Intelligent monitoring system for ring main unit fault
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANDONG AOLAIEN INTELLIGENT TECH CO LTD
- Filing Date
- 2026-06-09
- Publication Date
- 2026-07-10
Smart Images
Figure CN122361981A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power equipment condition monitoring technology, specifically a ring main unit fault intelligent monitoring system. Background Technology
[0002] Ring main units (RMS) are widely used in power distribution networks to perform functions such as power distribution, line protection, and load interruption. As a key component of the RMS, the reliability of the circuit breaker's opening and closing actions directly affects the operational safety of the power distribution equipment. During long-term operation, circuit breakers are susceptible to faults such as core jamming, contact deterioration, and transmission mechanism jamming due to factors such as accumulated operating frequency, mechanical wear, changes in lubrication, changes in contact condition, and operating temperature variations.
[0003] Existing ring main unit fault monitoring methods typically focus on the acquisition and judgment of single state quantities. For example, the state of the opening and closing coils can be judged based on the coil current waveform, or the contact temperature can be judged based on the contact heating situation, or the mechanism action can be judged based solely on the opening and closing time.
[0004] However, existing technologies still have the following limitations: 1. Iron core jamming usually causes synchronous abnormalities in coil current characteristics and operating timing characteristics; contact deterioration usually manifests as prominent abnormal temperature rise in contacts or conductive joints; jamming of the transmission mechanism is more likely to manifest as abnormal mechanism operating timing and inconsistencies between abnormal current and abnormal operation. If only a single threshold or a single state quantity is relied upon for judgment, it is easy to cause misjudgment of fault type or inaccurate warning level.
[0005] 2. The operating characteristics of a circuit breaker are affected by the cumulative number of operations and operating temperature conditions. Even if the circuit breaker is in a normal state, its opening and closing time, coil current waveform, mechanism stroke curve and temperature rise characteristics may differ at different life stages and temperature conditions. If a fixed reference template is used for comparison, the accuracy of anomaly identification may be reduced. Summary of the Invention
[0006] To overcome the shortcomings in the background art, embodiments of the present invention provide an intelligent monitoring system for ring main unit faults, which can effectively solve the problems involved in the background art.
[0007] The objective of this invention can be achieved through the following technical solution: a ring main unit fault intelligent monitoring system, comprising: a status acquisition module, a feature coupling module, a joint discrimination module, and an early warning output module.
[0008] The status acquisition module is connected to the feature coupling module, the feature coupling module is connected to the joint discrimination module, and the joint discrimination module is connected to the early warning output module.
[0009] The status acquisition module acquires the current waveform of the opening and closing coils of the ring main unit circuit breaker, the opening and closing time, the stroke curve of the operating mechanism, the temperature data of the moving and stationary contacts and joints, and the cumulative value of the number of operations.
[0010] The feature coupling module extracts current features based on the current waveform of the opening and closing coils, extracts action features based on the opening time, closing time and the stroke curve of the operating mechanism, extracts temperature rise features based on the temperature data of the moving and stationary contacts and joints, and uses the temperature rise features to perform time-series correction on the action features to form an electromechanical-thermal coupling feature vector.
[0011] The joint discrimination module matches the corresponding historical normal action templates based on the cumulative value of the number of actions and temperature data. It then quantifies and compares the electromechanical-thermal coupling feature vector with the reference features corresponding to the template to obtain the current offset, action timing deviation, and temperature rise anomaly. Based on the coupling anomaly relationship between the three, it determines the mechanical fault type, which includes core jamming fault, contact deterioration fault, and transmission mechanism jamming fault.
[0012] The early warning output module outputs the abnormality level based on the type of mechanical fault.
[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention obtains the current waveform of the opening and closing coil, the opening and closing time, the stroke curve of the operating mechanism, the temperature data of the moving and stationary contacts and the joint, and the cumulative value of the number of actions through the status acquisition module. It comprehensively characterizes the operating status of the ring network cabinet circuit breaker from the aspects of coil drive, mechanism movement, contact heating and mechanical life accumulation, and provides a multi-source status data basis for fault identification.
[0014] (2) The present invention constructs a coupled discrimination relationship based on current offset, action timing deviation and temperature rise anomaly, and obtains the discrimination value of iron core jamming fault, contact deterioration fault and transmission mechanism jamming fault respectively. The mechanical fault type is determined by the fault type corresponding to the main discrimination value, which is conducive to improving the accuracy of fault type identification.
[0015] (3) The present invention matches the corresponding historical normal action templates based on the cumulative value of the number of actions and temperature data, so that the current action features can be quantitatively compared with the reference features under the same or similar operating conditions, thereby reducing misjudgment caused by differences in the number of actions and temperature. Attached Figure Description
[0016] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.
[0017] Figure 1 This is a schematic diagram of the module connection of the present invention;
[0018] Figure 2 This is a flowchart of the method for forming electromechanical-thermal coupling feature vectors according to the present invention;
[0019] Figure 3 This is a flowchart of the method for obtaining current offset, timing deviation and temperature rise anomaly according to the present invention. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] Reference Figure 1 As shown, the present invention provides an intelligent monitoring system for ring main unit faults, including: a status acquisition module, a feature coupling module, a joint discrimination module, and an early warning output module.
