A method and system for condition assessment and fault early warning of oil-immersed current transformers

Abstract

The application discloses a kind of oil-immersed current transformer state evaluation and fault early warning method and system, the method includes: dividing hydrogen production fault type and constructing case database;Judge transformer abnormal state, if only chromatographic data is abnormal, then evaluate insulating oil gas production data, if hydrogen is overproof, acetylene is not overproof, and accurate evaluation conclusion is not obtained by existing evaluation standard, then according to the evaluation condition to be screened from case database Similar cases, obtain data set;Hydrogen content logarithm and cumulative occurrence rate are calculated, the hydrogen content logarithm-cumulative occurrence rate curve of each type of fault is analyzed by fitting, and the characteristic hydrogen content of different fault types is obtained;The hydrogen content value of the evaluation condition to be compared with the characteristic hydrogen content of different fault types respectively, and the state evaluation result of oil-immersed current transformer is obtained and fault early warning is carried out.The application can realize the accurate evaluation and early warning of single hydrogen overproof or only hydrogen and total hydrocarbon overproof condition.

Description

Technical Field

[0001] This invention belongs to the field of oil-immersed current transformer operation detection and analysis technology, and relates to a method and system for oil-immersed current transformer status assessment and fault early warning. Background Technology

[0002] Oil-immersed current transformers are crucial equipment for metering and protection in power systems, widely used due to their excellent dynamic and thermal stability and good heat dissipation. However, with the expansion of their application, various problems have gradually emerged during operation. For example, their small size, low oil volume, concentrated electric field strength, and the use of external insulating porcelain bushings without explosion-proof devices make them prone to internal gas accumulation and explosions. Therefore, high requirements are placed on product quality and operation and maintenance management. Manufacturing defects, incomplete factory testing, and extreme changes in the operating environment can easily lead to equipment failure; if fault handling is not timely, it may cause equipment burnout or explosion, seriously threatening power grid safety.

[0003] Dissolved gas composition analysis technology in insulating oil has become an important means of operation and maintenance monitoring for oil-immersed current transformers because it enables fault diagnosis while the power grid remains uninterrupted. According to the "Guidelines for Dissolved Gas Analysis and Judgment in Transformer Oil" (DL / T 722-2014), hydrocarbons such as hydrogen, acetylene, and methane are characteristic gases that reflect abnormal equipment operation, and their content exceeding the warning value requires close attention.

[0004] In actual operation and maintenance, it is common to encounter situations where a single component of hydrogen exceeds the warning value (referred to as "single hydrogen exceedance," the warning value is usually 150 μL / L), while other hydrocarbon gases do not show a significant increase, the electrical performance of the equipment is normal, and the chromatographic data is stable in the short term. This is especially true for oil-immersed current transformers, where the oil volume is low and the field strength is concentrated, meaning even a small amount of gas production can lead to a rapid increase in hydrogen concentration. Furthermore, the sources of hydrogen are complex, potentially stemming from fault factors such as partial discharge and overheating, or from non-fault factors such as material release and external infiltration. In such cases, traditional criteria such as single threshold and three-ratio methods are insufficient to effectively identify the fault type. Misjudgment or omission can lead to increased maintenance costs due to over-maintenance or potential safety accidents due to overlooking potential faults, posing significant challenges for maintenance personnel in handling such situations. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for condition assessment and fault early warning of oil-immersed current transformers. It focuses on cases where hydrogen levels in the oil of oil-immersed current transformers exceed standards, particularly single hydrogen atoms. By studying the content of dissolved gas components in the current transformer oil, a defect early warning strategy for power oil-filled equipment based on the analysis of hydrogen component content in the insulating oil is established. This effectively solves operational and maintenance decision-making dilemmas, prevents equipment failures, and ensures the safe and stable operation of the power grid.

[0006] The present invention adopts the following technical solution.

[0007] The first aspect of this invention proposes a method for condition assessment and fault early warning of an oil-immersed current transformer, comprising: Step 1: Pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and construct a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the warning value. Step 2: Make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, proceed to step 3. Step 3: Evaluate the gas generation data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion is obtained through the existing evaluation standards, then proceed to Step 4. Step 4: Based on the operating conditions to be evaluated for the oil-immersed current transformer, select similar cases from the case database to obtain a dataset; Step 5: Calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. Step 6: Compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types to obtain the condition assessment result of the oil-immersed current transformer and issue a fault warning.

[0008] Preferably, the fault types include non-faulty hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production, and high-energy discharge hydrogen production.

[0009] Preferably, the case database includes equipment information, chromatographic detection information, and fault information; the equipment information includes equipment model and voltage level; the chromatographic detection information includes the content of hydrogen, acetylene, methane, ethane, ethylene, carbon monoxide, and carbon dioxide; and the fault information includes fault type, cause analysis, and handling measures.

