Method and system for monitoring electrical power equipment fault based on Internet

Abstract

The invention discloses a method and system for monitoring faults of electrical power equipment based on the Internet, relates to the technical field of electricities, and aims to monitor the change condition of monitored oil temperature and the concentration state of combustible gas in oil in real time based on an internal oil circulating cooling system of a transformer, construct an oil load abnormal coefficient Xyc, and determine the fault of the electrical power equipment based on the Xyc. A load current Ifz is constructed according to a transformer load current change condition, the load current Ifz is associated with an oil load abnormity coefficient Xyc to calculate an oil load abnormity association index Zyf, a preliminary early warning signal is sent out, and the multi-dimensional state information of a winding, an iron core and a wiring terminal is collected and combined with the oil load abnormity association index Zyf, so that an oil load abnormity early warning result is obtained. According to the method, a fault comprehensive evaluation index Zgz is generated through fitting of a fault evaluation model, then the fault comprehensive evaluation index Zgz is compared with a preset evaluation threshold G, whether a fault risk exists or not is judged, a corresponding grade early warning instruction is generated, and the accuracy and response efficiency of transformer fault detection are achieved.

Description

Technical Field

[0001] This invention relates to the field of electrical technology, specifically to a method and system for monitoring faults in electrical power equipment based on the Internet. Background Technology

[0002] The electrical field is central to energy transmission and utilization in modern society, encompassing multiple stages from power generation to transmission and distribution. Among these, transformers, as core components of voltage conversion and energy transmission in power systems, directly impact the safety and reliability of the power grid. Oil circulation cooling systems, a crucial component of large transformers, function by using circulating oil to remove heat from the transformer's interior, preventing damage to the windings and core due to overheating. Furthermore, circulating oil possesses excellent electrical insulation properties; by providing a sealed cover to the active parts of the transformer, the circulating oil further enhances the insulation strength of the windings, core, and terminals. Extending the service life of transformers and improving operational safety are crucial. Therefore, in the transformer's oil circulation cooling system, the state of the circulating oil directly reflects winding overheating, uneven core heating, and arcing at the terminals. However, transformers are susceptible to load changes and component aging during long-term operation, easily leading to a series of potential faults. When these deficiencies are not addressed in a timely manner, they increase the operational risks of the transformer and may even cause transformer shutdowns, large-scale power outages, and fires, resulting in severe economic losses and social impact. Therefore, to effectively address these issues, transformer equipment fault monitoring technology in the electrical field has gradually become a research hotspot.

[0003] Although existing methods for monitoring transformer equipment faults have made some progress, they still have significant shortcomings and deficiencies. First, traditional monitoring methods mainly rely on periodic maintenance and manual inspections, which are insufficient to meet real-time monitoring needs and have limited ability to capture early fault characteristics. In addition, most existing technologies rely on single-parameter indicators and lack multi-dimensional information fusion analysis, resulting in inaccurate judgments on fault causes and risk levels. For example, monitoring oil temperature alone cannot distinguish the root cause of cooling system efficiency decline from winding overheating, nor can it accurately quantify complex faults such as localized overheating of the core and arc discharge at the terminals. At the same time, with the rapid growth of transformer operating data, traditional monitoring systems also face problems such as slow response, data loss, and overly simplified algorithm models in processing real-time data. Therefore, there is an urgent need for an internet-based method for monitoring electrical power equipment faults, transforming traditional manual inspections into intelligent and internet-based methods. By combining multi-sensor deployment, real-time data acquisition, and comprehensive fault assessment, this method provides strong technical support for the efficient operation of transformer equipment, while also extending the lifespan of transformer equipment and preventing sudden accidents and disasters. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for monitoring electrical equipment faults based on the Internet, thus solving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring electrical equipment faults based on the Internet, comprising the following steps;

[0006] S1. Based on the oil circulation cooling system equipped inside the transformer, the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system are monitored in real time, and several sets of oil load anomaly coefficients Xyc at monitoring time points are constructed.

[0007] S2. Based on the changes in transformer load current, the load current values ​​Ifz at each monitoring time point are constructed and correlated with the oil load anomaly coefficient Xyc at the corresponding monitoring time point. The influence of the load current value Ifz on the abnormal state of the oil load in the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf and issue a preliminary early warning signal.

[0008] S3. After receiving the preliminary warning signal, collect relevant heating characteristic information data of the winding in the active part of the transformer, and monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and construct relevant heat distribution information data and relevant arc state information data respectively.

[0009] S4. Correlate the relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization, and combined with the fault assessment model algorithm, fit and construct the fault comprehensive assessment index Zgz.

[0010] S5. Pre-set the evaluation threshold G and compare it with the comprehensive fault evaluation index Zgz to determine whether there is a fault risk in the current transformer and generate a corresponding level warning instruction.

[0011] Preferably, step S1 specifically includes:

[0012] S11. Based on the oil circulation cooling system equipped inside the transformer, multiple sets of oil temperature sensors are deployed at the top of the oil tank and the oil pipeline in the oil circulation cooling system. An online oil gas monitoring device is connected to the oil circulation cooling system and connected under high pressure and sealing. Both the oil temperature sensors and the online oil gas monitoring device are equipped with signal transmission units. Combined with wireless network communication technology, the multiple sets of oil temperature sensors and the online oil gas monitoring device are wirelessly connected to an Internet data platform. The Internet data platform is used to preprocess the received relevant data before storage.

[0013] S12. Using multiple sets of deployed oil temperature sensors and online oil gas monitoring devices, the changes in oil temperature and the concentration of combustible gas in the oil in the oil circulation cooling system are monitored in real time, relevant oil circulation status information data is constructed, and the relevant oil circulation status information data is sent to the Internet data platform. The relevant oil circulation status information data includes the oil temperature value Tyw and the combustible gas concentration value Cnd at each monitoring time point within the monitoring period.

[0014] S13. Preprocess the relevant oil cycle status information data received from the cloud computing data platform, including data denoising, outlier filtering, time series alignment and normalization, and store the preprocessed relevant oil cycle status information data in the Internet data platform.

[0015] Preferably, step S1 further includes:

[0016] S14. Extract features from the relevant oil circulation status information data stored in the Internet data platform to construct several sets of oil load anomaly coefficients Xyc at monitoring time points. The oil load anomaly coefficient Xyc at the i-th monitoring time point is used as the basis for this calculation. i For example, it is obtained in the following way:

[0017]

[0018] In the formula, Tyw i Let Cnd represent the oil temperature value at the i-th monitoring time point. i Let α1 and α2 represent the combustible gas concentration value at the i-th monitoring time point, and α1 and α2 represent the weight values.

