computer storage medium includes computer instructions, which, when executed on a computer, cause the computer to execute the method described above.

[0025] An electronic device, comprising:

[0026] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein:

[0027] When the processor executes the program, the method described is implemented.

[0028] Compared with the prior art, the present invention has the following advantages: the present invention realizes real-time monitoring, signal processing, feature extraction, intelligent identification of discharge severity and graded warning of transformer winding inter-turn discharge phenomenon by comprehensively using a variety of monitoring technologies and data analysis methods, which is of great significance for ensuring the safe operation of the power system, reducing the probability of faults and improving the efficiency of equipment maintenance and repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] By reading the detailed description of the preferred specific embodiments below, various other advantages and benefits of the present invention will become clear to those of ordinary skill in the art. The drawings in the specification are only for the purpose of illustrating the preferred embodiments and are not considered to be limitations of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work. Moreover, the same reference numerals are used to represent the same components throughout the drawings.

[0030] In the attached picture:

[0031] Figure 1 It is a schematic diagram of the architecture of the inter-turn discharge warning system of the power transformer winding;

[0032] Figure 2 It is the component architecture diagram of the signal acquisition module;

[0033] Figure 3 This is the component architecture diagram of the data cleaning module;

[0034] Figure 4 Build a modular component architecture diagram for the graph;

[0035] Figure 5 The component architecture diagram of the discharge severity identification module;

[0036] Figure 6 This is the component architecture diagram of the hierarchical alarm module.

[0037] The present invention is further explained below in conjunction with the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0038] The specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0039] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the present invention. The scope of protection of the present invention shall be determined by the attached claims.

[0040] To facilitate understanding of the embodiments of the present invention, further explanation will be given below by taking specific embodiments as examples in conjunction with the accompanying drawings, and each of the accompanying drawings does not constitute a limitation on the embodiments of the present invention.

[0041] like Figures 1 to 6 As shown, the power transformer winding turn-to-turn discharge warning system includes:

[0042] A signal acquisition module, which is used to collect multi-physical discharge signals generated between turns of power transformer windings in real time;

[0043] A data cleaning module, which is used to perform preprocessing operations on the collected multi-physical discharge signals;

[0044] A spectrum construction module, which is used to extract characteristic parameters of the multi-physical discharge signal after cleaning and construct a PRPD distribution spectrum of the corresponding multi-physical discharge signal;

[0045] A discharge severity identification module, which is used to generate an identification result of whether discharge occurs between winding turns according to the characteristic parameters, and analyze the PRPD spectrum to identify the severity of the discharge between winding turns;

[0046] The graded alarm module is used to automatically trigger the corresponding alarm mechanism according to the inter-turn discharge identification results, and trigger different levels of alarm responses according to the severity of the discharge.

[0047] In a preferred embodiment of the power transformer winding inter-turn discharge warning system, the signal acquisition module includes a sensor array unit and a data acquisition unit, the sensor array unit includes a high-frequency current sensor subunit for real-time acquisition of current signals generated between turns of the power transformer winding, a voltage sensor subunit for real-time acquisition of voltage signals generated between turns of the power transformer winding, an ultrasonic sensor subunit for real-time acquisition of ultrasonic signals generated between turns of the power transformer winding, and an ultra-high frequency sensor subunit for real-time acquisition of ultra-high frequency signals generated between turns of the power transformer winding, and the data acquisition unit includes a high-precision data acquisition card subunit for ensuring synchronous acquisition and calibration of multiple physical signals.

[0048] In a preferred embodiment of the power transformer winding interturn discharge warning system, the data cleaning module includes an abnormal value detection unit and a signal denoising unit, the abnormal value detection unit includes an abnormal value identification subunit for performing abnormal value identification on the multi-physical discharge signal based on a support vector machine algorithm and an abnormal value removal subunit for removing the identified abnormal values, and the signal denoising unit includes a Kalman filter subunit for performing noise removal on the signal after the abnormal values ​​have been removed. The abnormal value detection unit structure includes a sensor array composed of multiple current, voltage, ultrasonic signal, and ultra-high frequency signal sensors, which is used to perform multi-signal joint abnormal value identification on the current, voltage, ultrasonic signal, and ultra-high frequency signal based on a support vector machine algorithm, and its abnormal value judgment criterion is that at least one signal exceeds the empirical safety threshold.

