A method for diagnosing insulation of a converter bushing group under non-uniform voltage
By constructing a nonlinear normalized conversion model and a dynamic lateral baseline, the problems of false alarms and missed judgments in traditional bushing insulation diagnosis methods under non-uniform voltages are solved, and accurate diagnosis and high reliability assessment of bushing group insulation status are achieved.
Patent Information
- Application Number
- CN202610565694.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-27
- Publication Date
- 2026-06-23
Smart Images

Figure CN122260058A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of high voltage testing and power equipment condition monitoring technology, and more specifically, relates to a method for diagnosing the insulation of converter transformer bushing groups under non-uniform voltage. Background Technology
[0002] The insulation condition of the grid-side bushings of converter transformers is directly related to the safety of ultra-high voltage direct current transmission systems. In recent years, to improve on-site maintenance efficiency and reduce the risks of high-altitude operations, the technology of reduced decoupling induction voltage testing has been gradually promoted on engineering sites. This technology involves inducing a test high voltage on the grid side by low-voltage excitation on the valve side while retaining most of the electrical connections on the grid side. However, when conducting overall parallel voltage testing on the six converter transformers of the converter station pole system, due to the inherent "grid-side star-valve-side delta" interconnection topology of the pole system, single-phase excitation will induce a significant non-uniform voltage distribution characteristic of two high and four low voltages on the grid side (i.e., the energized phase bears the full voltage, while the non-energized phase bears only about half the voltage), causing the bushings of each phase to be in completely different voltage conditions.
[0003] This unique non-uniform voltage condition severely undermines traditional bushing insulation diagnostic methods. On one hand, the leakage current of each bushing varies significantly due to different terminal voltages, making direct absolute value comparison impossible. Furthermore, traditional dielectric loss testing heavily relies on high-precision phase extraction, which is easily distorted under strong electromagnetic interference and multiple units operating in parallel. On the other hand, the bushing insulation dielectric exhibits non-linear leakage conductance characteristics under different electric field strengths. Forcibly calculating the leakage current at half voltage using a simple linear ratio introduces significant theoretical errors, leading to false alarms or missed diagnoses. Therefore, there is an urgent need to develop a new method that is independent of complex phase dependence, effectively eliminates non-uniform voltage differences, and enables accurate joint diagnosis of bushing group insulation conditions. Summary of the Invention
[0004] The purpose of this invention is to propose an insulation diagnosis method for converter transformer bushing groups under non-uniform voltage conditions. This method aims to overcome the limitations of traditional insulation assessment techniques, such as lateral incomparability due to uneven voltage distribution in parallel pressurization of systems with few decoupling stages, and the susceptibility to distortion in dielectric loss phase measurements under strong interference environments. This invention completely eliminates the reliance on high-precision phase synchronization and innovatively constructs a nonlinear normalized conversion model and dynamic lateral baseline based on the correlation of physical voltage levels. This effectively eliminates the difference in absolute amplitude of leakage conductance under non-uniform electric fields, providing a standardized data processing model and technical guarantee for accurate and rapid screening and highly reliable joint diagnosis of insulation defects in bushings of multiple converter transformers in a converter station pole system.
[0005] To achieve the above objectives, the present invention provides the following technical solution:
[0006] S1: Based on the star-delta interconnection topology of six converter transformers in the valve hall, under the single-phase pressurization condition on the valve side, the operating voltage signal and the leakage current signal of the six grid-side bushings of the whole system are collected synchronously, and the effective voltage value and effective current value of each are calculated respectively to construct a pure amplitude dataset that is independent of phase dependence.
[0007] S2: Based on the induced voltage distribution pattern of two high and four low, the effective value of the measured voltage of each bushing is extracted and compared, and the six bushings are divided into a reference group that bears the full voltage and a group to be converted that bears half voltage.
