Oil-immersed transformer fault early warning method and system

Abstract

The application provides a kind of oil-immersed transformer fault early warning method and system, it is related to the technical field of oil-immersed transformer, comprising: obtaining various operating data generated in the continuous operation process of oil-immersed transformer;Various operating data are divided according to the preset operation cycle, and the periodic characteristic parameters of various operating data in each operation cycle are extracted;According to the periodic characteristic parameters of various operating data in the same operation cycle, the cooperative change relationship between various state characteristics is analyzed to generate a comprehensive influence result;Evolution analysis is carried out on the comprehensive influence result corresponding to the continuous multiple operation cycles to determine whether the comprehensive influence result meets the preset risk evolution condition;If it is satisfied, it is determined that the oil-immersed transformer has a fault risk, and the fault early warning information is output. Through the above scheme, the problem that the existing early warning method appears false alarm or miss report, and it is difficult to reflect the overall operation risk level of transformer in time and accurately is overcome.

Description

Technical Field

[0001] This application relates to the technical field of oil-immersed transformers, and in particular to a fault early warning method and system for oil-immersed transformers. Background Technology

[0002] As a key piece of equipment in the power system, the operating status of oil-immersed transformers directly affects the safety and stability of the power grid. Currently, transformers continuously generate various types of operating data during operation, including temperature, electrical quantities, load, and oil parameters. The industry typically monitors this data using threshold judgments, single-parameter trend analysis, or empirical models to achieve fault early warning.

[0003] However, these methods often rely on changes in a single operating parameter or isolated moments as the basis for judgment, lacking a systematic analysis of the interrelationships between different operating data within the same operating cycle. They also struggle to characterize the evolution of operating states across multiple operating cycles, resulting in limited ability to identify early, progressive faults. Under fluctuating operating conditions or the combined effects of multiple factors, existing early warning methods may produce false alarms or misses, failing to reflect the overall operational risk level of the transformer in a timely and accurate manner.

[0004] Therefore, how to integrate the synergistic changes among multiple types of operational data during continuous operation and realize reliable fault risk early warning based on their evolution over time remains an urgent problem to be solved in existing technologies. Summary of the Invention

[0005] The technical problem to be solved by the present invention is that in the case of fluctuations in operating conditions or the combined effect of multiple factors, the existing early warning methods may produce false alarms or omissions, making it difficult to reflect the overall operating risk level of the transformer in a timely and accurate manner. Therefore, the present invention provides a fault early warning method and system for oil-immersed transformers.

[0006] In view of this, a first aspect of the present invention provides a fault early warning method for an oil-immersed transformer, comprising: acquiring various types of operating data generated during the continuous operation of the oil-immersed transformer; dividing the various types of operating data according to a preset operating cycle, and extracting periodic characteristic parameters of various types of operating data within each operating cycle; analyzing the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle, so as to generate a comprehensive impact result; performing evolutionary analysis on the comprehensive impact result corresponding to multiple consecutive operating cycles, so as to determine whether the comprehensive impact result meets a preset risk evolution condition; if it meets the condition, determining that the oil-immersed transformer has a fault risk, and outputting fault early warning information.

[0007] Preferably, the various types of operating data include at least two of the following: temperature data, oil status data, and load-related data.

[0008] Preferably, the step of dividing various types of operating data according to a preset operating cycle and extracting the periodic feature parameters of various types of operating data in each operating cycle includes: dividing various types of operating data into operating cycles according to a preset operating cycle to obtain several operating parameters after cycle division; determining the periodic benchmark corresponding to the same type of operating data in each operating cycle; calculating the offset of the operating parameters relative to the periodic benchmark in each operating cycle; and generating periodic feature parameters based on the offset.

[0009] Preferably, the step of analyzing the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle to generate a comprehensive impact result includes: performing correlation analysis on different categories of periodic characteristic parameters within the same operating cycle, and determining, based on the analysis results, whether there is a synergistic change relationship of synchronous change, superimposed change, or mutual reinforcement among the periodic characteristic parameters within the operating cycle; determining the synergistic change type corresponding to each periodic characteristic parameter within the operating cycle based on the synergistic change relationship, and calculating the coupling strength corresponding to each synergistic change type; and comprehensively characterizing the degree of influence of the synergistic change type and its corresponding coupling strength on each category of periodic characteristic parameters to generate a comprehensive impact result.

