A method and device for fault diagnosis and monitoring of power transformer based on Internet of Things

Abstract

The invention discloses a method and device for fault diagnosis and monitoring of power transformers based on the Internet of Things, and relates to the technical field of transformer state monitoring and fault diagnosis. The method for fault diagnosis and monitoring of power transformers based on the Internet of Things includes obtaining real-time diagnostic data of power transformers; sending the real-time diagnostic data of power transformers to a central database based on Internet of Things communication, parsing the real-time diagnostic data of power transformers and storing them in a set category library by classification; performing fault diagnosis on power transformers based on the real-time diagnostic data of power transformers after parsing and classification storage, and obtaining a bushing discharge fault diagnosis index and a core vibration fault index; judging whether the power transformer has a bushing discharge fault or a core vibration fault by the bushing discharge fault diagnosis index and the core vibration fault index, and solving the problem that the prior art is inconvenient to diagnose and monitor the bushing discharge fault and the core vibration fault of the transformer at the same time, and the detection is not comprehensive enough.

Description

Technical Field

[0001] The present invention relates to the technical field of transformer status monitoring and fault diagnosis, and in particular to a method and device for fault diagnosis and monitoring of a power transformer based on the Internet of Things. Background Art

[0002] Transformers are key equipment in power distribution systems, and their operating status is directly related to the safety and stability of the power system. Transformer vibration is one of the main causes of failures. Theoretical analysis shows that when power transformers are in stable operation, the magnetostriction of silicon steel sheets will cause core vibration, and the windings will also vibrate under the electric field force of the load current. During the long-term operation of the transformer, these vibrations will affect the mechanical stability of the transformer itself, leading to transformer failure or shortening the service life of the transformer.

[0003] The Chinese patent application with publication number CN116243214A discloses an online monitoring system and method for vibration data of power transformers, including: a multi-dimensional sensor module assembled with the transformer, used to collect the operation data of the transformer in real time; an Internet of Things computing device connected to the multi-dimensional sensor module, used to pre-process the operation data to obtain several vibration data groups corresponding to different parts of the transformer according to the distribution position of the multi-dimensional sensor module; and a cloud service platform, the cloud service platform includes: a data processing center, used to analyze and process the vibration data group to obtain fault information corresponding to different operating parameters and generate diagnostic reports; a dispatching operation center, used to locate the corresponding transformer according to the fault information. The Internet of Things computing device cooperates with the data processing center to calculate and analyze the various types of collected operating parameters, so that the vibration data and fault analysis are more accurate, and real-time signal acquisition and fault judgment can be achieved with higher timeliness.

[0004] The Chinese patent application with publication number CN115079051A discloses a method and system for detecting bushing discharge faults of transformers, including: S1, determining the number of bushing discharge events that occur in the bushing of the transformer within a specified detection period; S2, judging whether the ratio of the number of bushing discharge events to the total number of detections within the detection period exceeds a set value, and if it exceeds the set value, it is judged that the bushing of the transformer has a discharge fault. The present invention has the advantages of no need to shut down the transformer, no need to contact live equipment, high detection safety, and convenient testing. Moreover, the method for judging the transformer bushing discharge fault of the present invention can be conveniently combined with the Internet of Things and information technology, and can realize real-time monitoring and early warning of the transformer bushing discharge state, which is conducive to improving the intelligent operation and inspection level and safe operation level of the transformer.

[0005] However, the prior art is inconvenient to diagnose and monitor the bushing discharge fault and the core vibration fault of the transformer at the same time, and the detection is not comprehensive enough. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention provides a method and device for diagnosing and monitoring power transformer faults based on the Internet of Things, which solves the problem that the prior art is inconvenient to diagnose and monitor the bushing discharge fault and the core vibration fault of the transformer at the same time, and the detection is not comprehensive enough.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a power transformer fault diagnosis and monitoring method based on the Internet of Things, comprising the following steps: real-time monitoring of the power transformer and obtaining real-time diagnostic data of the power transformer; sending the real-time diagnostic data of the power transformer to a central database based on the Internet of Things communication, and parsing the real-time diagnostic data of the power transformer received by the central database and storing them in a set category library according to the classification; diagnosing the power transformer fault based on the parsed and classified real-time diagnostic data of the power transformer, and obtaining a bushing discharge fault diagnosis index and a core vibration fault index; judging whether the power transformer has a bushing discharge fault or a core vibration fault through the bushing discharge fault diagnosis index and the core vibration fault index.

[0008] Furthermore, the real-time diagnostic data of the power transformer includes bushing discharge characteristic data and core vibration characteristic data. The process of sending the real-time diagnostic data of the power transformer to the central database based on the Internet of Things communication is as follows: preliminarily analyze the bushing discharge characteristic data to obtain bushing discharge ultrasonic data, bushing discharge partial discharge data and bushing discharge surrounding gas data; preliminarily analyze the core vibration characteristic data to obtain core vibration data, core acoustic wave data and magnetic field data around the core; encrypt and compress the preliminarily analyzed bushing discharge characteristic data and core vibration characteristic data; mark the encrypted and compressed preliminarily analyzed bushing discharge characteristic data and core vibration characteristic data with a timestamp and a device ID to obtain a marked data packet; and send the marked data packet to the central database based on the Internet of Things communication.

