Dry-type power transformer monitoring and diagnosis system

Abstract

The invention discloses a dry-type power transformer monitoring and diagnosis system, which relates to the technical field of power equipment monitoring and comprises the following steps: acquiring real-time monitoring data; the monitoring module receives and processes the real-time monitoring data from the data acquisition module, constructs a transformer monitoring model according to the real-time monitoring data, and generates a transformer monitoring coefficient; processing the historical data of the transformer based on a deep learning algorithm to obtain a transformer monitoring threshold value, performing comparative analysis on the transformer monitoring coefficient and the transformer monitoring threshold value, and performing intelligent diagnosis on the health state of the transformer according to an analysis result; and according to the fault diagnosis result, analyzing the insulating material performance coefficient and the temperature fluctuation coefficient of the transformer, and judging the fault diagnosis effect. The problems that an existing dry-type power transformer monitoring system is limited in monitoring parameter, low in fault diagnosis accuracy and incapable of responding in real time are solved, various working states of the dry-type power transformer can be efficiently and comprehensively monitored, and fault diagnosis can be carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment monitoring, and more specifically, to a dry-type power transformer monitoring and diagnosis system. Background Art

[0002] Dry-type power transformers are air-cooled transformers that are safer and more reliable than oil-immersed transformers. Monitoring and diagnostic systems are important means to ensure that dry-type power transformers are in good operating condition.

[0003] Deficiencies of existing technologies: Dry-type power transformers are widely used in power systems, especially in urban power grids and industrial electricity. With the increase in the load of the power system and the complexity of the transformer working environment, dry-type power transformers are prone to overload, temperature rise, insulation aging and other faults during operation, which may cause equipment damage or paralysis of the power system in severe cases. Therefore, real-time monitoring and fault diagnosis of dry-type power transformers are of great significance. The existing dry-type power transformer monitoring system has problems such as limited monitoring parameters, low fault diagnosis accuracy and inability to respond in real time. It is impossible to detect potential faults in time, thus affecting the stability of the power system. Therefore, there is a need for a technical solution that can efficiently and comprehensively monitor the various working conditions of dry-type power transformers and perform fault diagnosis.

[0004] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a dry-type power transformer monitoring and diagnosis system, which solves the problems raised in the above-mentioned background technology by monitoring and diagnosing the transformer.

[0006] To achieve the above object, the present invention provides the following technical solutions: A dry-type power transformer monitoring and diagnostic system, comprising: A data acquisition module is used to acquire real-time monitoring data, including winding temperature and partial discharge power, and measure the input and output voltage waveforms of the transformer, and analyze the harmonic content in the waveform to obtain the voltage harmonic amplitude; The data processing module is used to receive and process the real-time monitoring data from the data acquisition module, perform data filtering and normalization, and build a transformer monitoring model based on the real-time monitoring data to generate a transformer monitoring coefficient; The fault diagnosis module is used to analyze and process the historical data of the transformer based on the deep learning algorithm to obtain the transformer monitoring threshold, compare and analyze the transformer monitoring coefficient with the transformer monitoring threshold, and perform intelligent diagnosis of the transformer health status based on the analysis results; The evaluation module is used to obtain the insulation material performance coefficient according to the working temperature and health status of the insulation material in combination with the insulation resistance of the transformer, and to obtain the temperature fluctuation coefficient according to the winding temperature. In combination with the insulation material performance coefficient, the fault diagnosis effect is judged based on fuzzy reasoning.

[0007] In a preferred embodiment, the specific process of obtaining the winding temperature is as follows: Acquire real-time temperature data through a temperature sensor, the real-time temperature data including ambient temperature, temperature before and after the transformer is operated, and subtract the temperature before and after the transformer is operated from each other to obtain the temperature rise caused by the transformer operation; Obtain the current and resistance of the transformer winding, and calculate the load loss of the transformer according to the current and resistance of the transformer winding; Obtain the no-load loss of the transformer, and calculate the loss power of the transformer based on the no-load loss and load loss of the transformer; combine the ambient temperature and the temperature before and after the operation of the transformer to obtain the winding temperature of the transformer.

