A Fault Prediction and Control Method for Low-Voltage Switches Based on Contact Electrical Parameters

By constructing a polarity-zoned arcing energy sequence and adjusting the priority of control actions, the progressive degradation problem in the prediction and control of low-voltage switch contact faults was solved, achieving efficient monitoring and predictive adjustment of contact status and suppressing contact degradation.

CN123122030APending Publication Date: 2026-11-17浙江长征电气股份有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611629873.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing low-voltage switch condition monitoring methods are insufficient to effectively identify the progressive degradation trend of contact faults, especially during AC load switching due to differences in the polarity distribution of arcing energy and the coupling effect between control actions and the arcing process. This makes it difficult to adjust control actions in a timely manner to suppress contact degradation.

Method used

By collecting contact electrical parameters, coil drive parameters, and power supply phase reference signals of low-voltage switches, a polarity-zoned arcing energy sequence is constructed, the contact arcing bias index is calculated, fault risk prediction results are generated, and the priority and parameters of candidate control actions are adjusted according to the risk results to suppress contact degradation.

Benefits of technology

It achieves efficient prediction and control of localized contact erosion and contact performance degradation trends, enabling targeted adjustments before the failure risk develops into obvious failure, reducing the contact arcing bias index in subsequent control cycles and suppressing degradation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122030A_ABST
    Figure CN123122030A_ABST
Patent Text Reader

Abstract

This invention relates to the field of low-voltage electrical appliance condition monitoring and fault prediction control technology, specifically to a low-voltage switch fault prediction control method based on contact electrical parameters. This method collects relevant data on the opening and closing of the low-voltage switch according to the control cycle, obtains the contact degradation state vector, and calculates the contact arcing bias index. Based on this, it predicts the contact fault risk, establishes and executes predictive control actions, and generates a negative feedback control quantity for arcing energy by combining the evaluation results before and after execution. It then updates the action control priority and selects the target subsequent control action to reduce the contact arcing bias index and suppress contact degradation. This invention obtains the contact degradation state vector and contact arcing bias index by collecting contact electrical parameters, coil drive parameters, power supply phase reference signals, and contact action states within the control cycle, thereby achieving fault risk prediction, control action evaluation, and priority updating, thus reducing the subsequent contact arcing bias index and suppressing contact degradation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of low-voltage electrical appliance condition monitoring and fault prediction control technology, and more specifically, to a low-voltage switch fault prediction control method based on contact electrical parameters. Background Technology

[0002] As a fundamental electrical component in low-voltage power distribution, load switching, and electrical equipment start-up and shutdown control, the operational reliability of low-voltage switches is closely related to the condition of their contacts. During the opening and closing processes of low-voltage switches, changes in contact gap, load current, and AC power phase collectively affect the arcing process. Under long-term operating conditions, phenomena such as ablation, oxidation, material migration, contact pressure attenuation, and increased bounce may occur on the contact surface, leading to increased contact resistance, increased contact voltage drop, prolonged arcing time, and decreased operational stability. Existing low-voltage switches typically ensure operational safety through mechanical life design, electrical life verification, temperature rise monitoring, tripping protection, operation count recording, and regular maintenance, meeting basic protection and maintenance requirements under normal operating conditions.

[0003] With the increasing complexity of load types, switching frequencies, and operating environments in low-voltage switchgear applications, contact degradation exhibits more pronounced periodicity, phase correlation, and gradual accumulation characteristics. Particularly during AC load switching, arcing energy is not always uniformly distributed between the positive and negative half-cycles. Coupling relationships exist between the actual opening and closing phase angles, coil drive advance, release delay, contact bounce, and changes in contact electrical parameters. When arcing energy is consistently biased towards the same power polarity region, the arc erosion experienced by localized contact areas may accumulate unevenly, making it difficult to fully reflect contact degradation trends based solely on the number of operations, temperature rise, or instantaneous protection status.

[0004] Existing condition monitoring methods primarily focus on outcome-based information such as whether low-voltage switches have completed opening and closing actions, whether overload or short-circuit protection has occurred, and whether contact temperatures have exceeded limits. Some methods also incorporate voltage, current, or coil drive signals to analyze the operational process. While these methods are helpful in identifying abnormal actions and assessing operational status, there is still room for further refinement when it comes to predicting contact faults in advance. For example, threshold judgments based solely on electrical parameters for a single control cycle are insufficient to reflect the joint evolution of contact resistance, contact voltage drop, arcing time, arcing energy, bounce time, and actual opening and closing phase angles over multiple consecutive control cycles. Similarly, evaluating only the total arcing energy does not readily reveal the continuous bias relationship between the arcing energy in the positive and negative half-cycles.

