An artificial intelligence-based partial discharge data monitoring system and method

Abstract

The application discloses a kind of local discharge data supervision system and method based on artificial intelligence, belong to local discharge data supervision technical field.The application includes: artificial intelligence data expert module, local discharge data analysis module, supervision module and time early warning module;The output of the artificial intelligence data expert module is connected with the input of the local discharge data analysis module;The output of the local discharge data analysis module is connected with the input of the supervision module;The output of the supervision module is connected with the input of the time early warning module.The application can provide deep local discharge type analysis, improve detection capability as far as possible, while researching and developing local discharge spectrum model construction library, form local discharge online spectrum library construction, meet typical local discharge type spectrum library and local discharge fault case construction, during local discharge actual diagnosis, realize its case matching, training and analysis.

Description

Technical Field

[0001] This invention relates to the field of partial discharge data monitoring technology, specifically to a partial discharge data monitoring system and method based on artificial intelligence. Background Technology

[0002] Partial discharge is a localized and repetitive breakdown and extinction phenomenon that occurs within the insulation of electrical equipment under a certain applied voltage. The occurrence of partial discharge within the insulation is accompanied by phenomena such as light, heat, noise, electrical pulses, increased dielectric loss, and electromagnetic radiation. Currently, methods for detecting partial discharge generally include the following: pulsed current method, broadband detection method, narrowband detection method, ultrasonic detection method, and TEV detection method. However, current detection methods can only detect the generation of partial discharge and roughly determine its type. For example, the TEV detection method uses a capacitive sensor to couple the surface voltage to the sensor, which can quantitatively test the partial discharge status of equipment in operating switchgear. It can also locate the partial discharge source by the time difference of the same discharge source to different locations, and determine whether the partial discharge is internal or surface discharge. However, it cannot analyze deeper discharge types, such as internal discharge, which is located inside the insulation of electrical equipment. It is impossible to find the cause of the partial discharge, so it is impossible to take precautions in the production process. At the same time, in the process of monitoring partial discharge data, there is a lack of calibration values ​​for hazard handling, resulting in a lack of intelligent time monitoring. Currently, there are no relevant practices in China for in-depth research on partial discharge. Summary of the Invention

[0003] The purpose of this invention is to provide an artificial intelligence-based partial discharge data monitoring system and method to solve the problems mentioned in the background art.

[0004] To address the aforementioned technical problems, this invention provides the following technical solution: a partial discharge data monitoring method based on artificial intelligence, comprising the following steps:

[0005] S1. Construct an artificial intelligence data expert database, which includes historical test data on partial discharge. The historical test data is grouped and stored according to different periods of the same electrical equipment test specimen, and the partial discharge location of the electrical equipment test specimen is marked. The artificial intelligence data expert database is a computer intelligent program system with specialized knowledge and experience. The database used in the background is equivalent to the human brain, with rich knowledge reserves. The knowledge data and knowledge reasoning technology in the database are used to simulate experts solving problems.

[0006] S2. The artificial intelligence data expert database forms the trend of partial discharge data changes based on historical test data and constructs a partial discharge data analysis model.

[0007] S3. Based on real-time detection of partial discharge data changes, generate partial discharge prediction and analysis results, and feed them back to the regulatory port;

[0008] S4. The monitoring port is equipped with a time warning model. Based on the analysis of different dangerous time data, it outputs the monitoring time threshold and records it in the artificial intelligence data expert database. If the monitoring time threshold is exceeded and no feedback information is received from the technician, an emergency marker time is generated and pushed to the administrator port.

[0009] According to the above technical solution, the historical test data of partial discharge includes:

[0010] Six parameter data;

[0011] The six parameter data include: the interval between two partial discharges and the apparent discharge charge of each partial discharge, the test voltage of the partial discharge, the partial discharge energy, the partial discharge initiation voltage, and the partial discharge extinction voltage.

[0012] The apparent discharge charge refers to the amount of charge that causes an instantaneous change in voltage across the sample when the charge is instantaneously injected into the two ends of the partial discharge sample, which is equal to the instantaneous change in voltage caused by the partial discharge itself.

