Defect detection method based on circuit breaker bundle collet

Abstract

The present application belongs to the technical field of circuit breaker defect detection, and relates to a defect detection method based on a circuit breaker bundle-shaped chuck, specifically comprising: collecting total contact voltage drop during closing and each contact finger shunt current, calculating contact resistance of each contact finger at each time and generating a time sequence; intercepting a synchronous window in a current stable interval, calculating inter-plate dispersion of this closing operation, and contact resistance fluctuation amplitude of each contact finger in the window; comparing the current inter-plate dispersion with an adaptive dynamic baseline to determine abnormalities, outputting a clamping force degradation identifier when the number of abnormal accumulations or continuous deviations exceeds the tolerance limit; positioning the degradation target position according to the fluctuation amplitude, and determining the degradation level according to the incremental accumulation rate of the inter-plate resistance dispersion in continuous operation, and pushing to the operation and maintenance terminal. The present application breaks through the masking of parallel shunt compensation effect, realizes online detection, accurate positioning and graded early warning of single contact finger defects.

Claims

1. A method for defect detection based on a circuit breaker bundle collet, characterized in that, include: The total contact voltage drop between the moving and stationary contacts and the real-time shunt current value of each contact finger are obtained when the circuit breaker is closed and under load. The total contact voltage drop is divided by the shunt current value of each contact finger at each sampling time to generate the contact resistance time series of each contact finger. Within the current stable conduction interval, a synchronous observation window is extracted. The median of the instantaneous contact resistance dispersion of all fingers at each sampling time within the window is calculated as the inter-finger resistance dispersion. The contact resistance fluctuation amplitude of each finger within the window is also calculated. The inter-chip resistance dispersion of this closing operation is compared with the adaptively updated dynamic baseline by statistical deviation to determine whether an abnormal event has occurred. The dynamic baseline is established and adaptively updated in the following manner: In the initial preset number of closing operations after the circuit breaker is put into operation, the current level is divided according to the total closing current, and the inter-chip resistance dispersion obtained in each closing operation under the current current level is recorded as a dispersion sequence according to the closing operation time sequence. For each current range, the first inter-chip resistance dispersion is used as the initial center value, the initial fluctuation variance is set to a preset initial variance value, and the initial observation count is set to one. Each time an inter-chip resistance dispersion is added, the observation count is incremented by one. The center value is recursively corrected based on the difference between the added dispersion and the center value before the update. The fluctuation variance is recursively corrected based on the deviation between the added dispersion and the updated variance. The updated center value, fluctuation variance, and observation count are used to overwrite the original storage of the corresponding current range, serving as the dynamic baseline for the next comparison. When the cumulative number of abnormal events reaches the cumulative criterion threshold, or when the upward offset of the dispersion of consecutive preset number of closing operations exceeds the single tolerance limit, the clamping force degradation status indicator is output. Mark at least one contact finger with the largest contact resistance fluctuation amplitude when the output clamping force degradation status is identified as the degradation target location, and determine the degradation level based on the incremental accumulation rate of inter-finger resistance dispersion in continuous operation, and push it to the operation and maintenance terminal. The determination of the degradation level based on the incremental accumulation rate of inter-chip resistance dispersion during continuous operation includes: From the first recorded abnormal event, track the inter-chip resistance dispersion in each subsequent closing operation and extract the incremental sequence of inter-chip resistance dispersion with the number of closing operations; The ratio of the cumulative increment value of the increment sequence to the cumulative number of closing operations is calculated as the increment accumulation rate; The incremental accumulation rate is compared with the discrete growth rate range obtained by the same type of bundled clamp in accelerated aging test. The discrete growth rate range is divided by a first growth rate boundary corresponding to the stress relaxation rate of the clamping spring and a second growth rate boundary corresponding to the wear rate of the contact surface of the finger, wherein the first growth rate boundary is smaller than the second growth rate boundary. Rate ranges below the first growth rate threshold are mapped to the first degradation level, rate ranges between the first and second growth rate thresholds are mapped to the second degradation level, and rate ranges above the second growth rate threshold are mapped to the third degradation level. The degradation level corresponding to the rate range in which the current incremental cumulative rate is located is used as the degradation level determined in this instance.

2. The circuit breaker bundle collet based defect detection method of claim 1, wherein, The generation of the contact resistance time series for each finger includes: The effective current lower limit is set based on the sum of the lower limit of the current sensing unit's range and the noise floor. At each sampling time, the instantaneous contact resistance corresponding to the finger whose shunt current value is less than the effective current lower limit is marked as invalid; For a contact finger whose shunt current value is greater than or equal to the effective lower current limit, the total contact voltage drop is divided by the shunt current value at the sampling time, and the quotient is taken as the instantaneous contact resistance of the contact finger at the sampling time. The instantaneous contact resistance value of each finger at each sampling time is associated with the finger number and the sampling time sequence number and stored to form a contact resistance time series.