[0022] The status acquisition module is connected to the feature coupling module, the feature coupling module is connected to the joint discrimination module, and the joint discrimination module is connected to the early warning output module.
[0023] The status acquisition module acquires the current waveform of the opening and closing coils of the ring main unit circuit breaker, the opening and closing time, the stroke curve of the operating mechanism, the temperature data of the moving and stationary contacts and joints, and the cumulative value of the number of operations.
[0024] Considering that the current waveform of the opening and closing coil can reflect the movement state of the iron core, the opening and closing time and the stroke curve of the operating mechanism characterize the action characteristics of the mechanism, the temperature of the moving and stationary contacts and joints reflects contact heating, and the cumulative value of the number of actions is used to assess the mechanical wear of the circuit breaker and provide a basis for matching the corresponding historical normal template, the comprehensive electromechanical and thermal data provide a basis for three-state coupling analysis, accurately identifying iron core jamming, contact deterioration and transmission mechanism jamming faults.
[0025] The status acquisition module includes current sensors respectively installed in the circuit breaker's tripping coil circuit and closing coil circuit. When the circuit breaker receives a tripping or closing control command, the current sensors acquire the current change signal of the corresponding coil to form a tripping coil current waveform or a closing coil current waveform. The moment when the current signal first exceeds a preset current threshold is determined as the energization start moment.
[0026] The timing controller receives the coil energizing signal, the auxiliary switch state flipping signal, and the contact position signal respectively. When the coil energizing signal has a rising edge, the timing controller starts timing. During the opening process, the auxiliary switch state flipping signal is used as the timing endpoint to obtain the opening time. During the closing process, the contact position signal is used as the timing endpoint to obtain the closing time.
[0027] An angular displacement sensor installed on the main drive shaft of the operating mechanism continuously collects the rotation angle change data of the main drive shaft during the circuit breaker opening and closing processes, and generates the operating mechanism stroke curve according to the acquisition time sequence.
[0028] Temperature sensors are installed at the moving contact, stationary contact, conductive joint, and the environment inside the cabinet to collect the temperatures of the moving contact, stationary contact, conductive joint, and the environment inside the cabinet. The temperature rise value of each measuring point is calculated by subtracting the temperature from the ambient temperature inside the cabinet at the same sampling time. The temperature rise value reflects the contact heating of the contacts and joints.
[0029] The counter is connected to the coil energizing signal and the contact state transition signal. When the coil current is detected to be greater than the preset operating current threshold and a complete contact opening or closing state transition signal is received, a valid operation is confirmed to be completed, and the number of operations is accumulated to obtain the cumulative value of the number of operations.
[0030] It should be noted that the preset current threshold and preset operating current threshold are obtained through experimental calibration. Specifically, under normal circuit breaker operation, multiple opening and closing tests are performed, and background current data during the coil de-energized phase, operating current waveform after coil energization, and contact state transition signals are collected simultaneously.
[0031] The background current data during the unenergized phase of the coil is filtered, and the maximum fluctuation value of the background current is extracted as the noise upper limit. The initial energized current of the coil during multiple normal opening and closing tests is statistically analyzed to obtain the mean and standard deviation of the background current. The sum of the mean background current and three times the standard deviation of the background current is used as the preset current threshold.
[0032] The coil current waveforms corresponding to the effective opening and closing actions that complete the contact state transition are statistically analyzed, the peak current in each effective action waveform is extracted, and the lower limit of the peak current in the effective action sample is used as the preset action current threshold.
[0033] The feature coupling module extracts current features based on the current waveform of the opening and closing coils, extracts action features based on the opening time, closing time and the stroke curve of the operating mechanism, extracts temperature rise features based on the temperature data of the moving and stationary contacts and joints, and uses the temperature rise features to perform time-series correction on the action features to form an electromechanical-thermal coupling feature vector.
[0034] Considering that a single physical quantity characteristic is difficult to point to a specific fault, coupling the electromechanical and thermal characteristics of coil current, travel time, and contact temperature rise can characterize the three-state coordinated changes and abnormal correlations, providing input for identifying fault types by utilizing the coupling relationship of current offset, timing deviation, and temperature rise anomaly, and accurately distinguishing between core jamming, contact deterioration, and transmission mechanism jamming faults.
[0035] The extraction of current features based on the current waveform of the opening and closing coils is specifically as follows: After the current sensor acquires the original current sequence of the opening coil and the original current sequence of the closing coil, the original current sequence is first subjected to peak removal and smoothing filtering to remove sampling peaks and high-frequency interference; then, the continuous sampling data before the start of energization and in the state where the coil is not energized is selected, and its average current is calculated as the baseline current. The baseline current is then subtracted from the current value of each sampling point in the original current sequence to obtain the current waveform after baseline correction.
