On-line intelligent monitoring pressure relief valve of transformer

Abstract

The invention discloses a transformer on-line intelligent monitoring pressure relief valve which comprises a pressure relief valve body used for transformer on-line monitoring, a control terminal is installed on the surface of the pressure relief valve body and used for regulating and controlling the pressure relief valve body, and a control system comprises a data acquisition module, a data analysis module and a strategy feedback module. According to the on-line intelligent monitoring pressure relief valve for the transformer, collected data are classified, the collected data are extracted, an internal parameter change model is established, an influence rule of external temperature change on internal oil temperature is determined, a parameter safety threshold value is extracted based on historical data, and various parameter conditions are analyzed; whether the transformer has a fault problem is determined according to an analysis result, the abnormal condition of the transformer is determined in time by adopting multi-parameter fusion analysis, intelligent and safe operation of a distribution network is ensured, precise maintenance is realized, resource waste caused by blind maintenance is avoided, and core support is provided for power system safety and the like.

Description

Technical Field

[0001] The present invention relates to the field of monitoring and control technology, in particular to an online intelligent monitoring pressure relief valve for a transformer. Background Art

[0002] In smart grids, transformers are core equipment and are prone to risks such as current overload and heat dissipation failure due to long-term high-load operation. To ensure the intelligent and safe operation of the distribution network, the transformer online intelligent monitoring pressure relief valve was developed. The product uses the Internet of Things and sensor technology to address the safety pain points of transformer operation in substations.

[0003] The reference patent name is: An intelligent online monitoring device for substation transformers (patent publication number: CN114113856A, patent publication date: 2022-03-01), including an installation box, a transformer, an oil level monitoring device, a transformer partial discharge monitoring device and an intelligent terminal receiver. The oil level monitoring device includes an oil tank, an oil pipe, a float, a spiral plate, a rotating shaft and a spring. The oil tank is fixedly installed in the inner cavity of the transformer, the oil pipe is fixedly installed on the top of the inner cavity of the oil tank, the float is placed in the inner cavity of the oil pipe, the spiral plate is installed in the inner cavity of the oil pipe through the rotating shaft, and the rotating shaft passes through the oil pipe and the oil tank. The spring is fixedly installed at the other end of the rotating shaft, and the other end of the spring is fixedly installed on one side of the inner cavity of the transformer. A varistor is provided on the surface of the spring. The oil level monitoring device is used to monitor the oil level of the transformer, and the oil tank is used to hold oil.

[0004] Based on the description in the above-mentioned document, in the existing online monitoring process of transformers, under long-term load operation, abnormal temperature, unbalanced air pressure or excessively high oil level in the transformer will cause transformer operation failure. However, when monitoring operation abnormalities, the monitoring value is often compared with the threshold. That is, when the threshold is exceeded, the transformer has already failed. However, sometimes no abnormality occurs when the monitoring value is exceeded, resulting in inaccurate monitoring results. For this reason, the present invention provides an online intelligent monitoring pressure relief valve for transformers. Summary of the Invention

[0005] In response to the deficiencies in the prior art, the present invention provides an online intelligent monitoring pressure relief valve for transformers, which solves the problem that in the existing online monitoring process of transformers, abnormal temperature, air pressure imbalance or excessively high oil level in the transformer under long-term load operation may cause transformer operation failures. However, when monitoring operation abnormalities, the monitoring value is often compared with the threshold. If the threshold is exceeded, the transformer has failed. However, sometimes no abnormality occurs when the monitoring value exceeds the threshold, resulting in inaccurate monitoring results.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a transformer online intelligent monitoring pressure relief valve, comprising a pressure relief valve body for online monitoring of the transformer, wherein a control terminal is mounted on the surface of the pressure relief valve body for regulating the pressure relief valve body, and a display instrument is mounted on the outside of the control terminal for displaying the monitored parameter values, an inspection window is provided above the arc surface of the pressure relief valve body for viewing the oil level in the transformer, and the control terminal implements data analysis and processing and generates a control strategy through a control system; The control system includes a data acquisition module, a data analysis module and a strategy feedback module; The data acquisition module monitors the oil level, oil temperature and oil pressure parameters through a variety of sensors, and realizes data transmission through the communication module; The data analysis module classifies and processes the collected data, extracts the collected data to establish an internal parameter change model, determines the rules for the impact of external temperature changes on internal oil temperature, and extracts parameter safety thresholds based on historical data. By analyzing the conditions of various parameters, it determines whether the transformer has a fault problem based on the analysis results; The strategy feedback module derives the required control strategy based on the analysis results and the fault conditions, displays the generated parameters, and converts the strategy into instructions and transmits them to the control unit of the pressure relief valve body for control.