[0008] In a preferred embodiment, the process of obtaining the partial discharge power is as follows: Determine the timing of partial discharge events through time-frequency analysis techniques; And the instantaneous values ​​of voltage and current during partial discharge are collected in real time according to the voltage sensor and the current sensor; The partial discharge power is calculated based on the instantaneous values ​​of voltage and current and the time of the partial discharge event.

[0009] In a preferred embodiment, the specific steps of obtaining the voltage harmonic amplitude are as follows: Connect the input port of the harmonic analyzer to the input and output terminals of the transformer respectively; And set the frequency range of the harmonic analyzer and configure the sampling frequency of the harmonic analyzer to obtain accurate waveforms and harmonic components; Collect voltage waveforms at the input and output ends of the transformer respectively, and use the spectrum analysis function of the harmonic analyzer to view the spectrum distribution of the input and output waveforms; The time domain signal is converted into a frequency domain signal to display the components of different frequencies, and the amplitudes of different harmonic frequencies are calculated to obtain the voltage harmonic amplitude.

[0010] In a preferred embodiment, the transformer monitoring threshold acquisition step is as follows: Build a deep learning model and process the historical data of the transformer according to the deep learning algorithm. Divide the historical data into training set, validation set and test set, and train and optimize the deep learning model through parameter adjustment and cross-validation. Identify the normal operating range and determine the monitoring threshold by modeling historical data; For the predicted output exceeding the threshold, it is determined to be a potential abnormal state.

[0011] In a preferred embodiment, the steps of comparing and analyzing the transformer monitoring coefficient with the transformer monitoring threshold and performing intelligent diagnosis on the transformer health status according to the analysis result are as follows: Compare and analyze the transformer monitoring coefficient and the transformer monitoring threshold; If the transformer monitoring coefficient is greater than the transformer monitoring threshold, it is detected that the transformer is faulty and the fault needs to be diagnosed and the transformer needs to be repaired immediately; If the transformer monitoring coefficient is less than the transformer monitoring threshold, it is detected that the transformer is in normal condition and does not need to be repaired immediately, and continuous monitoring is performed.

[0012] In a preferred embodiment, the specific steps of obtaining the performance coefficient of the insulation material of the transformer are as follows: Use an insulation resistance meter to measure the insulation resistance of the transformer regularly; Use online monitoring equipment to monitor the health of transformer insulation materials in real time; And obtain the working temperature of the insulating material; combine the insulation resistance of the transformer and the health status of the insulating material to obtain the performance coefficient of the insulating material.

[0013] In a preferred embodiment, the specific steps of obtaining the temperature fluctuation coefficient are as follows: The winding temperature of the transformer is obtained regularly within a certain period, and the average winding temperature within the period is calculated based on the winding temperature of the transformer. The temperature fluctuation coefficient is calculated based on the average winding temperature combined with the winding temperature of the transformer.

[0014] In a preferred embodiment, the specific steps of determining the fault diagnosis effect are as follows: The insulation material performance coefficient and temperature fluctuation coefficient are defined as input variables and divided into different fuzzy sets respectively; The fault diagnosis effect is defined as the output variable and divided into fuzzy sets; Formulate fuzzy rules to describe the influence of insulation material performance coefficient and temperature fluctuation coefficient on fault diagnosis effect; Perform fuzzy reasoning based on fuzzy rules and evaluate the fault diagnosis effect.

[0015] The technical effects and advantages of the dry-type power transformer monitoring and diagnosis system of the present invention are as follows: 1. The present invention monitors the dry-type power transformer in real time, continuously measures and analyzes key parameters such as temperature, voltage, current, load, and oil temperature, and obtains the working status of the transformer in real time. The safety of transformer operation is improved, and early warning can be given before a fault occurs, reducing the impact of sudden faults on the power system. Equipment damage or reduced efficiency caused by long-term operation in an abnormal state is avoided. The system can comprehensively diagnose the operating status of the transformer and identify potential faults and operating problems by comprehensively analyzing a variety of monitoring data (such as temperature, current, vibration, etc.). Through comprehensive diagnosis of multiple parameters, the source of the problem can be identified more accurately. The operating efficiency and reliability of the transformer are improved. By comparing historical data with real-time data, potential hidden dangers are discovered and processed accordingly. The system can analyze historical data and optimize the monitoring model. The system will continuously improve the diagnostic accuracy through self-learning and pattern recognition capabilities. The system has adaptive capabilities. With the accumulation of operating data, the system's diagnostic capabilities are gradually enhanced, and more potential fault modes can be identified. Reduce manual intervention, realize automatic monitoring, and improve the intelligence level of the system. The system uses advanced data mining and pattern recognition technology to detect abnormal conditions during transformer operation and perform automatic diagnosis. It improves the accuracy and timeliness of diagnosis, and can issue early warnings before transformer failure occurs, preventing equipment failure from expanding and avoiding system shutdown or major accidents. It realizes remote diagnosis, and can remotely diagnose and view data of transformers through the network, reducing the workload of manual inspections and manual judgments.