[0010] Preferably, the characteristic hydrogen content corresponding to the different fault types includes: the hydrogen content value [H] when the cumulative number of F1 type fault cases reaches 90% saturation. F1 The hydrogen content threshold [H]0(Fn) for faults of type * and Fn, where n takes the value of 2 to 5, and faults of types F1 to F5 correspond to non-faulty hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production and high-energy discharge hydrogen production, respectively.

[0011] Preferably, in step 5, the logarithm of hydrogen content and the cumulative occurrence rate of various faults are calculated based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curves of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained, as follows: ① Obtain the hydrogen content [Hi] of each case i in the dataset, take its logarithm and record it as lg[Hi]; ② Based on the number of cases of each type of fault and the hydrogen content in the dataset, calculate the cumulative occurrence rate CI(Fn,i) of each type of fault = For each type of fault, the minimum and maximum cumulative occurrence rates CI(Fn,i) are selected to obtain the minimum CI for the corresponding fault type. min (Fn) and the maximum value CI max (Fn); Wherein, Fn is the fault type number, and F1 to F5 faults correspond to non-fault hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production and high-energy discharge hydrogen production, respectively. , This represents the number of cases of type Fn faults in the dataset where the hydrogen content is less than or equal to [Hi], and the total number of cases of type Fn faults in the dataset. ③ Plot the cumulative occurrence rate curves of various faults, lg[Hi]-CI(Fn,i), with lg[Hi] as the horizontal axis and CI(Fn,i) as the vertical axis; ④ For F1 type faults, fit the lg[Hi]-CI(F1,i) curve to obtain the F1 fitting curve; In the portion of the F1-fit curve that satisfies 0 ≤ CI(F1,i) ≤ 1, take CI(F1,i) equal to the maximum value CI. max The x-coordinate of 90% of the feature points of (F1) is denoted as [H]. F1 *,[H] F1 *This represents the hydrogen content value when the cumulative number of F1 type fault cases reaches 90% saturation. ⑤ For faults of types F2 to F5, fit the lg[Hi]-CI(Fn,i) curve to obtain the fitting curve for fault type Fn, where n takes the value of 2 to 5; ⑥ Select the interval on the Fn type fault fitting curve obtained in ⑤ that meets the condition of 0≤CI(Fn,i)≤1 and has monotonically increasing characteristics as the effective interval for equipment condition evaluation; If a point CI(Fn,i)=0 can be found in the effective interval, then the x-coordinate value of that point is recorded as lg[H]0; If there is no point where CI(Fn,i)=0, then the x-coordinate value corresponding to the minimum value of CI(Fn,i) in the effective interval is recorded as lg[H]0(Fn); Let CI(Fn,i) corresponding to lg[H]0(Fn) be denoted as CI0(Fn); The hydrogen content threshold [H]0(Fn) for fault type Fn is obtained from lg[H]0(Fn), where n takes the value of 2 to 5.

[0012] Preferably, for F1 type faults, the logistic function is used to fit the lg[Hi]-CI(F1,i) curve; For faults of types F2 to F5, the lg[Hi]-CI(Fn,i) curve is first fitted using the linear fitting method. If the linear fitting goodness requirement cannot be met, then a second-order polynomial fitting is used.

[0013] Preferably, in step 5, if there is insufficient case data for the same fault type and the logarithm-cumulative occurrence rate curve of hydrogen content for the corresponding fault type cannot be fitted, then the probability of occurrence of the corresponding fault type is recorded as 0. For the hydrogen content threshold [H]0(Fn) of the obtained Fn type fault, if 0 < [H]0(Fn) < hydrogen attention value, then the dataset is re-filtered.

[0014] Preferably, in step 6, the hydrogen content value [H] of the oil-immersed current transformer under the evaluated operating condition is compared with the characteristic hydrogen content corresponding to different fault types to obtain the state assessment result of the oil-immersed current transformer and to provide fault warning, specifically including: ①If [H] < [H] F1 *And for all [H]0(Fn), [H]<[H]0(Fn) is satisfied, where n takes the value of 2~5, then it is determined to be non-faulty hydrogen production, and the oil-immersed current transformer is in the observation state; ②If [H] < [H] F1 *And there exists [H]0(Fn) satisfying [H]>[H]0(Fn), where n takes the value 2~5, then it is determined to be non-faulty hydrogen production and the corresponding Fn-type faulty hydrogen production, and the oil-immersed current transformer is in a state of alert; ③If [H] > [H] F1 *If two or fewer [H]0(Fn) satisfy [H]>[H]0(Fn), where n is 2~5, then it is determined to be the fault type corresponding to satisfying [H]>[H]0(Fn), and the oil-immersed current transformer is in an abnormal state; ④ If [H] > [H] F1 *If three or more [H]0(Fn) satisfy [H]>[H]0(Fn), where n is 2~5, then the fault type corresponding to satisfying [H]>[H]0(Fn) is determined, and the oil-immersed current transformer is in a severe state.