[0007] Preferably, the multiple sensors include an internal temperature sensor, an external temperature sensor, a pressure sensor and an oil level sensor. The internal temperature sensor is used to monitor the oil temperature inside the transformer, the external temperature sensor is used to monitor the temperature outside the transformer, the pressure sensor is used to monitor the internal oil pressure of the transformer, and the oil level sensor is used to monitor the oil level status inside the transformer.

[0008] Preferably, the communication module transmits data through wired transmission and wireless transmission. The wired transmission is through the RS485 or Modbus communication interface, and the transformer status data is connected to the communication module in a wired manner. The wireless transmission is through a combination of one or more Bluetooth, WiFi, Lora and 4G wireless interfaces to realize the access of the transformer status data to the communication module.

[0009] Preferably, the data analysis module performs classification processing operations on the collected data as follows: Set the classification template, which consists of column title bar, row title bar and parameter result bar; The column title bar contains different types of parameter names, the row title bar contains the time node of parameter collection, and the intersection of the column title bar and the row title bar is the parameter result bar with the corresponding title contents combined; The collected data is mapped to the classification template, and the content features are extracted through the recognition window, and the matching with the corresponding content of the column title bar and row title is achieved. After the content of the column title bar and row title bar matches, the parameter results are filled into the corresponding parameter result bar to form a classification template with complete parameter results.

[0010] Preferably, the operation of extracting content features by the recognition window is: The collected data is traversed from beginning to end using a recognition window with two character recognition amounts, and the content features obtained by recognition are extracted and matched with the title content in the classification template; If the content features obtained by recognition are identical to some characters of the title content in the classification template, the character recognition amount of the recognition window is expanded until the content features obtained by recognition are completely identical to the title content features in the classification template, that is, the recognition window is initialized after the currently determined content features and the next content feature extraction operation is started. The initialized recognition window is a recognition window with two character recognition amounts, and after determining the contents of the column title bar and the row title bar, the corresponding parameter results are filled into the parameter result bar; During the recognition process, when a punctuation mark or a space is encountered, the content feature extraction operation of the current recognition window is terminated, and a new initialization recognition window is opened from the character after the punctuation mark or the space.

[0011] Preferably, the operation of extracting the collected data and establishing the internal parameter change model in the data analysis module is: The corresponding structure of the current pressure relief valve and the control unit of the related processing are digitally restored to form an initial model of the internal parameter change model; Then, the operation of the initial model is realized according to the dynamic operation of the control unit of the corresponding structure and related processing, and the historical data is extracted to form a training set and a test set in proportion. The initial model is introduced into the training set for learning, and the initial model is introduced into the test set for testing and optimization and adjustment are realized to form the required internal parameter change model.

[0012] Preferably, the data analysis module determines the influence of the external temperature change on the internal oil temperature as follows: By extracting the same external temperature at adjacent time nodes and marking it as T a The transformer internal oil temperature data corresponding to time node a is extracted and marked as S a , and sort the internal oil temperature data according to the order of time nodes; The time node sequence is used as the horizontal axis category, and the internal oil temperature data at the time node is used as the vertical axis category. The horizontal and vertical axes intersect to form a coordinate axis. The internal oil temperature data at the time node are connected to form an oil temperature change curve. The influence of the external temperature on the internal oil temperature is determined based on the trend of the oil temperature change curve. The current external temperature T is obtained based on the oil temperature data that tends to be horizontal in the subsequent change curve. a The lower internal oil temperature impact data is marked as R; The different external temperature data when the internal oil temperature influence data is R are extracted, and the influence interval is formed by the minimum and maximum values ​​of the current different external temperature data and marked as [Tr min , Tr max ], and Tr refers to the external temperature data when the internal oil temperature influence data is R. When the real-time collected external temperature belongs to the influence interval [Tr min , Tr max ] the internal oil temperature R can be predicted.