[0019] Preferably, step S2 specifically includes:

[0020] S21. Based on the Hall current sensor deployed at the transformer load end, the load current change of the transformer is monitored in real time to obtain the load current value Ifz at each monitoring time point within the monitoring period. Based on the load current value Ifz at each monitoring time point within the monitoring period and the corresponding oil load anomaly coefficient Xyc, and combined with a statistical averaging algorithm, the influence of the load current value Ifz on the abnormal oil load state of the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf. The oil load anomaly correlation index Zyf is obtained using the following formula:

[0021]

[0022] In the formula, Ifz i Let Xyc represent the load current value at the i-th monitoring time point. i Let be the oil load anomaly coefficient at the i-th monitoring time point. This represents the average load current during the monitoring period. This represents the average value of the oil load anomaly coefficient during the monitoring period, where i = 1, 2, 3, ..., n, and n represents the number of monitoring time points.

[0023] Preferably, step S2 further includes:

[0024] S22. A correlation threshold L is preset, and the correlation threshold L is compared and analyzed with the oil load anomaly correlation index Zyf to determine whether the load current has an impact on the oil load anomaly state of the oil circulation cooling system. The specific content is as follows:

[0025] If the oil load anomaly correlation index Zyf > the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system.

[0026] If the oil load anomaly correlation index Zyf < the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system.

[0027] If the oil load anomaly correlation index Zyf = the correlation threshold L, it means that the current transformer load current has no impact on the oil load anomaly state of the oil circulation cooling system, and no additional warning command will be issued at this time.

[0028] Preferably, step S3 specifically includes:

[0029] S31. After receiving the initial warning signal, deploy several sets of sensors on the active part of the transformer. Among them, the several sets of sensors include embedded temperature sensors, Hall current sensors, high-frequency electromagnetic sensors, online resistance monitoring devices, multi-point thermocouple sensors and high-frequency voltage sensors.

[0030] S32. Based on several sets of deployed sensors, collect relevant heating characteristic information data of the windings in the active part of the current transformer. The relevant heating characteristic information data includes the winding temperature value Trz, the winding current value Irz, and the winding insulation resistance value Rrz. According to the current operation process of the transformer, monitor the heat distribution state on the surface of the iron core and construct relevant heat distribution information data. The relevant heat distribution information data includes the temperature value Ttx of each monitoring point on the surface of the iron core.

[0031] S33. Based on the degree of oil electrolysis caused by the arc reaction at the terminals in the active part of the current transformer, monitor the arc reaction status at the terminals and obtain relevant arc status information data, including arc power Pdh, voltage amplitude Vdh, and contact resistance value Rdh.

[0032] Preferably, step S4 specifically includes:

[0033] S41. Extract features from the acquired relevant thermal distribution information data and, using a statistical mean-calculation algorithm, obtain the average temperature of the iron core surface. This data is then correlated with the relevant heating characteristic information of the winding and the relevant heat distribution information of the iron core. After feature extraction and dimensionless processing, a thermal anomaly coefficient Xbj is obtained through fitting. The thermal anomaly coefficient Xbj is obtained by the following formula:

[0034]

[0035] In the formula, Ttx j Let represent the temperature value at the j-th point on the surface of the iron core, where j = 1, 2, 3, ..., m, m represents the number of monitoring points on the surface of the iron core, Trz represents the winding temperature value, Irz represents the winding current value, Rrz represents the winding insulation resistance value, and β1 and β2 both represent weight values.

[0036] S42. Analyze the relevant arc state information data, and after feature extraction, construct the arc response anomaly coefficient Xdh. The arc response anomaly coefficient Xdh is obtained by the following formula:

[0037]

[0038] In the formula, Pdh represents the arc power, Vdh represents the voltage amplitude, Rdh represents the contact resistance value, and log(1+Rdh) represents the nonlinear processing of the contact resistance value Rdh in the form of a logarithmic function.

[0039] Preferably, step S4 further includes:

[0040] S43. By inputting the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj, and the arc response anomaly coefficient Xdh into the fault assessment model, and after normalization, a comprehensive fault assessment index Zgz is fitted and constructed. The comprehensive fault assessment index Zgz is obtained by the following formula:

[0041]

[0042] In the formula, ω1, ω2 and ω3 represent the weight values ​​of the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj and the arc reaction anomaly coefficient Xdh, respectively, and A represents the first correction constant.

[0043] Preferably, the specific steps of S5 include:

[0044] S51. A pre-set evaluation threshold G is compared with the comprehensive fault evaluation index Zgz to determine whether the current transformer has a fault risk, and a corresponding level warning instruction is generated, the specific content of which is as follows:

[0045] If the comprehensive fault assessment index Zgz is greater than or equal to the assessment threshold G, i.e. Zgz≥G, it indicates that there is a fault risk in the current transformer and a first-level early warning instruction is issued. The content includes: notifying the equipment maintenance personnel that there is a fault risk in the current transformer, generating and sending a maintenance suggestion report, including a suggestion to immediately shut down the transformer, further inspect the abnormal components in the active part of the transformer on site, and replace the circulating oil in the oil circulation cooling system.

[0046] If the comprehensive fault assessment index Zgz is less than the assessment threshold G, i.e., Zgz < G, it indicates that there is no fault risk in the current transformer. A level two early warning instruction is obtained, which includes: notifying the equipment maintenance personnel that there is no fault risk in the current transformer, recording the relevant monitoring data collected at present, and continuing to monitor the current operating status of the transformer.

[0047] The system for monitoring electrical equipment faults based on the Internet includes a correlation analysis module, a preliminary early warning module, a data acquisition module, a fault assessment module, and a fault level determination module.

[0048] The correlation analysis module is used to monitor the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system equipped inside the transformer in real time, and to construct several sets of oil load anomaly coefficients Xyc at monitoring time points.

[0049] The preliminary early warning module is used to construct the load current value Ifz at each monitoring time point based on the change of transformer load current, correlate it with the oil load anomaly coefficient Xyc at the corresponding monitoring time point, analyze the degree of influence of the load current value Ifz on the abnormal state of oil load in the oil circulation cooling system, generate the oil load anomaly correlation index Zyf, and issue a preliminary early warning signal.

[0050] The data acquisition module is used to collect relevant heating characteristic information data of the winding in the active part of the transformer after receiving the preliminary warning signal, and to monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and to construct relevant heat distribution information data and relevant arc state information data respectively.

[0051] The fault assessment module is used to correlate relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization processing, and combined with the fault assessment model algorithm, a comprehensive fault assessment index Zgz is fitted and constructed.