[0015] A second aspect of this invention provides a condition assessment and fault early warning system for an oil-immersed current transformer, comprising: The type classification and database construction module is used to pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and to build a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the attention value. The preliminary operating condition judgment module is used to make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, it will enter the gas production data evaluation module. The gas production data evaluation module is used to evaluate the gas production data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion can be obtained through the existing evaluation standards, then it will enter the data filtering module. The data filtering module is used to filter similar cases from the case database based on the operating conditions of the oil-immersed current transformer to be evaluated, and obtain a dataset. The fitting analysis module is used to calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. The condition assessment and early warning module is used to compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types, so as to obtain the condition assessment result of the oil-immersed current transformer and provide fault early warning.

[0016] A third aspect of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0017] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0018] Compared with the prior art, the beneficial effects of the present invention include at least the following: This invention combines the physicochemical principles of gas generation in insulating oil of power equipment, existing evaluation standards, and statistical data from a large number of cases to pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and has good applicability and operability.

[0019] This invention constructs a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the warning value, and performs data screening and status assessment based on this case database, which can improve the pertinence and accuracy of the assessment from the data source.

[0020] This invention establishes a quantitative relationship between equipment hydrogen content and failure rate (the cumulative failure rate CI - the logarithm of hydrogen content lg[H] fitting curve), providing a data-driven accurate prediction method for equipment condition assessment. It can achieve accurate condition assessment and early warning for operating conditions where only hydrogen exceeds the standard or only hydrogen and total hydrocarbon exceed the standard (hydrogen exceeds the standard, acetylene does not exceed the standard).

[0021] This invention proposes a [H] model based on the cumulative occurrence rate (CI). F1 The invention provides evaluation parameters and algorithms for hydrogen content characteristics such as * and [H]0(Fn), enabling the determination or prediction of the occurrence of a certain type of fault under actual operating conditions based on statistical data and the corresponding hydrogen content threshold for each type of fault. This invention calculates the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on a dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curves of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. The hydrogen content value of the oil-immersed current transformer under the evaluation condition is compared with the characteristic hydrogen content corresponding to different fault types to obtain the state assessment results of the oil-immersed current transformer and provide a fault early warning evaluation model. The evaluation parameters (characteristic hydrogen content, cumulative occurrence rate) can be calculated and evaluation results can be given by writing a program. Furthermore, the evaluation model supports programming by developers or generation by large AI models for direct use by operators.

[0022] The scalability of the dataset in this invention allows for the addition of existing operational condition data, thus enhancing the method's relevance and accuracy to actual operating conditions. Attached Figure Description

[0023] Figure 1 This is a flowchart of a method for condition assessment and fault early warning of an oil-immersed current transformer according to the present invention.

[0024] Figure 2 This is a data interface for the case library in an embodiment of the present invention.

[0025] Figure 3 The examples in this invention include the cumulative fault occurrence rate lg[H]-CI, the fitted curve, and the evaluation parameters for the case dataset. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0027] Embodiment 1 of the present invention provides a method for condition assessment and fault early warning of an oil-immersed current transformer, such as... Figure 1 As shown, the method includes: Step 1: Pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and construct a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the warning value. More preferably, combining the chemical mechanism of gas production in oil-filled equipment with existing research, hydrogen production failure cases are collected, and chromatographic data and failure types of key cases are statistically recorded to establish a case database. The classification method for hydrogen production failures of the equipment is as follows: Based on industry evaluation standards (GB / T7252—2001, DL / T722—2014, DL / T1690—2017), the physicochemical principles of gas generation in insulating oil of power equipment, and numerous operational case records, a classification method for hydrogen generation fault types in oil-immersed current transformers related to hydrogen content in insulating oil was established after comprehensive analysis and demonstration, as shown in Table 1.

[0028] Table 1. Classification Method for Hydrogen Generation Faults in Oil-Immersed Current Transformers

[0029] Considering non-faulty hydrogen generation such as hydrogen production from equipment material adsorption, incomplete exhaust gas emission, and catalytic dehydrogenation of metal components, which are characterized by relatively small hydrogen production and a saturation upper limit, their impact on the safe operation of instrument transformers is minimal. Therefore, the aforementioned hydrogen generation is distinguished from overheating and discharge-related hydrogen generation and classified separately as a non-faulty hydrogen generation type, providing a classification basis for more accurate substantive fault determination.

[0030] The case database system solution is as follows: A database of cases where the hydrogen content in the insulating oil of oil-immersed current transformers exceeds the warning value (hereinafter referred to as the "hydrogen exceedance case database" or "case database") is constructed. The main database structure and data characteristics are as follows: The case library structure includes: Equipment information: equipment model, voltage level, etc.; Chromatographic detection information: content of hydrogen, acetylene, methane, ethane, ethylene and other characteristic gases; Fault information: fault type, cause analysis and handling measures, etc.

[0031] The case data in the case library must meet the following requirements: ① To ensure the applicability of this method, the case sources should include publicly released or reported literature, as well as field operation records in the areas where this method is used. The operation data should account for more than 50% of the total case library.

[0032] Meanwhile, case data should be obtained randomly (when the data volume is large) or in its entirety (when the data volume is small).