[0013] The test voltage for partial discharge refers to the specified voltage applied during the test procedure, under which the sample does not exhibit partial discharge exceeding the specified value;

[0014] The partial discharge energy refers to the energy consumed by the partial discharge pulse;

[0015] The partial discharge initiation voltage refers to the voltage applied to the test sample, starting from the lowest value at which no partial discharge has ever been measured, and gradually increasing until the lowest voltage at which this discharge value is observed in the test circuit.

[0016] The partial discharge extinction voltage refers to the lowest voltage at which the voltage applied to the test sample gradually decreases from a relatively high value of the partially discharge that has been measured until the discharge value can no longer be observed in the test measurement circuit.

[0017] According to the above technical solution, the partial discharge includes internal partial discharge, surface discharge, and corona discharge;

[0018] Internal partial discharge refers to a localized discharge occurring in the insulation system of an electrical equipment test sample when the electric field strength in a local area reaches the breakdown field strength of the dielectric. However, the discharge does not penetrate between the two conductors to which the voltage is applied, and the entire insulation system maintains its insulation performance. This type of discharge occurs within the insulation body and is called internal discharge. Internal partial discharge includes internal discharge within composite dielectrics and internal discharge within single dielectrics. Internal discharge within composite dielectrics refers to partial discharge occurring between insulating layers with different dielectric properties in a non-uniform insulating medium, such as partial discharge in the intermediate dielectric layer of gas-solid composite insulation or liquid-solid composite insulation. Internal discharge within single dielectrics refers to partial discharge occurring within the same dielectric layer in a non-uniform insulating medium, such as partial discharge in the pores of solid insulation or in the bubbles of liquid insulation.

[0019] The surface discharge refers to the occurrence of surface discharge when a dielectric medium has an electric field component parallel to the surface in an electric field, and the electric field component reaches the breakdown field strength. The corona discharge is the discharge that occurs when the electric field strength near the surface of a conductor reaches the breakdown field strength of a gas under extremely non-uniform electric field conditions.

[0020] According to the above technical solution, the construction of the partial discharge data analysis model includes:

[0021] The location of the partial discharge in the sample is used to determine whether it is an internal discharge. If it is an internal discharge, the type of internal discharge is further determined. If it is not an internal discharge, the corresponding discharge type is directly output to the monitoring port.

[0022] To obtain historical test data of partial discharge, take N consecutive sets of data under the same internal partial discharge type of the same test sample, denoted as A1 to AN. Each set of data includes the corresponding six parameter data. The interval between two partial discharges in the first set of data is denoted as unit 1. Any set of data is denoted as a set.

[0023] Take any parameter data from the set except the interval between two partial discharges, and form a data regression model with the interval between two partial discharges to form N sets of data. The data regression model refers to a linear regression model formed by taking the interval between two partial discharges as the vertical axis and taking any parameter data except the interval between two partial discharges as the horizontal axis.

[0024] The goodness of fit of the data is determined based on the five sets of data regression models formed. The determination of goodness of fit includes:

[0025] Using the actual interval between two partial discharges as input, the regression simulation value is output based on the data regression model. The absolute value of the difference between the regression simulation value and the actual value is calculated. If the absolute value of the difference exceeds the system's set threshold, it is determined to be a value that does not conform to the model. The non-conforming values ​​in any set of data are summarized to form a set of numbers K1 to KN under N sets of data, where K1 to KN represent the number of non-conforming values ​​in each set of data.

[0026] Using K1 to KN as the horizontal axis and the interval between two partial discharges as the vertical axis, a fitting model is formed as the partial discharge data analysis model. The partial discharge data of electrical equipment under actual measurement is obtained, the number of values ​​that do not conform to the model is calculated, and a partial discharge data analysis model under actual measurement is formed. The similarity of the partial discharge data analysis model under actual measurement is compared with the two fitting models of internal partial discharge, and the internal partial discharge type with higher similarity is output to the regulatory port.