[0052] The level determination module is used to pre-set the evaluation threshold G, compare it with the comprehensive fault evaluation index Zgz, in order to determine whether there is a fault risk in the current transformer, and generate a corresponding level warning instruction.

[0053] This invention provides a method and system for monitoring electrical equipment faults based on the Internet, which has the following beneficial effects:

[0054] (1) By leveraging the deep integration of sensor deployment, wireless communication, and cloud platform, comprehensive real-time monitoring of transformer equipment operation status is achieved to comprehensively assess transformer fault risk level. Through multi-dimensional information collection of oil circulation cooling system and transformer active part, potential abnormal characteristics during operation can be accurately captured and a dynamic fault comprehensive assessment index Zgz can be generated. This method improves the accuracy of fault risk identification and response speed through dynamic normalization processing, feature extraction, and multi-dimensional data correlation analysis. At the same time, through the classification and early warning of multi-level fault risk levels, detailed and targeted maintenance suggestions are provided, which can not only prevent small faults from expanding into major accidents, but also reduce unnecessary equipment maintenance frequency and cost. Especially in the long-term operation of transformers, this method effectively overcomes the lag and limitations of traditional manual inspection methods and realizes intelligent, precise, and dynamic equipment fault management. The large-scale application of this method can not only improve the overall operation stability and efficiency of the power grid, but also extend the service life of transformer equipment and effectively reduce the social and economic losses caused by equipment failure.

[0055] (2) By deploying multiple sets of sensors in key locations of the transformer's oil circulation cooling system and active part, the oil temperature and combustible gas concentration can be monitored in real time. The collected multidimensional data is wirelessly connected to the Internet data platform through the signal transmission unit. After preprocessing such as data denoising, outlier filtering, and time series alignment, the integrity and accuracy of the data are ensured. The oil load anomaly correlation index Zyf is constructed using the correlation analysis process to dynamically reveal the relationship between the abnormal state of the oil load in the oil circulation cooling system and the load current fluctuation. Through real-time monitoring and feature extraction, abnormal signals can be captured in the early stage of the fault and a preliminary warning signal can be generated. The preliminary warning signal includes identifying the changing trend of abnormal parameters, providing technical support for rapid response in transformer operation. Compared with the traditional method of relying on manual inspection, this method shortens the fault identification time and effectively solves the problem that the traditional method is difficult to capture early fault signals in time, thereby avoiding the evolution of small faults into large faults and providing a valuable time window for subsequent maintenance.

[0056] (3) The key operating parameters of the active part of the transformer are further monitored through the data acquisition module, including the heating characteristics of the winding, the heat distribution of the iron core and the arc reaction of the terminal. After these data are associated with the preliminary warning signal, the fault comprehensive assessment index Zgz is generated by dynamic normalization processing and fault assessment model algorithm. The fault risk is classified according to the comparison and analysis between the fault comprehensive assessment index Zgz and the pre-set assessment threshold G. Corresponding maintenance suggestions are generated for different levels, including equipment shutdown, cooling oil replacement, and local inspection of the winding and iron core. At the same time, it can combine multi-dimensional parameters to accurately quantify the impact of complex faults, including insulation aging caused by winding heating, heat accumulation risk caused by local overheating of the iron core, and flammable gas generation induced by arc reaction of the terminal. This method based on multi-dimensional information fusion and dynamic assessment improves the accuracy and timeliness of fault location, thereby helping maintenance personnel to formulate maintenance plans more efficiently, reduce the frequency of unnecessary shutdown maintenance, and effectively avoid equipment damage and large-scale power grid paralysis caused by sudden accidents. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the method for monitoring electrical equipment faults based on the Internet according to the present invention;

[0058] Figure 2 This is a system block diagram for monitoring electrical equipment faults based on the Internet, according to the present invention. Detailed Implementation

[0059] 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.

[0060] Example 1

[0061] Please see Figure 1 This invention provides a method for monitoring electrical equipment faults based on the Internet, comprising the following steps;

[0062] S1. Based on the oil circulation cooling system equipped inside the transformer, the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system are monitored in real time, and several sets of oil load anomaly coefficients Xyc at monitoring time points are constructed.

[0063] S2. Based on the changes in transformer load current, the load current values ​​Ifz at each monitoring time point are constructed and correlated with the oil load anomaly coefficient Xyc at the corresponding monitoring time point. The influence of the load current value Ifz on the abnormal state of the oil load in the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf and issue a preliminary early warning signal.

[0064] S3. After receiving the preliminary warning signal, collect relevant heating characteristic information data of the winding in the active part of the transformer, and monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and construct relevant heat distribution information data and relevant arc state information data respectively.

[0065] S4. Correlate the relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization, and combined with the fault assessment model algorithm, fit and construct the fault comprehensive assessment index Zgz.

[0066] S5. Pre-set the evaluation threshold G and compare it with the comprehensive fault evaluation index Zgz to determine whether there is a fault risk in the current transformer and generate a corresponding level warning instruction.

[0067] This embodiment proposes an internet-based method for monitoring electrical equipment faults, overcoming several shortcomings of traditional fault monitoring technologies. It achieves real-time, efficient, and multi-dimensional monitoring of transformer equipment. By collecting dynamic changes in oil temperature, gas concentration, and load current in the oil circulation cooling system, an oil load anomaly correlation index Zyf is constructed. Furthermore, winding heating characteristics, core heat distribution, and terminal arc status information are incorporated into the comprehensive analysis to generate a fault comprehensive assessment index Zgz. This method combines multi-dimensional information fusion analysis with intelligent fault diagnosis, enabling rapid identification of potential early fault characteristics and precise location of the root causes of complex faults, including winding overheating, partial core short circuits, and terminal arc anomalies. Through dynamic threshold assessment and multi-level early warning, it effectively distinguishes the severity of faults and provides targeted maintenance suggestions. This effectively solves the problems of relying on single parameters, difficulty in quantifying complex faults, and response lag in traditional technologies. This method improves the accuracy and timeliness of fault monitoring, avoiding transformer outages, large-scale power outages, and serious accidents caused by undetected hidden dangers, ensuring the stability and safety of the power system, and providing strong technical support for the intelligent operation and maintenance of modern power grids.

[0068] Example 2

[0069] Please refer to Figure 1 Specifically: S1 includes the following steps:

[0070] S11. Based on the oil circulation cooling system equipped inside the transformer, multiple sets of oil temperature sensors are deployed at the top of the oil tank and the oil pipeline in the oil circulation cooling system. An online oil gas monitoring device is connected to the oil circulation cooling system and connected under high pressure and sealing. Both the oil temperature sensors and the online oil gas monitoring device are equipped with signal transmission units. Combined with wireless network communication technology, the multiple sets of oil temperature sensors and the online oil gas monitoring device are wirelessly connected to an Internet data platform. The Internet data platform is used to preprocess the received relevant data before storage.