[0049] In a preferred embodiment of the power transformer winding interturn discharge warning system, the spectrum construction module includes an extractor unit for performing time domain waveform and spectrum analysis on multi-physical discharge to extract time-frequency domain characteristic parameter features of multi-physical discharge signals and a spectrum constructor unit for generating a PRPD distribution spectrum of multi-physical discharge signals.

[0050] In the described early warning system for interturn discharge of power transformer windings, the discharge severity identification module includes a model training unit and a real-time identification unit for storing a learning model that has been trained in advance. The model training unit uses typical abnormal values ​​of current, voltage, ultrasonic signal, and ultra-high frequency signal in historical data sets to complete the training of the long short-term memory neural network model in advance, and trains the long short-term memory neural network model through at least 20 rounds, 1000 iterations per round, to ensure that the accuracy of the model in identifying whether interturn discharge occurs and the severity is above 95% when working. The real-time identification unit includes an interturn discharge real-time identification subunit for inputting the characteristic parameters into the model training unit to realize the identification of the occurrence of interturn discharge and a discharge severity real-time identification subunit for inputting the PRPD spectrum into the model training unit to realize the identification of the severity of interturn discharge.

[0051] In a preferred embodiment of the power transformer winding interturn discharge early warning system, the historical data set includes normal operation data and known fault data, and the discharge severity is divided into two levels: mild and severe.

[0052] In a preferred embodiment of the power transformer winding interturn discharge warning system, the hierarchical alarm module includes a hierarchical warning communication unit that sets a three-level warning mechanism and automatically triggers it based on the identification results, and a warning tracking and recording unit for tracking the progress of alarm event processing and recording alarm logs.

[0053] A method for early warning of interturn discharge of power transformer windings comprises the following steps:

[0054] Real-time collection of multi-physics discharge signals generated between turns of power transformer windings;

[0055] Performing preprocessing operations on the multi-physical discharge signal to clean and remove noise;

[0056] Extract characteristic parameters of the multi-physical discharge signal after cleaning and construct the PRPD distribution map of the corresponding signal;

[0057] Identify the generation of winding inter-turn discharge according to the characteristic parameters, and analyze the generated multi-physical discharge signal PRPD spectrum to identify the severity of the winding inter-turn discharge;

[0058] According to the inter-turn discharge identification results, the corresponding alarm mechanism is automatically triggered, and different levels of alarm responses are triggered according to the severity of the discharge.

[0059] A computer storage medium includes computer instructions, which, when executed on a computer, cause the computer to execute the method described above.

[0060] An electronic device, comprising:

[0061] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein:

[0062] When the processor executes the program, the method described is implemented.

[0063] In one embodiment, a system for intelligently identifying and grading the severity of inter-turn discharge of power transformer windings includes the following modules:

[0064] The signal acquisition module A01 is used to collect multi-physical discharge signals generated between turns of the power transformer winding in real time;

[0065] The data cleaning module A02 is used to perform preprocessing operations on the collected original multi-physics discharge signals;

[0066] A spectrum construction module A03 is used to extract characteristic parameters of the multi-physical discharge signal after cleaning and construct a PRPD distribution spectrum of the corresponding signal;

[0067] The discharge severity identification module A04 is used to identify the generation of winding inter-turn discharge based on the extracted characteristic parameters, and analyze the generated multi-physical discharge signal PRPD spectrum to identify the severity of the winding inter-turn discharge;

[0068] The graded alarm module A05 is used to automatically trigger the corresponding alarm mechanism according to the inter-turn discharge identification result, and trigger different levels of alarm response according to the severity of the discharge.