[0009] Furthermore, the process of parsing and classifying the real-time diagnostic data of the power transformer received by the central database is as follows: checking the consistency between the timestamp of the received marked data packet and the device ID, if they are inconsistent, it is determined that the data transmission is wrong; if the timestamp is consistent with the device ID, the marked data packet is decrypted and decompressed, and the bushing discharge ultrasonic data, bushing discharge partial discharge data, bushing discharge surrounding gas data, core vibration data, core acoustic wave data and core surrounding magnetic field data are classified and stored in the set category library.

[0010] Furthermore, the process of obtaining the casing discharge fault diagnosis index is as follows: extracting casing discharge ultrasonic data, casing discharge partial discharge data and casing discharge surrounding gas data from the category library in the central database; performing fusion analysis on the casing discharge ultrasonic data, casing discharge partial discharge data and casing discharge surrounding gas data, and obtaining the casing discharge fault diagnosis index based on the established casing discharge fusion analysis model.

[0011] Furthermore, the process of obtaining the casing discharge fault diagnosis index based on the established fusion analysis model is as follows: analyzing the casing discharge ultrasonic data to obtain the casing discharge ultrasonic intensity; analyzing the casing discharge local discharge data to obtain the energy of the maximum frequency in the frequency spectrum characteristics of the casing local discharge; analyzing the casing discharge surrounding gas data to obtain the casing discharge surrounding gas concentration, wherein the casing discharge surrounding gas concentration includes hydrogen concentration and ozone concentration; based on the established casing discharge fusion analysis model, the casing discharge ultrasonic intensity, the energy of the maximum frequency in the frequency spectrum characteristics of the casing local discharge and the casing discharge surrounding gas concentration are processed to obtain the casing discharge fault diagnosis index.

[0012] Furthermore, the process of obtaining the core vibration fault index is as follows: extracting the core vibration data, the core acoustic wave data and the magnetic field data around the core from the category library in the central database; performing fusion analysis on the core vibration data, the core acoustic wave data and the magnetic field data around the core, and obtaining the core vibration fault index based on the established core vibration fusion analysis model.

[0013] Furthermore, the process of obtaining the core vibration fault index based on the established core vibration fusion analysis model is as follows: analyze the core vibration data to obtain the core vibration characteristic data, and the core vibration characteristic data includes the core vibration spectrum offset and the core vibration maximum vibration amplitude; analyze the core acoustic wave data to obtain the core vibration acoustic emission energy; analyze the magnetic field data around the core to obtain the magnetic field variation index around the core; based on the established core vibration fusion analysis model, process the core vibration characteristic data, the core vibration acoustic emission energy and the magnetic field variation index around the core to obtain the core vibration fault index.

[0014] Furthermore, the process of determining whether a power transformer has a bushing discharge fault or a core vibration fault is as follows: determining whether a bushing discharge fault diagnosis index is greater than or equal to a set discharge fault threshold; if so, the power transformer has a bushing discharge fault; determining whether a core vibration fault index is greater than or equal to a set vibration fault threshold; if so, the power transformer has a core vibration fault; if the bushing discharge fault diagnosis index is less than the set discharge fault threshold and the core vibration fault index is less than the set vibration fault threshold, then evaluating the stability of the current power transformer based on the bushing discharge fault diagnosis index and the core vibration fault index.

[0015] Further, the process of evaluating the stability of the current power transformer based on the bushing discharge fault diagnosis index and the core vibration fault index is as follows: the stability index of the power transformer is obtained based on the bushing discharge fault diagnosis index and the core vibration fault index, and the calculation formula of the stability index is as follows:

[0016]

[0017] In the formula, Wd is the stability index, e is the natural constant, Fz is the bushing discharge fault diagnosis index, α1 is the bushing discharge weight factor, Dz is the core vibration fault index, and α2 is the core vibration weight factor; it is judged whether the stability index is greater than the set stability threshold, and if so, the stability of the power transformer is warned.

[0018] A power transformer fault diagnosis and monitoring device based on the Internet of Things comprises a processor and a memory, wherein computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, the processor executes the power transformer fault diagnosis and monitoring method based on the Internet of Things as described above, and further comprises a diagnostic data acquisition module, a classification storage module, a diagnostic module and an evaluation module, wherein: the diagnostic data acquisition module is used to monitor the power transformer in real time and obtain the real-time diagnostic data of the power transformer; the classification storage module is used to send the real-time diagnostic data of the power transformer to a central database based on the Internet of Things communication, and parse the real-time diagnostic data of the power transformer received by the central database and classify and store it in a set category library; the diagnostic module is used to diagnose the fault of the power transformer based on the parsed and classified stored real-time diagnostic data of the power transformer, and obtain the bushing discharge fault diagnosis index and the core vibration fault index; the evaluation module is used to judge whether the power transformer has a bushing discharge fault or a core vibration fault through the bushing discharge fault diagnosis index and the core vibration fault index.

