A high-voltage cabinet disconnection mismatch compensation method and system

Abstract

The application belongs to the technical field of high-voltage switch device control, and discloses a high-voltage cabinet breaking mismatch compensation method and system, the method comprising: identifying a second breaking event after first associated reclosing in an automatic operation sequence, generating complete sequence identification and operation stage identification, and acquiring mechanical residual state, coil thermal state and power supply state before the second breaking; determining a target driving candidate and generating a candidate latch record according to a candidate response model, a prediction error upper bound, a model application range, a power supply capacity and a driving safety boundary; immediately conducting a rated breaking branch when a second breaking command arrives, and controlling an additional excitation branch according to the candidate latch record check result; making the current candidate latch record invalid when the second breaking is completed, the associated time limit expires or the next reclosing command arrives; the application can compensate for the second breaking response mismatch caused by short interval operation, and limit the breaking driving parameters and mechanism motion parameters within the set safety boundary.

Description

Technical Field

[0001] This application relates to the field of high-voltage switchgear control technology, and more specifically, to a high-voltage switchgear disconnection mismatch compensation method and system. Background Technology

[0002] High-voltage switchgear typically uses circuit breakers to quickly disconnect fault currents and can be configured with automatic reclosing functionality. In a practical automatic operation sequence, the circuit breaker operates in the order O-0.3s-CO-180s-CO, where O represents a tripping operation, C represents a closing operation, and CO represents a tripping operation immediately following a closing operation. According to the order of operation, the first tripping is recorded as O1; after 0.3s, the closing and tripping operations in the first CO group are recorded as C1 and O2 respectively; after 180s, the closing and tripping operations in the second CO group are recorded as C2 and O3 respectively. If the momentary fault has not been cleared, the circuit breaker performs O2 after C1, constituting a short-interval second tripping operation after automatic reclosing; O3 is the third tripping operation in subsequent phases.

[0003] Existing circuit breakers typically determine the driving parameters and mechanical operating characteristics of the trip coil based on the rated control voltage, specified environmental conditions, and the initial state after the mechanism has settled. In actual operation, the trip control usually supplies power to the trip coil according to the pre-set rated driving mode after receiving the protection trip command. For trip operations that occur independently and are spaced far apart, the circuit breaker operating mechanism, buffer components, trip coil, and DC power supply system can usually recover to a near-calibrated initial state before the next trip, thus the rated driving mode can basically meet the trip response requirements.

[0004] However, in the short-interval second tripping scenario corresponding to O1-0.3s-C1O2, the mechanical vibration, motion inertia, buffer rebound, contact overtravel changes, tripping coil temperature rise, and DC power supply voltage fluctuations generated by O1 and C1 may not have subsided yet, causing the actual initial state of O2 at the beginning to differ from the static initial state on which the rated operating characteristics are based. Since the existing technology still executes O2 according to a fixed rated driving mode, it is easy to cause a mismatch between the actual mechanical separation response of O2 and the expected response. C2O3 after 180s belongs to the subsequent operation stage, and the residual state data and candidate latch records generated in the first operation stage cannot be used directly.

[0005] In view of this, this application proposes a high-voltage switchgear disconnection mismatch compensation method and system to solve the above problems. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this application is to provide a high-voltage switchgear tripping mismatch compensation method and system. This method identifies the second tripping event within the first group of CO in the O-short interval-CO-long interval-CO automatic operation sequence, obtains residual state data before the second tripping, predicts the tripping response corresponding to each drive candidate, and determines the target drive candidate by combining the model applicability results, power supply capacity, and drive safety boundaries. Upon arrival of the second tripping command, the rated tripping is immediately executed, and additional excitation is controlled based on the verification results of the candidate latching records in the current operation stage. This achieves the technical effects of compensating for the second tripping response mismatch caused by short interval operations, preventing the use of candidate latching records across operation stages, and limiting the tripping drive parameters and mechanism motion parameters within the set safety boundaries.

[0007] This application provides the following technical solution: a high-voltage switchgear disconnection mismatch compensation method, comprising: Acquire experimentally calibrated candidate drive data, candidate response model data, model applicability data, prediction error upper bound data, power supply capacity data, and drive safety boundary data; Based on the control command time stamp, auxiliary contact status, and mechanism stroke, identify the associated first tripping event, first associated reclosing event, and second tripping event, and generate a complete sequence identifier and operation stage identifier; before the second tripping command arrives, align the collected data with the time stamp, and then calculate the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data to generate the second tripping residual state data; The residual state data of the second trip is input into the candidate response model of each drive candidate. Based on the directional relationship between the directional mechanical residue and the mechanism motion data, the separation response time and mechanism motion speed prediction data are generated. Conservative prediction data is generated by combining the upper bound of the prediction error. The residual state of the second trip, the conservative prediction and the power supply recovery data are compared with the model's applicable range, drive safety boundary and power supply capacity data, respectively, to obtain safe and executable results and select the target drive candidate. Candidate latch records are periodically generated, including complete sequence identifier, operation stage identifier, second trip residual state, conservative prediction, safe and executable result, and target drive candidate. When the second trip command arrives, the consistency, timeliness, completeness, and safe and executable result of the most recent candidate latch record with the current complete sequence identifier and operation stage identifier are verified to obtain a valid candidate result, and the rated trip branch supplies power to the trip coil. When the target drive candidate is an enhanced drive candidate and the valid candidate result is valid, the controlled additional excitation branch is provided with additional current; otherwise, it remains off. When the second trip is completed, the second trip association duration expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

[0008] A high-voltage switchgear disconnection mismatch compensation system, implementing the aforementioned high-voltage switchgear disconnection mismatch compensation method, includes: The data acquisition module is used to acquire experimentally calibrated candidate drive data, candidate response model data, model applicability range data, prediction error upper bound data, power supply capacity data, and drive safety boundary data. The status analysis module is used to identify the associated first tripping event, first associated reclosing event, and second tripping event based on the control command time stamp, auxiliary contact status, and mechanism stroke, and generate complete sequence identifiers and operation stage identifiers; before the second tripping command arrives, the collected data is time-stamped, and then the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data are calculated to generate the second tripping residual status data. The target selection module is used to input the residual state data of the second trip into the candidate response model of each drive candidate, generate separation response time and mechanism motion speed prediction data based on the directional relationship between the directional mechanical residue and the mechanism motion data, and generate conservative prediction data by combining the upper bound of the prediction error; compare the residual state of the second trip, the conservative prediction and the power supply recovery data with the model's applicable range, drive safety boundary and power supply capacity data respectively, obtain safe and executable results and select target drive candidates; The mismatch compensation module is used to periodically generate candidate latch records containing complete sequence identifiers, operation stage identifiers, second trip residual states, conservative predictions, safe and executable results, and target drive candidates. When the second trip command arrives, it verifies the consistency, timeliness, completeness, and safe and executable results between the most recent candidate latch record and the current complete sequence identifier and operation stage identifier, obtains the candidate valid result, and supplies power to the trip coil from the rated trip branch. When the target drive candidate is an enhanced drive candidate and the candidate valid result is valid, it provides additional current to the controlled additional excitation branch; otherwise, it remains off. When the second trip is completed, the second trip association duration expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

[0009] The technical effects and advantages of the high-voltage switchgear disconnection mismatch compensation method and system disclosed in this application are as follows: This application obtains the mechanical residual state, coil thermal state, and power supply state before the second trip after automatic reclosing. It predicts the mechanical separation response time, peak tripping speed, and contact separation speed corresponding to each drive candidate, and determines the target drive candidate based on the model's applicability, power supply capacity, and drive safety boundaries. Upon arrival of the second tripping command, the rated tripping is executed immediately, and additional excitation is only applied when the latching record verification passes. This compensates for the mismatch in the second tripping response caused by the residual state during short-interval operations, reduces the possibility of the mechanical separation response time exceeding the allowable upper limit, and limits coil current, voltage, excitation energy, and mechanism movement speed within set boundaries without altering protection criteria, delaying the second tripping command, or using additional excitation as an execution condition for the rated tripping branch. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of a high-voltage switchgear disconnection mismatch compensation method according to this application; Figure 2 This is a schematic diagram of the automatic operation sequence of O-0.3s-CO-180s-CO in this application; Figure 3 This is a schematic diagram of the second tripping scenario in this application; Figure 4 This is a schematic diagram of the process for driving conservative prediction of candidates and screening of target candidates in this application; Figure 5 This is a schematic diagram of the candidate latch verification, failure handling and second tripping drive execution process in the first operation stage of this application. Figure 6 This is a schematic diagram of a high-voltage switchgear disconnection mismatch compensation system according to this application. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] Example 1: Please see Figure 1 , Figure 2 and Figure 3 As shown, this embodiment provides a high-voltage switchgear tripping mismatch compensation method, applied to a high-voltage switchgear with automatic reclosing function. The high-voltage switchgear includes a circuit breaker, a relay protection device, an automatic reclosing device, and a circuit breaker controller. The circuit breaker controller is communicatively connected to the rated tripping branch, the controlled additional excitation branch, the input selection switch, the mechanism travel sensor, the mechanism vibration sensor, the coil temperature sensor, the coil voltage sensor, the coil total current sensor, the additional branch current sensor, the DC bus voltage sensor, the local energy storage voltage sensor, the mechanism energy storage status sensor, the buffer status sensor, the contact overtravel detection device, and the linkage mechanism auxiliary contact group.