[0010] Preferably, the step of performing correlation analysis on different categories of periodic characteristic parameters within the same operating cycle, and determining whether there is a synchronous change, superimposed change, or mutually reinforcing synergistic change relationship among the periodic characteristic parameters within the operating cycle based on the analysis results, includes: determining whether there is a synchronous change relationship based on the consistency of the change direction and the correlation of the change magnitude of various periodic characteristic parameters within the operating cycle; if there is no synchronous change relationship, determining that there is no synergistic change relationship among the periodic characteristic parameters, and terminating the determination of synergistic change relationship within the current operating cycle; if there is a synchronous change relationship, determining whether a superimposed change relationship is formed based on the superposition effect of the corresponding offsets of various periodic characteristic parameters; if a superimposed change relationship is formed, determining the synergistic change type corresponding to the periodic characteristic parameter as a superimposed change relationship; if no superimposed change relationship is formed, determining the synergistic change type corresponding to the periodic characteristic parameter as a synchronous change relationship; when the superimposed change relationship causes the comprehensive offset of the corresponding periodic characteristic parameter to exceed the influence range of a single characteristic change, determining that there is a mutually reinforcing synergistic change relationship among various periodic characteristic parameters.

[0011] Preferably, the step of performing evolutionary analysis on the comprehensive impact results corresponding to multiple consecutive operating cycles to determine whether the comprehensive impact results meet the preset risk evolution conditions includes: serializing each comprehensive impact result according to the operating cycle to construct an evolutionary sequence, extracting evolutionary feature parameters based on the evolutionary sequence, and generating a risk evolution quantity using the evolutionary feature parameters; comparing and analyzing the risk evolution quantity with the preset risk evolution conditions; determining that the preset risk evolution conditions are met when the risk evolution quantity meets the preset trend condition or the cumulative change condition; and determining that the preset risk evolution conditions are not met when the risk evolution quantity does not meet the preset trend condition or the cumulative change condition.

[0012] Preferably, the steps of serializing each of the comprehensive impact results according to the operating cycle to construct an evolutionary sequence, extracting evolutionary feature parameters based on the evolutionary sequence, and generating a risk evolution quantity using the evolutionary feature parameters include: aligning each comprehensive impact result in time according to the time sequence of the operating cycle to generate a periodic state sequence, and smoothing the periodic state sequence to obtain an evolutionary sequence; extracting at least one evolutionary feature parameter from the evolutionary sequence to characterize the changing trend, rate of change, and degree of cumulative change of the comprehensive impact results; and quantifying the overall change characteristics of the evolutionary sequence based on the evolutionary feature parameter to generate a risk evolution quantity.

[0013] A second aspect of this invention provides a fault early warning system for an oil-immersed transformer, comprising: a data acquisition module for acquiring various types of operating data generated during the continuous operation of the oil-immersed transformer; a data division module for dividing various types of operating data according to a preset operating cycle and extracting periodic characteristic parameters of various types of operating data within each operating cycle; a data analysis module for analyzing the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle, so as to generate a comprehensive impact result; a data judgment module for performing evolution analysis on the comprehensive impact result corresponding to multiple consecutive operating cycles, so as to determine whether the comprehensive impact result meets a preset risk evolution condition; and a fault early warning module for determining that the oil-immersed transformer has a fault risk if the condition is met, and outputting fault early warning information.

[0014] A third aspect of the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the computer program to implement the oil-immersed transformer fault early warning method described above.

[0015] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, the computer program causing a processor, when executed by a processor, to perform the oil-immersed transformer fault early warning method as described in any of the above embodiments.

[0016] The technical solution of this invention has the following advantages: accurate identification and timely early warning. By triggering fault early warning output when the risk evolution conditions are met, the oil-immersed transformer is identified from the potential risk stage in advance, enabling maintenance personnel to intervene in abnormal operating conditions before obvious faults occur. This improves the operational safety and fault early warning for oil-immersed transformers, overcoming the problem that existing early warning methods often produce false alarms or omissions and fail to reflect the overall operational risk level of the transformer in a timely and accurate manner when operating conditions fluctuate or multiple factors are at play. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the fault early warning method for oil-immersed transformers provided in an embodiment of the present invention; Figure 2 A schematic block diagram of the structure of the oil-immersed transformer fault early warning system provided in an embodiment of the present invention; Figure 3 A schematic block diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0019] Figure label: 10. Oil-immersed transformer fault early warning system; 11. Data acquisition module; 12. Data partitioning module; 13. Data analysis module; 14. Data judgment module; 15. Fault early warning module; 20. Electronic equipment; 21. Memory; 22. Processor. Detailed Implementation

[0020] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0021] In the description of this invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0022] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] Example 1: This application provides an example of a fault early warning method for oil-immersed transformers. By performing periodic processing, collaborative change analysis, and cross-period evolution analysis on various operating data generated during the continuous operation of the oil-immersed transformer, it is possible to achieve early identification and stability warning of operating state risks of the oil-immersed transformer. Specifically, this fault early warning method for oil-immersed transformers includes: S1 acquires various operating data generated during the continuous operation of the oil-immersed transformer.