[0019] The present invention has the following beneficial effects:

[0020] (1) This power transformer fault diagnosis and monitoring method based on the Internet of Things uses the Internet of Things technology to collect and transmit key operating data of the transformer in real time, such as bushing discharge and core vibration, to ensure the immediacy and continuity of information flow. It also uses a fusion analysis model to integrate multiple monitoring data, such as ultrasonic intensity, discharge spectrum characteristics, and surrounding gas concentration, to provide more comprehensive and accurate fault diagnosis.

[0021] (2) This IoT-based power transformer fault diagnosis and monitoring method can provide early warning before a problem becomes a serious fault by comparing the fault index with a preset threshold, thereby improving the safety and reliability of the transformer and the entire power grid, diagnosing and handling potential faults in a timely manner, and significantly reducing unexpected downtime and related losses caused by faults.

[0022] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The present invention is a flow chart of the power transformer fault diagnosis and monitoring method based on the Internet of Things.

[0024] Figure 2 This is an exponential function image of bushing discharge fault diagnosis in the power transformer fault diagnosis and monitoring method based on the Internet of Things of the present invention.

[0025] Figure 3 This is an image of the core vibration fault index function of the power transformer fault diagnosis and monitoring method based on the Internet of Things of the present invention.

[0026] Figure 4 This is a structural connection diagram of the power transformer fault diagnosis and monitoring device based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0027] The embodiment of the present application realizes the construction of a comprehensive fault warning system by integrating real-time monitoring, data transmission, classified storage, fault analysis and diagnosis, etc. through a power transformer fault diagnosis and monitoring method and device based on the Internet of Things.

[0028] The problems in the embodiments of this application are generally solved as follows:

[0029] The Internet of Things technology is used to achieve real-time monitoring and data analysis of power transformers, thereby achieving fast and accurate fault diagnosis. The key data on bushing discharge and core vibration are collected through various sensors installed on the power transformer and sent to the central database in real time. In the database, these data are parsed and classified for easy access and processing. Using pre-defined algorithms, the discharge fault index and vibration fault index are calculated based on the classified data, and finally these indices are used to determine whether the transformer has a specific fault.

[0030] See also Figure 1The embodiment of the present invention provides a technical solution: a method for fault diagnosis and monitoring of power transformers based on the Internet of Things, comprising the following steps: real-time monitoring of power transformers and obtaining real-time diagnostic data of power transformers; sending the real-time diagnostic data of power transformers to a central database based on Internet of Things communication, parsing the real-time diagnostic data of power transformers received by the central database and storing them in a set category library according to classification; diagnosing faults of power transformers based on the parsed and classified real-time diagnostic data of power transformers, and obtaining a bushing discharge fault diagnosis index and a core vibration fault index; judging whether the power transformer has a bushing discharge fault or a core vibration fault through the bushing discharge fault diagnosis index and the core vibration fault index.

[0031] Through real-time monitoring and instant data processing, faults can be quickly discovered and diagnosed, greatly reducing fault response time. Real-time data analysis and accurate fault index calculation improve the accuracy of fault judgment. The Internet of Things technology is used to safely and efficiently transmit data to the central database to ensure data integrity and timeliness. The data is parsed and classified and stored in the database according to parameter type and source, simplifying data retrieval and processing processes.

[0032] Specifically, the real-time diagnostic data of the power transformer includes bushing discharge characteristic data and core vibration characteristic data. The process of sending the real-time diagnostic data of the power transformer to the central database based on the Internet of Things communication is as follows: perform a preliminary analysis on the bushing discharge characteristic data to obtain bushing discharge ultrasonic data, bushing discharge partial discharge data and bushing discharge surrounding gas data; perform a preliminary analysis on the core vibration characteristic data to obtain core vibration data, core acoustic wave data and magnetic field data around the core; encrypt and compress the bushing discharge characteristic data and core vibration characteristic data after preliminary analysis; mark the encrypted and compressed bushing discharge characteristic data and core vibration characteristic data after preliminary analysis with a timestamp and a device ID to obtain a marked data packet; and send the marked data packet to the central database based on the Internet of Things communication.

[0033] In this implementation scheme, a preliminary analysis is performed on the bushing discharge characteristic data and the core vibration characteristic data. The bushing discharge characteristic data includes ultrasonic data (directly indicating the intensity of the discharge activity), partial discharge data (providing a detailed characteristic analysis of the discharge), and surrounding gas data (reflecting the impact of the discharge on the environment, especially the thermal decomposition products of the insulating material). Core vibration characteristic data: including core vibration data (monitoring the health of the structure), acoustic wave data (capturing the sound caused by tiny cracks or damage), and surrounding magnetic field data (changes in the magnetic field can indicate deformation or displacement of the core structure). Preliminary analysis helps to extract the most critical information from the raw data, reducing the amount of data transmitted while retaining the important features required for decision-making.

[0034] Encryption ensures the security of data during transmission, preventing data leakage or malicious tampering, and compression reduces the size of data packets, optimizes storage requirements, and speeds up data transmission. Using timestamps and device IDs to mark data ensures data traceability and clarity of source, and using IoT technology to achieve real-time data transmission ensures the timeliness and real-time nature of data, allowing remote monitoring and management, and improving system flexibility and responsiveness.