[0071] S12. Using multiple sets of deployed oil temperature sensors and online oil gas monitoring devices, the changes in oil temperature and the concentration of combustible gases in the oil circulation cooling system are monitored in real time to construct relevant oil circulation status information data, and the relevant oil circulation status information data is sent to the Internet data platform. The relevant oil circulation status information data includes the oil temperature value Tyw and the combustible gas concentration value Cnd at each monitoring time point within the monitoring period; wherein, transformer failure will cause the circulating oil in the oil circulation cooling system to heat up rapidly and generate combustible gases during the electrolysis process of arc discharge;

[0072] It should be noted that the oil temperature value Tyw represents the real-time temperature of the oil in the oil circulation cooling system and is an important parameter reflecting the working status of the oil circulation cooling system. Abnormal temperature usually indicates internal overheating problems. The combustible gas concentration value Cnd represents the concentration level of dissolved combustible gases in the oil and is an important indicator of abnormal arc reaction at the terminal. Electrolytic gases are generated due to arc discharge. The oil temperature value Tyw and the combustible gas concentration value Cnd are obtained in real time by an oil temperature sensor and an online gas monitoring device in the oil circulation cooling system. The online gas monitoring device in the oil is a device used to monitor the concentration of dissolved gases in the circulating oil in real time. It extracts a small amount of oil sample from the oil circulation system through a dedicated sampling pipeline and uses gas chromatography to separate and measure the composition and concentration of combustible gases.

[0073] S13. Preprocess the relevant oil cycle status information data received from the cloud computing data platform, including data denoising, outlier filtering, time series alignment and normalization, and store the preprocessed relevant oil cycle status information data in the Internet data platform.

[0074] Specifically, the S1 steps also include:

[0075] S14. Extract features from the relevant oil circulation status information data stored in the Internet data platform to construct several sets of oil load anomaly coefficients Xyc at monitoring time points. The oil load anomaly coefficient Xyc at the i-th monitoring time point is used as the basis for this calculation. i For example, it is obtained in the following way:

[0076]

[0077] In the formula, Tyw i Let Cnd represent the oil temperature value at the i-th monitoring time point. i Let α1 and α2 represent the combustible gas concentration value at the i-th monitoring time point, and α1 and α2 represent the weight values.

[0078] In this embodiment, multiple sets of oil temperature sensors and online oil gas monitoring devices are deployed, combined with wireless network communication technology, to collect real-time information on oil temperature changes and combustible gas concentrations. The data is then processed through an internet data platform for noise reduction, outlier filtering, time series alignment, and normalization to generate high-quality oil circulation status information. Furthermore, through feature extraction, an oil load anomaly coefficient Xyc, reflecting the oil temperature value Tyw and combustible gas concentration value Cnd at each monitoring time point, is constructed. This enables dynamic quantification and anomaly assessment of the transformer's oil circulation cooling system status. Compared to traditional methods relying on manual inspection and single-parameter monitoring, this method effectively solves the problem of... Traditional monitoring methods suffer from insufficient real-time performance, limited data processing capabilities, and difficulty in integrating multi-dimensional information. This new method can accurately capture oil temperature rises and abnormal gas concentrations in oil circulation cooling systems caused by winding overheating, localized core heating, and arcing at the terminals. It provides reliable data support for subsequent multi-level fault analysis. Its innovation lies in introducing dynamic correlation analysis between oil temperature and gas concentration into oil circulation cooling system monitoring, making the monitoring results more comprehensive and targeted. This not only improves the efficiency of identifying anomalies in the oil circulation cooling system but also reduces the risk of equipment downtime and operational disruptions due to undetected hazards, providing important protection for transformer operation safety and power grid stability.

[0079] Example 3

[0080] Please refer to Figure 1 Specifically: The specific steps of S2 include:

[0081] S21. Based on the Hall current sensor deployed at the transformer load end, the load current change of the transformer is monitored in real time to obtain the load current value Ifz at each monitoring time point within the monitoring period. Based on the load current value Ifz at each monitoring time point within the monitoring period and the corresponding oil load anomaly coefficient Xyc, and combined with a statistical averaging algorithm, the influence of the load current value Ifz on the abnormal oil load state of the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf. The oil load anomaly correlation index Zyf is obtained using the following formula:

[0082]

[0083] In the formula, Ifz i Let Xyc represent the load current value at the i-th monitoring time point. i Let be the oil load anomaly coefficient at the i-th monitoring time point. This represents the average load current during the monitoring period. This represents the average value of the oil load anomaly coefficient during the monitoring period, where i = 1, 2, 3, ..., n, and n represents the number of monitoring time points.

[0084] Specifically, the S2 steps also include:

[0085] S22. A correlation threshold L is preset, and the correlation threshold L is compared and analyzed with the oil load anomaly correlation index Zyf to determine whether the load current has an impact on the oil load anomaly state of the oil circulation cooling system. The specific content is as follows:

[0086] If the oil load anomaly correlation index Zyf > the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system.

[0087] If the oil load anomaly correlation index Zyf < the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system.

[0088] If the oil load abnormality correlation index Zyf = the correlation threshold L, it means that the current transformer load current has no impact on the oil load abnormality state of the oil circulation cooling system, and no additional warning command will be issued at this time.

[0089] The active part of the transformer refers to the internal components of the transformer, which mainly include the core, windings, and terminals.

[0090] In this embodiment, the implementation of step S2 effectively solves the problem of the difficulty in accurately analyzing the impact of load current on the oil circulation cooling system in traditional transformer fault monitoring. A Hall current sensor is used to monitor changes in transformer load current in real time. Combined with statistical algorithms, the load current value Ifz and the oil load anomaly coefficient Xyc are dynamically correlated to generate the oil load anomaly correlation index Zyf. This achieves a quantitative assessment of the relationship between load current and oil circulation anomaly. By setting a correlation threshold L to classify the oil load anomaly correlation index Zyf, the potential impact on oil cooling performance under different load conditions can be clearly identified, thereby accurately triggering a preliminary warning signal. This process not only quickly captures... This method improves monitoring efficiency for early-stage transformer anomalies and provides reliable diagnostic information under complex load current fluctuations. Compared to traditional methods that rely on a single parameter, this method overcomes the limitations of a single parameter through load-oil circulation coupling analysis, avoiding missed or false alarms of potential faults. Simultaneously, the system dynamically optimizes the initial warning signal based on the oil load anomaly correlation index Zyf, ensuring the timeliness and relevance of the warning. This provides a scientific basis for subsequent in-depth fault analysis of the core, windings, and active components at the terminals, improving the accuracy and reliability of transformer equipment fault monitoring and effectively addressing the dual bottlenecks of data analysis capability and response speed in traditional monitoring.