[0036] The energization start time is set to time zero. For the opening process, the time corresponding to the auxiliary switch state flip signal obtained from the timer controller is taken as the action completion time. For the closing process, the time corresponding to the contact position signal obtained from the timer controller is taken as the action completion time.
[0037] The waveform segments from the start of energization to the completion of the tripping action in the tripping coil current waveform and the waveform segments from the start of energization to the completion of the closing action in the closing coil current waveform after baseline correction are respectively extracted as the corresponding effective action waveforms.
[0038] Feature extraction is performed on the effective action waveforms of opening and closing respectively. The effective action waveforms are represented as a set of sampling points arranged in the order of sampling time, where each sampling point includes the sampling time and the corresponding current value.
[0039] In the set of sampling points, the maximum value of the current is extracted as the peak current, and the time of the sampling point corresponding to the peak current is taken as the peak occurrence time.
[0040] The average current is obtained by summing the current values of all sampling points in the effective action waveform and dividing the sum of the current values of all sampling points by the number of sampling points. The product of the average current of adjacent sampling points and the corresponding time interval is accumulated to obtain the current integral.
[0041] The difference between the current value at each sampling point and the average current is calculated, and each difference is squared. Then, all the squared differences are summed and divided by the number of sampling points to obtain the current variance. Finally, the square root of the current variance is performed to obtain the standard deviation of the effective action waveform, and this standard deviation is used as the current fluctuation.
[0042] Peak current, peak occurrence time, average current, current integral, and current fluctuation are used as current characteristics.
[0043] The extraction of motion features based on opening time, closing time, and operating mechanism stroke curve specifically involves smoothing the operating mechanism stroke curves of opening and closing actions.
[0044] The first and last sampling points in the stroke curve of the operating mechanism are determined according to the sampling time sequence. The first sampling point is the sampling point in the stroke curve of the operating mechanism that is closest to the energization start time of the corresponding action, and the last sampling point is the sampling point in the stroke curve of the operating mechanism that is closest to the action completion time of the corresponding action.
[0045] For the tripping action, the completion time is the time corresponding to the auxiliary switch state flip signal; for the closing action, the completion time is the time corresponding to the contact position signal.
[0046] Extract the sampling time and angle value corresponding to the first sampling point, and use them as the start time and start angle of the action, respectively; extract the sampling time and angle value corresponding to the last sampling point, and use them as the completion time and termination angle of the action, respectively.
[0047] This provides the start and end times and start and end angles of the opening and closing actions under the same time base, laying the foundation for subsequent calculations of the mechanism's action duration, total stroke, maximum action speed, average action speed, and speed fluctuation.
[0048] For any opening or closing action, calculate the time difference between the completion time and the start time of the action, and use the time difference as the duration of the mechanism action.
[0049] Calculate the difference between the ending angle and the starting angle, and take the absolute value of the difference as the total travel.
[0050] After obtaining the smoothed stroke curve of the operating mechanism, each sampling point between the start time and the completion time of the action is selected according to the sampling time sequence, and the ratio of the angle difference between two adjacent sampling points to the sampling time difference is calculated in turn to obtain the corresponding discrete angular velocity value. The discrete angular velocity values are arranged in time sequence to form an angular velocity curve.
[0051] To avoid affecting the consistency of speed characteristics between opening and closing actions due to opposite directions of angular change, the absolute value of the discrete angular velocity is taken and used as the action speed value for the corresponding sampling interval.
[0052] In the angular velocity curve, the maximum value of the motion velocity is extracted as the maximum motion velocity; the average motion velocity is obtained by summing all motion velocity values from the start time to the completion time of the motion and dividing by the number of motion velocity values.
[0053] Calculate the difference between each motion speed value and the average motion speed, square each difference, sum all the squared results, divide by the number of motion speed values, and then perform a square root operation to obtain the standard deviation of the angular velocity curve, and use the standard deviation as the velocity fluctuation.
[0054] The opening time, closing time, mechanism action duration, total stroke, maximum action speed, average action speed, and speed fluctuation are used as action characteristics.
[0055] It should be noted that using the absolute values of time difference and angle difference to characterize the mechanism's action duration and total stroke amount respectively can avoid the problem of inconsistent stroke amount signs caused by the opposite direction of angle change between opening and closing actions, thus making the stroke characteristics under different action directions comparable.
[0056] The extraction of temperature rise characteristics based on temperature data of moving and stationary contacts and joints specifically involves: adding sampling time markers to the temperature data of each measuring point collected by temperature sensors at moving contacts, stationary contacts, conductive joints and the environment inside the cabinet, forming a time-stamped temperature data sequence.
[0057] Each temperature data sequence is filtered to remove abrupt interference points, and the removed data is then smoothed to obtain the moving contact temperature sequence, the stationary contact temperature sequence, the conductive joint temperature sequence, and the ambient temperature sequence.
[0058] Using the sampling time of temperature data at each measuring point as the reference time, the ambient temperature data in the ambient temperature sequence corresponding to the reference time are matched; when the sampling time of the ambient temperature data is inconsistent with the reference time, linear interpolation is performed based on the ambient temperature data before and after the reference time to obtain the ambient temperature at the corresponding reference time.