[0008] S3: Using the measured effective voltage value of the reference group as the full voltage reference, a nonlinear amplitude correction coefficient is introduced to map and convert the effective value of the leakage current of the group to be converted at half voltage to the full voltage reference, thereby eliminating the absolute amplitude difference caused by the non-uniform distribution of excitation voltage.
[0009] S4: Based on the normalized leakage current amplitude of the six bushings, the extreme value removal and averaging algorithm is adopted. After removing the maximum and minimum values in the six normalized current data, the arithmetic mean of the remaining four valid data is calculated to construct a dynamic transverse current baseline that reflects the overall health level of the bushing group in this batch.
[0010] S5: Calculate the percentage deviation of the normalized leakage current amplitude of a single bushing relative to the baseline of the group lateral current; when the percentage deviation of the bushing amplitude exceeds the preset diagnostic threshold, it is determined that the corresponding bushing has an abnormal insulation condition.
[0011] Preferably, step S1 includes: acquiring steady-state voltage and current waveform data of six bushings containing a complete power frequency cycle based on a unified synchronous triggering mechanism; after removing transient interference, directly calculating and extracting the effective voltage and effective current values of each bushing based on the time-domain waveform data as pure amplitude features that are independent of phase dependence.
[0012] Preferably, in step S2, the grouping rule for dividing the six bushings into a baseline group and a conversion group is as follows: the target test voltage is set to U. set Compare the measured effective voltage values of the six bushings, and set the measured effective voltage value within U... set Two bushings within the allowable deviation range are divided into a reference group; the effective value of the measured voltage is attenuated and set at 0.5U. set The remaining four bushings within the allowable deviation range are classified as the group to be converted.
[0013] Preferably, in step S3, the effective value of the leakage current of the group to be converted under half voltage is mapped to the nonlinear normalized model under the full voltage reference, which is expressed as formula (1):
[0014] (1);
[0015] In the formula, I n-i I represents the normalized leakage current amplitude after conversion for the i-th bushing; i U is the measured effective value of the leakage current of the i-th bushing obtained in step S1; i U is the effective value of the measured voltage of the i-th bushing obtained in step S1; base γ is the arithmetic mean of the measured effective values of the reference group voltage, which serves as the full voltage reference; γ is a preset nonlinear amplitude correction coefficient.
[0016] The formula for determining the value of the voltage level-related nonlinear correction exponent γ is as follows:
[0017] (2);
[0018] U N The rated voltage level of the grid-side bushing of the converter transformer under test; U ref γ0 is the preset reference voltage level; γ0 is the basic capacitive response constant, representing the pure capacitive characteristics of the insulating medium under ideal conditions; μ is the insulation structure influence coefficient, used to dynamically quantify the degree of nonlinear leakage current distortion caused by the increase of the rated voltage level of the equipment and the increase of the insulation thickness.
[0019] Preferably, in step S4, the process of constructing the dynamic transverse current baseline includes:
[0020] The normalized leakage current amplitude sets of the six bushings are sorted in ascending order to obtain the sequence {I1, I2, I3, I4, I5, I6}, where I1 is the minimum value and I6 is the maximum value. After removing the maximum and minimum values, the dynamic transverse current baseline I... ref Calculated using formula (3):
[0021] (3).
[0022] Preferably, in step S5, the percentage of amplitude deviation D of a single bushing is calculated. i Described by formula (4):
[0023] (4);
[0024] In the formula, I norm-i I is the normalized leakage current amplitude of a single bushing. ref For dynamic transverse current baseline; when the amplitude deviation percentage D i When the amplitude deviation is greater than the first preset threshold and less than or equal to the second preset threshold, an insulation degradation warning signal is output; when the amplitude deviation percentage D i When the value exceeds the second preset threshold, it is determined that the corresponding bushing has an insulation abnormality and alarm location information is output.