[0013] Preferably, the operation of extracting parameter safety thresholds based on historical data and analyzing various parameter conditions in the data analysis module is: Set periodic nodes, extract internal oil temperature data at the time nodes, and determine anomalies based on the extreme values ​​and changes in internal oil temperature and the corresponding thresholds; Extract internal oil pressure data at a specific time point and determine anomalies based on the extreme values ​​and changes in internal oil pressure and the corresponding thresholds. Extract the internal oil level data at the time node, and determine the anomaly based on the extreme value of the internal oil level and the comparison of the oil level change with the corresponding threshold.

[0014] Preferably, the operation of determining the abnormality based on the comparison of the extreme value of the internal oil temperature and the oil temperature change with the corresponding threshold value is: The oil temperature safety threshold range at a single time node under the historical data extraction is [P b , P c ], and extract the oil temperature change threshold value of the cycle time under the historical data as Q d The oil temperature value at the sequential time node is compared with the oil temperature safety threshold interval. If the oil temperature value is less than the oil temperature safety threshold interval or greater than the oil temperature safety threshold interval, and the duration is y, the current oil temperature value is abnormal; Then calculate the oil temperature change rate within the cycle time, the calculation formula is: F=[(U1-U2) / U2]×100%, F is the oil temperature change rate, U1 is the last node of the cycle time, U2 is the initial node of the cycle time, and the oil temperature change rate is compared with the oil temperature change threshold Q d Compare, and F>Q d When , the current oil temperature change is abnormal, and each time a time node is updated, the initial node of the previous oil temperature change rate is removed and the oil temperature change rate monitoring is continued; And by relying on the comparison and analysis principle of the extreme value of oil temperature and the change of oil temperature, the extreme value of oil pressure, the change of oil pressure, the extreme value of oil level and the change of oil level can be determined, thereby determining the abnormal situation.

[0015] Preferably, the strategy feedback module derives the required control strategy based on the analysis results and the fault situation as follows: When an abnormal oil temperature is detected, the oil temperature is adjusted by generating a temperature adjustment strategy; When abnormal oil pressure and oil temperature are detected, the temperature control strategy is generated to give priority to the control. Otherwise, the pressure relief valve is controlled by generating a pressure control strategy to perform the control operation. When an abnormal oil level is detected, if the oil pressure is also abnormal at the same time, a pressure regulating strategy is generated to control the pressure relief valve for regulation operation, otherwise a single oil level control strategy is generated.

[0016] The present invention provides an online intelligent monitoring pressure relief valve for transformers. Compared with the prior art, it has the following advantages: 1. The transformer's online intelligent monitoring pressure relief valve classifies and processes the collected data, extracts the collected data to establish an internal parameter change model, determines the rules for the impact of external temperature changes on internal oil temperature, and extracts parameter safety thresholds based on historical data. By analyzing the various parameter conditions, it determines whether the transformer has a fault problem based on the analysis results. Multi-parameter fusion analysis is used to promptly determine transformer abnormalities, ensure the intelligent and safe operation of the distribution network, achieve precise maintenance, avoid the waste of resources caused by blind maintenance, and provide core support for power system safety.

[0017] 2. The transformer online intelligent monitoring pressure relief valve transmits information through wired and wireless transmission to achieve remote monitoring and big data analysis, reducing the workload and frequency of manual inspections. It also sets classification templates to classify data to retain valid data. After classification, it is convenient for subsequent model establishment or data analysis, while reducing resource usage, and can complete detection operations more efficiently, ensuring better operation of the intelligent monitoring pressure relief valve.

[0018] 3. The transformer's online intelligent monitoring pressure relief valve extracts internal oil temperature data at time nodes by setting periodic nodes, and determines anomalies based on the extreme internal oil temperature and oil temperature changes compared with the corresponding thresholds. It also determines anomalies based on the extreme internal oil pressure and oil pressure changes compared with the corresponding thresholds. It also determines anomalies based on the extreme internal oil level and oil level changes compared with the corresponding thresholds. In this way, equipment hidden dangers can be discovered in advance based on parameters and precise maintenance can be carried out, which can avoid further deterioration of equipment failures, extend equipment service life, reduce the error rate of abnormality judgment, and achieve more accurate online monitoring of transformers. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a three-dimensional diagram of the external structure of the intelligent monitoring pressure relief valve of the present invention; Figure 2 It is a principle block diagram of the control system of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1-Figure 2 , the present invention provides two technical solutions: Embodiment 1: A transformer online intelligent monitoring pressure relief valve includes a pressure relief valve body 1 for transformer online monitoring. A control terminal 2 is mounted on the surface of the pressure relief valve body 1 for regulating the pressure relief valve body 1. A display instrument 3 is mounted on the outside of the control terminal 2 for displaying the monitored parameter values. An inspection window 4 is provided above the arc surface of the pressure relief valve body 1 for viewing the oil level in the transformer. The control terminal 2 performs data analysis and processing and generates a control strategy through a control system. The control system includes a data acquisition module, a data analysis module and a strategy feedback module; The data acquisition module monitors the oil level, oil temperature and oil pressure parameters through a variety of sensors, and realizes data transmission through the communication module; The data analysis module classifies and processes the collected data, extracts the collected data to establish an internal parameter change model, determines the rules for the impact of external temperature changes on internal oil temperature, and extracts parameter safety thresholds based on historical data. By analyzing the conditions of various parameters, it determines whether the transformer has a fault problem based on the analysis results; The strategy feedback module derives the required control strategy based on the analysis results and the fault conditions, displays the generated parameters, and converts the strategy into instructions and transmits them to the control unit of the pressure relief valve body 1 for control.