[0091] Example 4

[0092] Please refer to Figure 1 Specifically: The specific steps of S3 include:

[0093] S31. After receiving the initial warning signal, deploy several sets of sensors on the active part of the transformer. Among them, the several sets of sensors include embedded temperature sensors, Hall current sensors, high-frequency electromagnetic sensors, online resistance monitoring devices, multi-point thermocouple sensors and high-frequency voltage sensors.

[0094] S32. Based on several deployed sensors, relevant heating characteristic information data of the windings in the active part of the current transformer is collected. The relevant heating characteristic information data includes the winding temperature value Trz, the winding current value Irz, and the winding insulation resistance value Rrz. According to the current operation of the transformer, the heat distribution state on the surface of the iron core is monitored to construct relevant heat distribution information data. The relevant heat distribution information data includes the temperature value Ttx of each monitoring point on the surface of the iron core. The iron core and windings are important components of the transformer. The iron core is a magnetic circuit part composed of multiple layers of silicon steel sheets, which is used to concentrate and guide magnetic flux. When the transformer is working, the current in the input winding generates an alternating magnetic field. The iron core transfers electromagnetic energy to the output winding through the magnetic flux, thereby completing the voltage rise and fall conversion. The two work together to ensure the normal operation of the transformer. The windings are inductor coils wound with insulated copper wire, divided into high-voltage and low-voltage windings, which are used to realize the input and output of electrical energy through electromagnetic induction.

[0095] It should be noted that the winding temperature value Trz reflects the real-time temperature of the transformer winding during operation and is a key parameter for assessing whether the winding is within its safe operating range. Overheating of the winding can lead to insulation aging, decreased electrical performance, and even winding damage. This temperature is collected using an embedded temperature sensor, which is typically embedded directly inside the winding or installed around it to ensure real-time temperature changes are captured. The winding current value Irz represents the current intensity of the winding during operation, directly reflecting load conditions and current fluctuations. Excessive current can cause increased winding heating, potentially inducing overheating faults. This current is collected in real-time using a Hall effect current sensor. The sensor is installed in the current loop of the winding to accurately measure the current value. The winding insulation resistance value Rrz is an important indicator for evaluating the insulation performance of the winding. A decrease in insulation resistance is caused by physical aging and high temperature factors, which can easily lead to short circuits and partial discharge problems in the winding. It is obtained through an online resistance monitoring device, which uses high-precision resistance testing technology to detect changes in the resistance value of the winding insulation layer. The temperature value Ttx at each monitoring point on the iron core surface reflects the thermal distribution state of the iron core and is used to determine whether there are local overheating and short circuit problems in the iron core. Abnormal thermal distribution of the iron core will lead to decreased efficiency and structural damage. It is collected by arranging multi-point thermocouple sensors at multiple locations on the iron core surface.

[0096] S33. Based on the degree of oil electrolysis caused by the arc reaction at the terminals in the active part of the current transformer, monitor the arc reaction status at the terminals and obtain relevant arc status information data, including arc power Pdh, voltage amplitude Vdh, and contact resistance value Rdh.

[0097] It should be noted that the arc power Pdh reflects the energy released by the terminal during arc discharge and is an important parameter for judging the intensity of arc discharge and its impact on equipment. High arc power usually means strong discharge, which will aggravate the electrolytic reaction of oil and the deterioration of contact points, thereby inducing local overheating and insulation damage. Specifically, it is obtained through a high-frequency electromagnetic sensor, which can capture the high-frequency electromagnetic wave signal generated during arc discharge and calculate the arc power Pdh based on the signal strength and duration. The voltage amplitude Vdh represents the instantaneous voltage value of the terminal during arc discharge, reflecting the potential change characteristics of arc discharge. Abnormal voltage amplitude indicates poor contact and transient high voltage caused by discharge in the circuit. This is specifically collected by a high-frequency voltage sensor installed near the terminal to monitor voltage changes in real time and accurately capture the amplitude of transient voltage during discharge. The contact resistance value Rdh is an important indicator for evaluating the contact performance of the terminal. Increased contact resistance leads to an increase in arc discharge frequency and accelerates damage to the contact point and the electrolytic reaction of the oil. This is specifically obtained by an online resistance monitoring device, which monitors the resistance changes of the terminal contact point through current injection or resistance measurement technology.

[0098] In this embodiment, by implementing step S3, the fault monitoring capability of the active part of the transformer is improved, effectively overcoming the limitation of traditional monitoring technologies in detecting multi-dimensional fault characteristics. After receiving the initial warning signal, by deploying multiple sets of sensors, including embedded temperature sensors, Hall current sensors, high-frequency electromagnetic sensors, online resistance monitoring devices, multi-point thermocouple sensors, and high-frequency voltage sensors, refined monitoring of the operating status of the windings, core, and terminals is achieved. Specifically, the characteristic data of winding temperature value Trz, winding current value Irz, and winding insulation resistance value Rrz are collected, which not only reflects the local heating of the windings but also provides data support for the dynamic analysis of the winding heating degree. At the same time, by accurately acquiring the temperature value Ttx of each monitoring point on the core surface through multi-point thermocouple sensors, the relevant thermal analysis of the core can be constructed. By collecting information data, potential problems of localized overheating in the core are revealed. In addition, high-frequency electromagnetic sensors and online resistance monitoring devices are used to collect arc power Pdh, voltage amplitude Vdh, and contact resistance Rdh at the terminals, capturing the arc reaction state and oil electrolysis phenomenon at the terminals, thereby locating the root cause of poor contact and arc discharge. Compared with traditional methods that rely on single and low-frequency monitoring methods, the comprehensive collection and correlation analysis of multi-dimensional data can more comprehensively cover the operating status of the active part of the transformer, providing an efficient and accurate monitoring method for potential faults in windings, core, and terminals. This not only effectively solves the problems of insufficient data collection coverage and incomplete fault feature capture in traditional methods, but also provides a solid data foundation for subsequent comprehensive assessment and early warning, and lays the foundation for accurate location and scientific assessment of transformer fault risks.