[0025] Compared with the prior art, the present invention has the following beneficial effects:
[0026] This invention abandons the reliance of traditional diagnostics on high-precision phase measurement. Addressing the non-uniform voltage conditions caused by parallel pressurization in systems with fewer decoupling stages, it innovatively constructs a nonlinear amplitude normalization model associated with voltage levels. For the first time, it achieves lateral isomorphic benchmarking of bushing leakage current characteristics under different field strengths. Simultaneously, by combining a dynamic lateral baseline to remove extreme values with a deviation diagnostic mechanism that introduces background tolerance, it naturally offsets the interference of field environmental parameter drift and effectively avoids false alarms caused by minor fluctuations. Thus, with extremely low hardware requirements and extremely high field robustness, it achieves accurate and rapid screening and highly reliable joint diagnosis of insulation defects in converter transformer bushing groups. Attached Figure Description
[0027] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation
[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Further details of the technical solutions of the present invention will be described below.
[0029] Example 1
[0030] The present invention provides a method for diagnosing the insulation of converter transformer bushing groups under unequal voltage conditions, such as... Figure 1 As shown, taking the induction pressure test during the annual overhaul of a ±800kV converter station pole system as an example, the specific implementation steps are as follows:
[0031] S1: Based on the star-delta interconnection topology of six converter transformers in the valve hall, under the single-phase pressurization condition on the valve side, the operating voltage signal and the leakage current signal of the six grid-side bushings of the whole system are collected synchronously, and the effective voltage value and effective current value of each are calculated respectively to construct a pure amplitude dataset that is independent of phase dependence.
[0032] In this embodiment of the invention, synchronous amplitude data under non-uniform voltage conditions of the bushing group is acquired. Under single-phase pressurization conditions on the valve side, based on a unified system clock synchronization triggering mechanism, a signal adapter directly screwed to the end screen interface of the grid-side bushing and a high-voltage side voltage divider are used to synchronously acquire steady-state waveform data of the operating voltage and end screen leakage current of the grid-side bushings of the six converter transformers (denoted as A1, A2, B1, B2, C1, C2) in the entire system. A signal data window containing an integer number of complete power frequency cycles is extracted. After removing transient interference, the effective voltage value (U) of each bushing is directly calculated and extracted based on the time-domain waveform. i ) and the effective value of current (I) i This step completely eliminates the small phase angle extraction process, which is highly susceptible to on-site interference, and constructs a pure amplitude feature dataset that is completely independent of phase dependence.
[0033] S2: Based on the induced voltage distribution pattern of two high and four low, the effective value of the measured voltage of each bushing is extracted and compared, and the six bushings are divided into a reference group that bears the full voltage and a group to be converted that bears half voltage.
[0034] In this embodiment of the invention, voltage grouping and feature mapping preparation based on topological features are implemented. The target induced voltage on the network side for this experiment is set as U. set =10kV. The effective values of the measured voltages of the six bushings obtained from S1 are extracted and compared. Due to the valve-side angle closed-loop voltage division effect, the effective values of the measured voltages of bushings A1 and A2 are approximately 10kV, and they are classified as the baseline group; the effective values of the measured voltages of bushings B1, B2, C1, and C2 are attenuated to approximately 5kV, and they are uniformly classified as the group to be converted.
[0035] S3: Using the measured effective voltage value of the reference group as the full voltage reference, a nonlinear amplitude correction coefficient is introduced to map and convert the effective value of the leakage current of the group to be converted at half voltage to the full voltage reference, thereby eliminating the absolute amplitude difference caused by the non-uniform distribution of excitation voltage.
[0036] In this embodiment of the invention, a normalized leakage current model based on voltage response characteristics is constructed. The arithmetic mean of the measured effective voltage values of the reference group (A1, A2) is taken as the full-voltage reference reference U. base The effective value of the leakage current of the group to be converted at half voltage is mapped and converted to the full voltage reference. Considering the difference in the thickness of the insulating paper layer of bushings at different voltage levels, this embodiment constructs a normalized model that includes nonlinear compensation for field strength:
[0037] (1);
[0038] In the formula, I n-i I represents the normalized leakage current amplitude after conversion for the i-th bushing; iU is the measured effective value of the leakage current of the i-th bushing obtained in step S1; i U is the effective value of the measured voltage of the i-th bushing obtained in step S1; base γ is the arithmetic mean of the measured effective voltage values of the "reference group", which serves as the full voltage reference; γ is a preset nonlinear amplitude correction coefficient.