[0005] Furthermore, the control actions of low-voltage switches typically involve parameters such as the opening and closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowable conditions, and opening and closing frequency. If the control actions are executed only according to a fixed strategy, or only after a fault occurs, it becomes difficult to adjust the subsequent opening and closing control direction in a timely manner according to changes in the contact degradation state. Especially after a predictive control action is executed, without a comparative evaluation of the contact degradation state vector and contact arcing bias index before and after execution, it is difficult to determine whether the control action has caused an increase in total arcing energy, aggravated polarity bias, arcing phase angle lock-up, or an expansion of coil contact phase angle drift. It is also difficult to make targeted corrections to the action control priority of candidate control actions.

[0006] Therefore, considering the gradual nature of low-voltage switch contact degradation, the polarity distribution of arcing energy, and the coupling effect between control actions and the arcing process, it is necessary to propose a method that can collect contact electrical parameters, coil drive parameters, power supply phase reference signals, and contact action status according to the control cycle, and combine data from multiple consecutive control cycles to predict contact fault risk and adjust control actions, so as to identify, evaluate, and suppress the arcing bias trend before the fault risk develops into obvious failure. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the present invention aims to provide a low-voltage switch fault prediction and control method based on contact electrical parameters.

[0008] To achieve the above objectives, the present invention provides the following technical solution: A fault prediction and control method for low-voltage switches based on contact electrical parameters executes the following steps according to the control cycle: Step 1: Obtain the opening and closing control command of the low-voltage switch at the time of the current control cycle, collect the contact electrical parameters, coil drive parameters, power supply phase reference signal and contact operation status of the low-voltage switch, and form the current control cycle data. Step 2: Obtain the contact degradation state vector of the current control cycle based on the current control cycle data, and construct the polarity partition arcing energy sequence based on the control cycle data corresponding to multiple consecutive control cycles; calculate the contact arcing bias index of the current control cycle based on the polarity partition arcing energy sequence. Step 3: Generate contact failure risk prediction results based on the contact degradation state vector and contact arc ignition bias index in multiple consecutive control cycles; establish a candidate control action set based on the contact failure risk prediction results; select a candidate control action from the candidate control action set as the predicted control action, and execute the predicted control action. Step 4: After the predictive control action is executed, obtain the control action evaluation result based on the contact degradation state vector and the contact arc ignition bias index before and after the predictive control action is executed, and generate the arc energy negative feedback control quantity based on the control action evaluation result. Step 5: Update the action control priority of each candidate control action in the candidate control action set according to the negative feedback regulation of arc energy, and select the target subsequent control action from the updated candidate control action set to reduce the contact arc ignition bias index in the subsequent control cycle and suppress contact degradation.

[0009] In one embodiment, the current control cycle is any of the opening and closing control cycles of the low-voltage switch to be evaluated, the contact electrical parameters include contact voltage and contact current, and the coil drive parameters include coil voltage and coil current.

[0010] In one embodiment, the polarity partitioned arc energy sequence includes positive half-cycle arc energy and negative half-cycle arc energy divided according to power supply polarity within multiple consecutive control cycles; the contact degradation state vector of the current control cycle includes contact resistance, contact voltage drop, arc time, arc energy, bounce time, actual pull-in phase angle, and actual break-out phase angle; wherein, the arc energy includes at least one of the current cycle arc energy and the cycle cumulative arc energy.

[0011] In one embodiment, the contact ignition bias index is obtained by fusing the polarity amplitude imbalance subquantity, the polarity continuous bias subquantity, the polarity trend bias subquantity, and the polarity phase angle coupling subquantity according to the polarity data confidence weight; wherein, the polarity phase angle coupling subquantity is determined based on the contact action phase angle, the positive half-cycle arc energy, and the negative half-cycle arc energy; the contact action phase angle is determined based on at least one of the actual engagement phase angle and the actual disengagement phase angle; and the polarity data confidence weight is determined based on the power supply phase reference signal, the contact current, and the contact action state.

[0012] In one embodiment, the polarity amplitude imbalance subquantitative is determined based on the normalized energy difference between the arc energy in the positive half-cycle and the arc energy in the negative half-cycle within multiple consecutive control cycles, the polarity persistence subquantitative is determined based on the degree to which the arc energy continuously biases towards the same power source polarity within multiple consecutive control cycles, and the polarity trend bias subquantitative is determined based on the increasing trend of the normalized energy difference as the control cycle progresses.

[0013] In one embodiment, the candidate control action set includes multiple candidate control actions, each of which includes at least one control action parameter among the following: opening / closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowance condition, and opening / closing frequency.

[0014] In one embodiment, the control action evaluation result includes the control action response deviation vector, polarity bias change, and control action negative contribution index; the control action response deviation vector is composed of the total arc energy excess change, polarity bias excess change, arc phase angle locking amplification, bouncing arc coupling excess, coil contact phase angle drift, and phase angle contact resistance drift, with the polarity bias excess change serving as the polarity bias change.