[0024] By uniformly collecting monitoring data generated during the operation of oil-immersed transformers, multi-source operational data reflecting the thermal state, dielectric state, and load state of the oil-immersed transformers are synchronized and aligned in time to ensure the comparability of different types of operational data in the time dimension. The acquisition process is completed based on a fixed sampling period, and missing values ​​and abnormal abrupt values ​​are removed or corrected to form a continuous and stable operational data sequence.

[0025] The various types of operating data include at least two of the following: temperature data, oil condition data, and load-related data. Temperature data includes at least one of winding temperature and top oil temperature; oil condition data includes at least one of dissolved gas content and water content in the oil; load-related data includes at least one of load current and load rate, and all types of operating data are recorded according to a unified timestamp.

[0026] S2 divides various types of operational data according to preset operational cycles and extracts the periodic characteristic parameters of various types of operational data within each operational cycle.

[0027] Based on the operating characteristics of oil-immersed transformers, a fixed operating cycle is set, and continuous operating data is periodically segmented so that the data in each operating cycle can independently reflect the operating status characteristics within that time period.

[0028] If the continuous running data is divided into 24-hour periods, multiple independent running period data segments can be formed.

[0029] In another example, step S2 can preferably be performed as follows: By dividing various types of operational data into operational periods through a preset operational cycle, operational parameters after several period divisions are obtained.

[0030] The various types of operational data are segmented according to the operational cycle, and the corresponding set of operational parameters is extracted in each operational cycle. The set of operational parameters is used to characterize the overall change level of the operational data of that category in that operational cycle.

[0031] For example, the average value, maximum value, and variation of the winding temperature within the same operating cycle can be extracted to form the operating parameters of the winding temperature within that operating cycle.

[0032] Determine the cycle benchmark corresponding to the same category of operational data in each operational cycle, calculate the offset of the operational parameters relative to the cycle benchmark in each operational cycle, and generate cycle feature parameters based on the offset to characterize the state change characteristics of the corresponding category of operational data in that operational cycle.

[0033] The statistical results of the corresponding category of operating data within the historical stable operating cycle are used as the cycle benchmark. By calculating the difference or proportional offset of the operating parameters in the current operating cycle relative to the cycle benchmark, standardized cycle characteristic parameters are formed to eliminate the impact of the difference in the units of different categories of operating data.

[0034] The average winding temperature under historical stable operating conditions is used as the periodic reference. The deviation ratio of the average winding temperature in the current operating cycle relative to the periodic reference is calculated and used as the periodic characteristic parameter of temperature data in the current operating cycle.

[0035] S3 analyzes the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle, in order to generate comprehensive impact results.

[0036] By taking the periodic characteristic parameters corresponding to different categories of operating data within the same operating cycle as the analysis object, and by jointly analyzing the direction of change, magnitude of change and interrelationship of each periodic characteristic parameter within the operating cycle, we can identify whether different state characteristics exhibit intrinsically related change behaviors within the same operating cycle, thereby determining whether there is a synergistic change relationship between various state characteristics, and forming a comprehensive impact result that reflects the overall change characteristics of the operating state of the oil-immersed transformer within the operating cycle.

[0037] Within the same operating cycle, the periodic characteristic parameters corresponding to temperature data, oil condition data, and load-related data are uniformly organized. Based on the changes of each periodic characteristic parameter relative to its respective periodic baseline within the operating cycle, it is determined whether there are multiple state characteristics that deviate from the normal operating state simultaneously.

[0038] In another example, step S3 can preferably be performed as follows: Correlation analysis is performed on different categories of periodic characteristic parameters within the same operating cycle, and the results are used to determine whether there are synchronous changes, superimposed changes, or mutually reinforcing synergistic changes in each periodic characteristic parameter within the operating cycle.

[0039] For different types of periodic characteristic parameters obtained within the same operating cycle, a set of periodic characteristic parameters is constructed with the operating cycle as the unit. The consistency of change direction, correlation of change magnitude, and offset superposition characteristics of each periodic characteristic parameter in the set are analyzed item by item to distinguish the change relationship types between each periodic characteristic parameter.