[0099] Example 5

[0100] Please refer to Figure 1 Specifically: The specific steps of S4 include:

[0101] S41. Extract features from the acquired relevant thermal distribution information data and, using a statistical mean-calculation algorithm, obtain the average temperature of the iron core surface. This data is then correlated with the relevant heating characteristic information of the winding and the relevant heat distribution information of the iron core. After feature extraction and dimensionless processing, a thermal anomaly coefficient Xbj is obtained through fitting. The thermal anomaly coefficient Xbj is obtained by the following formula:

[0102]

[0103] In the formula, Ttx j Let represent the temperature value at the j-th point on the surface of the iron core, where j = 1, 2, 3, ..., m, m represents the number of monitoring points on the surface of the iron core, Trz represents the winding temperature value, Irz represents the winding current value, Rrz represents the winding insulation resistance value, and β1 and β2 both represent weight values.

[0104] S42. Analyze the relevant arc state information data, and after feature extraction, construct the arc response anomaly coefficient Xdh. The arc response anomaly coefficient Xdh is obtained by the following formula:

[0105]

[0106] In the formula, Pdh represents the arc power, Vdh represents the voltage amplitude, Rdh represents the contact resistance value, and log(1+Rdh) represents the nonlinear processing of the contact resistance value Rdh in the form of a logarithmic function.

[0107] The specific steps in S4 also include:

[0108] S43. By inputting the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj, and the arc response anomaly coefficient Xdh into the fault assessment model, and after normalization, a comprehensive fault assessment index Zgz is fitted and constructed. The comprehensive fault assessment index Zgz is obtained by the following formula:

[0109]

[0110] In the formula, ω1, ω2 and ω3 represent the weight values ​​of the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj and the arc reaction anomaly coefficient Xdh, respectively, and A represents the first correction constant.

[0111] In this embodiment, by implementing step S4, a scientific and efficient fault assessment system is constructed based on the comprehensive analysis of multi-dimensional data, effectively compensating for the shortcomings of traditional monitoring technologies in quantifying and comprehensively assessing fault causes. First, by extracting features from the relevant thermal distribution information data of the iron core and the relevant heating characteristic information data of the windings, and combining statistical mean algorithms and dimensionless processing, the thermal anomaly coefficient Xbj is accurately calculated. This dynamically quantifies the coupling effect between local overheating of the iron core and heating of the windings, providing a scientific basis for identifying internal thermal anomalies in the transformer. Simultaneously, by extracting features and performing nonlinear processing on the relevant arc state information data of the terminals, an arc response anomaly coefficient Xdh is constructed. This arc response anomaly coefficient Xdh incorporates the arc power Pdh, voltage amplitude Vdh, and contact resistance value Rdh. The correlation revealed the potential impact of terminal arc reaction on equipment operation. Finally, the oil load anomaly correlation index Zyf, thermal anomaly coefficient Xbj, and arc reaction anomaly coefficient Xdh were input into the fault assessment model. After normalization, a comprehensive fault assessment index Zgz was generated. The comprehensive fault assessment index Zgz is based on multi-parameter fusion, which effectively overcomes the one-sidedness of single-parameter judgment in traditional methods and can dynamically reflect the overall fault risk of transformers. This assessment method based on multi-dimensional comprehensive analysis not only improves the accuracy of fault diagnosis, but also lays the foundation for early identification and rapid response to complex faults. It effectively solves the problems of insufficient data correlation, single assessment model, and inaccurate fault location in traditional monitoring methods, and provides new technical support for the efficient operation and safe management of transformers.

[0112] Example 6

[0113] Please refer to Figure 1 Specifically: The S5 steps include:

[0114] S51. A pre-set evaluation threshold G is compared with the comprehensive fault evaluation index Zgz to determine whether the current transformer has a fault risk, and a corresponding level warning instruction is generated, the specific content of which is as follows:

[0115] If the comprehensive fault assessment index Zgz is greater than or equal to the assessment threshold G, i.e. Zgz≥G, it indicates that there is a fault risk in the current transformer and a first-level early warning instruction is issued. The content includes: notifying the equipment maintenance personnel that there is a fault risk in the current transformer, generating and sending a maintenance suggestion report, including a suggestion to immediately shut down the transformer, further inspect the abnormal components in the active part of the transformer on site, and replace the circulating oil in the oil circulation cooling system.

[0116] If the comprehensive fault assessment index Zgz is less than the assessment threshold G, i.e., Zgz < G, it indicates that there is no fault risk in the current transformer. A level two early warning instruction is obtained, which includes: notifying the equipment maintenance personnel that there is no fault risk in the current transformer, recording the relevant monitoring data collected at present, and continuing to monitor the current operating status of the transformer.

[0117] In this embodiment, by implementing step S5, the intelligent early warning mechanism for transformer fault monitoring is further improved, enhancing the accuracy and timeliness of fault risk management. By dynamically comparing the comprehensive fault assessment index Zgz with the preset assessment threshold G, the current fault risk status of the transformer can be quickly determined, and corresponding level early warning instructions can be generated. When the comprehensive fault assessment index Zgz ≥ the assessment threshold G, a level one early warning instruction is issued, immediately notifying maintenance personnel to respond and providing a detailed maintenance suggestion report, including shutting down the transformer, focusing on inspecting active components such as windings, core, and terminals, and specific measures for replacing circulating oil to ensure that the fault does not worsen. When the comprehensive fault assessment index Zgz < the assessment threshold G, the system provides a level one early warning. When the value is G, a level-two early warning command is issued, indicating that there is no obvious risk to the equipment at present, and the current data is recorded in detail. It is recommended to continuously monitor and optimize the operating status. Unlike the traditional fault judgment method that relies on a single parameter, the overall health status of the transformer is dynamically quantified by the fault comprehensive assessment index Zgz, avoiding false alarms and missed alarms caused by unreasonable single threshold settings. At the same time, by classifying the fault risk level and providing targeted maintenance suggestions, a seamless connection between early warning and maintenance optimization is achieved, which effectively reduces the risk of equipment damage caused by sudden faults, provides a solid guarantee for the safe operation of the transformer, improves the stability of the power grid operation, and effectively avoids economic losses and social impacts caused by the spread of faults.

[0118] Example 7

[0119] Please refer to Figure 1 and Figure 2 Specifically: a system for monitoring electrical equipment faults based on the Internet, including a correlation analysis module, a preliminary early warning module, a data acquisition module, a fault assessment module, and a fault level determination module;

[0120] The correlation analysis module is used to monitor the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system equipped inside the transformer in real time, and to construct several sets of oil load anomaly coefficients Xyc at monitoring time points.