[0027] According to the above technical solution, the time-warning model includes:

[0028] Obtain the partial discharge type provided by the regulatory port, and take the average value of the danger time value under the same partial discharge type of the test sample as the calibration value P. The danger time value refers to the duration between the time point of insulation failure of electrical equipment and the time point of the start of partial discharge.

[0029] The calibration value P is used as the regulatory time threshold and recorded in the artificial intelligence data expert database. If no feedback information is received from the technician after the regulatory time threshold is exceeded, an emergency marker is generated and pushed to the administrator's port.

[0030] An artificial intelligence-based partial discharge data monitoring system includes: an artificial intelligence data expert module, a partial discharge data analysis module, a monitoring module, and a time warning module;

[0031] The AI ​​data expert module is used to construct an AI data expert database, which includes historical test data on partial discharge. This historical test data is grouped and stored according to different periods of the same electrical equipment sample, and the partial discharge location of the electrical equipment sample is marked. The partial discharge data analysis module is used to call the historical test data from the AI ​​data expert database to form a trend of partial discharge data changes and construct a partial discharge data analysis model. The monitoring module, based on real-time detected changes in partial discharge data, forms a predictive analysis result of partial discharge and feeds it back to the monitoring port. Simultaneously, the monitoring port is equipped with a time warning model, which, based on different dangerous time data analysis, outputs a monitoring time threshold and records it in the AI ​​data expert database. The time warning module is used to generate an emergency marker and push it to the administrator port if no feedback information is received from the technician after the monitoring time threshold is exceeded.

[0032] The output of the artificial intelligence data expert module is connected to the input of the partial discharge data analysis module; the output of the partial discharge data analysis module is connected to the input of the monitoring module; and the output of the monitoring module is connected to the input of the time warning module.

[0033] According to the above technical solution, the artificial intelligence data expert module includes an analysis unit and a storage unit;

[0034] The analysis unit is used to build an artificial intelligence data expert database; the storage unit is used to store historical test data of partial discharge.

[0035] The output of the analysis unit is connected to the input of the storage unit.

[0036] According to the above technical solution, the partial discharge data analysis module includes a data retrieval unit and a model construction unit;

[0037] The data retrieval unit is used to retrieve historical test data from the artificial intelligence data expert database to form a partial discharge data change trend; the model building unit constructs a partial discharge data analysis model based on the partial discharge data change trend.

[0038] The output of the data retrieval unit is connected to the input of the model building unit.

[0039] According to the above technical solution, the monitoring module includes a model analysis unit and a time threshold unit;

[0040] The model analysis unit generates a prediction analysis result of partial discharge based on real-time detected changes in partial discharge data, and feeds it back to the monitoring port; the time threshold unit is equipped with a time warning model, which analyzes different dangerous time data, outputs a monitoring time threshold, and records it in the artificial intelligence data expert database.

[0041] The output of the model analysis unit is connected to the input of the time threshold unit.

[0042] According to the above technical solution, the time warning module includes a feedback unit and a warning unit;

[0043] The feedback unit is used to receive feedback information data from technicians; the early warning unit is used to generate an emergency marker and push it to the administrator port if no feedback information data is received from technicians after the monitoring time threshold has been exceeded.

[0044] The output of the feedback unit is connected to the input of the early warning unit.

[0045] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: This invention can provide in-depth partial discharge type analysis, thereby quickly determining the fault location and type when partial discharge occurs, maximizing detection capabilities, and further analyzing deeper discharge types after determining whether the partial discharge is internal or surface discharge; simultaneously, during the partial discharge data monitoring process, calibration values ​​for hazard handling are formed, improving the effectiveness of intelligent time-based monitoring; furthermore, a partial discharge atlas model construction library is researched and developed, forming an online partial discharge atlas library, satisfying the construction of typical partial discharge type atlas libraries and partial discharge fault cases, and enabling case matching, training, and analysis during actual partial discharge diagnosis, providing important theoretical basis for production and evaluation. Attached Figure Description

[0046] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0047] Figure 1 This is a flowchart illustrating an artificial intelligence-based partial discharge data monitoring method according to the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Please see Figure 1 In this first embodiment, an artificial intelligence-based partial discharge data monitoring method is provided. First, an artificial intelligence data expert database is constructed, which includes historical test data of partial discharge. The historical test data is grouped and stored according to different periods of the same electrical equipment test specimen, and the partial discharge location of the electrical equipment test specimen is marked.