[0016] 2. The present invention displays the monitoring data through a visual interface, so that the operator can clearly understand the real-time operating status of the transformer. Through the graphical interface, the user can intuitively view the equipment status and easily identify potential problems. Support real-time data query and historical data analysis, provide strong support for decision-makers, and improve the efficiency of management and maintenance. The design of the monitoring system takes into account the adaptability of different working environments and can work stably in a variety of complex and harsh environments. It can operate stably under conditions such as high temperature, high humidity, and harsh electromagnetic environment to ensure the accuracy of data and the continuity of the system. It is suitable for different types of dry-type power transformers and has strong versatility and scalability. The system can store historical data for a long time and generate detailed analysis reports. It provides a complete equipment operation history record, which helps to track and analyze the use of transformers for a long time. The automatically generated report can provide intuitive reports for equipment managers and maintenance personnel to help them make effective maintenance decisions. Through accurate monitoring and diagnosis of the transformer operation status, the system can accurately predict the maintenance cycle of the equipment and the parts that need to be replaced. Through a comprehensive assessment of the transformer status, unnecessary repairs and premature replacement of parts can be avoided, reducing maintenance costs. Extend equipment service life, reduce downtime and the frequency of emergency repairs, thereby reducing downtime losses and production losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a schematic diagram of the structure of a dry-type power transformer monitoring and diagnosis system. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Embodiment 1, Figure 1 The invention provides a dry-type power transformer monitoring and diagnosis system.

[0020] A data acquisition module is used to acquire real-time monitoring data, including winding temperature and partial discharge power, and measure the input and output voltage waveforms of the transformer, and analyze the harmonic content in the waveform to obtain the voltage harmonic amplitude; The specific process of obtaining the winding temperature is as follows: Install a temperature sensor and obtain real-time temperature data through the temperature sensor, wherein the real-time temperature data includes the ambient temperature and the temperature before and after the transformer is operated, and the temperature rise generated by the transformer operation is obtained by subtracting the temperature before and after the transformer is operated; Obtain the current and resistance of the transformer winding, and calculate the load loss of the transformer according to the current and resistance of the transformer winding; Obtain the no-load loss of the transformer, and calculate the transformer power loss based on the no-load loss and load loss of the transformer; combine the ambient temperature and the temperature before and after the operation of the transformer to obtain the winding temperature of the transformer. The specific calculation formula is as follows: ; Where T is the winding temperature, is the ambient temperature, I is the current of the transformer winding, R is the resistance of the transformer winding, is the load loss of the transformer, is the no-load loss of the transformer, K is the transformer design parameter; The real-time winding temperature is compared with the preset temperature alarm value. Once the preset temperature alarm value is exceeded, an alarm is immediately sounded, and the work is stopped, and maintenance personnel carry out repairs.

[0021] The process of obtaining the partial discharge power is as follows: Determine the timing of partial discharge events through time-frequency analysis techniques; And the instantaneous values ​​of voltage and current during partial discharge are collected in real time according to the voltage sensor and the current sensor; The partial discharge power is calculated based on the instantaneous values ​​of voltage and current and the time of the partial discharge event. The specific calculation formula is as follows: ; In the formula, is the partial discharge power, is the voltage and instantaneous value, is the instantaneous value of the current, t is the time of the partial discharge event; The partial discharge power is compared with the preset power alarm value; once the preset power alarm value is exceeded, an alarm is immediately sounded and the operation is stopped.

[0022] The occurrence of partial discharge is an early sign of insulation damage inside the transformer. Through high-precision partial discharge detection, if the partial discharge is significantly reduced or eliminated after adjustment, it means that the insulation condition of the transformer has improved.