[0033] ②The insulating oil chromatography test and fault information data in the case are valid; ③ The total number of cases should be 200 or more, of which the number of cases involving hydrogen exceeding the warning value should be 100 or more.

[0034] Step 2: Make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, proceed to step 3. More preferably, the following preliminary judgment is made on the working condition to be evaluated: Collect information on the operating conditions to be evaluated, including equipment model, operating voltage, and current operating status, to make a preliminary judgment on any abnormal operating conditions of the equipment. (1) If only the chromatographic detection data is abnormal in the working condition, it is considered that only the chromatographic data is abnormal, and then you can proceed to step 3; (2) If abnormalities occur in relative dielectric loss, capacitance detection, or equipment appearance, they are considered other abnormal situations and are beyond the scope of application of the method of this invention. Other indicators and standards need to be used for evaluation, and this method should not be used as the main judgment means.

[0035] Step 3: Further evaluate the gas generation data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion is obtained through the existing evaluation standards, then proceed to Step 4. More preferably, a routine evaluation of the gas generation data of the insulating oil is performed, specifically as follows: Routine evaluations are conducted according to current industry evaluation standards (GB / T7252—2001, DL / T722—2014, DL / T1690—2017), such as whether the content of hydrogen, total hydrocarbons, acetylene, and gas production rate exceed the standards, and the evaluation results are obtained using the three-ratio method.

[0036] If the hydrogen level exceeds the standard but the acetylene level does not, and no accurate evaluation conclusion is obtained through existing evaluation standards (GB / T7252—2001, DL / T722—2014, DL / T1690—2017), then proceed to step 4.

[0037] Step 4: Based on the operating conditions to be evaluated for the oil-immersed current transformer, select similar cases from the case database to obtain a dataset; More preferably, based on the operating conditions to be evaluated, similar records in the case library are screened using conditions such as hydrogen content, acetylene content, voltage level, and whether total hydrocarbons exceed the standard, to obtain Dataset 1. If the number of cases in Dataset 1 is insufficient or the obtained evaluation parameters are invalid, the data range can be expanded by lowering the screening conditions (e.g., successively ignoring conditions such as voltage level and whether total hydrocarbons exceed the standard), to obtain Datasets 2, 3, ... etc.

[0038] Step 5: Calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. More preferably, based on the dataset selected from the case library, a series of parameters such as cumulative occurrence rate and hydrogen content characteristic value are used to build a hydrogen content-based equipment condition assessment model (fault type evaluation model) to obtain the cumulative occurrence rate curve of faults and characteristic hydrogen content values ​​[H]. F1 * and [H]0(Fn) provide a basis for fault analysis and early warning of single-hydrogen exceedance conditions in oil-immersed current transformers. The algorithm for evaluation parameters based on cumulative occurrence rate in the state evaluation model of oil-immersed current transformers is as follows: (1) Cumulative occurrence rate Within a certain range of hydrogen content values, count the number of cases of a certain type of failure (denoted as N), and the total number of cases of that type in the database (denoted as N). t ).

[0039] The cumulative occurrence rate CI is defined as the ratio of N to Nt, representing the percentage of a certain type of failure case that occurs cumulatively within a certain range [H].

[0040] (1) (2) Algorithm for characteristic hydrogen content ① Record the hydrogen content in the case record as [H] (μL / L), take its logarithm and record it as lg[H]; ② For records in the case library that meet the criteria, classify and count the number of cases by fault type, and calculate the cumulative occurrence rate (CI) for each fault type, denoted as CI(F1), CI(F2), ..., etc. The minimum and maximum cumulative occurrence rates for each fault type are represented by CI. min and CI max express.

[0041] ③ Plot the cumulative occurrence rate curve lg[H]-CI of fault types with lg[H] as the horizontal axis and the cumulative occurrence rate CI as the vertical axis.

[0042] ④ The logistic function is used to fit the lg[H]-CI curve of the non-faulty gas generation type F1 to obtain the F1 fitted curve. In the portion of this fitted curve that satisfies 0≤CI≤1, CI is taken as the maximum value CI. max The x-coordinate value of 90% of the feature points is denoted as [H]. F1 *. [H] F1* This represents the hydrogen content value when the cumulative number of F1 cases reaches 90% saturation. It indicates that when the hydrogen content value reaches or exceeds this value, the CI value of F1 cases is close to the maximum value, the proportion of new cases is extremely small, and the number remains basically stable.

[0043] ⑤ Fit the lg[H]-CI curves for fault types F2~F5 (hereinafter collectively referred to as Fn) respectively (satisfying R... 2 (≥0.9). Linear fitting method is preferred. If the goodness requirement cannot be met, second-order polynomial fitting is selected to obtain the fitting curve of fault type Fn.

[0044] ⑥ Select the interval on the fitted curve that meets the condition of 0≤CI≤1 and has a monotonically increasing characteristic as the effective interval for equipment condition evaluation. This curve segment can represent the quantitative relationship between the equipment hydrogen content [H] and the cumulative failure rate CI. Within the effective interval of the fitted curve, find the point where CI=0 and denote its x-coordinate as lg[H]0; if no point where CI=0 exists, denote the x-coordinate of the point with the minimum CI value within the interval as lg[H]0. The CI value corresponding to lg[H]0 is denoteed as CI0.