[0040] For example, within a certain operating cycle, when the periodic characteristic parameters of temperature data show an increasing trend, the periodic characteristic parameters of oil condition data show a deteriorating trend, and the periodic characteristic parameters of load-related data also show an increasing trend, the relationship between the changes of the three types of periodic characteristic parameters is jointly judged to identify whether there is a synergistic change relationship.

[0041] Furthermore, the step of performing correlation analysis on different categories of periodic characteristic parameters within the same operating cycle, and determining whether there are synchronous changes, superimposed changes, or mutually reinforcing synergistic changes in each periodic characteristic parameter within the operating cycle based on the analysis results, can preferably be: Based on the consistency of the direction of change and the correlation of the magnitude of change of various periodic characteristic parameters within the operating cycle, it is determined whether there is a synchronous change relationship.

[0042] By comparing the offset direction of different types of periodic characteristic parameters relative to their corresponding periodic benchmarks within the same operating cycle, it is determined whether each periodic characteristic parameter simultaneously increases or simultaneously decreases. Furthermore, based on the correlation analysis results of the offsets of each periodic characteristic parameter, it is confirmed whether their changes have directional consistency.

[0043] When the offset of the periodic characteristic parameter of temperature data is positive, the offset of the periodic characteristic parameter of oil status data is positive, and the offset of the periodic characteristic parameter of load-related data is also positive, it is determined that there is a synchronous change relationship with the same direction of change among the three types of periodic characteristic parameters in this operating cycle.

[0044] If there is no synchronous change relationship, it is determined that there is no coordinated change relationship between the periodic characteristic parameters, and the determination of coordinated change relationship within the current operating cycle is terminated.

[0045] When at least two types of periodic characteristic parameters show opposite directions of change within the same operating cycle, or when their directions of change are not consistent, it is directly determined that there is no cooperative change relationship between the periodic characteristic parameters within that operating cycle, and the comprehensive impact result corresponding to that operating cycle is set to a state of no cooperative impact.

[0046] When the periodic characteristic parameters of temperature data show an increasing trend, while the periodic characteristic parameters of oil state data show a stable or decreasing trend, it is determined that there is no cooperative change relationship between the periodic characteristic parameters of different categories.

[0047] If a synchronous change relationship exists, the superposition effect of the offsets corresponding to various periodic characteristic parameters is used to determine whether a superposition change relationship has been formed.

[0048] Numerical superposition analysis is performed on the offsets corresponding to the periodic characteristic parameters that have synchronous change relationships to determine whether the offsets of multiple periodic characteristic parameters form a significant cumulative effect within the same operating cycle.

[0049] For example, the periodic characteristic parameter offsets of temperature data, oil condition data, and load-related data are summed, and the summation result is compared with the offset range of a single category of periodic characteristic parameter to determine whether an overlapping change relationship exists.

[0050] If a superimposed change relationship is formed, then the cooperative change type corresponding to the periodic characteristic parameter is determined as a superimposed change relationship.

[0051] When the offsets of multiple periodic feature parameters are superimposed and significantly higher than the offset level of any single periodic feature parameter, the type of coordinated change within that operating cycle is determined to be a superimposed change relationship.

[0052] When the sum of the offsets of the three types of periodic characteristic parameters exceeds the maximum offset of any one of the types of periodic characteristic parameters, the collaborative change type within that operating cycle is marked as a superimposed change relationship.

[0053] If no superimposed change relationship is formed, the cooperative change type corresponding to the periodic characteristic parameter is determined as a synchronous change relationship.

[0054] When the characteristic parameters of each cycle have the same direction of change, but the superposition of their offsets does not show a significant cumulative effect, the cooperative change type within the operating cycle is determined to be a synchronous change relationship.

[0055] When multiple types of periodic characteristic parameters rise simultaneously, but their superposition offset remains within the normal variation range of a single type of periodic characteristic parameter, it is determined to be a synchronous change relationship.

[0056] When the superimposed changes cause the combined offset of the corresponding periodic feature parameters to exceed the influence range of a single feature change, it is determined that there is a mutually reinforcing synergistic change relationship between various periodic feature parameters.

[0057] By comparing the superimposed overall offset with the maximum normal variation range of each single-category periodic characteristic parameter within the historical operating cycle, it can be determined whether the superimposed change further amplifies the overall deviation of the operating state.

[0058] When the combined offset exceeds the historical normal offset limit of each of the temperature data, oil status data, and load-related data, it is determined that there is a mutually reinforcing cooperative change relationship.