[0035] Specifically, the process of parsing and classifying the real-time diagnostic data of power transformers received by the central database is as follows: check the consistency of the timestamp of the received marked data packet with the device ID. If they are inconsistent, it is considered that the data transmission is wrong; if the timestamp is consistent with the device ID, the marked data packet is decrypted and decompressed, and the bushing discharge ultrasonic data, bushing discharge partial discharge data, bushing discharge surrounding gas data, core vibration data, core acoustic wave data and core surrounding magnetic field data are classified and stored in the set category library, wherein the category library includes the bushing discharge ultrasonic database, the bushing discharge partial discharge database, the bushing discharge surrounding gas database, the core vibration database, the core acoustic wave database and the core surrounding magnetic field database, and the bushing discharge ultrasonic data, the bushing discharge partial discharge data, the bushing discharge surrounding gas data, the core vibration data, the core acoustic wave data and the core surrounding magnetic field data are stored in the corresponding category library respectively. Organized data storage facilitates efficient data analysis, and the separately stored data is easy to process using specialized analysis tools and algorithms, so as to identify trends and anomalies more quickly.

[0036] In this implementation, before any further processing of the data is performed, the consistency of the timestamp and device ID of the data packet is first verified to ensure that the source of the data is correct and that no errors or tampering occurred during the data transmission process. After receiving the encrypted and compressed data, decryption and decompression operations are performed in order to restore the original state of the data packet so that the data can be correctly read and processed.

[0037] The data contains a variety of monitoring indicators, such as casing discharge ultrasonic data, partial discharge data, gas data, core vibration data, sound wave data and magnetic field data. These data are stored in preset category libraries according to their types. Such classification not only helps to organize and manage data, but also facilitates query and analysis of specific types of data. The structured storage of the database can also optimize the data access path, reduce query time and improve data processing performance.

[0038] Specifically, the process of obtaining the casing discharge fault diagnosis index is as follows: extract the casing discharge ultrasonic data, casing discharge partial discharge data and casing discharge surrounding gas data from the category library in the central database; perform fusion analysis on the casing discharge ultrasonic data, casing discharge partial discharge data and casing discharge surrounding gas data, and obtain the casing discharge fault diagnosis index based on the established casing discharge fusion analysis model.

[0039] The process of obtaining the casing discharge fault diagnosis index based on the established fusion analysis model is as follows: analyzing the casing discharge ultrasonic data to obtain the casing discharge ultrasonic intensity; analyzing the casing discharge local discharge data to obtain the energy of the maximum frequency in the frequency spectrum characteristics of the casing local discharge; analyzing the casing discharge surrounding gas data to obtain the casing discharge surrounding gas concentration, wherein the casing discharge surrounding gas concentration includes hydrogen concentration and ozone concentration; based on the established casing discharge fusion analysis model, the casing discharge ultrasonic intensity, the energy of the maximum frequency in the frequency spectrum characteristics of the casing local discharge and the casing discharge surrounding gas concentration are processed to obtain the casing discharge fault diagnosis index.

[0040] The necessary data is extracted from the corresponding category library of the central database: casing discharge ultrasonic data, partial discharge data and surrounding gas data. This purposeful extraction ensures that the analyzed data is highly relevant and has a specific purpose. The ultrasonic data, partial discharge spectrum characteristics and gas concentration data are fused and processed using a comprehensive analysis model. This fusion analysis takes into account multiple aspects of the discharge phenomenon and improves the accuracy of diagnosis. The established casing discharge fusion analysis model is applied. The casing discharge fusion analysis model includes algorithms such as neural networks, decision trees or other machine learning techniques to process the extracted data and calculate the casing discharge fault diagnosis index.

[0041] In addition to the above, the casing discharge fusion analysis model can also obtain the casing discharge fault diagnosis index through the following calculation formula, and the data is processed scalarly before calculation:

[0042]

[0043] Where, Fz is the bushing discharge fault diagnosis index, U is the bushing discharge ultrasonic intensity, and an ultrasonic sensor is used to monitor the acoustic wave activity near the transformer bushing. The discharge event will cause the generation of ultrasonic waves. The ultrasonic intensity can be used as a direct indicator of the discharge activity, because electrical discharge will cause acoustic vibrations in the gas. S(f) is the energy of the maximum frequency in the spectrum characteristics of the bushing local discharge. The discharge signal is recorded by a local discharge monitoring device (such as a PD detector) and a spectrum analysis is performed. Discharges of different types and severity will produce different spectrum characteristics. Spectral analysis can help identify and classify discharge types. H2 is the hydrogen concentration, and O3 is the ozone concentration. A gas analyzer is used to monitor the gas composition in the transformer or around the bushing, especially hydrogen and ozone. These gases are products of insulation material degradation caused by discharge. Gas analysis can provide indirect information about the discharge intensity and degree of damage, because the more severe the discharge, the more decomposition gas is usually produced.