[0121] The preliminary early warning module is used to construct the load current value Ifz at each monitoring time point based on the change of transformer load current, correlate it with the oil load anomaly coefficient Xyc at the corresponding monitoring time point, analyze the degree of influence of the load current value Ifz on the abnormal state of oil load in the oil circulation cooling system, generate the oil load anomaly correlation index Zyf, and issue a preliminary early warning signal.

[0122] The data acquisition module is used to collect relevant heating characteristic information data of the winding in the active part of the transformer after receiving the preliminary warning signal, and to monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and to construct relevant heat distribution information data and relevant arc state information data respectively.

[0123] The fault assessment module is used to correlate relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization processing, and combined with the fault assessment model algorithm, a comprehensive fault assessment index Zgz is fitted and constructed.

[0124] The level determination module is used to pre-set the evaluation threshold G, compare it with the comprehensive fault evaluation index Zgz, in order to determine whether there is a fault risk in the current transformer, and generate a corresponding level warning instruction.

[0125] In this embodiment, a modular design effectively solves the problems of poor real-time performance, limited data analysis, and difficulty in quantifying complex faults in traditional transformer fault monitoring. A correlation analysis module monitors the changes in oil temperature and gas concentration in the transformer oil circulation cooling system in real time, constructing an oil load anomaly coefficient Xyc to dynamically reveal the operating status of the cooling system. A preliminary early warning module correlates the load current Ifz with the oil load anomaly coefficient Xyc to generate an oil load anomaly correlation index Zyf, achieving a comprehensive analysis of load and oil cooling performance, triggering a preliminary early warning signal, and ensuring rapid detection of early faults. A data acquisition module uses multi-sensor technology to collect multi-dimensional information from the windings, core, and terminals, covering key parameters such as temperature, current, and arc status, providing more comprehensive data support for subsequent evaluation. The fault assessment module uses normalization processing and model fitting... This system integrates multi-dimensional information into a comprehensive fault assessment index Zgz, enabling precise quantification of fault states. The fault level determination module compares the comprehensive fault assessment index Zgz with a preset assessment threshold G, meticulously classifying warning commands and providing scientific maintenance recommendations. This achieves closed-loop management of the entire process from data acquisition to fault assessment. It dynamically correlates the state of the oil circulation cooling system with the active components of the transformer, effectively solving the problems of isolated data, delayed assessment, and difficulty in quantifying complex faults in traditional monitoring technologies. Through modular division of labor and multi-dimensional data integration, it can quickly and accurately locate faults and provide detailed maintenance recommendations, improving the operational reliability and lifespan of transformers. It effectively avoids large-scale power outages, equipment damage, and economic losses caused by fault expansion, providing technical support for intelligent monitoring and efficient operation and maintenance of power systems.

[0126] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring electrical equipment faults based on the Internet, characterized in that: Includes the following steps; S1. Based on the oil circulation cooling system equipped inside the transformer, the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system are monitored in real time, and several sets of oil load anomaly coefficients Xyc at monitoring time points are constructed. S2. Based on the changes in transformer load current, the load current values ​​Ifz at each monitoring time point are constructed and correlated with the oil load anomaly coefficient Xyc at the corresponding monitoring time point. The influence of the load current value Ifz on the abnormal state of the oil load in the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf and issue a preliminary early warning signal. S3. After receiving the preliminary warning signal, collect relevant heating characteristic information data of the winding in the active part of the transformer, and monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and construct relevant heat distribution information data and relevant arc state information data respectively. S4. Correlate the relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization, and combined with the fault assessment model algorithm, fit and construct the fault comprehensive assessment index Zgz. S5. Pre-set the evaluation threshold G and compare it with the comprehensive fault evaluation index Zgz to determine whether there is a fault risk in the current transformer and generate a corresponding level warning instruction.

2. The method for monitoring electrical equipment faults based on the Internet according to claim 1, characterized in that: The specific steps in S1 include: S11. Based on the oil circulation cooling system equipped inside the transformer, multiple sets of oil temperature sensors are deployed at the top of the oil tank and the oil pipeline in the oil circulation cooling system. An online oil gas monitoring device is connected to the oil circulation cooling system and connected under high pressure and sealing. Both the oil temperature sensors and the online oil gas monitoring device are equipped with signal transmission units. Combined with wireless network communication technology, the multiple sets of oil temperature sensors and the online oil gas monitoring device are wirelessly connected to an Internet data platform. The Internet data platform is used to preprocess the received relevant data before storage. S12. Using multiple sets of deployed oil temperature sensors and online oil gas monitoring devices, the changes in oil temperature and the concentration of combustible gas in the oil in the oil circulation cooling system are monitored in real time, relevant oil circulation status information data is constructed, and the relevant oil circulation status information data is sent to the Internet data platform. The relevant oil circulation status information data includes the oil temperature value Tyw and the combustible gas concentration value Cnd at each monitoring time point within the monitoring period. S13. Preprocess the relevant oil cycle status information data received from the cloud computing data platform, including data denoising, outlier filtering, time series alignment and normalization, and store the preprocessed relevant oil cycle status information data in the Internet data platform.

3. The method for monitoring electrical equipment faults based on the Internet according to claim 2, characterized in that: The specific steps in S1 also include: S14. Extract features from the relevant oil circulation status information data stored in the Internet data platform to construct several sets of oil load anomaly coefficients Xyc at monitoring time points. The oil load anomaly coefficient Xyc at the i-th monitoring time point is used as the basis for this calculation. i For example, it is obtained in the following way: In the formula, Tyw i Let Cnd represent the oil temperature value at the i-th monitoring time point. i Let α1 and α2 represent the combustible gas concentration value at the i-th monitoring time point, and α1 and α2 represent the weight values.

4. The method for monitoring electrical equipment faults based on the Internet according to claim 3, characterized in that: The specific steps in S2 include: S21. Based on the Hall current sensor deployed at the transformer load end, the load current change of the transformer is monitored in real time to obtain the load current value Ifz at each monitoring time point within the monitoring period. Based on the load current value Ifz at each monitoring time point within the monitoring period and the corresponding oil load anomaly coefficient Xyc, and combined with a statistical averaging algorithm, the influence of the load current value Ifz on the abnormal oil load state of the oil circulation cooling system is analyzed to generate the oil load anomaly correlation index Zyf. The oil load anomaly correlation index Zyf is obtained using the following formula: In the formula, Ifz i Let Xyc represent the load current value at the i-th monitoring time point. i Let be the oil load anomaly coefficient at the i-th monitoring time point. This represents the average load current during the monitoring period. This represents the average value of the oil load anomaly coefficient during the monitoring period, where i = 1, 2, 3, ..., n, and n represents the number of monitoring time points.