[0013] The rated tripping branch supplies power directly to the tripping coil according to the original protection control logic of the circuit breaker. The controlled additional excitation branch includes an input selection switch, a current source type excitation conversion unit, a current limiting unit, an additional excitation switch, an additional branch current sensor, a reverse current blocking unit, and a hardware fault shutdown unit. The input selection switch is used to connect the controlled additional excitation branch to a shared DC power supply or a local energy storage unit. After the rated tripping branch is activated, the controlled additional excitation branch adds an additional current in the same direction to the tripping coil.

[0014] The rated tripping branch and the controlled additional excitation branch are respectively equipped with current detection channels and switch control channels. The rated tripping branch does not use the compensation calculation result, the input selection switch status, or the status of the controlled additional excitation branch as the execution condition. When the controlled additional excitation branch is turned off, the circuit breaker controller loses power, or the local energy storage unit fails, the rated tripping branch still supplies power to the tripping coil according to the tripping command output by the relay protection device.

[0015] The complete automatic operation sequence described in this embodiment includes O-0.3s-CO-180s-CO; the first trip O1 represents the first trip command output by the relay protection device and the circuit breaker tripping process caused by the first trip command in the automatic operation sequence; the first associated reclosing C1 represents the first closing process executed according to the first reclosing interval after the first trip is completed; the second trip O2 represents the first trip command output by the relay protection device to the same circuit breaker after the first associated reclosing command and the circuit breaker tripping process caused by the first trip command; the subsequent reclosing C2 represents the next closing process executed after the set subsequent operation interval expires; the third trip O3 represents the tripping process executed after C2.

[0016] The first operation phase includes O1, C1, and O2, and ends when O2 is completed, the upper limit of the second tripping association time expires, or the next reclosing command arrives; the period from the completion of O2 to the arrival of the C2 command is the waiting period for subsequent operations, and the candidate latch records generated in the first operation phase remain invalid during the corresponding waiting period; the second operation phase includes O2, C2, and O3; in this embodiment, the second tripping mismatch and candidate response model is for O2 in the first operation phase; the second tripping residual state data, target drive candidates, and candidate latch records generated in the first operation phase shall not be used for O3 in the second operation phase.

[0017] This implementation does not change the relay protection criteria, does not block or delay the protection tripping command, and does not replace the rated tripping branch with the enhanced drive candidate. It only performs controlled additional excitation on the enhanced drive candidate that has been calibrated offline, predicted online and screened for safety when the rated tripping branch is put into operation.

[0018] The candidate set for tripping drive is denoted as . ,in, This indicates that only the rated drive candidate of the rated trip branch is executed. This indicates a shared DC enhanced drive candidate that executes the rated trip branch and receives additional excitation from a shared DC power supply. This indicates a local energy storage enhancement drive candidate that executes the rated trip branch and is provided with additional excitation by the local energy storage unit.

[0019] The methods include: Acquire experimentally calibrated candidate drive data, candidate response model data, model applicability data, prediction error upper bound data, power supply capacity data, and drive safety boundary data.

[0020] In specific implementations, the methods for obtaining candidate driver data, candidate response model data, model applicability range data, prediction error upper bound data, power supply capacity data, and driver safety boundary data include: The automatic reclosing device sends the complete automatic operation sequence, the first reclosing interval setting value, the first reclosing interval allowable deviation, the subsequent operation interval setting value, the subsequent operation interval allowable deviation, and the maximum reclosing waiting time to the circuit breaker controller; for the O-0.3s-CO-180s-CO automatic operation sequence, the first reclosing interval setting value is 0.3s, and the subsequent operation interval setting value is 180s.

[0021] The circuit breaker controller records the longest time allowed to identify the first associated reclosing command after the first trip as the upper limit of the first reclosing associated time. The upper limit of the first reclosing associated time The second tripping associated time limit is obtained by summing the first reclosing interval setting value and the corresponding allowable deviation in the forward time; the circuit breaker controller records the longest time from the effective edge of the first associated reclosing command to the allowable receipt of the tripping command within the first group of COs as the upper limit of the second tripping associated time. The upper limit of the second tripping associated duration. It is obtained by summing the upper limit of the closing action duration, the upper limit of the permanent fault protection action duration, the upper limit of the opening command transmission duration, and the upper limit of the positive timestamp error.

[0022] The circuit breaker controller records the longest duration from the completion of the first trip to the last operation event in the complete automatic operation sequence as the complete sequence association duration. The complete sequence association duration is obtained by summing the first reclosing interval of 0.3s, the allowable operation duration of the first group of COs, the subsequent operation interval of 180s, the allowable operation duration of the second group of COs, and the corresponding allowable deviation of the forward time. The complete sequence association duration is only used for recording and grouping the complete operation sequence and is not used to determine the second trip event.

[0023] Measurement error calibration is performed on the voltage, current, temperature, stroke, vibration, pressure, capacitance, and time channels used for subsequent calculations and safety screening. The calibration uses a reference measuring device with a range and accuracy higher than that of the calibrated channel, and covers the lower boundary, intermediate value, and upper boundary of the actual working range of the corresponding channel at the minimum allowable ambient temperature, room temperature of 20℃ to 25℃, and maximum allowable ambient temperature, respectively.

[0024] No. Error protection value of each calibration point Obtained using the following formula: ; Upper bound of measurement error for the corresponding channel Obtained using the following formula: ;in, Indicates the calibration point number. Indicates the calibrated channel in the 1st... Readings at each calibration point Indicates the reference measurement value. This indicates the upper bound of the uncertainty of the reference measuring device. Representing physical quantities The upper bound of the measurement error for the corresponding channel.

[0025] Thus, the upper bound of the travel measurement error is obtained respectively. Upper limit of speed measurement error Upper bound of DC bus voltage measurement error Upper limit of measurement error for total coil current Shared upper limit of DC input current measurement error Upper limit of measurement error for additional branch current Upper bound of local energy storage voltage measurement error Upper limit of temperature measurement error Upper bound of vibration measurement error Upper bound of capacitance measurement error Upper bound of time scale measurement error .

[0026] The command timescale, switch input timescale, and sampling timescale all use the same hardware timing reference. At least twenty common pulses are simultaneously input to the command input channel, switch input channel, and reference timing device. The maximum absolute value of the difference between the controller's recorded time and the reference time, plus the upper bound of the uncertainty of the reference timing device, is used as the reference time. ,when A valid time-stamped result is obtained within one state update cycle.

[0027] Under the conditions that the circuit breaker is in a static closed state, the rated opening branch coil current and the additional branch current are both zero, the energy storage motor is stopped, and the DC bus is supplying power normally, at least one thousand sample values ​​are collected according to the state update cycle, and the upper limit of the static current noise is obtained according to the following formula. ： ;in, Indicates the number of the static sampling point. Indicates the first One static sampling time express The total coil current at time t.

[0028] The vibration signal of the mechanism is subjected to the same causal bandpass filtering and envelope calculation as in the online phase, and the upper bound of the static vibration envelope is obtained according to the following formula. : ;in, express The vibration envelope value at time t.

[0029] Stable closing state is denoted as The criteria for determining a stable closing state include: the mechanism travel being in the closing position section for ten consecutive state update cycles; the auxiliary contact group of the linkage mechanism maintaining the closing state for ten consecutive state update cycles; the absolute values ​​of the total coil current and the additional branch current not exceeding the upper limit of the corresponding static current noise; the energy storage motor being in a stopped state; and the mechanism vibration envelope not exceeding the upper limit of the noise level for ten consecutive state update cycles. When all conditions are met simultaneously Otherwise .

[0030] The stable power supply status of the DC bus is denoted as The difference between the maximum and minimum DC bus voltage values ​​within ten consecutive state update cycles shall not exceed Furthermore, after performing a least-squares linear fit between time and voltage on the ten bus voltage samples, the product of the absolute value of the fitting slope and the duration of the ten state update cycles does not exceed [a certain value]. Otherwise .

[0031] The direction of the mechanism's travel coordinate from the closed position to the open position is defined as the positive direction. Under power outage maintenance conditions, the operating mechanism is slowly driven to complete no less than five closing-to-open and opening-to-closing operations, and the mechanism travel is recorded when the auxiliary contact group of the linkage mechanism is stably switched to the closed state and the open state, respectively.