[0059] Based on the cooperative change relationship, determine the cooperative change type of each periodic characteristic parameter within the operating cycle, and calculate the coupling strength corresponding to each cooperative change type.

[0060] Based on synchronous change relationships, superimposed change relationships, or mutually reinforcing synergistic change relationships, and combined with the proportion of each periodic characteristic parameter offset in the comprehensive offset, the corresponding coupling strength is calculated to quantify the degree of synergistic influence between different state characteristics.

[0061] By calculating the proportion of the offset of each periodic characteristic parameter to the overall offset, the corresponding coupling strength value is obtained, which is used to reflect the degree of contribution of different types of operational data to the coordinated change.

[0062] The influence of the type of coordinated change and its corresponding coupling strength on the periodic characteristic parameters of each category is comprehensively characterized to generate a comprehensive influence result that reflects the overall deviation of the operating state of the oil-immersed transformer within the operating cycle.

[0063] By combining the type of coordinated change with the corresponding coupling strength, a comprehensive impact result is formed to describe the degree of deviation of the overall operating state of the oil-immersed transformer within the operating cycle.

[0064] For example, mutually reinforcing synergistic changes and high coupling strength are jointly characterized as high-level comprehensive influence results to reflect significant abnormal changes in the operating status of oil-immersed transformers during the operating cycle.

[0065] S4 performs an evolution analysis on the comprehensive impact results corresponding to multiple consecutive operating cycles to determine whether the comprehensive impact results meet the preset risk evolution conditions.

[0066] By taking the comprehensive impact results generated in multiple adjacent and consecutive operating cycles as the object of time series analysis, and by analyzing the changing patterns of the comprehensive impact results in the operating time dimension, the evolution process of the oil-immersed transformer's operating state from a stable state to an abnormal state can be identified, thereby avoiding accidental misjudgments caused by relying solely on the comprehensive impact results of a single operating cycle for risk assessment.

[0067] When the overall impact result is temporarily higher within a certain operating cycle, but quickly falls back to the normal range in the next operating cycle, the change is determined to be a transient operating disturbance and is not used as a basis for risk evolution.

[0068] In another example, step S4 can preferably be performed as follows: The results of each comprehensive impact are serialized according to the operating cycle to construct an evolution sequence of the comprehensive impact results over time. Evolutionary characteristic parameters are extracted based on the evolution sequence to characterize the trend and rate of change of the comprehensive impact results, and risk evolution parameters are generated using the evolutionary characteristic parameters.

[0069] According to the chronological order of the operating cycles, the comprehensive impact results corresponding to each operating cycle are sorted by time to form a sequence of comprehensive impact results arranged continuously according to the operating cycle. Based on this sequence, the direction, magnitude and rate of change of the comprehensive impact results are calculated, thereby quantifying the overall evolution characteristics of the comprehensive impact results in multiple operating cycles.

[0070] For example, if the overall impact results corresponding to five consecutive operating cycles show an increasing trend cycle by cycle, and the increment between adjacent operating cycles gradually amplifies, then this change is identified as a continuously escalating risk evolution process.

[0071] Furthermore, the steps of serializing each comprehensive impact result according to the operating cycle to construct an evolution sequence of the comprehensive impact result over operating time, and extracting evolutionary characteristic parameters to characterize the trend and rate of change of the comprehensive impact result based on the evolutionary sequence, and generating risk evolution quantity using the evolutionary characteristic parameters, can be preferably: The results of each comprehensive impact are time-aligned according to the time sequence of the operation cycle to generate a continuous periodic state sequence. The periodic state sequence is then smoothed to eliminate abnormal interference caused by short-term operational fluctuations, thereby obtaining a stable evolution sequence for evolution analysis.

[0072] By aligning the start and end times of each operating cycle, the comprehensive impact results corresponding to different operating cycles are compared on the same time scale. Furthermore, the cycle state sequence is smoothed using a moving average algorithm or an exponential smoothing algorithm to reduce the impact of short-term fluctuations in operating data on the evolution judgment results.

[0073] For example, the combined impact results of seven consecutive operating cycles are processed using a three-cycle moving average.

[0074] Extract at least one evolutionary characteristic parameter from the evolutionary sequence to characterize the trend, rate of change, and degree of cumulative change of the combined impact results.

[0075] Based on the numerical changes of the comprehensive impact results in the stable evolution sequence, the slope of the linear change of the comprehensive impact results along the operating cycle dimension, the rate of change between adjacent operating cycles, and the cumulative offset within multiple operating cycles are calculated as evolutionary characteristic parameters describing the degree of operational state evolution. Linear regression analysis of the stable evolution sequence yields the trend slope of the comprehensive impact results as a function of the operating cycle, which is used to characterize the directionality of risk development.