[0050] Secondly, a trend analysis of partial discharge data changes is generated based on historical test data, wherein the historical test data of partial discharge includes:

[0051] Six parameter data;

[0052] The six parameter data include: the interval between two partial discharges and the apparent discharge charge of each partial discharge, the test voltage of the partial discharge, the partial discharge energy, the partial discharge initiation voltage, and the partial discharge extinction voltage.

[0053] The apparent discharge charge refers to the amount of charge that causes an instantaneous change in voltage across the sample when the charge is instantaneously injected into the two ends of the partial discharge sample, which is equal to the instantaneous change in voltage caused by the partial discharge itself.

[0054] The test voltage for partial discharge refers to the specified voltage applied during the test procedure, under which the sample does not exhibit partial discharge exceeding the specified value;

[0055] The partial discharge energy refers to the energy consumed by the partial discharge pulse;

[0056] The partial discharge initiation voltage refers to the voltage applied to the test sample, starting from the lowest value at which no partial discharge has ever been measured, and gradually increasing until the lowest voltage at which this discharge value is observed in the test circuit.

[0057] The partial discharge extinction voltage refers to the lowest voltage at which the voltage applied to the test sample gradually decreases from a relatively high value of the partially discharge that has been measured until the discharge value can no longer be observed in the test measurement circuit.

[0058] The partial discharge includes internal partial discharge, surface discharge, and corona discharge;

[0059] Internal partial discharge refers to a localized discharge occurring in the insulation system of an electrical equipment test sample when the electric field strength in a local area reaches the breakdown field strength of the dielectric. However, the discharge does not penetrate between the two conductors to which the voltage is applied, and the entire insulation system maintains its insulation performance. This type of discharge occurs within the insulation body and is called internal discharge. Internal partial discharge includes internal discharge in composite dielectrics and internal discharge in single dielectrics. Internal discharge in composite dielectrics refers to partial discharge occurring between insulation layers with different dielectric properties in a non-uniform insulating medium. Internal discharge in single dielectrics refers to partial discharge occurring within the same dielectric layer in a non-uniform insulating medium.

[0060] The surface discharge refers to the occurrence of surface discharge when a dielectric medium has an electric field component parallel to the surface in an electric field, and the electric field component reaches the breakdown field strength. The corona discharge is the discharge that occurs when the electric field strength near the surface of a conductor reaches the breakdown field strength of a gas under extremely non-uniform electric field conditions.

[0061] The construction of the partial discharge data analysis model includes:

[0062] The location of the partial discharge in the sample is used to determine whether it is an internal discharge. If it is an internal discharge, the type of internal discharge is further determined. If it is not an internal discharge, the corresponding discharge type is directly output to the monitoring port.

[0063] To obtain historical test data of partial discharge, take N consecutive sets of data under the same internal partial discharge type of the same test sample, denoted as A1 to AN. Each set of data includes the corresponding six parameter data. The interval between two partial discharges in the first set of data is denoted as unit 1. Any set of data is denoted as a set.

[0064] Take any parameter data from the set other than the interval between two partial discharges, and form a data regression model with N sets of data along with the interval between two partial discharges. The data regression model refers to a linear regression model with the interval between two partial discharges as the ordinate and any parameter data other than the interval between two partial discharges as the abscissa. For example, linear regression is used.

[0065] This results in five fitted lines, each with a different slope and constant term;

[0066] The goodness of fit of the data is determined based on the five sets of data regression models formed. The determination of the goodness of fit includes:

[0067] Using the actual interval between two partial discharges as input, the regression simulation value is output based on the data regression model. The absolute value of the difference between the regression simulation value and the actual value is calculated. If the absolute value of the difference exceeds the system's set threshold, it is determined to be a value that does not conform to the model. The non-conforming values ​​in any set of data are summarized to form a set of numbers K1 to KN under N sets of data, where K1 to KN represent the number of non-conforming values ​​in each set of data.