[0032] Closing Position Section Section where the circuit breaker is in place and the movement section of the mechanism Obtained using the following formula: ; ; ;in, Indicates the power outage operation number. Indicates the first The mechanism travels during the stable switching of the auxiliary contact to the closed state in this operation. Indicates the first The mechanism travels during the stable switching of the auxiliary contact to the open state in this operation. Indicates the upper boundary of the interval. Indicates the lower boundary of the interval.

[0033] Closing center position Obtained using the following formula: A low-voltage current-limiting DC source is used to connect to each of the three-phase main circuits, and the three-phase conduction current and mechanism stroke are collected synchronously. After the rated trip branch operates, the first... The phase conduction current dropped to the first time. When the stable conduction current before phase tripping is below 5% and remains within the corresponding range for three consecutive sampling cycles, the corresponding time is recorded as the 1st twentieth ... Actual contact separation time And obtain the first by means of journey interpolation Actual contact separation position .

[0034] No. The latest actual contact separation time of the three phases in this operation and the latest actual contact separation position of the three phases Obtained using the following formula: ; ;in, Indicate the phases, A, B, and C represent the three-phase main circuits respectively. Indicates time The corresponding organizational itinerary.

[0035] The contact separation position test covers the minimum permissible ambient temperature, room temperature, maximum permissible ambient temperature, minimum effective rated opening input voltage, and rated input voltage. At least five effective operations are performed for each ambient temperature and input voltage boundary combination, and the conservative contact separation position is obtained according to the following formula. : The equivalent mechanical separation response time for the second tripping operation represents the time from the effective edge of the second tripping command to the first time the mechanism travels to reach its maximum value. The duration allows for an upper limit on the equivalent mechanical separation response time of the second trip. Subtract the smaller of the maximum tripping time specified in the circuit breaker product specifications and the maximum permissible mechanical response time for protection coordination. get.

[0036] The mechanism stroke is collected by a synchronous reference acquisition device at a frequency no less than ten times the online stroke sampling frequency. Local quadratic fitting is performed on the reference stroke data and the first derivative is calculated to obtain the reference mechanism speed. The maximum absolute value of the difference between the speed obtained by the controller according to the online speed calculation method and the speed of the reference mechanism at the same moment is added to the upper bound of the reference speed uncertainty, and these are used as the upper bounds of the peak speed measurement error of the opening motion. Upper bound of velocity measurement error corresponding to the conservative contact separation position .

[0037] The maximum absolute value of the difference between the equivalent mechanical separation response time of the second circuit breaker obtained by the controller and the corresponding time obtained by the reference timing device, plus the upper bound of the reference timing uncertainty, is used as the upper bound of the response time measurement error. .

[0038] Peak speed limit of the tripping motion The upper speed limit corresponding to the position of separation from the conservative contact. The speed is obtained by subtracting the upper limit of the corresponding speed measurement error from the smaller of the following values: the allowable speed of the operating mechanism design, the allowable input speed of the buffer, and the speed confirmed by the mechanical life test of the circuit breaker.

[0039] For any continuous state quantity Calibration scale Obtained using the following formula: ;in, and These represent the state variables during offline calibration and effective training operations, respectively. The maximum and minimum values, State variables Corresponding sensor range, State variables The upper bound of the measurement error.

[0040] Thus, the mechanical residual velocity calibration scale was obtained. Travel position calibration scale Mechanism speed calibration scale Vibration amplitude calibration scale Bus voltage calibration standards Busbar recovery slope calibration scale Institutional energy storage calibration standards Buffer state calibration scale and contact overtravel calibration scale .

[0041] A causal linear phase finite impulse response bandpass filter is used for the vibration data of the mechanism. The vibration spectrum of the mechanism is obtained through rated opening and short interval reclosing tests, and the static noise spectrum is obtained through static test. The narrowest continuous frequency band that is higher than the upper bound formed by the static noise power spectrum and the vibration measurement error and contains no less than 90% of the total energy of the vibration is used as the filter passband.

[0042] The filter order, filter coefficients, and sampling period are written into the calibration record and remain unchanged during online operation. The filter group delay is calculated according to the linear phase finite impulse response filter order and sampling period and verified by common pulse input test. The effective result of the filter parameters is obtained when the difference between the calculated group delay and the measured group delay does not exceed one sampling period. Bidirectional zero-phase filtering is not used in online calculation.

[0043] Each enhanced driver candidate record candidate identifier Target input power supply Additional target current trajectory Target current allowable error band Additional excitation duration Additional excitation duration upper limit Candidate maximum output current Conservative value of candidate input current increment The upper limit of the energy required for candidates Candidate lowest effective input voltage and candidate versions .

[0044] Target current allowable error band Obtained using the following formula: ;in, Indicates enhancement driver candidate Additional upper bound for current tracking error; in enhanced drive candidates In the load test, the measured current of the additional branch during the period from the effective edge of the additional excitation control command to the end of the additional excitation is aligned with the trajectory of the additional target current according to the sampling time. The absolute value of the difference between the two is calculated, and the upper limit of the measurement error of the additional branch current is subtracted. The non-negative value after that, and the maximum value among the non-negative values ​​corresponding to all effective load tests and all sampling times is taken as ; This indicates the upper limit of the measurement error for the additional branch current.

[0045] At least five load tests are performed on each enhanced drive candidate at the minimum allowable ambient temperature, room temperature, maximum allowable ambient temperature, minimum candidate input voltage, and rated input voltage, from the effective edge of the additional excitation control command to the first entry of the additional branch current. And maintained for three consecutive sampling periods Duration within and The maximum value of the sum is used as the upper bound of the current establishment time. .

[0046] For the Secondary candidate load test, total peak current protection value of the coil Maximum voltage protection value at the coil terminal Minimum voltage protection value at coil terminals Total excitation energy protection value of coil They are obtained according to the following formulas: ; ; ; ;in, Indicates the first Secondary candidate Total coil current during load test This indicates the voltage at the corresponding coil terminals. Indicates the first Each sampling time, This indicates the time interval between adjacent sampling times.

[0047] Conservative value of total peak current of candidate coil Conservative value of the highest voltage at the candidate coil terminals Conservative value of the lowest voltage at the candidate coil terminals Conservative value of total excitation energy of candidate coils They are obtained according to the following formulas: ; ; ; Candidate In the Additional branch output energy protection value during subload test Obtained using the following formula: The upper bound for the required additional energy for a candidate is obtained according to the following formula: ;in, Indicates candidate In the In the secondary load test Additional branch current at any given time.

[0048] Shared DC-enhanced drive candidate Conservative value of input current increment Obtained using the following formula: ;in, and They represent the first The test included the shared DC power supply input current before and after the additional excitation was applied.

[0049] Drive safety boundaries include the upper limit of the total peak current of the coil. Upper limit of coil terminal voltage Lower limit of coil terminal voltage Upper limit of total excitation energy of coil Additional excitation duration upper limit Maximum output current of additional branch and the protection control circuit allows the minimum bus voltage .

[0050] The aforementioned drive safety boundaries were obtained based on the rated parameters of the trip coil, rated trip branch, controlled additional excitation branch, and input selection switch, respectively. They were verified by rated load tests at the minimum allowable ambient temperature, room temperature, and maximum allowable ambient temperature. The drive safety boundary results were obtained when no coil temperature rise exceeded the limit, insulation abnormality, overcurrent of switching devices, or output runaway occurred during all verification operations.

[0051] The protection control circuit allows a minimum bus voltage. It is obtained from the maximum value of the minimum operating voltage of each device in the relay protection device, circuit breaker controller, rated trip branch controller and input selector switch.

[0052] The second tripping command is issued while the closing coil of the first associated reclosing circuit is still energized or the operating mechanism is still in the closing motion section. Tripping is performed only by the rated tripping branch. The test covers the lower, middle, and upper boundaries of the allowable ranges for the minimum permissible ambient temperature, room temperature, maximum permissible ambient temperature, minimum effective input voltage, rated input voltage, and the first reclosing interval of 0.3s. Each boundary combination is executed at least five times. All tests are conducted within... Arrival Furthermore, the effective result of free tripping capability was obtained without any occurrence of mechanism failure to operate, closing retention, mechanical jamming, or auxiliary contact state conflict. .

[0053] Branch isolation tests were conducted under three conditions: single activation of the rated tripping branch, single activation of the controlled additional excitation branch, and simultaneous activation of both branches. When the controlled additional excitation branch was turned off, the absolute value of the additional branch current did not exceed [a certain value] for ten consecutive sampling cycles. When both branches are simultaneously activated, no excitation direction opposite to the target direction is detected for more than three consecutive sampling cycles. The current, and the peak current variation of the rated trip branch does not exceed the rated allowable deviation. When the sum of these parameters is obtained, effective results in circulation protection can be achieved. .