[0059] At the same reference time, the differences between the moving contact temperature, stationary contact temperature, and conductive joint temperature and the corresponding ambient temperature are calculated respectively, and the differences are used as the temperature rise values of the corresponding measuring points. When the difference is less than zero, it indicates that there is no effective temperature rise of the measuring point relative to the ambient temperature. In order to eliminate the influence of measurement noise on subsequent statistics, the temperature rise value of the corresponding measuring point is recorded as zero.
[0060] The temperature rise values of the moving contact, the stationary contact, and the conductive joint at the same sampling time are combined into a set of temperature rise values. When the conductive joint is equipped with multiple temperature measuring points, the temperature rise values corresponding to each conductive joint are included in the set of temperature rise values.
[0061] The temperature rise values in the set are compared, and the temperature rise value with the largest value is selected as the maximum temperature rise.
[0062] The average temperature rise is obtained by summing the temperature rise values in the set and dividing by the number of measuring points.
[0063] Extract the maximum and minimum temperature rise values from the set of temperature rise values, and use the difference between the two as the temperature rise imbalance.
[0064] The ratio of the temperature rise difference between adjacent sampling times at the same measuring point to the sampling time difference is taken as the temperature rise rate, and the maximum value of the temperature rise rate among all measuring points is taken as the characteristic value of the temperature rise rate.
[0065] The maximum temperature rise, average temperature rise, temperature rise imbalance, and temperature rise rate are used as temperature rise characteristics.
[0066] Reference Figure 2 As shown, the process of using temperature rise characteristics to perform time-series correction on action characteristics and forming an electromechanical-thermal coupling feature vector is as follows: obtain the maximum temperature rise, average temperature rise, and temperature rise imbalance corresponding to the same opening or closing action.
[0067] The local temperature rise coefficient is obtained by comparing the maximum temperature rise with the average temperature rise. This local temperature rise coefficient is used to characterize the prominence of a single measuring point relative to the overall heating level. The distribution deviation coefficient is obtained by comparing the temperature rise unevenness with the average temperature rise. This distribution deviation coefficient is used to characterize the degree of inconsistency in the heat distribution among the moving contact, stationary contact, and conductive joint.
[0068] When the average temperature rise is zero, it is determined that there is no effective temperature rise effect in the current operation cycle. The local temperature rise coefficient and the distribution deviation coefficient are both set to zero to avoid abnormal ratio calculation.
[0069] After obtaining the local temperature rise coefficient and the distribution deviation coefficient, the two are summed to obtain the time series correction coefficient.
[0070] Add the timing correction factor to 1 to obtain the timing correction reference factor.
[0071] Obtain the opening time, closing time, and mechanism action duration corresponding to the same opening or closing action, and divide the opening time by the timing correction reference coefficient to obtain the corrected opening time; divide the closing time by the timing correction reference coefficient to obtain the corrected closing time; divide the mechanism action duration by the timing correction reference coefficient to obtain the corrected mechanism action duration.
[0072] After obtaining the corrected opening time, the corrected closing time, and the corrected mechanism action duration, the time-based action features and the stroke-based action features are integrated accordingly.
[0073] For the tripping action, the corrected tripping time, the corrected mechanism action duration, total stroke, maximum action speed, average action speed, and speed fluctuation are used as the corrected tripping action characteristics; for the closing action, the corrected closing time, the corrected mechanism action duration, total stroke, maximum action speed, average action speed, and speed fluctuation are used as the corrected closing action characteristics.
[0074] The total stroke, maximum operating speed, average operating speed, and speed fluctuation are obtained from the stroke curve of the corresponding operating mechanism without temperature rise timing correction; the correction process only applies to the opening time, closing time, and mechanism operating duration to maintain the original physical meaning of the stroke amplitude characteristics and speed characteristics.
[0075] The process of energizing the primary coil and completing the contact state transition is taken as an operation cycle. The operation cycle is marked with the operation type, energization start time and operation completion time. The operation type includes opening operation and closing operation.
[0076] Within the same action cycle, the current characteristics, corrected action characteristics, and temperature rise characteristics corresponding to the action type are extracted; among them, the tripping action corresponds to the tripping coil current characteristics and tripping action characteristics, the closing action corresponds to the closing coil current characteristics and closing action characteristics, and the temperature rise characteristics are obtained by using the temperature rise data of each measuring point corresponding to the time of the action cycle.
[0077] The current characteristics, corrected operating characteristics, and temperature rise characteristics are arranged in a predetermined field order, and the corresponding operating type and cumulative number of operating times are added to provide information on the current life stage of the circuit breaker and form an electromechanical-thermal coupling feature vector corresponding to the operating cycle.