[0045] The hydrogen content threshold [H]0(Fn) can be obtained from lg[H]0. The physical meaning of [H]0(Fn) is the critical value of hydrogen content corresponding to a certain type of failure with a cumulative occurrence rate of 0. This threshold is not an absolute safety boundary, but a reference benchmark in a statistical sense where the probability of failure approaches zero.

[0046] (3) Algorithm description ①Based on the analysis of a large number of case data, within the statistical range (i.e., hydrogen concentration greater than the warning value), the lg[H]-CI curve of the F1 non-faulty hydrogen production type closely resembles an S-shaped curve, and its logistic function goodness of fit is above 0.9. Furthermore, according to the classification criteria for non-faulty hydrogen production types, its critical hydrogen content value [H]0 is theoretically close to the hydrogen saturation concentration in the oil under operating conditions; therefore, it is not calculated or considered in this method.

[0047] ② If the goodness of fit R of the fitting curve for a certain abnormality or fault type is 2 ≤0.9, or the obtained CI0>CI min If the data is insufficient to support a reliable assessment, it indicates that similar failure cases need to be added or the model parameters need to be recalibrated.

[0048] ③ The acetylene content in insulating oil has a significant impact on fault type assessment. Therefore, this method yields more accurate evaluation results for operating conditions where the acetylene content does not exceed the warning value. Appropriate case selection can be performed based on the characteristics of the data to be evaluated, and the evaluation parameters can be recalculated to improve the accuracy of the evaluation.

[0049] ④ The accuracy and universality of the case study data directly determine the effectiveness of this evaluation algorithm, and this needs to be ensured by the amount and source of the case study data.

[0050] (4) The numerical values ​​of the characteristic hydrogen content can vary as follows: ① If there is no lg[H]-CI data curve, that is, there are no cases of a certain type of fault in the dataset, or there are too few records, it means that the probability of the fault occurring under the evaluation conditions is very small, so its probability of occurrence is recorded as 0.

[0051] ② The hydrogen concentration of 0 < [H]0(Fn) < 150 μL / L is a warning value. According to existing standards (GB / T7252—2001, DL / T722—2014, DL / T1690—2017), this value is not valid for evaluation. This situation likely arises because the dataset contains many fault cases related to acetylene or other abnormal electrical indicators, with little correlation to hydrogen alone; the dataset needs to be re-screened.

[0052] ③[H]0≥150μL / L hydrogen gas attention value, which can be used for evaluation.

[0053] Step 6: Compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types to obtain the condition assessment result of the oil-immersed current transformer and issue a fault warning.

[0054] More preferably, the evaluation and early warning strategies in the state evaluation model of the oil-immersed current transformer are as follows: The characteristic value of hydrogen content [H] obtained by this method F1 * and [H]0 are used as the core criteria of the status evaluation model, and then combined with the equipment chromatographic monitoring data in actual working conditions, the equipment operating status is evaluated.

[0055] Specifically, the hydrogen content value [H] of the working condition to be evaluated is compared with the characteristic hydrogen content of various fault types, and the evaluation conclusion is obtained according to the following judgment method.

[0056] ①If [H] < [H] F1 *If [H] < [H]0(Fn), then the cumulative occurrence rate of fault Fn CI'(Fn) ≤ CI0; it can be determined that: there is non-faulty gas production, and fault Fn has not occurred; the operating condition is under observation, and the change in hydrogen content of the equipment should be continuously observed during routine chromatographic detection, and degassing treatment should be carried out when necessary.

[0057] ②If [H] < [H] F1*If [H] > [H]0(Fn), it can be determined that there are two types of hydrogen production failures: non-fault and Fn. At the same time, the CI' value corresponding to the [H] value to be evaluated can be obtained from the cumulative occurrence rate fitting curve. The probability / likelihood of hydrogen production failure Fn occurring under this operating condition can be obtained from the magnitude of the CI'(Fn) value. The operating condition is under attention, and the routine detection cycle should be shortened. Attention should be paid to changes in parameters such as equipment oil chromatography detection values ​​and gas production rate.

[0058] ③If [H] > [H] F1 If there are two or fewer [H]0(Fn) satisfying [H]>[H]0(Fn), it indicates that the main fault in hydrogen production is no longer the non-faulty gas production type, but mainly other fault types (the fault types corresponding to satisfying [H]>[H]0(Fn)). The operating condition is abnormal, and more indicator detection and evaluation should be carried out in a timely manner so as to obtain accurate conclusions by using other evaluation methods.

[0059] ④ If [H] > [H] F1 If there are 3 or more [H]0(Fn) satisfying [H]>[H]0(Fn), it indicates that the operating condition has exceeded the hydrogen content threshold for more than three faults, and the fault type has entered the high probability range. At this time, the operating condition to be evaluated can be evaluated as a serious state, and power outage and maintenance should be carried out as soon as possible.