3. The circuit breaker bundle collet based defect detection method of claim 1, wherein, The stable current conduction interval is determined by the following method: From the moment the closing operation is triggered, the total closing current is obtained by summing the real-time shunt current values ​​of each contact finger, and the absolute change of the total closing current between adjacent sampling moments is calculated. The product of the accuracy percentage of the current sensing unit and the full-scale value of the range is used as the current stability judgment threshold. When the absolute change is less than the current stability judgment threshold within a continuous preset number of sampling times, it is determined that the current has entered a stable conduction state. The period from the moment the circuit breaker enters the stable conduction state until the circuit breaker trips or the total closing current falls below a preset proportion of the stable conduction initiation current value is defined as the current stable conduction interval.

4. The defect detection method based on the circuit breaker bundle clamp according to claim 2, characterized in that, The inter-chip resistance dispersion is obtained in the following way: Within the synchronous observation window, the instantaneous contact resistance values ​​of each finger at each sampling time are collected into a set, and the instantaneous contact resistance values ​​marked as invalid values ​​in the set are removed. If the number of valid values ​​in the set at a certain sampling time is less than the minimum statistical coverage, then skip that sampling time. For sampling times that are not skipped, the mean of all valid values ​​in the set is calculated as the average contact resistance. The average contact resistance is subtracted from each valid value and the absolute value is taken. The mean of the resulting absolute value sequence is calculated to obtain the dispersion of the inter-finger contact resistance at the sampling time. The median of the inter-finger contact resistance dispersion corresponding to all sampling times that were not skipped within the window is taken as the inter-chip resistance dispersion.

5. The defect detection method based on circuit breaker bundle clamps according to claim 4, characterized in that, The calculation of the contact resistance fluctuation amplitude of each contact finger within the window includes: For each contact finger, the maximum and minimum values ​​of the instantaneous contact resistance at all sampling times that were not skipped within the synchronous observation window are taken, and the difference between the maximum and minimum values ​​is taken as the contact resistance fluctuation amplitude.

6. The defect detection method based on the circuit breaker bundle clamp according to claim 1, characterized in that, The determination of whether an abnormal event has occurred includes: Determine the current range based on the total closing current of this closing operation, and retrieve the currently stored center value and fluctuation variance for the corresponding current range. Calculate the absolute difference between the inter-chip resistance dispersion and the center value, and divide the absolute difference by the fluctuation variance to obtain the standardized deviation. Based on the statistical upper limit of the variation range of inter-piece resistance dispersion of the same type of bundled clamp under multiple normal conditions, the abnormal offset of the current range is set. The standardized deviation and the outlier offset are calculated by difference. Based on the result of the difference calculation, the probability ratio of this observation under the normal hypothesis and the outlier hypothesis is determined, and the logarithm of the probability ratio is taken as the single test increment. The single test increment is added to the previous cumulative test value to obtain a new cumulative test value. The new cumulative test value is compared with the upper decision boundary and the lower decision boundary, which are determined according to the preset allowable false alarm rate and allowable false negative rate. When the new cumulative check value is greater than the upper decision boundary, an abnormal event is determined to occur, and the cumulative check value is cleared to zero. When the new cumulative test value is less than the lower decision boundary, it is determined that no abnormal event has occurred. The cumulative test value is cleared to zero, and the current inter-chip resistance dispersion is taken as a normal value. The center value and fluctuation variance of the corresponding current range are updated. When the new cumulative test value is between the lower and upper decision limits, the current cumulative test value remains unchanged, and the system waits for the next closing operation to continue accumulating.

7. The defect detection method based on the circuit breaker bundle clamp according to claim 1, characterized in that, The single-time tolerance limit is confirmed in the following ways: Under the same current range, the sum of the center value in the dynamic baseline and the upper limit of the standard deviation of the inter-piece resistance dispersion obtained by the same type of bundled clamp in the factory test during multiple closing operations in the factory inspection report of the circuit breaker is used as the single tolerance limit. If the standard deviation upper limit is not recorded in the factory inspection report, then within the first preset operation and maintenance cycle after the circuit breaker is put into operation, the range of the inter-segment resistance dispersion obtained from the previous preset number of closing operations will be taken, and the sum of the range and the center value in the dynamic baseline will be used as the single tolerance limit.

8. The defect detection method based on the circuit breaker bundle clamp according to claim 1, characterized in that, The degradation target location includes the following marking process: In this closing operation that outputs the clamping force degradation status indicator, the contact resistance fluctuation amplitude of all contacts is read, and the contact resistance fluctuation amplitude of each contact is sorted from largest to smallest to construct a descending sequence of fluctuation amplitude. The first finger in the descending sequence is taken as the main degradation target. If the difference between the contact resistance fluctuation amplitude of the second finger in the descending sequence and the contact resistance fluctuation amplitude of the adjacent finger meets the jump criterion, the second finger is included in the degradation target range. The process is repeated down the descending sequence until the current adjacent difference does not meet the jump criterion. All tentacles included in the degradation target range before the recursion stops, together with the main degradation target, are used as the degradation target position.

Images