[0078] It should be noted that the theoretical basis for using temperature rise characteristics to correct the timing of action characteristics is that the opening and closing action times of a circuit breaker are affected not only by the state of the operating mechanism itself, but also by the operating thermal state. The temperature rise at the moving contact, stationary contact, and conductive joints can reflect the degree of heating of the conductive contact parts. However, localized temperature rise concentration and uneven temperature rise distribution can cause changes in contact pressure, changes in the conductive connection state, and thermal deformation of adjacent structural components, thus causing thermal state-related offsets in the opening time, closing time, and mechanism action duration collected under the same mechanical state.
[0079] It should also be noted that the timing correction coefficient is the sum of the local temperature rise coefficient and the distribution deviation coefficient. The physical basis for this is as follows: the influence of temperature rise on the timing of action mainly includes two aspects: first, when the temperature rise of a local measuring point is higher than the overall temperature rise level, it indicates that there is a local heat concentration in that part, which is prone to changes in contact pressure or local thermal deformation; second, when the temperature rise distribution of different measuring points is uneven, it indicates that the contacts, joints and adjacent structural components are heated inconsistently, which is prone to uneven changes in the mechanism's fit clearance and motion resistance.
[0080] Based on the above mechanism, the ratio of the maximum temperature rise to the average temperature rise is used as the local temperature rise coefficient to characterize the prominence of the local temperature rise relative to the overall temperature rise; the ratio of the temperature rise imbalance to the average temperature rise is used as the distribution deviation coefficient to characterize the degree of heating inconsistency between different measuring points.
[0081] Since both local temperature rise concentration and uneven temperature rise distribution can cause time-related action characteristics to shift in the same direction, and both are dimensionless relative quantities normalized to the average temperature rise, without introducing additional empirical weights, the local temperature rise coefficient and the distribution deviation coefficient are superimposed with equal weights as the equivalent thermal influence quantity of the current action cycle, i.e., the timing correction coefficient.
[0082] This weighted superposition does not necessarily require a strict linear physical relationship between the temperature rise effect and the action time. Instead, it performs a first-order equivalent approximation on the thermal influence factors within the normal operation and early abnormal range of the circuit breaker. When the local temperature rise concentration or uneven temperature rise distribution increases, the timing correction coefficient increases accordingly, thereby improving the comparability of the opening time, closing time and mechanism action time under different thermal conditions.
[0083] The joint discrimination module matches the corresponding historical normal action templates based on the cumulative value of the number of actions and temperature data. It then quantifies and compares the electromechanical-thermal coupling feature vector with the reference features corresponding to the template to obtain the current offset, action timing deviation, and temperature rise anomaly. Based on the coupling anomaly relationship between the three, it determines the mechanical fault type, which includes core jamming fault, contact deterioration fault, and transmission mechanism jamming fault.
[0084] Reference Figure 3 As shown, the process of quantizing and comparing the electromechanical-thermal coupling feature vector with the reference feature corresponding to the template to obtain the current offset, action timing deviation and temperature rise anomaly is as follows: After obtaining the electromechanical-thermal coupling feature vector corresponding to the current action cycle, the reference current feature, reference action feature and reference temperature rise feature under the same action type are read from the matched historical normal action template according to the opening or closing type of the current action.
[0085] Each feature value in the current electromechanical-thermal coupling feature vector is matched with the corresponding reference value in the historical normal action template. The difference between the current feature value and the corresponding reference value is calculated, and the absolute value of the difference is taken to obtain the single absolute deviation.
[0086] To eliminate the influence of comparisons of different feature dimensions on the results, the absolute deviation of a single item is divided by the absolute value of the corresponding reference value to obtain the single item deviation value. When the corresponding reference value is zero, the minimum non-zero value among other reference feature values belonging to the same feature category as the current feature item is used as the normalization benchmark. If all reference feature values in the same feature category are zero, the single item deviation value is directly set to zero.
[0087] The individual deviation values corresponding to each feature item are classified into current-related individual deviation values, time-related individual deviation values, and temperature rise-related individual deviation values according to the feature category. Among them, the current-related individual deviation values correspond to peak current, peak occurrence time, average current, current integral, and current fluctuation; the time-related individual deviation values correspond to the corrected opening time, corrected closing time, and corrected mechanism action duration; and the temperature rise-related individual deviation values correspond to the maximum temperature rise, average temperature rise, temperature rise imbalance, and temperature rise change rate.
[0088] Since the individual deviation value is a dimensionless relative deviation, the summation of the individual deviation values under the same category and the division by the number of characteristic items of that category yield the current offset, the action timing deviation, and the temperature rise anomaly, respectively.
[0089] It should be noted that the historical normal operation template was obtained through experimental calibration. Specifically, a sample circuit breaker of the same model as the circuit breaker of the ring main unit to be monitored was selected, and after confirming that its coil, operating mechanism, contacts and conductive joints were all in normal condition, multiple opening and closing tests were performed.
[0090] During the experiment, the test samples were classified according to the interval of the cumulative number of actions and the interval of the temperature data, and the current waveform of the opening and closing coil, the opening and closing time, the stroke curve of the operating mechanism, and the temperature data of the moving and stationary contacts and joints were collected simultaneously.