[0069] In one embodiment, the signal acquisition module A01 includes a sensor array unit A011 and a data acquisition unit A012. The sensor array unit A011 includes four subunits, namely a high-frequency current sensor subunit A0111, a voltage sensor subunit A0112, an ultrasonic sensor subunit A0113 and an ultra-high frequency sensor subunit A0114, which are used to collect multi-physical discharge signals generated between turns of power transformer windings in real time. The high-frequency current sensor subunit A0111 is used to collect current signals generated between turns of power transformer windings in real time. The voltage sensor subunit A0112 is used to collect voltage signals generated between turns of power transformer windings in real time. The ultrasonic sensor subunit A0113 is used to collect ultrasonic signals generated between turns of power transformer windings in real time. The ultra-high frequency sensor subunit A0114 is used to collect ultra-high frequency signals generated between turns of power transformer windings in real time. The data acquisition unit A012 includes a high-precision data acquisition card subunit A0121. The high-precision data acquisition card subunit A0121 is used to ensure the synchronous acquisition and calibration of multi-physical signals. The data cleaning module A02 includes an outlier detection unit A021 and a signal denoising unit A022. The outlier detection unit A021 includes two subunits, namely an outlier identification subunit A0211 and an outlier removal subunit A0212, which are used to pre-process the collected original multi-physics discharge signal. The outlier identification subunit A0211 is used to perform outlier identification on the collected original multi-physics discharge signal based on the support vector machine algorithm. The support vector machine algorithm identifies and removes abnormal data by constructing a hyperplane that maximizes the interval between normal data points. The basic solution formula of the hyperplane is: In the formula, is the weight vector, is the bias term, is a point in the feature space. The outlier removal subunit A0212 is used to remove the identified outliers to ensure the authenticity of the signal to avoid misjudgment. The signal denoising unit A022 includes a Kalman filter subunit A0221. The Kalman filter subunit A0221 is used to remove background noise, electromagnetic interference and other noise from the signal after the outliers have been removed. The Kalman filter algorithm includes two stages: prediction and update. In the prediction stage, the solution formula for state prediction is: , where is the state prediction at time k based on the information at the previous time, is the state transition matrix, is the state estimate at the previous moment, is the control input matrix, is the control input at time k. In the prediction stage, the solution formula for the prediction covariance is: , where is the covariance matrix of the state prediction at time k, is the covariance matrix of the state estimate at the previous moment, is the covariance matrix of the process noise.

[0070] In the update phase, the solution formula of Kalman gain is: , where is the Kalman gain matrix at time k, is the observation matrix, is the covariance matrix of the observation noise.

[0071] In the update phase, the solution formula for state update is: , where is the state estimate at time k, is the actual observed value at time k. In the update phase, the solution formula for updating the covariance is: , where is the covariance matrix of the state estimate at time k, is the identity matrix.

[0072] In one embodiment, the spectrum construction module A03 includes a feature extractor unit A031 and a spectrum constructor unit A032. The feature extractor unit A031 is used to perform time domain waveform and spectrum analysis on multi-physical discharges to extract time and frequency domain feature parameters of multi-physical discharge signals. The time and frequency domain feature parameters are divided into time domain feature parameters and frequency domain feature parameters. The time domain feature parameters include peak value, mean value, variance, entropy, skewness, and kurtosis. The frequency domain feature parameters include fundamental frequency, harmonics, bandwidth, frequency center, and peak factor. The spectrum constructor unit A032 is used to generate a PRPD distribution spectrum of multi-physical discharge signals. The discharge severity identification module A04 includes a model training unit A041 and a real-time identification unit A042. The model training unit A041 is used to store a learning model that has been trained in advance. Specifically, the model training unit A041 uses historical data sets, including normal operation data and known fault data, to complete the training of the long short-term memory neural network model in advance to ensure that the model can accurately and reliably identify whether inter-turn discharge occurs and its severity when working. The real-time identification unit A042 includes two subunits, namely, the inter-turn discharge real-time identification subunit A0421 and the discharge severity real-time identification subunit A0422. The inter-turn discharge real-time identification subunit A0421 is used to input the extracted characteristic parameters into the model training unit A041 to realize the identification of the generation of inter-turn discharge. The discharge severity real-time identification subunit A0422 is used to input the generated PRPD spectrum into the model training unit A041 to realize the identification of the severity of inter-turn discharge, and the discharge severity can be divided into two levels: mild and severe. The graded alarm module A05 includes a graded warning communication unit A051 and a warning tracking and recording unit A052. The hierarchical warning communication unit A051 sets a three-level warning mechanism and automatically triggers it according to the recognition result of the real-time identification unit. If no inter-turn discharge occurs, a green warning table is sent to inform the operation and maintenance personnel that the equipment is safe. If a slight inter-turn discharge occurs, a yellow warning table is sent to inform the operation and maintenance personnel to pay close attention to the operation of the equipment. If a serious inter-turn discharge occurs, a red warning table is sent to inform the operation and maintenance personnel to carry out maintenance immediately.