[0076] The overall change characteristics of the evolutionary sequence are quantitatively characterized based on evolutionary feature parameters to generate risk evolution quantities.

[0077] The extracted trends, rates of change, and cumulative degrees of change are combined according to preset weights to generate a single numerical form of risk evolution quantity, which is used to uniformly measure the degree of risk evolution of the current operating state of oil-immersed transformers.

[0078] For example, by setting the weight of the trend of change to 0.4, the weight of the rate of change to 0.3, and the weight of the cumulative degree of change to 0.3, and then performing a weighted summation on the corresponding evolutionary characteristic parameters, the risk evolution quantity can be obtained.

[0079] The risk evolution quantity is compared and analyzed with the preset risk evolution conditions.

[0080] The risk evolution amount corresponding to the current operating status is compared with the pre-set risk evolution threshold or evolution rule to determine whether the comprehensive impact result has entered the evolution stage that can be judged as a failure risk.

[0081] For example, the risk evolution level can be compared with a reference upper limit obtained based on historical stable operating cycles to determine whether the current risk evolution level exceeds the upper limit.

[0082] When the risk evolution quantity meets the preset trend condition or cumulative change condition, it is determined that the preset risk evolution condition is met.

[0083] If the trend of change corresponding to the risk evolution continues to point in an abnormal direction, or the degree of cumulative change exceeds the preset safety accumulation threshold, then the operating status of the oil-immersed transformer is considered to have formed a continuously evolving risk characteristic.

[0084] When the risk evolution quantity continues to increase over multiple consecutive operating cycles, and the increase rate exceeds the set minimum evolution threshold, the risk evolution condition is determined to be met.

[0085] When the risk evolution quantity does not meet the preset trend condition and the change accumulation condition, it is determined that the preset risk evolution condition is not met, so as to determine whether the comprehensive impact result meets the preset risk evolution condition.

[0086] If the risk evolution fluctuates over multiple operating cycles but does not show a clear trend of growth, and the cumulative offset is within a safe range, then it is determined that the current comprehensive impact result has not yet formed an effective risk evolution process.

[0087] For example, if the risk evolution amount fluctuates around a preset threshold and does not show a continuous growth trend, it is determined that the risk evolution condition is not met.

[0088] If the conditions are met, the oil-immersed transformer is determined to have a fault risk, and a fault warning message is output.

[0089] In this embodiment, various types of operational data generated during the continuous operation of the oil-immersed transformer are acquired and divided according to a preset operating cycle, thereby forming multiple independent and comparable operating cycles in the time dimension. Based on this, corresponding periodic feature parameters are extracted for each type of operational data in each operating cycle to characterize the overall operating state of the oil-immersed transformer in that operating cycle. Subsequently, based on the periodic feature parameters of various types of operational data in the same operating cycle, the synergistic change relationship between different state characteristics is analyzed to generate a comprehensive impact result reflecting the combined effect of multiple operating states. Furthermore, an evolutionary analysis is performed on the comprehensive impact result corresponding to multiple consecutive operating cycles to determine whether the comprehensive impact result meets the preset risk evolution conditions. When the determination result is met, it is determined that the oil-immersed transformer has a fault risk, and the corresponding fault warning information is output.

[0090] Compared with existing technologies, this technical solution avoids the randomness caused by judging based solely on instantaneous operating data by dividing various types of operating data into operating cycles and extracting cycle characteristic parameters. At the same time, by analyzing the synergistic changes among various state characteristics within the same operating cycle, the fault warning results can reflect the comprehensive impact of multiple operating states, rather than the abnormal changes of a single parameter. In addition, by performing evolutionary analysis on the comprehensive impact results of multiple consecutive operating cycles and combining them with preset risk evolution conditions, potential risk trends can be identified before the fault becomes apparent. This improves the foresight and reliability of fault warnings for oil-immersed transformers and overcomes the problem that existing warning methods produce false alarms or omissions and fail to reflect the overall operating risk level of the transformer in a timely and accurate manner when operating conditions fluctuate or multiple factors work together.

[0091] Example 2: This application example also provides an oil-immersed transformer fault early warning system 10, which specifically includes: a data acquisition module 11, a data partitioning module 12, a data analysis module 13, a data judgment module 14, and a fault early warning module 15.