[0044] The ultrasonic intensity directly responds to the acoustic activity, while the spectrum characteristics provide information about the type of discharge, and the gas concentration indirectly reflects the impact of the discharge on the material. Together, these parameters form a framework for comprehensive monitoring of discharge activity. A single parameter may be difficult to accurately determine the fault state, but when they are combined, they can provide more accurate fault diagnosis. For example, even if the ultrasonic intensity is low, if the hydrogen and ozone concentrations are high, it may indicate the presence of chronic or incipient discharge activity.

[0045] The spectral characteristics provide precise information about the discharge type, the gas concentration reflects the chemical impact of the discharge, and the ultrasonic intensity shows the physical characteristics of the discharge. By integrating information from multiple data sources through a fusion analysis model, the health status of the casing can be fully evaluated. Different types of data complement each other, providing a more comprehensive perspective and reducing the possibility of misdiagnosis.

[0046] The following table shows the data of the ultrasonic intensity of bushing discharge, the energy of the maximum frequency in the spectrum characteristics of bushing partial discharge, and the gas concentration around bushing discharge:

[0047] Table 1 Parameter data table of casing discharge fault diagnosis index

[0048]

[0049]

[0050] It can be seen from the table that when the ultrasonic intensity U and the spectrum characteristic S(f) of partial discharge are zero, the bushing discharge fault diagnosis index Fz is zero even if the concentrations of hydrogen and ozone are high. This shows that the fault index is dependent on the concentrations of hydrogen and ozone.

[0051] Figure 2The image of the casing discharge fault diagnosis index shows the effect of ultrasonic intensity U on the casing discharge fault diagnosis index Fz, where the color depth represents the hydrogen concentration. It can be seen from the figure that with the increase of U, Fz increases significantly, especially under high hydrogen concentration conditions. The change in color indicates that the hydrogen concentration is from low to high. Overall, the casing discharge fault diagnosis index Fz shows a significant positive correlation with the increase of ultrasonic intensity and hydrogen concentration, reflecting the importance of these parameters for identifying casing discharge faults.

[0052] Specifically, the process of obtaining the core vibration fault index is as follows: extract the core vibration data, core acoustic wave data and magnetic field data around the core from the category library in the central database; perform fusion analysis on the core vibration data, core acoustic wave data and magnetic field data around the core, and obtain the core vibration fault index based on the established core vibration fusion analysis model.

[0053] The process of obtaining the core vibration fault index based on the established core vibration fusion analysis model is as follows: analyze the core vibration data to obtain the core vibration characteristic data, and the core vibration characteristic data includes the core vibration spectrum offset and the core vibration maximum vibration amplitude; analyze the core acoustic wave data to obtain the core vibration acoustic emission energy; analyze the magnetic field data around the core to obtain the magnetic field variation index around the core; based on the established core vibration fusion analysis model, process the core vibration characteristic data, the core vibration acoustic emission energy and the magnetic field variation index around the core to obtain the core vibration fault index.

[0054] The core vibration data, core acoustic wave data, and magnetic field data around the core are extracted from a specific category library in the central database, which ensures that the data used in the analysis process is specifically targeted at diagnosing core vibration-related issues. The core vibration data is analyzed to obtain the core vibration spectrum offset and maximum vibration amplitude, which directly reflect the physical state of the core. The core acoustic wave data is analyzed to obtain the acoustic emission energy, which can indicate microcracks or damage inside the core. The magnetic field data around the core is analyzed to obtain the magnetic field variation index, which reflects the magnetic field changes of the core or its nearby structures.

[0055] Use the predefined core vibration fusion analysis model to integrate various types of data. The core vibration fusion analysis model can be based on machine learning or other statistical methods to evaluate the overall health of the core. By integrating multi-source data, a comprehensive core vibration fault index is provided. This index is more accurate and reliable than the analysis of a single data source.

[0056] In addition to the above, the core vibration fusion analysis model can also obtain the core vibration fault index through the following calculation formula, and the data is processed scalarly before calculation:

[0057]

[0058] Wherein, Dz is the core vibration fault index, Δf is the core vibration spectrum offset, the vibration signal of the core is captured by the vibration sensor, and the change of the main vibration frequency is detected by spectrum analysis. The spectrum offset can indicate the change of the core structure, such as looseness or deformation, A is the maximum vibration amplitude of the core vibration, the vibration sensor measures the maximum vibration amplitude of the core, the vibration amplitude is an intuitive parameter for evaluating the health status of the machine, excessive vibration may indicate that the core is poorly fixed or there are other structural problems, E is the core vibration acoustic emission energy, the acoustic emission sensor is used to monitor the sound wave energy emitted by the core, acoustic emission monitoring can capture the sound wave activity caused by tiny cracks or damage to the core structure, and is a sensitive early fault detection technology, M is the magnetic field variation index around the core, the magnetic field sensor is used to monitor the magnetic field strength and its changes around the core, the vibration or structural problems of the core may cause local unevenness or changes in the magnetic flux, and the magnetic field variation index can provide quantitative information about these changes.

[0059] While vibration amplitude and spectral shift provide direct information about the mechanical condition, acoustic emission and magnetic field variability indices provide insights into structural integrity and electromagnetic properties, and combined these parameters can provide a more comprehensive assessment of the core's health. For example, even if the vibration amplitude is not large, high acoustic emission energy and magnetic field variability may indicate the presence of subtle but severe internal damage.