[0022] By classifying and processing the collected data, extracting the collected data to establish an internal parameter change model, determining the rules for the impact of external temperature changes on internal oil temperature, and extracting parameter safety thresholds based on historical data, analyzing the various parameter conditions, and determining whether the transformer has a fault problem based on the analysis results, multi-parameter fusion analysis is used to promptly determine transformer abnormalities, ensure the intelligent and safe operation of the distribution network, achieve precise maintenance, avoid the waste of resources caused by blind maintenance, and provide core support for power system safety.

[0023] In an embodiment of the present invention, the multiple sensors include an internal temperature sensor, an external temperature sensor, a pressure sensor, and an oil level sensor. The internal temperature sensor is used to monitor the oil temperature inside the transformer, the external temperature sensor is used to monitor the temperature outside the transformer, the pressure sensor is used to monitor the internal oil pressure of the transformer, and the oil level sensor is used to monitor the oil level status inside the transformer.

[0024] In an embodiment of the present invention, the communication module transmits data through wired transmission and wireless transmission. The wired transmission is through the RS485 or Modbus communication interface, and the transformer status data is connected to the communication module in a wired manner. The wireless transmission is through a combination of one or more Bluetooth, WiFi, Lora and 4G wireless interfaces to realize the access of the transformer status data to the communication module.

[0025] Information is transmitted through wired and wireless transmission to achieve remote monitoring and big data analysis, reducing the workload and frequency of manual inspections. Classification templates are set to classify data to retain valid data. Classification facilitates subsequent model establishment or data analysis, while reducing resource usage, enabling more efficient detection operations and ensuring better operation of the intelligent monitoring pressure relief valve.

[0026] In the embodiment of the present invention, the data analysis module performs classification processing operations on the collected data as follows: Set the classification template, which consists of column title bar, row title bar and parameter result bar; The column title bar contains different types of parameter names, the row title bar contains the time node of parameter collection, and the intersection of the column title bar and the row title bar is the parameter result bar with the corresponding title contents combined; The collected data is mapped to the classification template, and the content features are extracted through the recognition window, and the matching with the corresponding content of the column title bar and row title is achieved. After the content of the column title bar and row title bar matches, the parameter results are filled into the corresponding parameter result bar to form a classification template with complete parameter results.

[0027] In the embodiment of the present invention, the operation of extracting content features by the recognition window is: The collected data is traversed from beginning to end using a recognition window with two character recognition amounts, and the content features obtained by recognition are extracted and matched with the title content in the classification template; If the content features obtained by recognition are identical to some characters of the title content in the classification template, the character recognition amount of the recognition window is expanded until the content features obtained by recognition are completely identical to the title content features in the classification template, that is, the recognition window is initialized after the currently determined content features and the next content feature extraction operation is started. The initialized recognition window is a recognition window with two character recognition amounts, and after determining the contents of the column title bar and the row title bar, the corresponding parameter results are filled into the parameter result bar; During the recognition process, when a punctuation mark or a space is encountered, the content feature extraction operation of the current recognition window is terminated, and a new initialization recognition window is opened from the character after the punctuation mark or the space.

[0028] In the embodiment of the present invention, the operation of extracting the collected data and establishing the internal parameter change model in the data analysis module is as follows: The corresponding structure of the current pressure relief valve and the control unit of the related processing are digitally restored to form an initial model of the internal parameter change model; Then, the operation of the initial model is realized according to the dynamic operation of the control unit of the corresponding structure and related processing, and the historical data is extracted to form a training set and a test set in proportion. The initial model is introduced into the training set for learning, and the initial model is introduced into the test set for testing and optimization and adjustment are realized to form the required internal parameter change model.