5. The method for monitoring electrical equipment faults based on the Internet according to claim 4, characterized in that: The specific steps in S2 also include: S22. A correlation threshold L is preset, and the correlation threshold L is compared and analyzed with the oil load anomaly correlation index Zyf to determine whether the load current has an impact on the oil load anomaly state of the oil circulation cooling system. The specific content is as follows: If the oil load anomaly correlation index Zyf > the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system. If the oil load anomaly correlation index Zyf < the correlation threshold L, it indicates that the current transformer load current has an impact on the oil load anomaly state of the oil circulation cooling system, and a preliminary warning signal is issued. At this time, further analysis will be carried out on the active part of the transformer covered by the oil circulation cooling system. If the oil load anomaly correlation index Zyf = the correlation threshold L, it means that the current transformer load current has no impact on the oil load anomaly state of the oil circulation cooling system, and no additional warning command will be issued at this time.

6. The method for monitoring electrical equipment faults based on the Internet according to claim 5, characterized in that: The specific steps of S3 include: S31. After receiving the initial warning signal, deploy several sets of sensors on the active part of the transformer. Among them, the several sets of sensors include embedded temperature sensors, Hall current sensors, high-frequency electromagnetic sensors, online resistance monitoring devices, multi-point thermocouple sensors and high-frequency voltage sensors. S32. Based on several sets of deployed sensors, collect relevant heating characteristic information data of the windings in the active part of the current transformer. The relevant heating characteristic information data includes the winding temperature value Trz, the winding current value Irz, and the winding insulation resistance value Rrz. According to the current operation process of the transformer, monitor the heat distribution state on the surface of the iron core and construct relevant heat distribution information data. The relevant heat distribution information data includes the temperature value Ttx of each monitoring point on the surface of the iron core. S33. Based on the degree of oil electrolysis caused by the arc reaction at the terminals in the active part of the current transformer, monitor the arc reaction status at the terminals and obtain relevant arc status information data, including arc power Pdh, voltage amplitude Vdh, and contact resistance value Rdh.

7. The method for monitoring electrical equipment faults based on the Internet according to claim 6, characterized in that: The specific steps of S4 include: S41. Extract features from the acquired relevant thermal distribution information data and, using a statistical mean-calculation algorithm, obtain the average temperature of the iron core surface. This data is then correlated with the relevant heating characteristic information of the winding and the relevant heat distribution information of the iron core. After feature extraction and dimensionless processing, a thermal anomaly coefficient Xbj is obtained through fitting. The thermal anomaly coefficient Xbj is obtained by the following formula: In the formula, Ttx j Let represent the temperature value at the j-th point on the surface of the iron core, where j = 1, 2, 3, ..., m, m represents the number of monitoring points on the surface of the iron core, Trz represents the winding temperature value, Irz represents the winding current value, Rrz represents the winding insulation resistance value, and β1 and β2 both represent weight values. S42. Analyze the relevant arc state information data, and after feature extraction, construct the arc response anomaly coefficient Xdh. The arc response anomaly coefficient Xdh is obtained by the following formula: In the formula, Pdh represents the arc power, Vdh represents the voltage amplitude, Rdh represents the contact resistance value, and log(1+Rdh) represents the nonlinear processing of the contact resistance value Rdh in the form of a logarithmic function.

8. The method for monitoring electrical equipment faults based on the Internet according to claim 7, characterized in that: The specific steps in S4 also include: S43. By inputting the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj, and the arc response anomaly coefficient Xdh into the fault assessment model, and after normalization, a comprehensive fault assessment index Zgz is fitted and constructed. The comprehensive fault assessment index Zgz is obtained by the following formula: In the formula, ω1, ω2 and ω3 represent the weight values ​​of the oil load anomaly correlation index Zyf, the thermal anomaly coefficient Xbj and the arc reaction anomaly coefficient Xdh, respectively, and A represents the first correction constant.

9. The method for monitoring electrical equipment faults based on the Internet according to claim 8, characterized in that: The specific steps of S5 include: S51. A pre-set evaluation threshold G is compared with the comprehensive fault evaluation index Zgz to determine whether the current transformer has a fault risk, and a corresponding level warning instruction is generated, the specific content of which is as follows: If the comprehensive fault assessment index Zgz is greater than or equal to the assessment threshold G, i.e. Zgz≥G, it indicates that there is a fault risk in the current transformer and a first-level early warning instruction is issued. The content includes: notifying the equipment maintenance personnel that there is a fault risk in the current transformer, generating and sending a maintenance suggestion report, including a suggestion to immediately shut down the transformer, further inspect the abnormal components in the active part of the transformer on site, and replace the circulating oil in the oil circulation cooling system. If the comprehensive fault assessment index Zgz is less than the assessment threshold G, i.e., Zgz < G, it indicates that there is no fault risk in the current transformer. A level two early warning instruction is obtained, which includes: notifying the equipment maintenance personnel that there is no fault risk in the current transformer, recording the relevant monitoring data collected at present, and continuing to monitor the current operating status of the transformer.

10. A system for monitoring electrical equipment faults via the Internet, used to implement the method for monitoring electrical equipment faults via the Internet as described in any one of claims 1 to 9, characterized in that: It includes a correlation analysis module, a preliminary early warning module, a data acquisition module, a fault assessment module, and a fault determination module; The correlation analysis module is used to monitor the oil temperature change and the concentration of combustible gas in the oil in the oil circulation cooling system equipped inside the transformer in real time, and to construct several sets of oil load anomaly coefficients Xyc at monitoring time points. The preliminary early warning module is used to construct the load current value Ifz at each monitoring time point based on the change of transformer load current, correlate it with the oil load anomaly coefficient Xyc at the corresponding monitoring time point, analyze the degree of influence of the load current value Ifz on the abnormal state of oil load in the oil circulation cooling system, generate the oil load anomaly correlation index Zyf, and issue a preliminary early warning signal. The data acquisition module is used to collect relevant heating characteristic information data of the winding in the active part of the transformer after receiving the preliminary warning signal, and to monitor the heat distribution state on the iron core surface and the arc reaction state at the terminal according to the current operation process of the transformer, and to construct relevant heat distribution information data and relevant arc state information data respectively. The fault assessment module is used to correlate relevant heating characteristic information data, relevant heat distribution information data, and relevant arc state information data with the oil load abnormality correlation index Zyf. After normalization processing, and combined with the fault assessment model algorithm, a comprehensive fault assessment index Zgz is fitted and constructed. The level determination module is used to pre-set the evaluation threshold G, compare it with the comprehensive fault evaluation index Zgz, in order to determine whether there is a fault risk in the current transformer, and generate a corresponding level warning instruction.

Images