[0068] Using K1 to KN as the horizontal axis and the interval between two partial discharges as the vertical axis, a fitting model is formed as the partial discharge data analysis model. The partial discharge data of electrical equipment under actual measurement is obtained, the number of values ​​that do not conform to the model is calculated, and a partial discharge data analysis model under actual measurement is formed. The similarity of the partial discharge data analysis model under actual measurement is compared with the two fitting models of internal partial discharge, and the internal partial discharge type with higher similarity is output to the regulatory port.

[0069] Obtain the partial discharge type provided by the regulatory port, and take the average value of the danger time value under the same partial discharge type of the test sample as the calibration value P. The danger time value refers to the duration between the time point of insulation failure of electrical equipment and the time point of the start of partial discharge.

[0070] The calibration value P is used as the regulatory time threshold and recorded in the artificial intelligence data expert database. If no feedback information is received from the technician after the regulatory time threshold is exceeded, an emergency marker is generated and pushed to the administrator's port.

[0071] In this second embodiment, an artificial intelligence-based partial discharge data monitoring system is provided. The system includes: an artificial intelligence data expert module, a partial discharge data analysis module, a monitoring module, and a time warning module.

[0072] The AI ​​data expert module is used to construct an AI data expert database, which includes historical test data on partial discharge. This historical test data is grouped and stored according to different periods of the same electrical equipment sample, and the partial discharge location of the electrical equipment sample is marked. The partial discharge data analysis module is used to call the historical test data from the AI ​​data expert database to form a trend of partial discharge data changes and construct a partial discharge data analysis model. The monitoring module, based on real-time detected changes in partial discharge data, forms a predictive analysis result of partial discharge and feeds it back to the monitoring port. Simultaneously, the monitoring port is equipped with a time warning model, which, based on different dangerous time data analysis, outputs a monitoring time threshold and records it in the AI ​​data expert database. The time warning module is used to generate an emergency marker and push it to the administrator port if no feedback information is received from the technician after the monitoring time threshold is exceeded.

[0073] The output of the artificial intelligence data expert module is connected to the input of the partial discharge data analysis module; the output of the partial discharge data analysis module is connected to the input of the monitoring module; and the output of the monitoring module is connected to the input of the time warning module.

[0074] The artificial intelligence data expert module includes an analysis unit and a storage unit;

[0075] The analysis unit is used to build an artificial intelligence data expert database; the storage unit is used to store historical test data of partial discharge.

[0076] The output of the analysis unit is connected to the input of the storage unit.

[0077] The partial discharge data analysis module includes a data retrieval unit and a model construction unit;

[0078] The data retrieval unit is used to retrieve historical test data from the artificial intelligence data expert database to form a partial discharge data change trend; the model building unit constructs a partial discharge data analysis model based on the partial discharge data change trend.

[0079] The output of the data retrieval unit is connected to the input of the model building unit.

[0080] The monitoring module includes a model analysis unit and a time threshold unit;

[0081] The model analysis unit generates a prediction analysis result of partial discharge based on real-time detected changes in partial discharge data, and feeds it back to the monitoring port; the time threshold unit is equipped with a time warning model, which analyzes different dangerous time data, outputs a monitoring time threshold, and records it in the artificial intelligence data expert database.

[0082] The output of the model analysis unit is connected to the input of the time threshold unit.

[0083] The time warning module includes a feedback unit and a warning unit;

[0084] The feedback unit is used to receive feedback information data from technicians; the early warning unit is used to generate an emergency marker and push it to the administrator port if no feedback information data is received from technicians after the monitoring time threshold has been exceeded.

[0085] The output of the feedback unit is connected to the input of the early warning unit.