[0039] The formula for determining the voltage level-related nonlinear correction exponent γ is as follows:
[0040] (2);
[0041] U N The rated voltage level of the grid-side bushing of the converter transformer under test; U ref γ0 is the preset reference voltage level; γ0 is the basic capacitive response constant, which is preferably set to 1 in this embodiment, representing the pure capacitive characteristics of the insulating medium under ideal conditions; μ is the insulation structure influence coefficient, which is empirically preferably set to 0.05 in this embodiment, and is used to dynamically quantify the degree of nonlinear leakage current distortion caused by the increase of the rated voltage level of the equipment and the increase of the insulation thickness.
[0042] Using this model, the measured current I of the four bushings in the group to be converted was... i By performing individual calculations, the normalized leakage current amplitude I can be obtained. norm-i As for the two bushings in the benchmark group, due to their U... i It is itself close to U base The normalized current after conversion is basically equal to the measured current. This effectively eliminates the absolute amplitude difference caused by the non-uniform distribution of the excitation voltage, achieving lateral isomorphism of the leakage current characteristics of the six bushings.
[0043] S4: Based on the normalized leakage current amplitude of the six bushings, the extreme value removal and averaging algorithm is adopted. After removing the maximum and minimum values in the six normalized current data, the arithmetic mean of the remaining four valid data is calculated to construct a dynamic transverse current baseline that reflects the overall health level of the bushing group in this batch.
[0044] In this embodiment of the invention, the set of normalized leakage current amplitudes of the six bushings calculated in S3, {I1, I2, I3, I4, I5, I6}, is arranged in ascending order of numerical value. Using an extreme value removal and averaging algorithm, after removing the minimum and maximum values, the arithmetic mean of the remaining four valid data points is calculated to obtain the dynamic transverse current baseline.
[0045] (3);
[0046] This baseline naturally offsets the population leakage conductivity drift caused by the current ambient temperature and humidity, and effectively eliminates extreme value interference from potentially faulty equipment.
[0047] S5: Calculate the percentage deviation of the normalized leakage current amplitude of a single bushing relative to the baseline of the group lateral current; when the percentage deviation of the bushing amplitude exceeds the preset diagnostic threshold, it is determined that the corresponding bushing has an abnormal insulation condition.
[0048] In this embodiment of the invention, the percentage D of amplitude deviation of a single bushing is calculated. i Described by formula (4):
[0049] (4);
[0050] In the formula, I norm-i I is the normalized leakage current amplitude of a single bushing. ref This is a dynamic lateral current baseline; the preset first diagnostic threshold is 15%, and the second diagnostic threshold is 30%. When the amplitude deviation percentage D... i When the amplitude deviation is greater than the first preset threshold and less than or equal to the second preset threshold, an insulation degradation warning signal is output; when the amplitude deviation percentage D i When the value exceeds the second preset threshold, it is determined that the corresponding bushing has an insulation abnormality and alarm location information is output.