[0015] In one embodiment, the negative contribution index of the control action is obtained by fusing the positive deviation component in the control action response deviation vector pointing to the direction of accelerated contact degradation after safety boundary normalization, contribution weight allocation and data confidence correction; when the negative contribution index of the control action exceeds the preset negative contribution judgment threshold and the polarity bias change exceeds the preset polarity bias threshold, it is determined that the predicted control action has the risk of increased contact arc ignition bias index and accelerated contact degradation.

[0016] In one embodiment, the arc energy negative feedback control amount is obtained by fusing the total arc energy negative feedback component, the polarity balance negative feedback component, the phase angle delocking negative feedback component, and the coil timing correction component; wherein, the total arc energy negative feedback component is determined based on the control action negative contribution index and the total arc energy excess change amount, the polarity balance negative feedback component is determined based on the contact arc ignition bias index and the polarity bias change amount, the phase angle delocking negative feedback component is determined based on the arc phase angle locking amplification amount, and the coil timing correction component is determined based on the coil contact phase angle drift amount.

[0017] In one embodiment, the action control priority is updated based on the arc energy negative feedback regulation amount, and the target subsequent control action is determined based on the updated action control priority; the target subsequent control action includes at least one of the following control action parameters: opening and closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowable conditions, and opening and closing frequency.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention constructs a polarity-zoned arc energy sequence based on control cycle data corresponding to multiple consecutive control cycles, and calculates the contact ignition and arc offset index for the current control cycle accordingly. This not only reflects the overall change in arc energy, but also identifies the cumulative offset of arc energy in different power polarity regions. By combining the contact degradation state vector with the contact ignition and arc offset index to generate contact failure risk prediction results, the predictive accuracy for local contact erosion, contact performance degradation, and accelerated degradation trends can be improved. Based on the contact failure risk prediction results, a set of candidate control actions is established. Predicted control actions are selected and executed. After the execution of the predicted control actions, the contact degradation state vector and the contact arcing bias index before and after execution are compared to obtain the control action evaluation results and generate the arcing energy negative feedback control quantity. The arcing energy negative feedback control quantity is further used to update the action control priority of each candidate control action and select the target subsequent control action. This allows control action parameters such as the opening and closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowable conditions, and opening and closing frequency to be adjusted according to the contact state changes. This helps to reduce the contact arcing bias index in subsequent control cycles and suppress contact degradation. Attached Figure Description

[0019] Figure 1 This is an overall flowchart of a low-voltage switch fault prediction and control method based on contact electrical parameters according to the present invention. Figure 2 This is a schematic diagram illustrating the generation of contact failure risk prediction results and the establishment of a candidate control action set in this invention. Figure 3 This is a schematic diagram illustrating the relationship between the candidate control action set, candidate control actions, and predicted control actions in this invention. Figure 4 This is a schematic diagram illustrating the acquisition of the control action evaluation results in this invention. Detailed Implementation

[0020] Reference Figures 1 to 4 A fault prediction and control method for low-voltage switches based on contact electrical parameters, which executes the following steps according to the control cycle: Step 1: Obtain the opening and closing control command time of the low-voltage switch in the current control cycle, collect the contact electrical parameters, coil drive parameters, power supply phase reference signal, and contact operation status of the low-voltage switch, and form the current control cycle data; the opening and closing control command time serves as the time reference for various types of sampled data in this control cycle, which facilitates the corresponding analysis of contact electrical parameters, coil drive parameters, power supply phase reference signal, and contact operation status in the same operation process; Contact electrical parameters reflect the voltage and current changes of the low-voltage switch contacts during the connection, disconnection, and arcing processes. Coil drive parameters reflect the excitation state of the drive mechanism on the contact action. The power supply phase reference signal provides the basis for distinguishing the positive half-cycle, negative half-cycle, and phase angle position. The contact action state reflects the actual contact action process. The above data together form the current control cycle data, providing the original data source for obtaining the subsequent contact degradation state vector, polarity zone arcing energy sequence, and contact arcing bias index.