[0054] The rated permissible deviation is obtained through individual activation tests of the rated trip branch. The test covers the minimum permissible ambient temperature, room temperature, and maximum permissible ambient temperature, as well as combinations of the minimum effective input voltage and the rated input voltage. Under each test condition combination, the controlled additional excitation branch is kept off, and the rated trip branch performs at least five effective trip operations, recording the peak current of the rated trip branch for each. The median of each peak current under the same test condition combination is used as the reference value of the rated peak current for the corresponding condition. The absolute difference between each peak current and the corresponding reference value of the rated peak current is calculated, and the maximum value of the absolute difference obtained from all test condition combinations is used as the rated permissible deviation. The change in the peak current of the rated trip branch in the branch isolation test is the absolute value of the difference between the peak current of the rated trip branch when both branches are activated simultaneously and the reference value of the rated peak current when the rated trip branch is activated alone under the same test conditions.

[0055] Perform at least twenty operation tests on the input selection switch and the auxiliary excitation switch respectively, and measure the time from the effective edge of the control command to the switch feedback aligning with the target state. The maximum value of the sum is used as the upper bound of the corresponding switch action time. The switch self-test result is obtained when all actions are completed within the specified action time and no contact sticking, feedback error, or inter-source short circuit occurs. .

[0056] Tripping coil calibration resistance function The temperature is obtained through isothermal resistance measurement. At least five temperature calibration points are set within the range of the minimum to the maximum allowable coil temperature. The difference between the maximum and minimum values ​​of the coil temperature over ten consecutive sampling periods does not exceed [a certain value]. The coil resistance is measured at regular intervals, and linear interpolation is used between adjacent temperature calibration points to obtain the temperature. The corresponding rated resistor.

[0057] Coil thermal conversion coefficient and thermal decay time constant The coil thermal response was obtained through testing. Excitation energy covering both the rated drive candidate and the enhanced drive candidate was input within each temperature range. Coil temperature rise data after excitation was recorded, and least-squares fitting was performed according to the following formula: ; in, Indicates time The coil temperature rise, This indicates the heat input to the coil during the test. Indicates the time when excitation ends. This represents the initial temperature rise conversion factor corresponding to a unit of heat generated by the coil. This represents the coil thermal decay time constant.

[0058] The root mean square error of the fit does not exceed Furthermore, the fitted temperature rise at all calibration times is not less than the measured temperature rise minus [the measured temperature rise]. At that time, the corresponding temperature range is obtained. and If the fitting conditions are not met, the temperature range is narrowed and refitted. If the fitting conditions are still not met after refitting, the enhanced drive candidate model containing the residual heat of the coil is not enabled.

[0059] Under the maximum configured communication load and the controller's maximum operating load, at least one thousand state update cycles are continuously executed. The upper bound of the data acquisition duration is defined as the sum of the maximum time from actual sampling to data readability on the necessary channel and a timing quantization unit. The upper bound of the processing time is defined as the sum of the maximum time from the start of calculating the residual state data of the second circuit breaker tripping to the generation of a complete candidate latch record and one timing quantization unit. .

[0060] Allowed candidate look-ahead duration And the upper limit of the current state sampling age Obtained using the following formula: ; ;in, This indicates the state update cycle.

[0061] The mechanical characteristic test of short-interval automatic reclosing in the first operation phase is performed using each drive candidate. The first operation phase is performed according to O1-0.3s-C1O2. The test conditions should at least cover the lowest ambient temperature, highest ambient temperature, room temperature, lowest effective input voltage, rated input voltage, and the lower boundary, intermediate value, and upper boundary of the allowable range corresponding to the 0.3s first reclosing interval, and each input boundary should appear in at least one test condition. The subsequent 180s operation interval and the second group of C2O3 are not written into the training samples, training convergence samples, error calibration samples, or safety verification samples of the second tripping candidate response model.

[0062] For each operating condition, at least ten training operations, five training convergence operations, five independent error calibration operations, and five independent safety verification operations are obtained. The training sample set, training convergence sample set, error calibration sample set, and safety verification sample set are formed by mutually independent operation sequences, and all samples formed by the same operation sequence are written into only one sample set.

[0063] The valid operation criteria include that the command timestamps of O1, C1, and O2 are complete, C1 is identified as the first associated reclosing event, O2 arrives before the upper limit of the second tripping associated time expires and before the next reclosing command arrives, the mechanism completes O1 and responds to O2 after C1, all necessary acquisition channels are valid, no undervoltage of the protection control circuit, abnormal switch feedback, or abnormal shutdown of the additional branch occurs, and the conduction status of the three-phase main circuit can determine the actual contact separation time corresponding to O2; if any criterion is not met, the corresponding operation is not written into the sample set of the second tripping candidate response model.

[0064] In addition, perform no less than five operation phase isolation verification tests according to the complete automatic operation sequence of O-0.3s-CO-180s-CO; each test should meet the following conditions: after O2 is completed, the candidate latch record of the first operation phase is set to invalid; when C2 arrives after 180s, a new operation phase identifier is established; O3 is not identified as the second tripping event of the first operation phase; and the candidate latch record of the first operation phase cannot pass the candidate validity verification corresponding to O3; if any condition is not met, the enhanced drive candidate is not enabled.

[0065] For the This is the second valid operation, starting from the moment of the second trip command. Forward in the interval The state calculation time is obtained according to the state update cycle. The candidate look-ahead duration is obtained according to the following formula: ;No. The actual second trip equivalent mechanical separation response time of the first effective operation Actual peak speed of tripping motion Speed ​​corresponding to the actual conservative contact separation position They are obtained according to the following formulas: ; ; ;in, Indicates the time from the second trip command. Until the tripping time is reached During the period of institutional speed The maximum value, Indicates the first The maximum value of the actual contact separation time of the three phases in one effective operation, that is, the latest actual contact separation time of the three phases. Indicates the first The first sampling moment when the mechanism travels into the tripping zone for the first time in a valid operation and remains within the tripping zone for three consecutive sampling cycles. Indicates the first During the second effective operation The corresponding mechanical speed.

[0066] For the One continuous state input component, normalized scale and Normalization Center Obtained using the following formula: ; ;in, Indicates the input component number of the continuous state. and They represent the first and second training samples, respectively. The maximum and minimum values ​​of each input component. Indicates the first The width of the allowed value range for each input component. Indicates the first Upper bound of measurement error for each input component.

[0067] No. Normalized values ​​of continuous state input components Obtained using the following formula: ; All normalized continuous state input components and direction relationship parameters The model state feature vectors are composed in a fixed order. ,in, This represents the residual state vector of the second circuit breaker trip.

[0068] Candidate response models include the equivalent mechanical separation response time model for the second trip. Peak speed model of circuit breaker tripping motion Velocity model corresponding to the conservative contact separation position The three models are calculated according to the following formulas: ; ; ;in, Indicates candidate The predicted value of the second trip equivalent mechanical separation response time. Indicates candidate The predicted peak speed of the tripping motion, Indicates candidate The conservative contact separation position corresponds to the predicted velocity value. , and These represent the intercepts of the three models, respectively. , and These represent the state coefficient vectors of the three models, respectively. , and These represent the candidate look-ahead duration coefficients for the three models, respectively.

[0069] The least squares fitting is performed using the samples corresponding to the training operations. When the column rank of the model design matrix is ​​equal to the number of coefficients to be determined and the number of independent training operations is not less than twice the number of coefficients to be determined, the model can be identified. If the model cannot be identified, training operations are added or input components that cannot form independent changes are deleted and then retrained.

[0070] The training convergence operation is not used to calculate model coefficients. After adding the training operation and refitting, for any model coefficient... The relative change of the coefficient is obtained according to the following formula. : ;in, This represents the model coefficients obtained after adding training operations. This represents the model coefficients obtained before the training operation. Represents model coefficients Numerical resolution.

[0071] All model coefficients A training convergence result is obtained when the root mean square error of the three outputs during the training convergence operation does not exceed one percent and the change in the root mean square error of the corresponding output does not exceed the upper bound of the measurement error of the corresponding output. When two consecutive training convergence results are obtained, the training operation is stopped and the model is frozen.

[0072] Independent error calibration is not used to select input components, change the model form, or calculate model coefficients; candidate... Corresponding positive error in response time Peak speed positive error Positive velocity error corresponding to the conservative contact separation position They are obtained according to the following formulas: ; ; Upper bound of one-sided prediction error for the three outputs , and They are obtained according to the following formulas: ; ; .

[0073] State vector to be evaluated and candidate The Normalized distance between training samples and the minimum normalized distance between it and all training samples They are obtained according to the following formulas: ; ;in, Indicates the number of continuous state input components. Represents the first state in the state vector to be evaluated. One input component, Indicates candidate The The th training sample One input component.

[0074] The maximum normalized distance between each independent error calibration sample and the training samples is used as the upper limit of the applicable distance for the model. When the input component of any independent error calibration sample exceeds the corresponding input calibration interval of the training sample, the corresponding training condition is supplemented and retraining is performed, without directly expanding the input calibration interval.