[0091] The cumulative action count is divided into intervals of 500 actions, and the ambient temperature data is divided into intervals of 5°C. If the current cumulative action count or temperature data falls between two intervals, the interval closest to the current value is selected as the matching interval; if the two endpoints are equal, the higher interval is selected. If the current value exceeds all calibrated intervals, the outermost interval is selected as the matching interval.
[0092] For each test sample, the same feature extraction and time-series correction process was used to obtain the corresponding current characteristics, corrected action characteristics, and temperature rise characteristics. Statistical analysis was performed on each characteristic under the same action type, the same action frequency range, and the same temperature range. The average value of each characteristic was used as the reference characteristic value, and the standard deviation of each characteristic was used as the reference discrete value.
[0093] The action type, action frequency range, temperature range, reference current characteristics, reference action characteristics, reference temperature rise characteristics, and corresponding reference discrete quantities are associated and stored to form a historical normal action template.
[0094] The process of determining the mechanical fault type based on the abnormal coupling relationship among the three is as follows: compare the magnitude of the current offset and the action timing deviation, and take the smaller value of the two as the electromechanical covariance abnormal quantity.
[0095] Calculate the difference between the current offset and the timing deviation, and take the absolute value of the difference to obtain the electromechanical mismatch.
[0096] Using electromechanical covariation anomaly as the main characterization quantity, and electromechanical mismatch and temperature rise anomaly as suppression quantities, the electromechanical covariation anomaly quantity is divided by 1 and the sum of electromechanical mismatch and temperature rise anomaly to obtain the core jamming fault discrimination value, which is used to characterize the degree of synchronous increase of coil drive anomaly and mechanism action timing anomaly.
[0097] The temperature rise anomaly is used as the main characterization quantity, and the electromechanical covariance anomaly quantity and the action timing deviation quantity are used as suppression quantities. The temperature rise anomaly quantity is divided by 1 and the sum of the electromechanical covariance anomaly quantity and the action timing deviation quantity to obtain the contact deterioration fault discrimination value, which is used to characterize the degree of dominance of contact or conductive joint heating anomaly relative to mechanism action anomaly.
[0098] The timing deviation and electromechanical mismatch are used as the main characterizing quantities, and the current offset is used as the suppression quantity. The sum of the timing deviation and electromechanical mismatch is divided by 1 and the sum of the current offset to obtain the jamming fault discrimination value of the transmission mechanism, which is used to characterize the degree of obvious abnormality in the timing of the mechanism's operation and the degree of inconsistency with the abnormality of the coil current.
[0099] The fault identification values for core jamming, contact deterioration, and transmission mechanism jamming are sorted from largest to smallest, and the largest value is taken as the main identification value. The fault type corresponding to the main identification value is then determined as the mechanical fault type.
[0100] When the core jamming fault discrimination value is the main discrimination value, it is determined to be a core jamming fault.
[0101] When the contact deterioration fault discrimination value is the main discrimination value, it is determined to be a contact deterioration fault.
[0102] When the main criterion value for transmission mechanism jamming fault is the criterion value, it is determined to be a transmission mechanism jamming fault.
[0103] The early warning output module outputs the abnormality level based on the type of mechanical fault.
[0104] The process of outputting the abnormality level according to the mechanical fault type is as follows: the main discrimination value is compared with the level range of the corresponding mechanical fault type. If the main discrimination value is less than the lower limit of the preset level range, the abnormality level is determined to be a level one abnormality.
[0105] If the main discriminant value is within the preset level range, the anomaly level is determined to be a level 2 anomaly.
[0106] If the main discrimination value is greater than the upper limit of the preset level range, the anomaly level is determined to be a level three anomaly.
[0107] It should be noted that the preset level range is obtained through experimental calibration. Specifically, a sample circuit breaker with the same model as the circuit breaker to be monitored is selected, and multiple opening and closing tests are conducted under normal conditions, core jamming fault, contact deterioration fault, and transmission mechanism jamming fault. At least three different degrees (slight, moderate, and severe) are set for each fault condition.
[0108] Following the same feature extraction, time-series correction, template comparison, and coupled discrimination steps as in the online monitoring process, the principal discriminant value corresponding to each test sample is calculated, and the corresponding mechanical fault type is recorded.
[0109] For the same type of mechanical fault, the tripping and closing actions are calibrated independently. The main discriminant values of all fault samples under this type are sorted in ascending order and divided into low-value, medium-value, and high-value groups, with each group having the same number of samples or differing by at most one. The low-value group corresponds to Level 1 anomaly (minor fault), the medium-value group corresponds to Level 2 anomaly (moderate fault), and the high-value group corresponds to Level 3 anomaly (serious fault).
[0110] Simultaneously, the principal discriminant value of each sample under normal conditions is calculated, and the maximum value is taken as the normal threshold.
[0111] Calculate the average of the maximum principal discriminant value in the low-value group and the minimum principal discriminant value in the median group, and compare it with the normal threshold. Select the larger of the two values as the lower limit of the preset level interval.
[0112] The average of the largest principal discriminant value in the median group and the smallest principal discriminant value in the high-value group is used as the upper limit of the preset level interval.