[0060] By integrating multiple monitoring data (physical vibration, acoustic signals, magnetic field changes), fusion analysis provides a comprehensive view of the core state, which can more accurately diagnose potential problems than analyzing any data type alone. Fusion analysis can eliminate the risk of misjudgment that may come from relying on a single data source. For example, a single vibration data may have difficulty distinguishing between normal operation and vibration caused by minor damage, while the combination of acoustic wave data and magnetic field data can provide more context and enhance the confidence of diagnosis.

[0061] The following table shows the core vibration characteristic data, core vibration acoustic emission energy and magnetic field variation index around the core:

[0062] Table 2 Core vibration fault index related parameter data table

[0063] Δf / Hz A / mm E M / Gauss Dz 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.5 0.0 0.0 0.0 0.0 25.0 0.0 0.0 0.0 0.0 37.5 0.0 0.0 0.0 0.0 50.0 0.0 0.0 0.0 25.0 0.0 0.0 0.0 0.0 25.0 12.5 0.0 0.0 0.0 25.0 25.0 0.0 0.0 0.0 25.0 37.5 0.0 0.0 0.0 25.0 50.0 0.0

[0064] It can be seen from the table that when the values ​​of core vibration spectrum offset, core vibration maximum vibration amplitude and core vibration acoustic emission energy are all low, the core vibration fault index remains zero even if the magnetic field variation index around the core is high, which means that in the absence of sufficient vibration and acoustic emission activities, magnetic field changes alone cannot cause high fault indexes.

[0065] Figure 3 The image of the core vibration fault index shows the influence of the core vibration spectrum offset on the core vibration fault index in the form of a line graph, where different lines represent different magnetic field variation indices. It can be seen from the figure that with the increase of the core vibration spectrum offset, the core vibration fault index gradually increases. Each line represents a specific magnetic field variation index, where a higher magnetic field variation index value corresponds to a higher core vibration fault index, reflecting the significant influence of the magnetic field variation index on the core vibration fault index. In addition, with the increase of the core vibration spectrum offset, the lines of different magnetic field variation indices show obvious separation, indicating that the contribution of the magnetic field variation index to the fault index increases at higher spectrum offsets. This shows that when the spectrum offset is significant, magnetic field monitoring is of great significance for diagnosing core vibration faults.

[0066] Specifically, the process of judging whether the power transformer has a bushing discharge fault or a core vibration fault is as follows: judging whether the bushing discharge fault diagnosis index is greater than or equal to the set discharge fault threshold, if so, the power transformer has a bushing discharge fault; judging whether the core vibration fault index is greater than or equal to the set vibration fault threshold, if so, the power transformer has a core vibration fault; if the bushing discharge fault diagnosis index is less than the set discharge fault threshold and the core vibration fault index is less than the set vibration fault threshold, then evaluating the stability of the current power transformer based on the bushing discharge fault diagnosis index and the core vibration fault index.

[0067] Set a fault threshold for bushing discharge and core vibration respectively. These thresholds are set based on historical data, empirical statistics, equipment performance standards and safety requirements to distinguish normal operating conditions from potential fault conditions. Compare the bushing discharge diagnostic index and the core vibration fault index with their corresponding fault thresholds. If any fault index exceeds its threshold, it indicates that the relevant component may be faulty. This comparison is direct and clear, which facilitates rapid fault diagnosis.

[0068] If the bushing discharge fault diagnosis index and the core vibration fault index are both below their corresponding thresholds, the overall stability of the transformer is further evaluated, which can be based on the absolute or relative values ​​of the fault index, as well as taking into account other operating parameters and environmental factors.

[0069] By setting clear fault thresholds, the fault detection process can be simplified, allowing operators or automatic monitoring systems to quickly determine the status of the equipment. This method reduces the complex judgment process and improves diagnostic efficiency. Once the fault index exceeds the threshold, an alarm is immediately triggered or corresponding maintenance measures are taken, which helps to quickly respond to possible equipment failures and reduce the risk of equipment damage and maintenance costs.

[0070] The process of evaluating the stability of the current power transformer based on the bushing discharge fault diagnosis index and the core vibration fault index is as follows: the stability index of the power transformer is obtained based on the bushing discharge fault diagnosis index and the core vibration fault index. The calculation formula of the stability index is as follows:

[0071]

[0072] In the formula, Wd is the stability index, e is the natural constant, Fz is the bushing discharge fault diagnosis index, α1 is the bushing discharge weight factor, Dz is the core vibration fault index, and α2 is the core vibration weight factor; it is judged whether the stability index is greater than the set stability threshold, and if so, the stability of the power transformer is warned.

[0073] In this implementation, the formula uses natural logarithm and exponential function to combine the bushing discharge fault diagnosis index and the core vibration fault index to calculate the stability index. The logarithm and exponential function provide a nonlinear way to integrate and smooth the fault index, making the result more sensitive and with better discrimination. The weight factor can be adjusted according to the design and operating environment of different transformers, providing good customizability to adapt to different operating conditions and safety standards.