[0075] Three conservative prediction values ​​in independent security verification operations , and They are obtained according to the following formulas: ; ; .

[0076] Candidate response model about Since it is an affine relation, we calculate it separately. and The candidate response model is validated when the predicted value is taken as the larger value, the actual output of all independent safety verification operations does not exceed the corresponding conservative predicted value, and the actual current, voltage, excitation energy, and additional excitation duration do not exceed the drive safety boundary. .

[0077] Candidate lowest effective input voltage The input voltage was obtained through candidate load testing. At each ambient temperature, the input voltage was gradually reduced from the rated input voltage, with at least five candidate excitations performed for each voltage level. The additional output current in each of the five tests was within [specific parameters]. Enter The system maintains this state until the additional excitation ends, the feedback from both the input selection switch and the additional excitation switch is normal, and no undervoltage shutdown occurs, to obtain a valid result for the corresponding voltage level. Then, the maximum value of the sum of the lowest effective voltage at the three ambient temperatures and the corresponding upper bound of the voltage measurement error is taken as the [value / percentage]. .

[0078] The upper limit of the equivalent internal resistance of the shared DC power supply is obtained through a load step test. A stable voltage is obtained when the DC bus is in a stable power supply state before the load step test. Determine the minimum DC bus voltage after the load is turned on. At the same moment, the increase in current relative to the steady-state input current before the step jump is read. .

[0079] when At that time, the first Upper bound of equivalent internal resistance in this experiment and the upper limit of the equivalent internal resistance of the shared DC power supply They are obtained according to the following formulas: ; .

[0080] Minimum effective capacitance of local energy storage unit The value after deducting manufacturing tolerances, temperature derating, and lifetime derating from the device's rated capacitance is subtracted from the measured capacitance at the minimum permissible ambient temperature. The smaller of the obtained values ​​is used.

[0081] Maximum effective capacitance of local energy storage unit The value obtained by adding positive manufacturing tolerance to the rated capacitance of the device, plus the value obtained by adding the actual capacitance measured at room temperature. The larger of the obtained values ​​is used.

[0082] The lowest energy conversion efficiency was obtained through local energy storage-enhanced drive candidate load testing. The lower bound of the additional output energy for this experiment Obtained using the following formula: The upper limit of the reduction in electricity consumption by local energy storage during the same period. Obtained using the following formula: ;in, and They represent the first The local energy storage voltage at the start and end of the additional excitation in this test; Indicates the first In the test of the candidate load for enhanced local energy storage drive, the first Each sampling time The corresponding coil terminal voltage; Indicates the first Sampling time in each local energy storage enhanced drive candidate load test The corresponding additional branch current; Indicates the sampling point number in the load test; This indicates the time interval between adjacent sampling times.

[0083] No. Lower bound of energy conversion efficiency in this experiment and minimum energy conversion efficiency They are obtained according to the following formulas: ; All tests meet the requirements. and The minimum energy conversion efficiency is obtained.

[0084] The maximum rate of drop in local energy storage voltage was obtained through a holding test with the controlled additional excitation branch off and the circuit breaker controller at maximum operating load, and maintained for no less than [value missing] under various ambient temperatures. It also collects the local energy storage voltage according to the status update cycle.

[0085] When adjacent valid sampling times satisfy At that time, the upper bound of the local decline rate and the maximum voltage drop rate of local energy storage They are obtained according to the following formulas: ; ;in, Indicates the number of adjacent valid sampling intervals; and These represent the start and end valid sampling times of the corresponding sampling interval, respectively. and These represent the local energy storage unit at time [time]. and time The energy storage voltage; This indicates the upper bound of the local energy storage voltage measurement error; Indicates the upper limit of time-scale measurement error; minimum effective voltage of local energy storage. Based on the lowest effective input voltage of the local energy storage enhanced drive candidate and the hardware undervoltage shutdown threshold, plus The larger of the obtained values ​​is used.

[0086] Based on the control command time stamp, auxiliary contact status, and mechanism stroke, the first tripping event, the first associated reclosing event, and the second tripping event are identified and associated, generating a complete sequence identifier and an operation stage identifier. Before the second tripping command arrives, the collected data is time-stamped, and then the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data are calculated to generate the second tripping residual state data.

[0087] In a specific implementation, the method for generating the second trip residual state data includes: After receiving the first tripping command in the complete automatic operation sequence, the circuit breaker controller records the time of the first tripping command and generates an operation stage identifier consisting of the high-voltage cabinet identifier, the circuit breaker identifier, the complete operation sequence number, the first tripping command count value, the time of the first tripping command, and the first operation stage number.

[0088] After the mechanism travels from the closing position section to the mechanism movement section, it enters the opening position section. When the auxiliary contact group of the linkage mechanism switches to the opening state, the completion result of the first opening action corresponding to O1 is obtained and the time of completion of the first opening action is recorded. The completion event of the first opening action is used as the short interval start event of the first operation stage.

[0089] After the first tripping action is completed, when the first closing command for the same circuit breaker is received within the upper limit of the first reclosing associated time, the first closing command is identified as the first associated reclosing event corresponding to C1; when the mechanism travel enters the closing position section and the linkage mechanism auxiliary contact group switches to the closing state, the result of the completion of the first associated reclosing action is obtained.

[0090] Starting from the effective edge of the first associated reclosing command, before the upper limit of the second tripping associated duration expires and before the next reclosing command arrives, when the first tripping command for the same circuit breaker is received, the first tripping command is determined as the second tripping event corresponding to O2 and the time of the second tripping command is recorded. The second tripping event is used as the short interval end event of the first operation phase.

[0091] After the second tripping command arrives, the mechanism travels from the closing position section or the mechanism movement section to the tripping position section, and when the auxiliary contact group of the linkage mechanism switches to the tripping state, the second tripping completion event is obtained and the second tripping completion time is recorded; the second tripping completion event serves as the normal end event of the first operation stage.

[0092] When any of the following conditions are met: the second trip is completed, the upper limit of the second trip associated time expires, or the next reclosing command arrives, the first operation phase ends and the candidate latch record of the first operation phase is invalidated; when the closing command corresponding to C2 is received 180s later, the second operation phase number is generated, and the O3 received after C2 is not certain to be the second trip event of the first operation phase.

[0093] When a closing command corresponding to C2 and a opening command corresponding to O3 are received within the complete sequence association time starting from the completion time of the first opening, the current complete sequence identifier is used, and the corresponding operation stage identifier and the current CO sequence number are updated respectively; when a new opening command or closing command is received after the complete sequence association time has expired, the current complete operation sequence ends, and a new complete sequence identifier is generated according to the new complete action sequence number.

[0094] Before the second tripping command arrives, the mechanism's travel is still in the [position missing]. Or the organization's itinerary has already entered However, the vibration envelope still exceeds... Seasonal Combination and Separation Action Overlap Marker ,exist and Enhanced driver candidates are not enabled at this time.

[0095] Analog channels use linear interpolation between two adjacent valid sample values. Digital channels maintain the previous valid state until a new state edge is detected. If channel data is missing, sampling times are reversed, the continuous sampling interval exceeds twice the normal sampling period, or a channel self-test failure occurs, the corresponding channel's valid flag is reset. Otherwise .

[0096] The vibration data of the mechanism is filtered online using a causal bandpass filter with frozen parameters, and the filter group delay is subtracted from the time scale of the filtered output to obtain the directional vibration acceleration. .

[0097] Mechanical residual velocity characteristics Obtained using the following formula: In the formula, Indicates the current state calculation time; Indicates the integration time; This represents the average directional vibration acceleration under stable closed-circuit conditions. This represents the time constant of mechanical vibration decay; This indicates the width of the window for calculating mechanical residuals.

[0098] Mechanical vibration decay time constant The vibration envelope was obtained after the circuit breaker was closed. The fitting start point was the moment of circuit breaker completion, and the first ten consecutive sampling periods of the vibration envelope did not exceed a certain threshold. The time at which the fitting endpoint is reached is used to subtract... The positive envelope is then fitted using least squares according to an exponential function. The result is obtained when the root mean square error of the fitting does not exceed the upper bound of the vibration envelope measurement error. Valid result.

[0099] Mechanical residual calculation window width Obtained using the following formula: ;in, This indicates the longest permissible time from the completion of the first associated reclosing to the arrival of the second tripping command. It also indicates the maximum permissible time between the completion of C1 action and the arrival of O2 command, which is obtained based on the upper limit of the second tripping associated time and the duration of the first associated reclosing action, excluding the 180s subsequent operation interval.

[0100] Vibration residual amplitude Obtained using the following formula: ;in, Representing an interval The maximum value of the vibration envelope within.