[0029] In the embodiment of the present invention, the data analysis module determines the influence of the external temperature change on the internal oil temperature as follows: By extracting the same external temperature at adjacent time nodes and marking it as T a The transformer internal oil temperature data corresponding to time node a is extracted and marked as S a , and sort the internal oil temperature data according to the order of time nodes; The time node sequence is used as the horizontal axis category, and the internal oil temperature data at the time node is used as the vertical axis category. The horizontal and vertical axes intersect to form a coordinate axis. The internal oil temperature data at the time node are connected to form an oil temperature change curve. The influence of the external temperature on the internal oil temperature is determined based on the trend of the oil temperature change curve. The current external temperature T is obtained based on the oil temperature data that tends to be horizontal in the subsequent change curve. a The lower internal oil temperature impact data is marked as R; The different external temperature data when the internal oil temperature influence data is R are extracted, and the influence interval is formed by the minimum and maximum values ​​of the current different external temperature data and marked as [Tr min , Tr max], and Tr refers to the external temperature data when the internal oil temperature influence data is R. When the real-time collected external temperature belongs to the influence interval [Tr min , Tr max ] the internal oil temperature R can be predicted.

[0030] In the embodiment of the present invention, the operation of extracting parameter safety thresholds based on historical data and analyzing various parameter conditions in the data analysis module is as follows: Set periodic nodes, extract internal oil temperature data at the time nodes, and determine anomalies based on the extreme values ​​and changes in internal oil temperature and the corresponding thresholds; Extract internal oil pressure data at a specific time point and determine anomalies based on the extreme values ​​and changes in internal oil pressure and the corresponding thresholds. Extract the internal oil level data at the time node, and determine the anomaly based on the extreme value of the internal oil level and the comparison of the oil level change with the corresponding threshold.

[0031] In the embodiment of the present invention, the operation of determining an abnormality based on the comparison of the extreme value of the internal oil temperature and the oil temperature change with the corresponding threshold value is as follows: The oil temperature safety threshold range at a single time node under the historical data extraction is [P b , P c ], and extract the oil temperature change threshold value of the cycle time under the historical data as Q d The oil temperature value at the sequential time node is compared with the oil temperature safety threshold interval. If the oil temperature value is less than the oil temperature safety threshold interval or greater than the oil temperature safety threshold interval, and the duration is y, the current oil temperature value is abnormal; Then calculate the oil temperature change rate within the cycle time, the calculation formula is: F=[(U1-U2) / U2]×100%, F is the oil temperature change rate, U1 is the last node of the cycle time, U2 is the initial node of the cycle time, and the oil temperature change rate is compared with the oil temperature change threshold Q d Compare, and F>Q d When , the current oil temperature change is abnormal, and each time a time node is updated, the initial node of the previous oil temperature change rate is removed and the oil temperature change rate monitoring is continued; And by relying on the comparison and analysis principle of the extreme value of oil temperature and the change of oil temperature, the extreme value of oil pressure, the change of oil pressure, the extreme value of oil level and the change of oil level can be determined, thereby determining the abnormal situation.

[0032] By setting periodic nodes, extracting internal oil temperature data at time nodes, and comparing the extreme values ​​of internal oil temperature and oil temperature changes with corresponding thresholds to determine anomalies, comparing the extreme values ​​of internal oil pressure and oil pressure changes with corresponding thresholds to determine anomalies, and comparing the extreme values ​​of internal oil level and oil level changes with corresponding thresholds to determine anomalies, equipment hidden dangers can be discovered in advance based on parameters and precise maintenance can be carried out, which can avoid further deterioration of equipment failures, extend equipment service life, reduce the error rate of abnormality judgment, and achieve more accurate online monitoring of transformers.

[0033] In the embodiment of the present invention, the strategy feedback module derives the required control strategy operations based on the analysis results and the fault situation: When an abnormal oil temperature is detected, the oil temperature is adjusted by generating a temperature adjustment strategy; When abnormal oil pressure and oil temperature are detected, the temperature control strategy is generated to give priority to the control. Otherwise, the pressure relief valve is controlled by generating a pressure control strategy to perform the control operation. When an abnormal oil level is detected, if the oil pressure is also abnormal at the same time, a pressure regulating strategy is generated to control the pressure relief valve for regulation operation, otherwise a single oil level control strategy is generated.