[0073] Specifically, the alarm mechanism is shown in the following table.

[0074]

[0075] The warning tracking and recording unit A052 is used to track the progress of alarm event processing, ensure that the problem is solved in time, and record the alarm log.

[0076] Although the embodiments of the present invention are described above in conjunction with the accompanying drawings, the present invention is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and instructive, rather than restrictive. A person of ordinary skill in the art can also make many forms under the guidance of this specification and without departing from the scope of protection of the claims of the present invention, all of which belong to the protection of the present invention.

Claims

1. A power transformer winding interturn discharge warning system, characterized in that: These include, A signal acquisition module, which is used to collect multi-physical discharge signals generated between turns of power transformer windings in real time; A data cleaning module, which is used to perform preprocessing operations on the collected multi-physical discharge signals; A spectrum construction module, which is used to extract characteristic parameters of the multi-physical discharge signal after cleaning and construct a PRPD distribution spectrum of the corresponding multi-physical discharge signal; A discharge severity identification module, which is used to generate an identification result of whether discharge occurs between winding turns according to the characteristic parameters, and analyze the PRPD spectrum to identify the severity of the discharge between winding turns; The graded alarm module is used to automatically trigger the corresponding alarm mechanism according to the inter-turn discharge identification results, and trigger different levels of alarm responses according to the severity of the discharge.

2. The power transformer winding interturn discharge warning system according to claim 1, characterized in that: Preferably, the signal acquisition module includes a sensor array unit and a data acquisition unit, the sensor array unit includes a high-frequency current sensor subunit for real-time acquisition of current signals generated between turns of a power transformer winding, a voltage sensor subunit for real-time acquisition of voltage signals generated between turns of a power transformer winding, an ultrasonic sensor subunit for real-time acquisition of ultrasonic signals generated between turns of a power transformer winding, and an ultra-high frequency sensor subunit for real-time acquisition of ultra-high frequency signals generated between turns of a power transformer winding, and the data acquisition unit includes a high-precision data acquisition card subunit for ensuring synchronous acquisition and calibration of multiple physical signals.

3. The power transformer winding interturn discharge warning system according to claim 1, characterized in that: The hierarchical alarm module includes a hierarchical alarm communication unit that sets a three-level alarm mechanism and automatically triggers it according to the identification result, and an alarm tracking and recording unit that is used to track the progress of alarm event processing and record alarm logs.

4. A method for early warning of interturn discharge of power transformer windings, characterized in that: It includes the following steps, Real-time collection of multi-physics discharge signals generated between turns of power transformer windings; Performing preprocessing operations on the multi-physical discharge signal to clean and remove noise; Extract characteristic parameters of the multi-physical discharge signal after cleaning and construct the PRPD distribution map of the corresponding signal; Identify the generation of winding inter-turn discharge according to the characteristic parameters, and analyze the generated multi-physical discharge signal PRPD spectrum to identify the severity of the winding inter-turn discharge; According to the inter-turn discharge identification results, the corresponding alarm mechanism is automatically triggered, and different levels of alarm responses are triggered according to the severity of the discharge.

5. A computer storage medium, characterized in that: The storage medium includes computer instructions, which, when executed on a computer, cause the computer to perform the method as claimed in claim 4 .

6. An electronic device, characterized in that: The electronic device comprises: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to claim 4 is implemented.

Images