[0092] The data acquisition module 11 is used to acquire various types of operating data generated by the oil-immersed transformer during continuous operation.

[0093] The data partitioning module 12 is used to partition various types of operating data according to a preset operating cycle and extract the periodic characteristic parameters of various types of operating data in each operating cycle.

[0094] The data analysis module 13 is used to analyze the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle, so as to generate comprehensive impact results.

[0095] The data judgment module 14 is used to perform evolution analysis on the comprehensive impact results corresponding to multiple consecutive operating cycles, so as to determine whether the comprehensive impact results meet the preset risk evolution conditions.

[0096] The fault warning module 15 is used to determine that there is a fault risk in the oil-immersed transformer if the conditions are met, and to output fault warning information.

[0097] In this embodiment, the processing steps of the oil-immersed transformer fault early warning method, including data acquisition, operation cycle division, periodic characteristic parameter extraction, collaborative change relationship analysis, and risk evolution condition judgment, are respectively carried out by the data acquisition module 11, data division module 12, data analysis module 13, data judgment module 14, and fault early warning module 15. This allows various types of operation data to be transmitted and processed between modules according to a unified data structure and processing order. As a result, at the system level, the system can realize the step-by-step processing of the oil-immersed transformer operation status from single-cycle analysis to multi-cycle evolution analysis, ensuring the consistency between the logic for generating comprehensive impact results and the logic for risk evolution judgment, and improving the stability and reliability of fault early warning results in continuous operation scenarios.

[0098] Example 3: This application example also provides an electronic device 20, including a memory 21 and a processor 22. The memory 21 stores a computer program that can run on the processor 22. When the processor 22 executes the computer program, it implements the oil-immersed transformer fault early warning method of Embodiment 1.

[0099] In this embodiment, by storing the computer program for implementing the oil-immersed transformer fault early warning method in the memory 21 and executing it by the processor 22 in the order of preset program instructions, the processor 22 can sequentially complete the acquisition of various types of operating data, the division of operating cycles, the calculation of cycle characteristic parameters, the analysis of cooperative change relationships, and the evolution analysis of comprehensive influence results. Thus, the processing flow of the oil-immersed transformer fault early warning method in Embodiment 1 is completely reproduced at the level of the electronic device 20, so that the fault early warning judgment can be stably executed in a programmed manner on the electronic device 20, avoiding the inconsistency caused by human experience judgment.

[0100] Example 4: This application also provides a computer-readable storage medium having a computer program stored thereon, which, when run by a processor, causes the processor to execute the oil-immersed transformer fault early warning method as described in Example 1.

[0101] In this embodiment, by storing the computer program for executing the oil-immersed transformer fault early warning method in a computer-readable storage medium, the computer program, when called by the processor, can drive the processor to complete the periodic processing of operating data, analysis of collaborative change relationships, and judgment of risk evolution conditions according to the predetermined data processing logic. This ensures that the oil-immersed transformer fault early warning method can be repeatedly called and deployed in the form of program instructions, thereby improving the portability and reusability of the fault early warning method in different operating environments.

[0102] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device and each module described above can be referred to the corresponding process in the aforementioned Embodiment 1, and will not be repeated here.

[0103] The above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A fault early warning method for oil-immersed transformers, characterized in that, include: Acquire various operating data generated during the continuous operation of oil-immersed transformers; The various types of operational data are divided according to the preset operational cycle, and the periodic characteristic parameters of the various types of operational data within each operational cycle are extracted. Based on the periodic characteristic parameters of various types of operational data within the same operational cycle, analyze the coordinated change relationship between various state characteristics to generate comprehensive impact results; An evolutionary analysis is performed on the comprehensive impact results corresponding to multiple consecutive operating cycles to determine whether the comprehensive impact results meet the preset risk evolution conditions; If the conditions are met, the oil-immersed transformer is determined to have a fault risk, and a fault warning message is output.

2. The fault early warning method for oil-immersed transformers according to claim 1, characterized in that, The various types of operational data include at least two of the following: temperature data, oil status data, and load-related data.

3. The fault early warning method for oil-immersed transformers according to claim 1, characterized in that, The step of dividing various types of operational data according to a preset operational cycle and extracting the periodic characteristic parameters of various types of operational data within each operational cycle includes: By dividing various types of operational data into operational periods through a preset operational period, several operational parameters after the period division are obtained. Determine the cycle benchmark corresponding to the same category of running data in each running cycle, calculate the offset of the running parameters relative to the cycle benchmark in each running cycle, and generate cycle feature parameters based on the offset.