[0023] The specific steps for obtaining the voltage harmonic amplitude are as follows: Connect the input port of the harmonic analyzer to the input and output terminals of the transformer respectively; And set the frequency range of the harmonic analyzer and configure the sampling frequency of the harmonic analyzer to obtain accurate waveforms and harmonic components; Collect voltage waveforms at the input and output ends of the transformer respectively, and use the spectrum analysis function of the harmonic analyzer to view the spectrum distribution of the input and output waveforms; The time domain signal is converted into a frequency domain signal to display the components of different frequencies, and the amplitudes of different harmonic frequencies are calculated to obtain the voltage harmonic amplitude; the specific calculation formula is as follows: ;Where f is the voltage harmonic amplitude, Z is the signal period, is the time domain signal, and h is the harmonic order.

[0024] The data processing module is used to receive and process the real-time monitoring data from the data acquisition module, perform data filtering and normalization, and build a transformer monitoring model based on the real-time monitoring data to generate a transformer monitoring coefficient; The data filtering and normalization processing steps are as follows: Kalman filtering is used to remove noise from real-time monitoring data, and statistical methods are used to detect and remove unreasonable data; The data is normalized by scaling it to the range [0, 1] using min-max normalization.

[0025] The benefits of data filtering and normalization of real-time monitoring data include the following aspects: 1. Improve data quality Data filtering: Real-time monitoring data is often affected by noise, interference or outliers. Data filtering can remove these invalid or inaccurate data and retain valid signals, thereby improving the accuracy and reliability of the data. For example, filtering can remove sensor errors or sudden fluctuations to ensure that subsequent analysis and processing are based on high-quality data.

[0026] Normalization: Monitoring data from different sources or types may have different dimensions and ranges. Normalization converts all data to a unified scale, which helps avoid unfair weight distribution due to large numerical differences between data and ensures fairness and consistency in data analysis.

[0027] 2. Improve model performance Data filtering: By removing noise and abnormal data, model training and prediction effects will be better. Especially in the process of machine learning and data mining, purifying data can effectively reduce errors, prevent the model from being affected by erroneous data, and improve prediction accuracy and reliability.

[0028] Normalization: For many machine learning algorithms (such as support vector machines, neural networks, K-means clustering, etc.), normalization can improve the convergence speed and computational efficiency of the algorithm, prevent the dimensional differences of different features from affecting the performance of the model, and ensure that each feature has the same influence during the training process.

[0029] 3. Speed ​​up calculations Data filtering: Remove irrelevant, duplicate or abnormal data points to reduce the amount of data, making subsequent processing (such as storage, transmission and calculation) more efficient and avoiding unnecessary waste of resources.

[0030] Normalization: Normalization usually makes data distribution more even, helps reduce overflow and precision problems during numerical processing, and optimizes the calculation process. Especially for high-dimensional data, it can significantly improve the calculation speed and stability of the algorithm.

[0031] 4. Enhance the stability and robustness of the model Data filtering: By removing inaccurate or abnormal data points, the model can avoid shock and instability when facing sudden data, and enhance the robustness of the system or model. This is especially important for real-time monitoring systems, because data fluctuations may have an adverse effect on real-time response.

[0032] Normalization: After data normalization, all features have the same scale, reducing the risk of some features dominating the entire model training process due to their large values, and enhancing the stability of the model under different input data conditions.

[0033] 5. Facilitate data fusion and comparison Data filtering: Cleaned data is easier to fuse and compare with other data from different sensors or sources, which helps improve the accuracy of data integration. Filtered data also reduces the interference of redundant information.

[0034] Normalization: By standardizing all data to the same range or scale, data from different sources can be seamlessly connected, simplifying the difficulty of multi-source data fusion. Normalization makes different types of monitoring data more comparable and compatible.

[0035] 6. Improve system responsiveness Data filtering: After removing unnecessary fluctuations and erroneous data in real-time monitoring, the system will respond to valid data more quickly and accurately, avoiding false alarms or response delays caused by erroneous data.

[0036] Normalization processing: In real-time monitoring systems, normalization makes the fluctuation range of various data features tend to be consistent, which can ensure that the system responds to data changes more stably and quickly.