[0060] Embodiment 2 of the present invention provides an evaluation example based on the oil-immersed current transformer condition assessment and fault early warning method of the present invention: (1) Establish a case database: According to the method of the present invention, a database of cases in which the hydrogen content of the insulating oil of oil-immersed current transformers exceeds the warning value is established, such as... Figure 2 As shown in the example. The case database contains 243 valid case records, of which 157 exceed the limit for single hydrogen, and 59 exceed the limits for both hydrogen and total hydrocarbons.

[0061] (2) Evaluation Case: Ten sets of working conditions to be evaluated were selected (see Table 2) and evaluated using this method.

[0062] Table 2 Chromatographic data of the operating conditions to be evaluated (gas content in μL / L)

[0063] First, make a preliminary judgment on the operating conditions in Table 2 based on the operating condition description: Groups 1-9 belong to the case of only abnormal chromatographic data and can proceed to the next evaluation step; while the operating condition in Group 10 belongs to "medium damage and abnormal chromatographic data" and is not applicable to the subsequent steps of the method of this invention.

[0064] Then, the gas production data of groups 1-9 were evaluated: the hydrogen content in all groups exceeded 150 μL / L; the acetylene content in group 1 exceeded the standard; and the total hydrocarbon content in group 9 exceeded the standard. According to the aforementioned conditions, group 1 is not applicable to the subsequent steps of the present invention. Groups 2-9 all meet the criteria of hydrogen exceeding the standard and acetylene not exceeding the standard. If an accurate evaluation conclusion cannot be obtained through existing evaluation standards, the subsequent steps of the present invention can be performed.

[0065] (3) Establish an evaluation model: Based on the overall characteristics of the gas content under the evaluation conditions (i.e., [H] ≥ 150 μL / L and [C₂H₂] < 1 μL / L), data matching these characteristics were selected from the case library, resulting in a dataset of 216 cases. Using this method, the lg[H]-CI fitting curves and characteristic hydrogen content values ​​[H] for each fault type were obtained. F1 * and [H]0(Fn), the specific process is as follows: ① Obtain the hydrogen content [Hi] of each case i in the dataset, take its logarithm and record it as lg[Hi]; ② Based on the number of cases of each type of fault and the hydrogen content in the dataset, calculate the cumulative occurrence rate CI(Fn,i) of each type of fault = For each type of fault, the minimum and maximum cumulative occurrence rates CI(Fn,i) are selected to obtain the minimum CI for the corresponding fault type. min (Fn) and the maximum value CI max (Fn); Wherein, Fn is the fault type number, and F1 to F5 faults correspond to non-fault hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production and high-energy discharge hydrogen production, respectively. , This represents the number of Fn-type fault cases in the dataset with hydrogen content less than or equal to [Hi] and the total number of Fn-type fault cases in the dataset. ③ Plot the cumulative occurrence rate curves of various faults, lg[Hi]-CI(Fn,i), with lg[Hi] as the horizontal axis and CI(Fn,i) as the vertical axis; ④ For F1 type faults, fit the lg[Hi]-CI(F1,i) curve to obtain the F1 fitting curve; In the portion of the F1-fit curve that satisfies 0 ≤ CI(F1,i) ≤ 1, take CI(F1,i) equal to the maximum value CI. max The x-coordinate of 90% of the feature points of (F1) is denoted as [H]. F1 *,[H] F1 *This represents the hydrogen content value when the cumulative number of F1 type fault cases reaches 90% saturation. ⑤ For faults of types F2 to F5, fit the lg[Hi]-CI(Fn,i) curve to obtain the fitting curve for fault type Fn, where n takes the value of 2 to 5; ⑥ Select the interval on the Fn type fault fitting curve obtained in ⑤ that meets the condition of 0≤CI(Fn,i)≤1 and has monotonically increasing characteristics as the effective interval for equipment condition evaluation; If a point CI(Fn,i)=0 can be found in the effective interval, then the x-coordinate value of that point is recorded as lg[H]0; If there is no point where CI(Fn,i)=0, then the x-coordinate value corresponding to the minimum value of CI(Fn,i) in the effective interval is recorded as lg[H]0(Fn); Let CI(Fn,i) corresponding to lg[H]0(Fn) be denoted as CI0(Fn); The hydrogen content threshold [H]0(Fn) for fault type Fn is obtained from lg[H]0(Fn), where n takes the value of 2 to 5.

[0066] The final lg[H]-CI fitting curves and characteristic hydrogen content values ​​[H] for each fault type were obtained. F1 *and [H]0(Fn) as Figure 3 As shown, from Figure 3 It can be seen that the [H] of non-faulty hydrogen production F1 was obtained from the case dataset. F1 * Hydrogen content thresholds for partial discharge, low-energy discharge, and high-energy discharge: [H]0(F3), [H]0(F4), and [H]0(F5).

[0067] In this dataset, there was only one case of overheating hydrogen production failure, with a cumulative occurrence rate of only 0.005. [H]0(F2) had no valid value and was therefore not considered.