[0086] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0087] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A partial discharge data monitoring method based on artificial intelligence, characterized in that: The method includes the following steps: S1. Construct an artificial intelligence data expert database, which includes historical test data of partial discharge. The historical test data is grouped and stored according to different periods of the same electrical equipment test specimen, and the partial discharge location of the electrical equipment test specimen is marked. S2. An artificial intelligence data expert database forms a partial discharge data change trend based on historical test data, and constructs a partial discharge data analysis model; the construction of the partial discharge data analysis model includes: The location of the partial discharge in the sample is used to determine whether it is an internal discharge. If it is an internal discharge, the type of internal discharge is further determined. If it is not an internal discharge, the corresponding discharge type is directly output to the monitoring port. To obtain historical test data of partial discharge, take N consecutive sets of data under the same internal partial discharge type of the same test sample, denoted as A1 to AN. Each set of data includes the corresponding six parameter data. The interval between two partial discharges in the first set of data is denoted as unit 1. Any set of data is denoted as a set. Take any parameter data from the set except the interval between two partial discharges, and form a data regression model with the interval between two partial discharges to form N sets of data. The data regression model refers to a linear regression model formed by taking the interval between two partial discharges as the vertical axis and taking any parameter data except the interval between two partial discharges as the horizontal axis. The goodness of fit of the data is determined based on the five sets of data regression models formed. The determination of the goodness of fit includes: Using the actual interval between two partial discharges as input, the regression simulation value is output based on the data regression model. The absolute value of the difference between the regression simulation value and the actual value is calculated. If the absolute value of the difference exceeds the system's set threshold, it is determined to be a value that does not conform to the model. The non-conforming values ​​in any set of data are summarized to form a set of numbers K1 to KN under N sets of data, where K1 to KN represent the number of non-conforming values ​​in each set of data. Using K1 to KN as the horizontal axis and the interval between two partial discharges as the vertical axis, a fitting model is formed as the partial discharge data analysis model. The partial discharge data of electrical equipment under actual measurement is obtained, the number of values ​​that do not conform to the model is calculated, and a partial discharge data analysis model under actual measurement is formed. The similarity of the partial discharge data analysis model under actual measurement is compared with the two fitting models of internal partial discharge, and the internal partial discharge type with higher similarity is output to the regulatory port. S3. Based on real-time detection of partial discharge data changes, generate partial discharge prediction and analysis results, and feed them back to the regulatory port; S4. The monitoring port is equipped with a time-based early warning model. Based on the analysis of different dangerous time data, it outputs a monitoring time threshold, which is recorded in the artificial intelligence data expert database. If the monitoring time threshold is exceeded and no feedback information is received from the technician, an emergency marker is generated and pushed to the administrator port. The time-based early warning model includes: Obtain the partial discharge type provided by the regulatory port, and take the average value of the danger time value under the same partial discharge type of the test sample as the calibration value P. The danger time value refers to the duration between the time point of insulation failure of electrical equipment and the time point of the start of partial discharge. The calibration value P is used as the regulatory time threshold and recorded in the artificial intelligence data expert database. If no feedback information is received from the technician after the regulatory time threshold is exceeded, an emergency marker is generated and pushed to the administrator's port.

2. The partial discharge data monitoring method based on artificial intelligence according to claim 1, characterized in that: The historical test data for partial discharge includes: Six parameter data; The six parameter data include: the interval between two partial discharges and the apparent discharge charge of each partial discharge, the test voltage of the partial discharge, the partial discharge energy, the partial discharge initiation voltage, and the partial discharge extinction voltage. The apparent discharge charge refers to the amount of charge that causes an instantaneous change in voltage across the sample when the charge is instantaneously injected into the two ends of the partial discharge sample, which is equal to the instantaneous change in voltage caused by the partial discharge itself. The test voltage for partial discharge refers to the specified voltage applied during the test procedure, under which the sample does not exhibit partial discharge exceeding the specified value; The partial discharge energy refers to the energy consumed by the partial discharge pulse; The partial discharge initiation voltage refers to the voltage applied to the test sample, starting from the lowest value at which no partial discharge has ever been measured, and gradually increasing until the lowest voltage at which this discharge value is observed in the test circuit. The partial discharge extinction voltage refers to the lowest voltage at which the voltage applied to the test sample gradually decreases from a relatively high value of the partially discharge that has been measured until the discharge value can no longer be observed in the test measurement circuit.