[0113] If the maximum principal discriminant value of the low-value group is equal to the minimum principal discriminant value of the median group, then the lower limit value is taken as this common value. It is stipulated that the principal discriminant value greater than the lower limit value is assigned to the median group, and the principal discriminant value equal to the lower limit value is assigned to the low-value group.
[0114] If the maximum principal discriminant value of the median group is equal to the minimum principal discriminant value of the high-value group, then the upper limit value is taken as this common value. It is stipulated that the principal discriminant value greater than the upper limit value is assigned to the high-value group, and the principal discriminant value equal to the upper limit value is assigned to the median group.
[0115] This yields preset grade ranges corresponding to core jamming faults, contact deterioration faults, and transmission mechanism jamming faults, respectively.
[0116] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.
Claims
1. A ring main unit fault intelligent monitoring system, characterized in that, include, The status acquisition module acquires the current waveform of the opening and closing coils of the ring main unit circuit breaker, the opening and closing time, the stroke curve of the operating mechanism, the temperature data of the moving and stationary contacts and joints, and the cumulative value of the number of operations. The feature coupling module extracts current features based on the current waveform of the opening and closing coils, extracts action features based on the opening time, closing time and the stroke curve of the operating mechanism, extracts temperature rise features based on the temperature data of the moving and stationary contacts and joints, and uses the temperature rise features to perform timing correction on the action features to form an electromechanical-thermal coupling feature vector. The joint discrimination module matches the corresponding historical normal action templates based on the cumulative value of the number of actions and temperature data. It quantifies and compares the electromechanical-thermal coupling feature vector with the reference features corresponding to the historical normal action templates to obtain the current offset, action timing deviation and temperature rise anomaly. Based on the coupling anomaly relationship between the three, it determines the mechanical fault type, which includes iron core jamming fault, contact deterioration fault and transmission mechanism jamming fault. The early warning output module outputs the abnormality level based on the type of mechanical fault.
2. The intelligent fault monitoring system for ring main unit according to claim 1, characterized in that, The status acquisition module includes: Current sensors, respectively installed in the circuit breaker's tripping and closing coil circuits, collect the current waveforms of the tripping and closing coils, and determine the moment when the current signal first exceeds a preset current threshold as the energization start moment. Connect the coil energizing signal, the auxiliary switch state flipping signal, and the contact position arrival signal to the timing controller. Use the rising edge of the coil energizing signal as the timing start point and the auxiliary switch state flipping signal as the timing end point to obtain the opening time. Use the contact position arrival signal as the timing end point to obtain the closing time. An angular displacement sensor installed on the main drive shaft of the operating mechanism continuously collects data on the rotation angle change of the main drive shaft during the circuit breaker's opening and closing processes, and generates the operating mechanism's stroke curve. Temperature sensors are installed at moving contacts, stationary contacts, conductive joints, and the environment inside the cabinet to collect temperature data at each measuring point and ambient temperature data. The counter, which is connected to the coil energizing signal and the contact state transition signal, accumulates one action count when it detects that the coil current is greater than the preset operating current threshold and the contact completes one opening or closing state transition, and obtains the cumulative action count value.
3. The intelligent fault monitoring system for ring main unit according to claim 2, characterized in that, The extraction of current features based on the current waveform of the opening and closing coils specifically involves: The current waveforms of the opening and closing coils are filtered and denoised, and the baseline is corrected. The corresponding effective action waveform is extracted based on the energization start time. The peak current, peak occurrence time and average current are extracted based on the effective action waveform. The product of the average current of adjacent sampling points and the corresponding time interval is accumulated to obtain the current integral. The standard deviation of the effective action waveform is used as the current fluctuation. Peak current, peak occurrence time, average current, current integral, and current fluctuation are used as current characteristics.
4. The intelligent monitoring system for ring main unit faults according to claim 2, characterized in that, The extraction of motion features based on opening time, closing time, and the stroke curve of the operating mechanism specifically includes: The stroke curves of the operating mechanism for opening and closing actions are smoothed respectively. The first and last sampling points of the operating mechanism stroke curves are extracted. The time and angle value corresponding to the first sampling point are used as the start time and start angle of the action, and the time and angle value corresponding to the last sampling point are used as the completion time and termination angle of the action. The difference between the completion time and the start time of the action is taken as the duration of the action, and the absolute value of the difference between the ending angle and the starting angle is taken as the total stroke. The stroke curve of the operating mechanism is differentiated to obtain the angular velocity curve. The maximum and average action speeds are extracted based on the angular velocity curves, and the standard deviation of the angular velocity curves is used as the speed fluctuation. The opening time, closing time, mechanism action duration, total stroke, maximum action speed, average action speed, and speed fluctuation are used as action characteristics.