4. The fault early warning method for oil-immersed transformers according to claim 1, characterized in that, The step of analyzing the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operating data within the same operating cycle to generate a comprehensive impact result includes: Correlation analysis is performed on different categories of periodic characteristic parameters within the same operating cycle, and the analysis results are used to determine whether there are synchronous changes, superimposed changes, or mutually reinforcing synergistic changes in each periodic characteristic parameter within the operating cycle. Based on the aforementioned collaborative change relationship, the collaborative change type corresponding to each periodic characteristic parameter within the operating cycle is determined, and the coupling strength corresponding to each collaborative change type is calculated. The influence of the cooperative change type and its corresponding coupling strength on the periodic characteristic parameters of each category is comprehensively characterized to generate a comprehensive influence result.

5. The fault early warning method for oil-immersed transformers according to claim 4, characterized in that, The step of performing correlation analysis on different categories of periodic characteristic parameters within the same operating cycle, and determining whether there are synchronous changes, superimposed changes, or mutually reinforcing synergistic changes in each periodic characteristic parameter within the operating cycle based on the analysis results, includes: Based on the consistency of the direction of change and the correlation of the magnitude of change of various periodic characteristic parameters within the operating cycle, it is determined whether there is a synchronous change relationship. If there is no synchronous change relationship, it is determined that there is no coordinated change relationship between the periodic feature parameters, and the determination of coordinated change relationship in the current operating cycle is terminated. If a synchronous change relationship exists, the superposition effect of the offsets corresponding to various periodic characteristic parameters is used to determine whether a superposition change relationship has been formed. If a superimposed change relationship is formed, then the cooperative change type corresponding to the periodic characteristic parameter is determined as a superimposed change relationship; If no superimposed change relationship is formed, the cooperative change type corresponding to the periodic characteristic parameter is determined as a synchronous change relationship; When the superimposed changes cause the combined offset of the corresponding periodic feature parameters to exceed the influence range of a single feature change, it is determined that there is a mutually reinforcing synergistic change relationship between various periodic feature parameters.

6. The fault early warning method for oil-immersed transformers according to claim 1, characterized in that, The step of performing evolutionary analysis on the comprehensive impact results corresponding to multiple consecutive operating cycles to determine whether the comprehensive impact results meet the preset risk evolution conditions includes: The comprehensive impact results are serialized according to the operating cycle to construct an evolutionary sequence, and evolutionary feature parameters are extracted based on the evolutionary sequence, and risk evolution quantity is generated using the evolutionary feature parameters; The risk evolution quantity is compared and analyzed with the preset risk evolution conditions; When the risk evolution quantity meets the preset change trend condition or change accumulation condition, it is determined that the preset risk evolution condition is met. When the risk evolution quantity does not meet the preset change trend condition and does not meet the change accumulation condition, it is determined that the preset risk evolution condition is not met.

7. The fault early warning method for oil-immersed transformers according to claim 6, characterized in that, The steps of serializing the comprehensive impact results according to the operating cycle to construct an evolutionary sequence, extracting evolutionary feature parameters based on the evolutionary sequence, and generating a risk evolution quantity using the evolutionary feature parameters include: The comprehensive impact results are time-aligned according to the time sequence of the operation cycle to generate a periodic state sequence, and the periodic state sequence is smoothed to obtain an evolution sequence. Extract at least one evolutionary feature parameter from the evolutionary sequence to characterize the trend, rate of change, and degree of cumulative change of the overall impact result; The overall change characteristics of the evolutionary sequence are quantitatively characterized based on the evolutionary feature parameters to generate a risk evolution quantity.

8. A fault early warning system for an oil-immersed transformer, characterized in that, include: The data acquisition module is used to acquire various types of operating data generated by the oil-immersed transformer during continuous operation. The data partitioning module is used to partition various types of operational data according to a preset operational cycle and extract the periodic characteristic parameters of various types of operational data within each operational cycle. The data analysis module is used to analyze the synergistic change relationship between various state characteristics based on the periodic characteristic parameters of various types of operational data within the same operating cycle, in order to generate comprehensive impact results; The data judgment module is used to perform evolution analysis on the comprehensive impact results corresponding to multiple consecutive operating cycles, so as to determine whether the comprehensive impact results meet the preset risk evolution conditions. The fault warning module is used to determine that the oil-immersed transformer has a fault risk if the conditions are met, and to output fault warning information.

9. An electronic device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the oil-immersed transformer fault early warning method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, causes the processor to perform the oil-immersed transformer fault early warning method as described in any one of claims 1 to 7.

Images