[0101] Get current location based on organization's itinerary. For the sampling times are respectively , and The current mechanism velocity is obtained by using the three most recent valid stroke samples and employing second-order backward difference. When the sampling intervals are not the same, perform a quadratic fitting of time and travel for the three sampling points and then fit the curve to... The first derivative at point is used as .

[0102] Travel position deviation and mechanism motion deviation They are obtained according to the following formulas: ; .

[0103] Before the first tripping command arrives, the circuit breaker controller saves the total coil current, coil temperature, DC bus voltage, mechanism energy storage status, buffer status, and contact overtravel status through a cyclic buffer. and Furthermore, if the ten sampling periods do not overlap with the bus voltage decay interval after the previous tripping, closing, operation of the energy storage motor, or stopping of the energy storage motor, the average value of the ten sampling periods is frozen and used as the bus reference voltage before tripping. Stable closing reference value for energy storage mechanism Buffer stable closing reference value and contact overtravel stable closing reference value .

[0104] The moment of the first trip command The total current of the coil first satisfies And the start time of ten consecutive sampling periods is recorded as The first trip coil thermal input Obtained using the following formula: ;in, Indicates the integration time. The corresponding trip coil temperature.

[0105] Current coil thermal residual Obtained using the following formula: When crossing the boundary of a temperature segment, the moment when the measured temperature enters the next temperature segment is taken as the segmentation moment. The residual coil heat at the end of the previous temperature segment is taken as the starting value of the next temperature segment, and the corresponding value of the next temperature segment is used. Continue calculating thermal decay.

[0106] Front bus voltage gap and bus voltage cumulative gap They are obtained according to the following formulas: ; ;in, This represents the DC bus voltage at the current state calculation moment; Indicates the integration time. The corresponding DC bus voltage; It represents the difference between the highest and lowest bus voltages covered by offline calibration, representing a fixed bus voltage scale.

[0107] Physical recovery slope of the most recent three valid bus voltage samples Obtained using the following formula: ;in, Indicates the first Each bus voltage sampling time Indicates the corresponding bus voltage. This represents the average value at three sampling times. This represents the average value of the three bus voltage samples.

[0108] Will Divide by the recovery slope calibration scale to obtain the bus voltage recovery slope. .

[0109] The lower bound of the local energy storage voltage is obtained based on the current local energy storage voltage, the upper bound of the local energy storage voltage measurement error, and the maximum rate of voltage drop of the local energy storage. : In the formula, Indicates the current state calculation time. The corresponding local energy storage voltage.

[0110] Institutional energy storage deviation Buffer state deviation and contact overtravel deviation They are obtained according to the following formulas: ; ; ;in, Indicates the current energy storage status of the institution. Indicates the current position or pressure of the buffer. This indicates the current contact overtravel detection value.

[0111] The contact overtravel detection value is obtained through an independent overtravel sensor or by querying a mechanically calibrated correspondence table between contact overtravel and mechanism stroke. If no buffer state sensor or contact overtravel detection channel is configured, the corresponding component is not written into the state vector, and the candidate response model that has been retrained, error-calibrated and safety-verified according to the same input structure is called.

[0112] Directional value of mechanical residual velocity characteristics Directional value of the deviation of the mechanism motion They are obtained according to the following formulas: ; ;in, This represents the upper bound of the normalized error of the mechanical residual velocity obtained from the propagation of vibration acceleration measurement error and integration operation error. This represents the upper bound of the normalized measurement error of the mechanism's motion deviation.

[0113] Directional relationship parameters Obtained using the following formula: For each input channel in the state vector representing the current value, read the most recent valid sampled value before the current state calculation time, and obtain the current state sampling age according to the following formula. : ;in, Indicates the current value input channel exist The most recent valid sampling time.

[0114] when The current state and results are obtained in real time; historical samples in the vibration integration window and coil heat integration window are not included. The calculation is performed, but if there is a sampling interruption within the corresponding historical window that exceeds twice the normal sampling period, the historical window is marked as intact. Otherwise .

[0115] Second trip residual state vector Generate according to the following formula: ; where superscript This represents the transpose of a vector.

[0116] The residual state data of the second trip is input into the candidate response model of each drive candidate. Based on the directional relationship between the directional mechanical residue and the mechanism motion data, the separation response time and mechanism motion speed prediction data are generated. Conservative prediction data is generated by combining the upper bound of the prediction error. The residual state of the second trip, the conservative prediction and the power supply recovery data are compared with the model's applicable range, drive safety boundary and power supply capacity data, respectively, to obtain safe and executable results and select the target drive candidate.

[0117] In a specific implementation, refer to Figure 4 The methods for generating safe executable results for each driver candidate and selecting a target driver candidate include: The circuit breaker controller will Input the candidate response model corresponding to each driving candidate, and then input the response model of each candidate driver. and Calculate the three output values, take the larger of the two endpoint prediction values, and sum them with the corresponding upper bound of the one-sided prediction error to obtain the result. , and .

[0118] Each input component of the state vector lies within the corresponding model input calibration interval. , , Furthermore, the necessary channel validity flags are all time-limited model applicable flags. Otherwise .

[0119] Candidate A safe mechanical response result is obtained when the following equations are simultaneously satisfied: ; ; Candidate The following equations must be satisfied simultaneously to obtain an electrical drive safety result: ; ; ; ; ; .

[0120] Shared DC-enhanced drive candidate Lower limit of bus voltage Obtained using the following formula: In the formula, This is the upper limit of the bus voltage measurement error; Candidate for shared DC enhanced drive The conservative value of the input current increment is the same parameter as the candidate conservative value of the input current increment obtained through the load step test. The shared DC power supply capability meets the candidate requirements when the upper limit of the equivalent internal resistance of the shared DC power supply is not lower than the corresponding candidate minimum effective input voltage, and is not lower than the minimum bus voltage allowed by the protection control circuit.

[0121] Shared DC-enhanced drive candidate Simultaneously meets the following seasonal shared DC power supply capability indicator Otherwise ; ; .

[0122] Lower bound of locally available energy storage Obtained using the following formula: Local energy storage enhances the driving force candidate Simultaneously satisfying the following seasonal local energy storage power supply capacity indicator Otherwise ; ; .

[0123] Enhanced driver candidates Simultaneously possesses The results show that the candidate response model has passed validation, the mechanical response safety results, the electrical drive safety results, and the corresponding input power supply capability flag is set to 1. , , The controlled additional excitation branch self-test pass result and the necessary channels are complete and valid, indicating that it is safe and executable. Otherwise .

[0124] Rated drive candidate satisfy and At that time, Identified as a target-driven candidate .

[0125] Rated drive candidate The response time condition is not met and there exists a condition that is met. When selecting enhancement driver candidates, choose the target driver candidate from the safe and executable enhancement driver candidates: Among them, the upper limit of energy is increased according to the requirements of the candidates. Prioritize selections from smallest to largest. When the response time is the same, conservative predictions are made for the duration. Choose from smallest to largest.

[0126] When the online available candidate set includes only the rated driver candidate and one enhanced driver candidate, energy sorting among multiple enhanced driver candidates is not performed. If the rated driver candidate does not meet the response time condition and the only enhanced driver candidate meets the safe execution condition, the only enhanced driver candidate is determined as the rated driver candidate. .

[0127] When no safe and executable enhancement driver candidate exists, Determined as The compensation is unavailable flag. And keep the rated trip branch operational.

[0128] Candidate latch records are periodically generated, including complete sequence identifier, operation stage identifier, second trip residual state, conservative prediction, safe and executable result, and target drive candidate. When the second trip command arrives, the consistency, timeliness, completeness, and safe and executable result of the most recent candidate latch record with the current complete sequence identifier and operation stage identifier are verified to obtain a valid candidate result, and the rated trip branch supplies power to the trip coil. When the target drive candidate is an enhanced drive candidate and the valid candidate result is valid, the controlled additional excitation branch is provided with additional current; otherwise, it remains off. When the second trip is completed, the second trip association duration expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

[0129] In a specific implementation, refer to Figure 5 Methods for generating candidate valid results include: After the first associated reclosing action begins and before the second tripping command arrives, the circuit breaker controller repeatedly obtains the residual state vector of the second tripping operation, conservative prediction data, safe and executable results, and target drive candidates according to the state update cycle, and generates a candidate latching record corresponding to the first operation stage at the end of each state update cycle. .

[0130] Candidate latch record Includes complete sequence identifier and operation stage identifier. Current CO sequence number, residual state vector of the second trip. Current status sampling age Complete history window marker State calculation time Record the generation time Candidate response model version, allowed candidate look-ahead duration Conservative prediction data for each driving candidate, model applicability indicators, power supply capability indicators, safety boundary comparison results, safety executability indicators, and target driving candidates. Target input power supply Additional target current trajectory Additional excitation duration Record valid flags and integrity check values. ;in, The operation stage identifier consists of the high-voltage switchgear identifier, circuit breaker identifier, complete action sequence number, current operation stage number, first trip command count value, and first trip command time. This indicates that the candidate latch records are latched in the order listed, excluding... Integrity check values ​​are obtained by performing cyclic redundancy check on each field outside the candidate latch record; after the candidate latch record is written, the cyclic redundancy check value is recalculated, and the recalculated value is compared with the value in the record. When consistent, the integrity check result is correct.