[0037] The transformer monitoring model is constructed based on the real-time monitoring data and the voltage harmonic amplitude, and the transformer monitoring coefficient calculation formula is generated as follows: ; Where B is the transformer monitoring coefficient, is the winding temperature, is the partial discharge power, f is the voltage harmonic amplitude, is the winding temperature weighting factor, is the partial discharge power weighting factor, is the voltage harmonic amplitude weighting factor, , Both are greater than 0.

[0038] The fault diagnosis module is used to analyze and process the historical data of the transformer based on the deep learning algorithm to obtain the transformer monitoring threshold, compare and analyze the transformer monitoring coefficient with the transformer monitoring threshold, and perform intelligent diagnosis of the transformer health status based on the analysis results; The steps for obtaining the transformer monitoring threshold are as follows: Build a deep learning model and process the historical data of the transformer according to the deep learning algorithm. Divide the historical data into training set, validation set and test set, and train and optimize the deep learning model through parameter adjustment and cross-validation. Identify the normal operating range and determine the monitoring threshold by modeling historical data; For the predicted output exceeding the threshold, it is determined to be a potential abnormal state.

[0039] The steps of comparing and analyzing the transformer monitoring coefficient with the transformer monitoring threshold and performing intelligent diagnosis on the transformer health status according to the analysis results are as follows: Compare and analyze the transformer monitoring coefficient and the transformer monitoring threshold; If the transformer monitoring coefficient is greater than the transformer monitoring threshold, it is detected that the transformer is faulty and the fault needs to be diagnosed and the transformer needs to be repaired immediately; If the transformer monitoring coefficient is less than the transformer monitoring threshold, it is detected that the transformer is in normal condition and does not need to be repaired immediately, and continuous monitoring is performed.

[0040] Through the above comparative analysis, intelligent diagnosis can bring the following benefits: Early warning: Timely identification of potential transformer failures and early warnings help formulate maintenance plans and avoid power outages or equipment damage due to equipment failures.

[0041] Optimized maintenance: Intelligent diagnosis based on monitoring data can implement accurate maintenance strategies and avoid unnecessary periodic inspections, thereby saving maintenance costs and human resources.

[0042] Improve safety: Through continuous monitoring and evaluation, we can enhance the safety management of transformer equipment, reduce the occurrence of equipment accidents, and thus protect the safety of personnel and property.

[0043] Extend equipment life: By promptly identifying and addressing problems, you can keep the transformer in optimal operating condition, thereby extending its service life and improving return on investment.

[0044] Data-driven decision support: The collected historical monitoring data and analysis results can provide reliable support for future decisions, such as optimizing the selection and operation of transformers.

[0045] The evaluation module is used to obtain the insulation material performance coefficient according to the working temperature and health status of the insulation material in combination with the insulation resistance of the transformer, and to obtain the temperature fluctuation coefficient according to the winding temperature. In combination with the insulation material performance coefficient, the fault diagnosis effect is judged based on fuzzy reasoning.

[0046] The specific steps for obtaining the performance coefficient of the insulation material of the transformer are as follows: Use an insulation resistance meter to measure the insulation resistance of the transformer regularly; Use online monitoring equipment to monitor the health of transformer insulation materials in real time; And obtain the working temperature of the insulating material; combine the insulation resistance of the transformer and the health status of the insulating material to obtain the insulation material performance coefficient. The specific calculation formula is as follows: ; In the formula, J is the performance coefficient of the transformer insulation material, is the insulation resistance of the transformer, Health factor of insulation material, is the operating temperature of the insulating material.

[0047] The specific steps for obtaining the temperature fluctuation coefficient are as follows: The transformer winding temperature is obtained regularly within a certain period, and the average winding temperature within the period is calculated based on the transformer winding temperature. The temperature fluctuation coefficient is calculated based on the average winding temperature combined with the transformer winding temperature. The specific calculation formula is as follows: ; In the formula is the temperature fluctuation coefficient, is the winding temperature recorded at the ith time, is the average winding temperature and n is the total number of temperatures.

[0048] The specific steps of determining the fault diagnosis effect are as follows: Step C1, defining the insulation material performance coefficient and the temperature fluctuation coefficient as input variables, and dividing them into different fuzzy sets respectively.

[0049] For example, "Low", "Medium", "High" for insulation material performance coefficient, "Low", "Medium", "High" for temperature fluctuation coefficient.