Claims
1. A method for diagnosing the insulation of converter transformer bushing groups under non-uniform voltage conditions, characterized in that, Includes the following steps: S1: Based on the star-delta interconnection topology of six converter transformers in the valve hall, under the single-phase pressurization condition on the valve side, the operating voltage signal and the leakage current signal of the six grid-side bushings of the whole system are collected synchronously, and the effective voltage value and effective current value of each are calculated respectively to construct a pure amplitude dataset that is independent of phase dependence. S2: Based on the induced voltage distribution pattern of two high and four low, the effective value of the measured voltage of each bushing is extracted and compared, and the six bushings are divided into a reference group that bears the full voltage and a group to be converted that bears half voltage. S3: Using the measured effective voltage value of the reference group as the full voltage reference, a nonlinear amplitude correction coefficient is introduced to map and convert the effective value of the leakage current of the group to be converted at half voltage to the full voltage reference, thereby eliminating the absolute amplitude difference caused by the non-uniform distribution of excitation voltage. S4: Based on the normalized leakage current amplitude of the six bushings, the extreme value removal and averaging algorithm is adopted. After removing the maximum and minimum values in the six normalized current data, the arithmetic mean of the remaining four valid data is calculated to construct a dynamic transverse current baseline that reflects the overall health level of the bushing group in this batch. S5: Calculate the percentage deviation of the normalized leakage current amplitude of a single bushing relative to the baseline of the group lateral current; when the percentage deviation of the bushing amplitude exceeds the preset diagnostic threshold, it is determined that the corresponding bushing has an abnormal insulation condition.
2. The method according to claim 1, characterized in that: Step S1 includes: acquiring steady-state voltage and current waveform data of six bushings containing a complete power frequency cycle based on a unified synchronous triggering mechanism; after removing transient interference, directly calculating and extracting the effective voltage and effective current values of each bushing based on the time-domain waveform data as pure amplitude features that are independent of phase dependence.
3. The method according to claim 1, characterized in that: In step S2, the grouping rule for dividing the six bushings into a baseline group and a conversion group is as follows: the target voltage for the test is set to U. set Compare the measured effective voltage values of the six bushings, and set the measured effective voltage value within U... set Two bushings within the allowable deviation range are divided into a reference group; the effective value of the measured voltage is attenuated and set at 0.5U. set The remaining four bushings within the allowable deviation range are classified as the group to be converted.
4. The method according to claim 1, characterized in that: In step S3, the effective value of the leakage current of the group to be converted under half voltage is mapped to the nonlinear normalized model under the full voltage reference, which is expressed as formula (1): (1); In the formula, I n-i I represents the normalized leakage current amplitude after conversion for the i-th bushing; i U is the measured effective value of the leakage current of the i-th bushing obtained in step S1; i U is the effective value of the measured voltage of the i-th bushing obtained in step S1; base The arithmetic mean of the measured effective voltage values of the reference group serves as the full voltage reference; γ is a preset nonlinear amplitude correction index. The formula for determining the voltage level-related nonlinear correction index γ is as follows: (2); U N The rated voltage level of the grid-side bushing of the converter transformer under test; U ref γ0 is the preset reference voltage level; γ0 is the basic capacitive response constant, representing the pure capacitive characteristics of the insulating medium under ideal conditions; μ is the insulation structure influence coefficient, used to dynamically quantify the degree of nonlinear leakage current distortion caused by the increase of the rated voltage level of the equipment and the increase of the insulation thickness.
5. The method according to claim 1, characterized in that: In step S4, the process of constructing a dynamic transverse current baseline includes: The normalized leakage current amplitude sets of the six bushings are sorted in ascending order to obtain the sequence {I1, I2, I3, I4, I5, I6}, where I1 is the minimum value and I6 is the maximum value. After removing the maximum and minimum values, the dynamic transverse current baseline I... ref Calculated using formula (3): (3)。 6. The method according to claim 1, characterized in that: In step S5, the percentage of amplitude deviation D for a single bushing is calculated. i Described by formula (4): (4); In the formula, I norm-i I is the normalized leakage current amplitude of a single bushing. ref For dynamic transverse current baseline; When the amplitude deviation percentage D i When the value is greater than the first preset threshold and less than or equal to the second preset threshold, an insulation degradation warning signal is output. When the amplitude deviation percentage D i When the value exceeds the second preset threshold, it is determined that the corresponding bushing has an insulation abnormality and alarm location information is output.