3. The partial discharge data monitoring method based on artificial intelligence according to claim 2, characterized in that: The partial discharge includes internal partial discharge, surface discharge, and corona discharge; Internal partial discharge refers to a localized discharge occurring in the insulation system of an electrical equipment test sample when the electric field strength in a local area reaches the breakdown field strength of the dielectric. However, the discharge does not penetrate between the two conductors to which the voltage is applied, and the entire insulation system maintains its insulation performance. This type of discharge occurs within the insulation body and is called internal discharge. Internal partial discharge includes internal discharge in composite dielectrics and internal discharge in single dielectrics. Internal discharge in composite dielectrics refers to partial discharge occurring between insulation layers with different dielectric properties in a non-uniform insulating medium. Internal discharge in single dielectrics refers to partial discharge occurring within the same dielectric layer in a non-uniform insulating medium. The surface discharge refers to the occurrence of surface discharge when a dielectric medium has an electric field component parallel to the surface in an electric field, and the electric field component reaches the breakdown field strength. The corona discharge is the discharge that occurs when the electric field strength near the surface of a conductor reaches the breakdown field strength of a gas under extremely non-uniform electric field conditions.

4. An artificial intelligence-based partial discharge data monitoring system, used to implement the artificial intelligence-based partial discharge data monitoring method as described in claim 1, characterized in that: The system includes: an artificial intelligence data expert module, a partial discharge data analysis module, a monitoring module, and a time warning module; The AI ​​data expert module is used to construct an AI data expert database, which includes historical test data on partial discharge. This historical test data is grouped and stored according to different periods of the same electrical equipment sample, and the partial discharge location of the electrical equipment sample is marked. The partial discharge data analysis module is used to call the historical test data from the AI ​​data expert database to form a trend of partial discharge data changes and construct a partial discharge data analysis model. The monitoring module, based on real-time detected changes in partial discharge data, forms a predictive analysis result of partial discharge and feeds it back to the monitoring port. Simultaneously, the monitoring port is equipped with a time warning model, which, based on different dangerous time data analysis, outputs a monitoring time threshold and records it in the AI ​​data expert database. The time warning module is used to generate an emergency marker and push it to the administrator port if no feedback information is received from the technician after the monitoring time threshold is exceeded. The output of the artificial intelligence data expert module is connected to the input of the partial discharge data analysis module; the output of the partial discharge data analysis module is connected to the input of the monitoring module; and the output of the monitoring module is connected to the input of the time warning module.

5. The partial discharge data monitoring system based on artificial intelligence according to claim 4, characterized in that: The artificial intelligence data expert module includes an analysis unit and a storage unit; The analysis unit is used to build an artificial intelligence data expert database; the storage unit is used to store historical test data of partial discharge. The output of the analysis unit is connected to the input of the storage unit.

6. The partial discharge data monitoring system based on artificial intelligence according to claim 4, characterized in that: The partial discharge data analysis module includes a data retrieval unit and a model construction unit; The data retrieval unit is used to retrieve historical test data from the artificial intelligence data expert database to form a partial discharge data change trend; the model building unit constructs a partial discharge data analysis model based on the partial discharge data change trend. The output of the data retrieval unit is connected to the input of the model building unit.

7. The partial discharge data monitoring system based on artificial intelligence according to claim 4, characterized in that: The monitoring module includes a model analysis unit and a time threshold unit; The model analysis unit generates a prediction analysis result of partial discharge based on real-time detected changes in partial discharge data, and feeds it back to the monitoring port; the time threshold unit is equipped with a time warning model, which analyzes different dangerous time data, outputs a monitoring time threshold, and records it in the artificial intelligence data expert database. The output of the model analysis unit is connected to the input of the time threshold unit.

8. The partial discharge data monitoring system based on artificial intelligence according to claim 4, characterized in that: The time warning module includes a feedback unit and a warning unit; The feedback unit is used to receive feedback information data from the technician; The early warning unit is used to generate an emergency marker and push it to the administrator's port if no feedback information is received from the technician after the monitoring time threshold has been exceeded. The output of the feedback unit is connected to the input of the early warning unit.

Images