[0074] In addition to the above mathematical model, the stability index can also be obtained in the following ways:

[0075] 1. Use historical fault data and operational data to train machine learning models, such as random forests, support vector machines, or neural networks, to predict transformer stability.

[0076] 2. Estimate the stability index by combining various operating parameters and fault indicators through statistical analysis methods, such as principal component analysis (PCA) or linear regression.

[0077] The bushing discharge weighting factor and core vibration weighting factor are mainly based on statistical analysis, historical data evaluation, expert input, or optimized through machine learning algorithms:

[0078] 1. Using linear regression or multiple regression analysis, with the bushing discharge diagnostic index and core vibration fault index as independent variables and the result of the fault as the dependent variable, the regression model will provide the coefficients of each independent variable, which can be directly used as weight factors to indicate the relative importance of each index in predicting faults.

[0079] 2. Use tree-based algorithms (such as random forests and gradient boosting machines) for model training to provide importance scores for each feature in the model. The scores can be directly used as weight factors to indicate the relative importance of each index in the prediction model.

[0080] A power transformer fault diagnosis and monitoring device based on the Internet of Things, such as Figure 4 As shown, it includes a processor and a memory, in which computer program instructions are stored. When the computer program instructions are executed by the processor, the processor executes the power transformer fault diagnosis and monitoring method based on the Internet of Things as described above, and also includes a diagnostic data acquisition module, a classification storage module, a diagnostic module and an evaluation module, wherein: the diagnostic data acquisition module is used to monitor the power transformer in real time and obtain the real-time diagnostic data of the power transformer; the classification storage module is used to send the real-time diagnostic data of the power transformer to the central database based on the Internet of Things communication, and parse the real-time diagnostic data of the power transformer received by the central database and classify and store it in a set category library; the diagnostic module is used to diagnose the fault of the power transformer based on the real-time diagnostic data of the power transformer after parsing and classification and storage, and obtain the bushing discharge fault diagnosis index and the core vibration fault index; the evaluation module is used to judge whether the power transformer has a bushing discharge fault or a core vibration fault through the bushing discharge fault diagnosis index and the core vibration fault index.

[0081] In this implementation, the diagnostic data acquisition module monitors the key operating parameters of the power transformer in real time, such as voltage, current, temperature, sound waves and vibration, and acquires these data in real time. The classification storage module transmits the real-time monitoring data to the central database through the Internet of Things communication technology. In the central database, the data is parsed and classified according to its properties and stored in a preset category library, such as bushing discharge data, core vibration data, etc., which is helpful for data organization, retrieval and subsequent analysis.

[0082] Using the stored and classified data, combined with the predefined analysis model (such as the aforementioned algorithm), the bushing discharge fault diagnosis index and the core vibration fault index are calculated. This step is the core of the fault diagnosis process and ensures the accuracy of the fault analysis. According to the calculated fault diagnosis index, the preset fault threshold is compared to determine whether the transformer has a bushing discharge fault or a core vibration fault. This step is the final decision-making process to ensure that the fault is identified and handled in a timely manner.

[0083] A computer-readable storage medium is used to store a program, and when the program is executed by a processor, the power transformer fault diagnosis and monitoring method based on the Internet of Things as described above is implemented.

[0084] In summary, this application has at least the following effects:

[0085] Through the Internet of Things technology, key operating data of the transformer, such as bushing discharge and core vibration, is collected and transmitted in real time to ensure the immediacy and continuity of information flow. A fusion analysis model is used to integrate multiple monitoring data, such as ultrasonic intensity, discharge spectrum characteristics, and surrounding gas concentration, to provide more comprehensive and accurate fault diagnosis.

[0086] By comparing the fault index with the preset threshold, an early warning can be issued before the problem becomes a serious fault, improving the safety and reliability of the transformer and the entire power grid, diagnosing and handling potential faults in a timely manner, and significantly reducing unexpected downtime and related losses caused by faults.