[0050] In step C2, the fault diagnosis effect is defined as the output variable, which is divided into fuzzy sets, for example, "High", "Low", for the fault diagnosis effect.

[0051] Step C3, formulate a set of fuzzy rules to describe the impact of different input variables on output variables. The definition of rules can be based on professional knowledge or obtained through data analysis and experiments. For example: The insulation material performance coefficient is marked as J and the temperature fluctuation coefficient is marked as , the fault diagnosis effect is marked as P and can be defined as Rule 1: IF (J is Low) AND ( is High) THEN (P is Low ) Rule 2: IF (J is High) AND ( is Low) THEN (P is High ) ... Step C4, perform fuzzy reasoning according to fuzzy rules and evaluate the fault diagnosis effect.

[0052] It should be noted that the division of fuzzy sets can be adjusted according to actual conditions. For example, although this embodiment takes three fuzzy sets as examples, in fact, the insulation material performance coefficient, temperature fluctuation coefficient, and fault diagnosis effect can be divided into more than three sets to facilitate better and more accurate identification.

[0053] Furthermore, for the judgment of the insulation material performance coefficient and the temperature fluctuation coefficient, the threshold value can be set according to the actual situation; when the insulation material performance coefficient is higher than 20, it is marked as "High", when the temperature fluctuation coefficient is higher than 0.7, it is marked as "High", and so on. I will not go into details here.

[0054] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0055] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0056] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0057] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0058] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0059] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dry-type power transformer monitoring and diagnostic system, characterized in that: It includes data acquisition module, data processing module, fault diagnosis module and evaluation module. There are links between modules: A data acquisition module is used to acquire real-time monitoring data, including winding temperature and partial discharge power, and measure the input and output voltage waveforms of the transformer, and analyze the harmonic content in the waveform to obtain the voltage harmonic amplitude; The data processing module is used to receive and process the real-time monitoring data and voltage harmonic amplitude from the data acquisition module, perform data filtering and normalization, and build a transformer monitoring model based on the real-time monitoring data and voltage harmonic amplitude to generate a transformer monitoring coefficient; The fault diagnosis module is used to analyze and process the historical data of the transformer based on the deep learning algorithm to obtain the transformer monitoring threshold, compare and analyze the transformer monitoring coefficient with the transformer monitoring threshold, and perform intelligent diagnosis of the transformer health status based on the analysis results; The evaluation module is used to obtain the insulation material performance coefficient according to the working temperature and health status of the insulation material in combination with the insulation resistance of the transformer, and to obtain the temperature fluctuation coefficient according to the winding temperature. In combination with the insulation material performance coefficient, the fault diagnosis effect is judged based on fuzzy reasoning.

2. A dry-type power transformer monitoring and diagnostic system according to claim 1, characterized in that: The specific process of obtaining the winding temperature is as follows: Acquire real-time temperature data through a temperature sensor, the real-time temperature data including ambient temperature, temperature before and after the transformer is operated, and subtract the temperature before and after the transformer is operated from each other to obtain the temperature rise caused by the transformer operation; Obtain the current and resistance of the transformer winding, and calculate the load loss of the transformer according to the current and resistance of the transformer winding; Obtain the no-load loss of the transformer, and calculate the transformer power loss based on the no-load loss and load loss of the transformer; combine the ambient temperature and the temperature before and after the operation of the transformer to obtain the winding temperature of the transformer. The specific calculation formula is as follows: ; Where T is the winding temperature, is the ambient temperature, I is the current of the transformer winding, R is the resistance of the transformer winding, is the load loss of the transformer, is the no-load loss of the transformer, and K is the transformer design parameter.

3. A dry-type power transformer monitoring and diagnostic system according to claim 2, characterized in that: The process of obtaining the partial discharge power is as follows: Determine the timing of partial discharge events through time-frequency analysis techniques; And the instantaneous values ​​of voltage and current during partial discharge are collected in real time according to the voltage sensor and the current sensor; The partial discharge power is calculated based on the instantaneous values ​​of voltage and current and the time of the partial discharge event. The specific calculation formula is as follows: ; In the formula, is the partial discharge power, is the voltage and instantaneous value, is the instantaneous value of the current and t is the time of the partial discharge event.