5. The intelligent fault monitoring system for ring main unit according to claim 1, characterized in that, The extraction of temperature rise features based on temperature data of moving and stationary contacts and joints specifically involves: The temperature data of moving and stationary contacts and joints and the ambient temperature data are filtered and aligned with the sampling time. The difference between the temperature of each measuring point and the corresponding ambient temperature is calculated as the temperature rise value of each measuring point. The maximum value of the temperature rise at each measuring point is extracted as the maximum temperature rise, the average value of the temperature rise at each measuring point is taken as the average temperature rise, and the difference between the maximum and minimum values of the temperature rise at each measuring point is taken as the temperature rise imbalance. The ratio of the temperature rise difference between adjacent sampling times at the same measuring point to the sampling time difference is taken as the temperature rise rate, and the maximum value of the temperature rise rate among all measuring points is taken as the characteristic value of the temperature rise rate. The maximum temperature rise, average temperature rise, temperature rise imbalance, and temperature rise rate are used as temperature rise characteristics.
6. The intelligent fault monitoring system for ring main unit according to claim 5, characterized in that, The process of using temperature rise characteristics to perform time-series correction on motion characteristics to form an electromechanical-thermal coupling feature vector is as follows: The ratio of the maximum temperature rise to the average temperature rise is used as the local temperature rise coefficient, and the ratio of the temperature rise imbalance to the average temperature rise is used as the distribution deviation coefficient. The sum of the local temperature rise coefficient and the distribution deviation coefficient is used as the time-series correction coefficient, and then added to 1 to obtain the time-series correction reference coefficient. Divide the tripping time and mechanism action duration by the timing correction reference coefficient to obtain the corrected tripping time and corrected mechanism action duration. Divide the closing time and mechanism action duration of the closing action by the timing correction reference coefficient to obtain the corrected closing time and corrected mechanism action duration; The corrected opening time, closing time, and mechanism action duration are combined with the total stroke, maximum action speed, average action speed, and speed fluctuation in the action characteristics to form the corrected action characteristics. The current characteristics, corrected action characteristics, and temperature rise characteristics are combined according to the same opening or closing action to form an electromechanical-thermal coupling characteristic vector.
7. The intelligent fault monitoring system for ring main unit according to claim 6, characterized in that, The process of quantizing and comparing the electromechanical-thermal coupling feature vector with the reference feature corresponding to the template to obtain the current offset, the action timing deviation, and the temperature rise anomaly is as follows: Extract the reference current characteristics, reference action characteristics, and reference temperature rise characteristics corresponding to the current opening or closing action from the matched historical normal action templates; The current feature, corrected action feature, and temperature rise feature in the electromechanical-thermal coupling feature vector are respectively matched with the corresponding reference current feature, reference action feature, and reference temperature rise feature according to the feature items. The absolute value of the difference between each feature value and the corresponding reference value is calculated, and the absolute value is divided by the corresponding reference value to obtain the individual deviation value. The current offset is obtained by averaging the individual deviation values corresponding to each current characteristic. The average of the individual deviation values of the opening or closing time and the duration of the mechanism action corresponding to the current action in the corrected action characteristics is used to obtain the action timing deviation. The temperature rise anomaly is obtained by averaging the individual deviation values corresponding to each temperature rise characteristic.
8. The intelligent fault monitoring system for ring main unit according to claim 7, characterized in that, The process of determining the type of mechanical fault based on the abnormal coupling relationship among the three is as follows: The smaller value between the current offset and the action timing deviation is taken as the electromechanical common variation anomaly, and the absolute value of the difference between the current offset and the action timing deviation is taken as the electromechanical mismatch. Divide the electromechanical common variable abnormality by 1 and add the electromechanical mismatch and temperature rise abnormality to obtain the core jamming fault discrimination value. The sum of the abnormal temperature rise divided by 1 and the electromechanical common variation abnormality and the action timing deviation is used to obtain the contact deterioration fault discrimination value. The sum of the timing deviation and the electromechanical mismatch is divided by 1 and then added to the current offset to obtain the jamming fault judgment value of the transmission mechanism.
9. The intelligent fault monitoring system for ring main unit according to claim 8, characterized in that, The process of determining the type of mechanical fault based on the abnormal coupling relationship among the three also includes: The fault identification values for core jamming, contact deterioration, and transmission mechanism jamming are sorted from largest to smallest, and the largest value is taken as the main identification value. The fault type corresponding to the main identification value is determined as the mechanical fault type. When the core jamming fault discrimination value is the main discrimination value, it is determined to be a core jamming fault; When the contact deterioration fault discrimination value is the main discrimination value, it is determined to be a contact deterioration fault; When the main criterion value for transmission mechanism jamming fault is the criterion value, it is determined to be a transmission mechanism jamming fault.
10. The intelligent fault monitoring system for ring main unit according to claim 9, characterized in that, The process of outputting the abnormality level based on the type of mechanical fault is as follows: The main discrimination value is compared with the preset level range of the corresponding mechanical fault type. If the main discrimination value is less than the lower limit of the preset level range, the abnormality level is determined to be a level one abnormality. If the main discrimination value is within the preset level range, the anomaly level is determined to be a level 2 anomaly. If the main discrimination value is greater than the upper limit of the preset level range, the anomaly level is determined to be a level three anomaly.