[0087] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0091] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0092] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for fault diagnosis and monitoring of power transformers based on the Internet of Things, characterized in that: The following steps are involved: Monitor power transformers in real time and obtain real-time diagnostic data of power transformers; The real-time diagnostic data of the power transformer is sent to the central database based on the Internet of Things communication, and the real-time diagnostic data of the power transformer received by the central database is analyzed and classified and stored in a set category library; Perform fault diagnosis on the power transformer based on the real-time diagnosis data of the power transformer that has been analyzed, classified and stored, and obtain the bushing discharge fault diagnosis index and the core vibration fault index; The bushing discharge fault diagnosis index and the core vibration fault index are used to judge whether the power transformer has a bushing discharge fault or a core vibration fault; The real-time diagnostic data of the power transformer includes bushing discharge characteristic data and core vibration characteristic data. The process of sending the real-time diagnostic data of the power transformer to the central database based on the Internet of Things communication is as follows: Conduct a preliminary analysis on the bushing discharge characteristic data to obtain bushing discharge ultrasonic data, bushing discharge partial discharge data and bushing discharge surrounding gas data; Conduct a preliminary analysis on the core vibration characteristic data to obtain core vibration data, core acoustic wave data, and magnetic field data around the core; Encrypt and compress the bushing discharge characteristic data and core vibration characteristic data after preliminary analysis; The encrypted and compressed bushing discharge characteristic data and the core vibration characteristic data after preliminary analysis are marked using a timestamp and a device ID to obtain a marked data packet; Sending the marked data packets to the central database based on IoT communication; The process of parsing and classifying the real-time diagnostic data of power transformers received by the central database is as follows: Check the consistency between the timestamp of the received marked data packet and the device ID. If they are inconsistent, it is considered that the data transmission is wrong. If the timestamp is consistent with the device ID, the marked data packet is decrypted and decompressed, and the casing discharge ultrasonic data, casing discharge partial discharge data, casing discharge surrounding gas data, core vibration data, core sound wave data and core surrounding magnetic field data are classified and stored in the set category library; The process of obtaining the bushing discharge fault diagnosis index is as follows: Extracting bushing discharge ultrasonic data, bushing discharge partial discharge data and bushing discharge surrounding gas data from the category library in the central database; The casing discharge ultrasonic data, casing discharge partial discharge data and casing discharge surrounding gas data are fused and analyzed, and the casing discharge fault diagnosis index is obtained based on the established casing discharge fusion analysis model; The process of obtaining the casing discharge fault diagnosis index based on the established fusion analysis model is as follows: Analyze the casing discharge ultrasonic data to obtain the casing discharge ultrasonic intensity; Analyze the bushing partial discharge data to obtain the energy of the maximum frequency in the spectrum characteristics of bushing partial discharge; Analyzing the gas data around the bushing discharge to obtain the gas concentration around the bushing discharge, wherein the gas concentration around the bushing discharge includes hydrogen concentration and ozone concentration; Based on the established casing discharge fusion analysis model, the casing discharge ultrasonic intensity, the energy of the maximum frequency in the frequency spectrum characteristics of casing partial discharge and the gas concentration around the casing discharge are processed to obtain the casing discharge fault diagnosis index. The process of obtaining the core vibration fault index is as follows: Extracting core vibration data, core acoustic wave data and magnetic field data around the core from the category library in the central database; The core vibration data, core acoustic wave data and magnetic field data around the core are fused and analyzed, and the core vibration fault index is obtained based on the established core vibration fusion analysis model; The process of obtaining the core vibration fault index based on the established core vibration fusion analysis model is as follows: Analyze the core vibration data to obtain core vibration characteristic data, wherein the core vibration characteristic data includes a core vibration spectrum offset and a core vibration maximum amplitude; Analyze the core acoustic wave data to obtain the core vibration acoustic emission energy; Analyze the magnetic field data around the core to obtain the magnetic field variation index around the core; Based on the established core vibration fusion analysis model, the core vibration characteristic data, core vibration acoustic emission energy and magnetic field variation index around the core are processed to obtain the core vibration fault index; The process of judging whether there is a bushing discharge fault or a core vibration fault in the power transformer is as follows: Determine whether the bushing discharge fault diagnosis index is greater than or equal to a set discharge fault threshold; if so, the power transformer has a bushing discharge fault; Determine whether the core vibration fault index is greater than or equal to a set vibration fault threshold, if so, the power transformer has a core vibration fault; If the bushing discharge fault diagnosis index is less than the set discharge fault threshold and the core vibration fault index is less than the set vibration fault threshold, the stability of the current power transformer is evaluated based on the bushing discharge fault diagnosis index and the core vibration fault index; The process of evaluating the stability of the current power transformer based on the bushing discharge fault diagnosis index and the core vibration fault index is as follows: The stability index of the power transformer is obtained based on the bushing discharge fault diagnosis index and the core vibration fault index. The calculation formula of the stability index is as follows: Where, Wd is the stability index, e is the natural constant, Fz is the bushing discharge fault diagnosis index, α1 is the bushing discharge weight factor, Dz is the core vibration fault index, and α2 is the core vibration weight factor; It is determined whether the stability index is greater than the set stability threshold. If so, an early warning is issued for the stability of the power transformer.

2. A power transformer fault diagnosis and monitoring device based on the Internet of Things, characterized in that: include: A processor and a memory, wherein computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, the processor executes the power transformer fault diagnosis and monitoring method based on the Internet of Things according to claim 1, further comprising a diagnostic data acquisition module, a classification storage module, a diagnostic module and an evaluation module, wherein: The diagnostic data acquisition module is used to monitor the power transformer in real time and acquire real-time diagnostic data of the power transformer; The classification storage module is used to send the real-time diagnosis data of the power transformer to the central database based on the Internet of Things communication, and parse the real-time diagnosis data of the power transformer received by the central database and classify and store it in a set category library; The diagnostic module is used to perform fault diagnosis on the power transformer based on the real-time diagnostic data of the power transformer that has been analyzed and classified and stored, and obtain the bushing discharge fault diagnosis index and the core vibration fault index; The evaluation module is used to determine whether the power transformer has a bushing discharge fault or a core vibration fault through a bushing discharge fault diagnosis index and a core vibration fault index.

Images