4. A dry-type power transformer monitoring and diagnostic system according to claim 3, characterized in that: The specific steps for obtaining the voltage harmonic amplitude are as follows: Connect the input port of the harmonic analyzer to the input and output terminals of the transformer respectively; And set the frequency range of the harmonic analyzer and configure the sampling frequency of the harmonic analyzer to obtain accurate waveforms and harmonic components; Collect voltage waveforms at the input and output ends of the transformer respectively, and use the spectrum analysis function of the harmonic analyzer to view the spectrum distribution of the input and output waveforms; The time domain signal is converted into a frequency domain signal to display the components of different frequencies, and the amplitudes of different harmonic frequencies are calculated to obtain the voltage harmonic amplitude; the specific calculation formula is as follows: ;Where f is the voltage harmonic amplitude, Z is the signal period, is the time domain signal and h is the harmonic frequency.

5. A dry-type power transformer monitoring and diagnostic system according to claim 4, characterized in that: The transformer monitoring model is constructed based on the real-time monitoring data and the voltage harmonic amplitude, and the transformer monitoring coefficient calculation formula is generated as follows: ; Where B is the transformer monitoring coefficient, is the winding temperature, is the partial discharge power, f is the voltage harmonic amplitude, is the winding temperature weighting factor, is the partial discharge power weighting factor, is the voltage harmonic amplitude weighting factor, , Both are greater than 0.

6. A dry-type power transformer monitoring and diagnostic system according to claim 5, characterized in that: The steps for obtaining the transformer monitoring threshold are as follows: Build a deep learning model and process the historical data of the transformer according to the deep learning algorithm. Divide the historical data into training set, validation set and test set, and train and optimize the deep learning model through parameter adjustment and cross-validation. Identify the normal operating range and determine the monitoring threshold by modeling historical data; For the predicted output exceeding the threshold, it is determined to be a potential abnormal state.

7. A dry-type power transformer monitoring and diagnostic system according to claim 6, characterized in that: The steps of comparing and analyzing the transformer monitoring coefficient with the transformer monitoring threshold and performing intelligent diagnosis on the transformer health status according to the analysis results are as follows: Compare and analyze the transformer monitoring coefficient and the transformer monitoring threshold; If the transformer monitoring coefficient is greater than the transformer monitoring threshold, it is detected that the transformer is faulty and the fault needs to be diagnosed and the transformer needs to be repaired immediately; If the transformer monitoring coefficient is less than the transformer monitoring threshold, it is detected that the transformer is in normal condition and does not need to be repaired immediately, and continuous monitoring is performed.

8. A dry-type power transformer monitoring and diagnostic system according to claim 7, characterized in that: The specific steps for obtaining the performance coefficient of the insulation material of the transformer are as follows: Use an insulation resistance meter to measure the insulation resistance of the transformer regularly; Use online monitoring equipment to monitor the health of transformer insulation materials in real time; And obtain the working temperature of the insulating material, and combine the insulation resistance of the transformer and the health status of the insulating material to obtain the insulation material performance coefficient. The specific calculation formula is as follows: ; In the formula, J is the performance coefficient of the transformer insulation material, is the insulation resistance of the transformer, Health factor of insulation material, is the operating temperature of the insulating material.

9. A dry-type power transformer monitoring and diagnostic system according to claim 8, characterized in that: The specific steps for obtaining the temperature fluctuation coefficient are as follows: The winding temperature of the transformer is obtained regularly within a certain period, and the average winding temperature within the period is calculated based on the winding temperature of the transformer; The temperature fluctuation coefficient is calculated based on the average winding temperature combined with the winding temperature of the transformer. The specific calculation formula is as follows: ; In the formula is the temperature fluctuation coefficient, is the winding temperature recorded at the ith time, is the average winding temperature and n is the total number of temperatures.

10. A dry-type power transformer monitoring and diagnostic system according to claim 9, characterized in that: The specific steps of determining the fault diagnosis effect are as follows: The insulation material performance coefficient and temperature fluctuation coefficient are defined as input variables and divided into different fuzzy sets respectively; The fault diagnosis effect is defined as the output variable and divided into fuzzy sets; Formulate fuzzy rules to describe the influence of insulation material performance coefficient and temperature fluctuation coefficient on fault diagnosis effect; Perform fuzzy reasoning based on fuzzy rules and evaluate the fault diagnosis effect.

Images