[0034] The difference between Example 2 and Example 1 is that: by setting the transformer with the same parameters, the existing pressure relief valve and the online intelligent monitoring pressure relief valve of the present invention are installed on the transformer to perform parameter monitoring operations, and the processing efficiency and the number of faults of the existing pressure relief valve and the online intelligent monitoring pressure relief valve of the present invention within the cycle time are recorded, and the recording results are shown in Table 1: Table 1 Record table

[0035] To sum up, although the existing pressure relief valve can be used for monitoring operations, it often requires staff to go to the site for inspection and maintenance. During the monitoring process through the intelligent monitoring pressure relief valve of the present invention, not only can remote viewing be performed, but the number of abnormalities is also reduced. It can also effectively predict the situation of failure after the abnormality and perform predictive maintenance in advance, so that the transformer online intelligent monitoring pressure relief valve can be better used in actual operations.

[0036] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0037] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0038] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A transformer online intelligent monitoring pressure relief valve, comprising a pressure relief valve body (1) for transformer online monitoring, characterized in that: A control terminal (2) is installed on the surface of the pressure relief valve body (1) for regulating the pressure relief valve body (1), and a display instrument (3) is installed outside the control terminal (2) for displaying the parameter values ​​obtained by monitoring. An inspection window (4) is provided above the arc surface of the pressure relief valve body (1) for checking the oil level in the transformer, and the control terminal (2) realizes data analysis and processing and generates a control strategy through a control system; The control system includes a data acquisition module, a data analysis module and a strategy feedback module; The data acquisition module monitors the oil level, oil temperature and oil pressure parameters through a variety of sensors, and realizes data transmission through the communication module; The data analysis module classifies and processes the collected data, extracts the collected data to establish an internal parameter change model, determines the rules for the impact of external temperature changes on internal oil temperature, and extracts parameter safety thresholds based on historical data. By analyzing the conditions of various parameters, it determines whether the transformer has a fault problem based on the analysis results; The strategy feedback module derives the required control strategy based on the analysis results and the fault conditions, displays the generated parameters, and converts the strategy into instructions to be transmitted to the control unit of the pressure relief valve body (1) for control.

2. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The multiple sensors include an internal temperature sensor, an external temperature sensor, a pressure sensor and an oil level sensor. The internal temperature sensor is used to monitor the oil temperature inside the transformer, the external temperature sensor is used to monitor the temperature outside the transformer, the pressure sensor is used to monitor the internal oil pressure of the transformer, and the oil level sensor is used to monitor the oil level status inside the transformer.

3. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The communication module transmits data through wired transmission and wireless transmission. Wired transmission uses RS485 or Modbus communication interface to connect transformer status data to the communication module in a wired manner. Wireless transmission uses one or more combinations of Bluetooth, WiFi, Lora and 5G wireless interfaces to realize the access of transformer status data to the communication module.

4. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The data analysis module performs classification processing operations on the collected data as follows: Set the classification template, which consists of column title bar, row title bar and parameter result bar; The column title bar contains different types of parameter names, the row title bar contains the time node of parameter collection, and the intersection of the column title bar and the row title bar is the parameter result bar with the corresponding title contents combined; The collected data is mapped to the classification template, and the content features are extracted through the recognition window, and the matching with the corresponding content of the column title bar and row title is achieved. After the content of the column title bar and row title bar matches, the parameter results are filled into the corresponding parameter result bar to form a classification template with complete parameter results.

5. The transformer online intelligent monitoring pressure relief valve according to claim 4 is characterized in that: The operation of extracting content features by the recognition window is as follows: The collected data is traversed from beginning to end using a recognition window with two character recognition amounts, and the content features obtained by recognition are extracted and matched with the title content in the classification template; If the content features obtained by recognition are identical to some characters of the title content in the classification template, the character recognition amount of the recognition window is expanded until the content features obtained by recognition are completely identical to the title content features in the classification template, that is, the recognition window is initialized after the currently determined content features and the next content feature extraction operation is started. The initialized recognition window is a recognition window with two character recognition amounts, and after determining the contents of the column title bar and the row title bar, the corresponding parameter results are filled into the parameter result bar; During the recognition process, when a punctuation mark or a space is encountered, the content feature extraction operation of the current recognition window is terminated, and a new initialization recognition window is opened from the character after the punctuation mark or the space.

6. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The operation of extracting the collected data and establishing the internal parameter change model in the data analysis module is as follows: The corresponding structure of the current pressure relief valve and the control unit of the related processing are digitally restored to form an initial model of the internal parameter change model; Then, the operation of the initial model is realized according to the dynamic operation of the control unit of the corresponding structure and related processing, and the historical data is extracted to form a training set and a test set in proportion. The initial model is introduced into the training set for learning, and the initial model is introduced into the test set for testing and optimization and adjustment are realized to form the required internal parameter change model.

7. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The data analysis module determines the influence of external temperature changes on internal oil temperature as follows: By extracting the same external temperature at adjacent time nodes and marking it as T a The transformer internal oil temperature data corresponding to time node a is extracted and marked as S a , and sort the internal oil temperature data according to the order of time nodes; The time node sequence is used as the horizontal axis category, and the internal oil temperature data at the time node is used as the vertical axis category. The horizontal and vertical axes intersect to form a coordinate axis. The internal oil temperature data at the time node are connected to form an oil temperature change curve. The influence of the external temperature on the internal oil temperature is determined based on the trend of the oil temperature change curve. The current external temperature T is obtained based on the oil temperature data that tends to be horizontal in the subsequent change curve. a The lower internal oil temperature impact data is marked as R; The different external temperature data when the internal oil temperature influence data is R are extracted, and the influence interval is formed by the minimum and maximum values ​​of the current different external temperature data and marked as [Tr min , Tr max ], and Tr refers to the external temperature data when the internal oil temperature influence data is R. When the real-time collected external temperature belongs to the influence interval [Tr min , Tr max ] the internal oil temperature R can be predicted.

8. The transformer online intelligent monitoring pressure relief valve according to claim 1 is characterized in that: The data analysis module extracts parameter safety thresholds based on historical data and analyzes various parameter conditions in the following operations: Set periodic nodes, extract internal oil temperature data at the time nodes, and determine anomalies based on the extreme values ​​and changes in internal oil temperature and the corresponding thresholds; Extract internal oil pressure data at a specific time point and determine anomalies based on the extreme values ​​and changes in internal oil pressure and the corresponding thresholds. Extract the internal oil level data at the time node, and determine the anomaly based on the extreme value of the internal oil level and the comparison of the oil level change with the corresponding threshold.

9. The transformer online intelligent monitoring pressure relief valve according to claim 8, characterized in that: The operation of determining an abnormality based on the extreme value of the internal oil temperature, the oil temperature change and the corresponding threshold value is as follows: The oil temperature safety threshold range at a single time node under the historical data extraction is [P b , P c ], and extract the oil temperature change threshold value of the cycle time under the historical data as Q d The oil temperature value at the sequential time node is compared with the oil temperature safety threshold interval. If the oil temperature value is less than the oil temperature safety threshold interval or greater than the oil temperature safety threshold interval, and the duration is y, the current oil temperature value is abnormal; Then calculate the oil temperature change rate within the cycle time, the calculation formula is: F=[(U1-U2) / U2]×100%, F is the oil temperature change rate, U1 is the last node of the cycle time, U2 is the initial node of the cycle time, and the oil temperature change rate is compared with the oil temperature change threshold Q d Compare, and F>Q d When , the current oil temperature change is abnormal, and each time a time node is updated, the initial node of the previous oil temperature change rate is removed and the oil temperature change rate monitoring is continued; And by relying on the comparison and analysis principle of the extreme value of oil temperature and the change of oil temperature, the extreme value of oil pressure, the change of oil pressure, the extreme value of oil level and the change of oil level can be determined, thereby determining the abnormal situation.

10. The transformer online intelligent monitoring pressure relief valve according to claim 8, characterized in that: The strategy feedback module derives the required control strategy operations based on the analysis results and the fault conditions: When an abnormal oil temperature is detected, the oil temperature is adjusted by generating a temperature adjustment strategy; When abnormal oil pressure and oil temperature are detected, the temperature control strategy is generated to give priority to the control. Otherwise, the pressure relief valve is controlled by generating a pressure control strategy to perform the control operation. When an abnormal oil level is detected, if the oil pressure is also abnormal at the same time, a pressure regulating strategy is generated to control the pressure relief valve for regulation operation, otherwise a single oil level control strategy is generated.

Images