[0131] Candidate latched records are updated as a whole. The new record replaces the previous complete record after all fields have been written and the integrity check has been passed. If the writing is interrupted, fields are missing, or the integrity check fails, the previous complete record is retained.

[0132] When any of the following conditions are met: completion of the second trip, expiration of the second trip association duration, arrival of the next reclosing command, change of the current CO sequence number, or update of the operation phase identifier, the record validity flag in the most recent complete candidate latch record will be set to invalid, and the controlled additional excitation branch will remain closed. The invalid record is only used for operation logs and online review, and will no longer participate in the candidate validity verification of subsequent trip events.

[0133] When the valid edge of the second tripping command arrives, the rated tripping branch immediately supplies power to the tripping coil. The circuit breaker controller reads the most recently generated complete candidate latch record before the arrival of the second tripping command and obtains the actual candidate look-ahead duration based on the difference between the time of the second tripping command and the state calculation time. A candidate is considered valid if the following conditions are met: the complete sequence identifier in the candidate latch record matches the current complete sequence identifier; the operation stage identifier matches the current operation stage identifier; the current CO sequence number is the first group CO; the record validity flag is valid; the actual candidate look-ahead duration is within the allowable candidate look-ahead duration range; the current state sampling age does not exceed the corresponding upper limit; the historical window is complete; the record fields are complete; the integrity check is correct; the candidate response model version matches the current valid version; the target-driven candidate has a safe and executable result; the input selection switch feedback matches the target input power supply; and the record generation time is earlier than the time of the second tripping command. Otherwise, make the candidate valid flag. And obtained invalid candidate results.

[0134] The target-driven candidate is an enhancement-driven candidate and At that time, the circuit breaker controller turns on the auxiliary excitation switch, and the current source type excitation conversion unit operates according to... Provide additional current, target drive candidate is rated drive candidate, Alternatively, if the enhanced drive candidate is not safe to execute, the controlled additional excitation branch shall be kept off, and the candidate valid flag shall not be used as the execution condition for the rated trip branch.

[0135] After the controlled additional excitation branch is activated, the total excitation energy of the accumulated coil is obtained according to the following formula: ;in, Indicates the integration time. The corresponding coil terminal voltage, Indicates the integration time. The corresponding total coil current.

[0136] The additional excitation switch shall be immediately shut off and the rated trip branch shall continue to operate according to the original protection control logic under any of the following conditions: , , , , , , , Organizational itinerary reached The following are examples of issues: the auxiliary contact group of the linkage mechanism switching to the open state; the feedback of the additional excitation switch not being consistent with the control state within the upper limit of the action duration; and the detection of non-target direction current exceeding three consecutive sampling cycles. ;in, This indicates that after the controlled additional excitation branch is put into operation, the time... The corresponding measured current of the additional branch.

[0137] When the controlled additional excitation branch is shut down due to any of the aforementioned conditions, the additional excitation backoff flag is activated. The rated trip branch continues to operate according to the trip command output by the relay protection device.

[0138] According to the organization's schedule, the first time to arrive The moment and The difference yields the actual second trip equivalent mechanical separation response time. The actual peak speed of the opening movement is obtained based on the maximum speed of the mechanism within the second opening movement range. According to the organization's itinerary The interpolation rate at that time yields the velocity corresponding to the actual conservative contact separation position. .

[0139] According to the candidate latch records Second trip command time The actual candidate look-ahead duration used for online review is obtained using the following formula. : .

[0140] Will and Input the corresponding candidate response model and obtain the online verification positive residuals of the three outputs according to the following formula. , and : ; ; Three model drift trigger boundaries , and They are obtained according to the following formulas: ; ; .

[0141] When any model drift trigger boundary is equal to zero, the corresponding enhancement driver candidate is not enabled. When any online verification positive residual is not less than the corresponding model drift trigger boundary, the corresponding enhancement driver candidate is stopped from entering online screening, offline verification prompts are generated, and the rated driver candidate is retained.

[0142] When the trip coil, operating mechanism, mechanism stroke sensor, controlled additional excitation branch, or local energy storage unit is replaced, an offline verification prompt is generated. When the replacement of the component affects the definition of the residual state vector of the second trip, the mechanical transmission relationship, the power supply topology, or the drive safety boundary, the enhanced drive candidate is stopped from entering the online screening. After completing the new mechanical characteristic test, model training, independent error calibration, independent safety verification, power supply capacity test, branch self-test, and drive safety verification, the enhanced drive candidate is reactivated.

[0143] The power outage verification test re-obtains the latest actual contact separation position of the three phases based on the ambient temperature, input voltage, number of tests, effective operation criteria, and calculation methods used during the initial conservative contact separation position calibration. If the re-obtained conservative contact separation position is not greater than the original conservative contact separation position, the original conservative contact separation position is retained. If the re-obtained conservative contact separation position is greater than the original conservative contact separation position, the re-obtained conservative contact separation position is adopted. The mechanical characteristic test, candidate response model training, independent error calibration, independent safety verification, and drive safety verification are then re-executed. The conservative contact separation position is not modified based on the results of a single online operation.

[0144] Example 2: Please see Figure 6 As shown, this embodiment provides a high-voltage switchgear disconnection mismatch compensation system, including: The data acquisition module is used to acquire experimentally calibrated candidate drive data, candidate response model data, model applicability range data, prediction error upper bound data, power supply capacity data, and drive safety boundary data.

[0145] The status analysis module is used to identify the associated first tripping event, first associated reclosing event, and second tripping event based on the control command time stamp, auxiliary contact status, and mechanism stroke, and generate complete sequence identifiers and operation stage identifiers. Before the second tripping command arrives, the collected data is time-stamped, and then the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data are calculated to generate the second tripping residual status data.

[0146] The target selection module is used to input the residual state data of the second tripping circuit breaker into the candidate response model of each drive candidate, generate separation response time and mechanism motion speed prediction data based on the directional relationship between the directional mechanical residue and the mechanism motion data, and generate conservative prediction data by combining the upper bound of the prediction error. The residual state of the second tripping circuit breaker, the conservative prediction and the power supply recovery data are compared with the model's applicable range, drive safety boundary and power supply capacity data, respectively, to obtain a safe and executable result and select the target drive candidate.

[0147] The mismatch compensation module is used to periodically generate candidate latch records containing complete sequence identifiers, operation stage identifiers, second trip residual states, conservative predictions, safe and executable results, and target drive candidates. When the second trip command arrives, it verifies the consistency, timeliness, completeness, and safe and executable results between the most recent candidate latch record and the current complete sequence identifier and operation stage identifier, obtains the candidate valid result, and supplies power to the trip coil from the rated trip branch. When the target drive candidate is an enhanced drive candidate and the candidate valid result is valid, it provides additional current to the controlled additional excitation branch; otherwise, it remains off. When the second trip is completed, the second trip association duration expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

[0148] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for compensating for high-voltage switchgear disconnection mismatch, characterized in that, include: Acquire experimentally calibrated candidate drive data, candidate response model data, model applicability data, prediction error upper bound data, power supply capacity data, and drive safety boundary data; Based on the control command time stamp, auxiliary contact status, and mechanism stroke, identify the associated first tripping event, first associated reclosing event, and second tripping event, and generate a complete sequence identifier and operation stage identifier; before the second tripping command arrives, align the collected data with the time stamp, and then calculate the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data to generate the second tripping residual state data; The residual state data of the second trip is input into the candidate response model of each drive candidate. Based on the directional relationship between the directional mechanical residue and the mechanism motion data, the separation response time and mechanism motion speed prediction data are generated. Conservative prediction data is generated by combining the upper bound of the prediction error. The residual state of the second trip, the conservative prediction and the power supply recovery data are compared with the model's applicable range, drive safety boundary and power supply capacity data, respectively, to obtain safe and executable results and select the target drive candidate. Periodically generate candidate latch records containing complete sequence identifiers, operation stage identifiers, second trip residual states, conservative predictions, safe and executable results, and target drive candidates; when the second trip command arrives, verify the consistency, timeliness, completeness, and safe and executable results of the most recent candidate latch record with the current complete sequence identifier and operation stage identifier, obtain the candidate valid results, and enable the rated trip branch to supply power to the trip coil; When the target drive candidate is an enhanced drive candidate and the candidate's valid result is valid, the controlled additional excitation branch is provided with additional current; otherwise, it remains off. When the second trip is completed, the second trip association time expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

2. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The method for obtaining the operation phase identifier and the second tripping event includes: reading the automatic reclosing action sequence, writing the complete action sequence number and the current operation phase number into the operation phase identifier; receiving the first tripping command, writing the high-voltage cabinet identifier, circuit breaker identifier, first tripping command count value, and first tripping command time into the operation phase identifier; obtaining the first tripping completion event when the mechanism travel enters the tripping position section and the linkage mechanism auxiliary contact group switches to the tripping state; identifying the first associated reclosing event of the same circuit breaker within the first reclosing association duration, and determining the first tripping command received after the first associated reclosing event, within the second tripping association duration, and before the arrival of the next reclosing command as the second tripping event; ending the current operation phase and updating the operation phase number when the second tripping is completed, the second tripping association duration expires, or the next reclosing command arrives.

3. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The method for obtaining the residual state data of the second tripping circuit includes: obtaining the mechanical residual velocity characteristics and vibration residual amplitude based on the mechanism vibration data; obtaining the stroke position deviation and mechanism motion deviation based on the mechanism stroke; and obtaining the directional relationship parameters based on the direction values ​​of the mechanical residual velocity characteristics and mechanism motion deviation; obtaining the coil heat residual amount based on the heat input of the first tripping coil; obtaining the current bus voltage gap, cumulative gap, and recovery slope based on the DC bus voltage; and obtaining the mechanism energy storage deviation, buffer state deviation, contact overtravel deviation, and current state sampling age; and arranging the aforementioned data in a fixed order to obtain the residual state data of the second tripping circuit.

4. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The method for obtaining candidate response models includes: collecting independent training operations, training convergence operations, error calibration operations, and safety verification operations; associating the second trip residual state data, candidate look-ahead duration, and the actual equivalent mechanical separation response duration, actual tripping peak speed, and actual conservative contact separation position corresponding speed at each state calculation time in the training operations as training samples; associating the corresponding data in the error calibration operations as independent error calibration samples; performing model fitting and identifiability verification for each drive candidate using the training samples; performing training convergence verification using the training convergence operation; calibrating the upper bound of the one-sided prediction error and the upper limit of the applicable distance of each output quantity using the independent error calibration samples; and performing conservative prediction and drive safety boundary verification using the safety verification operation to obtain the candidate response model and the model verification results.

5. The high-voltage switchgear disconnection mismatch compensation method according to claim 4, characterized in that, The method for calibrating the upper bound of the one-sided prediction error and the upper limit of the model's applicable distance includes: for each output quantity of each driving candidate, calculating the difference between the actual output quantity of each independent error calibration sample and the corresponding predicted output quantity of the candidate response model, taking the larger value between the difference and zero as the positive prediction residual, and taking the sum of the maximum value of each positive prediction residual and the upper bound of the measurement error of the corresponding output quantity as the upper bound of the one-sided prediction error of the corresponding output quantity; according to the normalization scale of each input component, calculating the distance between the second trip residual state data of each independent error calibration sample and each training sample of the corresponding driving candidate, taking the minimum distance as the normalized minimum distance of the corresponding independent error calibration sample, and taking the maximum value among the normalized minimum distances of each independent error calibration sample as the upper limit of the model's applicable distance of the corresponding driving candidate.

6. The high-voltage switchgear disconnection mismatch compensation method according to claim 5, characterized in that, The candidate response model follows an affine relationship with respect to the candidate look-ahead duration; The method for obtaining conservative prediction data includes: inputting the residual state data of the second tripping circuit breaker into the candidate response model corresponding to each drive candidate; calculating the equivalent mechanical separation response time of the second tripping circuit breaker, the peak speed of the tripping movement, and the speed corresponding to the conservative contact separation position with the candidate look-ahead time being zero and the allowed candidate look-ahead time, respectively; selecting the larger value among the predicted values ​​corresponding to the two candidate look-ahead times for each output quantity, and superimposing the corresponding upper bound of the single-sided prediction error to obtain conservative prediction data.

7. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The enhanced drive candidates include shared DC enhanced drive candidates powered by a shared DC power supply; The method for obtaining the power supply capability satisfaction result of the shared DC enhanced drive candidate includes: obtaining the lower bound of the bus voltage of the shared DC enhanced drive candidate based on the current bus voltage, the upper bound of the bus voltage measurement error, the negative part of the slope of the physical change of the bus voltage, the allowable candidate look-ahead time, the upper bound of the candidate input current, and the upper limit of the equivalent internal resistance of the shared DC power supply; when the lower bound of the bus voltage is not lower than the candidate minimum effective input voltage and the minimum allowable bus voltage of the protection control circuit, the shared DC power supply capability satisfaction result is obtained.

8. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The enhanced drive candidates include local energy storage enhanced drive candidates powered by local energy storage units; The method for obtaining the power supply capability satisfaction result of the local energy storage enhancement drive candidate includes: obtaining the local energy storage voltage lower bound based on the current local energy storage voltage, the upper bound of the local energy storage voltage measurement error, the maximum rate of decrease of the local energy storage voltage, and the allowable candidate look-forward duration; The local usable energy storage lower bound is obtained based on the minimum energy conversion efficiency, minimum effective capacitance, lower bound of local energy storage voltage, and minimum effective voltage of local energy storage. When the local energy storage lower bound is not lower than the candidate minimum effective input voltage and the local usable energy storage lower bound is not less than the candidate required increase in energy upper bound, the local energy storage power supply capability is satisfied.

9. The high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The method for obtaining target drive candidates includes: when the model of the rated drive candidate is applicable and the conservative prediction response time is not greater than the upper limit of the allowable equivalent mechanical separation response time, the rated drive candidate is used as the target drive candidate; otherwise, from the enhanced drive candidates that simultaneously meet the conditions of model verification, model applicability, drive safety boundary, power supply capability, free trip capability, branch self-test and circulating current protection, the candidate with the smaller upper limit of the required additional energy is selected first, and when the upper limits of energy are the same, the candidate with the shorter conservative prediction response time is selected as the target drive candidate; when there are no enhanced drive candidates that meet the above conditions, the rated drive candidate is used as the target drive candidate and a compensation unavailable flag is generated.

10. A high-voltage switchgear disconnection mismatch compensation method according to claim 1, characterized in that, The method for obtaining valid candidate results and executing target-driven candidates includes: periodically generating and updating candidate latch records, including complete sequence identifiers and operation stage identifiers, from the start of the first associated reclosing action to the arrival of the second tripping command; immediately activating the rated tripping branch when the second tripping command arrives, and reading the most recently generated complete candidate latch record before the arrival of the second tripping command; verifying the candidate latch records based on the complete sequence identifier, operation stage identifier, actual candidate look-ahead duration, current state sampling age, historical window integrity, record generation time, record field integrity, model version, integrity check value, safe executable result, and input selection switch feedback; obtaining valid candidate results when all verifications pass, and activating the controlled additional excitation branch when the target-driven candidate is an enhanced-driven candidate; setting the corresponding candidate latch record to invalid when the second tripping is completed, the second tripping associated duration expires, or the next reclosing command arrives.

11. A high-voltage switchgear disconnection mismatch compensation system, used to implement the high-voltage switchgear disconnection mismatch compensation method according to any one of claims 1-10, characterized in that, include: The data acquisition module is used to acquire experimentally calibrated candidate drive data, candidate response model data, model applicability range data, prediction error upper bound data, power supply capacity data, and drive safety boundary data. The status analysis module is used to identify the associated first tripping event, first associated reclosing event, and second tripping event based on the control command time stamp, auxiliary contact status, and mechanism stroke, and generate complete sequence identifiers and operation stage identifiers; before the second tripping command arrives, the collected data is time-stamped, and then the directional mechanical residual data, mechanism motion data, coil thermal residual data, and power supply recovery data are calculated to generate the second tripping residual status data. The target selection module is used to input the residual state data of the second trip into the candidate response model of each drive candidate, generate separation response time and mechanism motion speed prediction data based on the directional relationship between the directional mechanical residue and the mechanism motion data, and generate conservative prediction data by combining the upper bound of the prediction error; compare the residual state of the second trip, the conservative prediction and the power supply recovery data with the model's applicable range, drive safety boundary and power supply capacity data respectively, obtain safe and executable results and select target drive candidates; The mismatch compensation module is used to periodically generate candidate latch records containing complete sequence identifiers, operation stage identifiers, second trip residual states, conservative predictions, safe and executable results, and target drive candidates. When the second trip command arrives, it verifies the consistency, timeliness, completeness, and safe and executable results between the most recent candidate latch record and the current complete sequence identifier and operation stage identifier, obtains the candidate valid results, and enables the rated trip branch to supply power to the trip coil. When the target drive candidate is an enhanced drive candidate and the candidate's valid result is valid, the controlled additional excitation branch is provided with additional current; otherwise, it remains off. When the second circuit breaker is completed, the second circuit breaker association time expires, or the next reclosing command arrives, the current candidate latch record is invalidated.

Images