[0068] In addition, only 5 cases of high-energy discharge hydrogen production were found in the dataset, so the accuracy of the hydrogen content threshold [H]0 (F5) is low and may have a large error.

[0069] (4) Evaluation and Conclusion: The hydrogen content value [H] of the working condition to be evaluated is compared with the characteristic hydrogen content of various fault types. The working condition status is determined and fault warning is given according to this method, as shown in Table 3.

[0070] Table 3 Evaluation process and conclusions (gas content in μL / L)

[0071] As shown in Table 3, the evaluation model of this method for the nine actual operating conditions is consistent with the results of the actual operating records, which verifies that the evaluation of hydrogen content by this method has good accuracy.

[0072] Embodiment 3 of the present invention provides a condition assessment and fault early warning system for an oil-immersed current transformer, comprising: The type classification and database construction module is used to pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and to build a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the attention value. The preliminary operating condition judgment module is used to make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, it will enter the gas production data evaluation module. The gas production data evaluation module is used to evaluate the gas production data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion can be obtained through the existing evaluation standards, then it will enter the data filtering module. The data filtering module is used to filter similar cases from the case database based on the operating conditions of the oil-immersed current transformer to be evaluated, and obtain a dataset. The fitting analysis module is used to calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. The condition assessment and early warning module is used to compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types, so as to obtain the condition assessment result of the oil-immersed current transformer and provide fault early warning.

[0073] Embodiment 4 of the present invention provides a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0074] Embodiment 5 of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0075] Compared with existing technologies, this invention establishes an evaluation model based on a targeted case database, which improves the relevance and accuracy from the data source. Combining the physicochemical principles of gas generation in insulating oil of power equipment, existing evaluation standards, and statistical analysis of a large number of case data, a classification method for hydrogen generation faults in oil-immersed current transformers is proposed, which has good applicability and operability. An evaluation parameter and algorithm based on cumulative occurrence rate are proposed to establish a quantitative relationship between equipment hydrogen content and fault occurrence rate, providing a data-driven and accurate prediction method for equipment fault assessment. The evaluation model can be programmed to calculate evaluation parameters and provide evaluation results. Furthermore, the evaluation model supports programming by developers or generation by large AI models, facilitating direct use by operators.

[0076] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0077] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0078] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0079] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for condition assessment and fault early warning of an oil-immersed current transformer, characterized in that, include: Step 1: Pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and construct a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the warning value. Step 2: Make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, proceed to step 3. Step 3: Evaluate the gas generation data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion is obtained through the existing evaluation standards, then proceed to Step 4. Step 4: Based on the operating conditions to be evaluated for the oil-immersed current transformer, select similar cases from the case database to obtain a dataset; Step 5: Calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. Step 6: Compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types to obtain the condition assessment result of the oil-immersed current transformer and issue a fault warning.

2. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 1, characterized in that: The fault types include non-faulty hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production, and high-energy discharge hydrogen production.

3. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 1, characterized in that: The case database includes equipment information, chromatographic detection information, and fault information; the equipment information includes equipment model and voltage level; the chromatographic detection information includes the content of hydrogen, acetylene, methane, ethane, ethylene, carbon monoxide, and carbon dioxide; the fault information includes fault type, cause analysis, and handling measures.

4. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 1, characterized in that: The characteristic hydrogen content corresponding to the different fault types includes: the hydrogen content value [H] when the cumulative number of F1 type fault cases reaches 90% saturation. F1 The hydrogen content threshold [H]0(Fn) for faults of type * and Fn, where n takes the value of 2 to 5, and faults of types F1 to F5 correspond to non-faulty hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production and high-energy discharge hydrogen production, respectively.

5. A method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 1 or 4, characterized in that: In step 5, the logarithm of hydrogen content and the cumulative occurrence rate of various faults are calculated based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curves of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained, as follows: ① Obtain the hydrogen content [Hi] of each case i in the dataset, take its logarithm and record it as lg[Hi]; ② Based on the number of cases of each type of fault and the hydrogen content in the dataset, calculate the cumulative occurrence rate CI(Fn,i) of each type of fault = For each type of fault, the minimum and maximum cumulative occurrence rates CI(Fn,i) are selected to obtain the minimum CI for the corresponding fault type. min (Fn) and the maximum value CI max (Fn); Wherein, Fn is the fault type number, and F1 to F5 faults correspond to non-fault hydrogen production, overheating hydrogen production, partial discharge hydrogen production, low-energy discharge hydrogen production and high-energy discharge hydrogen production, respectively. , This represents the number of cases of type Fn faults in the dataset where the hydrogen content is less than or equal to [Hi], and the total number of cases of type Fn faults in the dataset. ③ Plot the cumulative occurrence rate curves of various faults, lg[Hi]-CI(Fn,i), with lg[Hi] as the horizontal axis and CI(Fn,i) as the vertical axis; ④ For F1 type faults, fit the lg[Hi]-CI(F1,i) curve to obtain the F1 fitting curve; In the portion of the F1-fit curve that satisfies 0 ≤ CI(F1,i) ≤ 1, take CI(F1,i) equal to the maximum value CI. max The x-coordinate of 90% of the feature points of (F1) is denoted as [H]. F1 *,[H] F1 *This represents the hydrogen content value when the cumulative number of F1 type fault cases reaches 90% saturation. ⑤ For faults of types F2 to F5, fit the lg[Hi]-CI(Fn,i) curve to obtain the fitting curve for fault type Fn, where n takes the value of 2 to 5; ⑥ Select the interval on the Fn type fault fitting curve obtained in ⑤ that meets the condition of 0≤CI(Fn,i)≤1 and has monotonically increasing characteristics as the effective interval for equipment condition evaluation; If a point CI(Fn,i)=0 can be found in the effective interval, then the x-coordinate value of that point is recorded as lg[H]0; If there is no point where CI(Fn,i)=0, then the x-coordinate value corresponding to the minimum value of CI(Fn,i) in the effective interval is recorded as lg[H]0(Fn); Let CI(Fn,i) corresponding to lg[H]0(Fn) be denoted as CI0(Fn); The hydrogen content threshold [H]0(Fn) for fault type Fn is obtained from lg[H]0(Fn), where n takes the value of 2 to 5.

6. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 5, characterized in that: For F1 type faults, the logistic function is used to fit the lg[Hi]-CI(F1,i) curve; For faults of types F2 to F5, the lg[Hi]-CI(Fn,i) curve is first fitted using the linear fitting method. If the linear fitting goodness requirement cannot be met, then a second-order polynomial fitting is used.

7. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 4, characterized in that: In step 5, if there is insufficient case data for the same fault type and the logarithm-cumulative occurrence rate curve of hydrogen content for the corresponding fault type cannot be fitted, then the probability of occurrence of the corresponding fault type is recorded as 0. For the hydrogen content threshold [H]0(Fn) of the obtained Fn type fault, if 0 < [H]0(Fn) < hydrogen attention value, then the dataset is re-filtered.

8. The method for condition assessment and fault early warning of an oil-immersed current transformer according to claim 4, characterized in that: In step 6, the hydrogen content value [H] of the oil-immersed current transformer under the evaluated operating condition is compared with the characteristic hydrogen content corresponding to different fault types to obtain the state assessment result of the oil-immersed current transformer and to provide fault warning, specifically including: ①If [H] < [H] F1 *And for all [H]0(Fn), [H]<[H]0(Fn) is satisfied, where n takes the value of 2~5, then it is determined to be non-faulty hydrogen production, and the oil-immersed current transformer is in the observation state; ②If [H] < [H] F1 *And there exists [H]0(Fn) satisfying [H]>[H]0(Fn), where n takes the value 2~5, then it is determined to be non-faulty hydrogen production and the corresponding Fn-type faulty hydrogen production, and the oil-immersed current transformer is in a state of alert; ③If [H] > [H] F1 *If two or fewer [H]0(Fn) satisfy [H]>[H]0(Fn), where n is 2~5, then it is determined to be the fault type corresponding to satisfying [H]>[H]0(Fn), and the oil-immersed current transformer is in an abnormal state; ④ If [H] > [H] F1 *If three or more [H]0(Fn) satisfy [H]>[H]0(Fn), where n is 2~5, then the fault type corresponding to satisfying [H]>[H]0(Fn) is determined, and the oil-immersed current transformer is in a severe state.

9. A condition assessment and fault early warning system for an oil-immersed current transformer, comprising the method described in any one of claims 1-8, characterized in that, The system includes: The type classification and database construction module is used to pre-classify the hydrogen generation fault types of oil-immersed current transformers related to the hydrogen content of insulating oil, and to build a case database of oil-immersed current transformers whose insulating oil hydrogen content exceeds the attention value. The preliminary operating condition judgment module is used to make a preliminary judgment on the operating condition of the oil-immersed current transformer to be evaluated. If only the chromatographic data is abnormal, it will enter the gas production data evaluation module. The gas production data evaluation module is used to evaluate the gas production data of the insulating oil of the oil-immersed current transformer. If the hydrogen exceeds the standard and the acetylene does not exceed the standard, and no accurate evaluation conclusion can be obtained through the existing evaluation standards, then it will enter the data filtering module. The data filtering module is used to filter similar cases from the case database based on the operating conditions of the oil-immersed current transformer to be evaluated, and obtain a dataset. The fitting analysis module is used to calculate the logarithm of hydrogen content and the cumulative occurrence rate of various faults based on the dataset. By fitting and analyzing the logarithm-cumulative occurrence rate curve of hydrogen content for various faults, the characteristic hydrogen content corresponding to different fault types is obtained. The condition assessment and early warning module is used to compare the hydrogen content value of the oil-immersed current transformer under the evaluation condition with the characteristic hydrogen content corresponding to different fault types, so as to obtain the condition assessment result of the oil-immersed current transformer and provide fault early warning.

10. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-8.

Images