[0021] The timing of the opening and closing control commands of the low-voltage switch in the current control cycle is obtained. The current control cycle is any opening and closing control cycle of the low-voltage switch to be evaluated. When the opening control command or closing control command is generated, the timing of the opening and closing control command is written into the time base record of the current control cycle, and the acquisition time range is set around the timing of the opening and closing control command. Within the acquisition time range, the contact electrical parameters, coil drive parameters, power supply phase reference signal, and contact operation status of the low-voltage switch are synchronously acquired. The contact electrical parameters include contact voltage and contact current. The contact voltage records the voltage changes across the contact during the closing, opening, and arcing phases, while the contact current records the current changes flowing through the contact circuit. The coil drive parameters include coil voltage and coil current. The coil voltage records the excitation voltage when the drive coil is controlled to open and close, while the coil current records the current waveform during the pull-in, holding, or release processes of the drive coil. The power supply phase reference signal establishes a phase reference based on the changes in the AC power supply's zero-crossing point, positive half-cycle, and negative half-cycle. The contact operation status records the actual operational position changes of the contacts from open to closed or from closed to open. Taking a low-voltage switch with a rated AC voltage of 220 volts and a rated current of 32 amperes as an example, when the closing control command is issued, the moment of the closing control command is recorded as the starting point of the current control cycle. Before and after this starting point, contact voltage, contact current, coil voltage, coil current, power supply phase reference signal, and contact operation status are continuously collected and integrated into the current control cycle data according to a unified time mark. This avoids time misalignment between various electrical parameters and contact operation status, and ensures that the subsequent contact degradation state vector and polarity zone arcing energy sequence have a consistent data source.

[0022] Step 2: Obtain the contact degradation state vector of the current control cycle based on the current control cycle data, and construct a polarity partitioned arcing energy sequence based on the control cycle data corresponding to multiple consecutive control cycles; calculate the contact arcing bias index of the current control cycle based on the polarity partitioned arcing energy sequence; convert the electrical response, drive response and action state collected in a single control cycle into state quantities that can reflect the contact degradation, and distinguish the arcing energy in multiple control cycles according to the power supply polarity; The contact degradation state vector records the state changes related to contact quality, arcing intensity, phase angle, and bounce characteristics during the current control cycle. The polarity-partitioned arcing energy sequence focuses on the distribution differences of arcing energy under different power supply polarities across multiple consecutive control cycles. The contact arcing bias index for the current control cycle is obtained from this sequence and can reflect whether the arcing energy is consistently biased towards a certain power supply polarity, and whether the degree of this bias is related to the contact degradation trend.

[0023] Record No. The sampling time for each control cycle is The sampling interval is The power supply phase angle is The contact action state is The power supply phase angle, power supply phase reference signal, actual engagement phase angle, actual disengagement phase angle, arc initiation phase angle, and target action phase angle all use the same phase reference, and are designed so that the phase angle value is not less than... and less than The range is normalized and recorded; the positive and negative half-cycles are divided according to the sine sign under this phase reference. The set of arc sampling points for each control cycle is denoted as When sampling point satisfy It is in a separation transition state or a bouncing state, and , At that time, sampling points Into .in, Contact voltage, For contact current, The threshold for determining arc voltage. The threshold for determining arcing current. and Calibration is based on the no-load noise, rated load current, contact breaking voltage waveform, and arc identification test data of the same model of low-voltage switch. Arcing time. satisfy No. Arc energy during the positive half-cycle of each control cycle satisfy No. Arc energy during the negative half-cycle of each control cycle satisfy Current cycle arc energy The energy of the positive half-cycle arc With negative half-cycle arc energy The sum of the contact degradation state vectors. include , , , , , and ,in, For contact resistance, For contact voltage drop, For bounce time, This is the actual pull-in phase angle. This is the actual phase angle. Take the statistical value of the ratio of contact voltage to contact current under the steady-state condition of contact closure. The statistical value of the absolute value of the contact voltage under stable contact closure conditions is taken. This statistical value is the arithmetic mean of all valid sampled values ​​within the corresponding stable conduction interval; the absolute value of the contact current is below the conduction current threshold. The sampling points are not included in the calculation of contact resistance and contact voltage drop. Calibration is based on the current sampling noise, minimum effective conduction current, and rated load current of the closed stable range of the same type of low-voltage switch.

[0024] For the control cycle data corresponding to multiple consecutive control cycles, the arcing time period identified in each control cycle is mapped to the positive half-cycle and negative half-cycle of the power supply phase reference signal, and the positive half-cycle arcing energy and negative half-cycle arcing energy are accumulated respectively to construct the polarity partition arcing energy sequence. Continuous control cycle number set The length is . The time weight is set based on the switching frequency of low-voltage switches, the rate of contact degradation, and the amount of historical control cycle data. The weight of a historical control cycle is determined based on the time distance between the historical control cycle and the current control cycle, with the weight of a historical control cycle closer to the current control cycle not lower than that of a more distant historical control cycle. Normalized energy difference per control cycle satisfy in, The constant for energy denominator is determined based on the upper limit of energy corresponding to the sampling noise of contact voltage and contact current, or the minimum effective arc energy in the arc identification test. When the value is positive, the arc energy is biased towards the positive half-cycle; When the value is negative, the arc energy is biased towards the negative half-cycle. Time weighting exist Internal normalization, weight summation is The direction of arc offset during ignition is as follows: The sign is determined; Inside Exceeding the bias determination threshold ,and When the same sign is maintained continuously, it is recorded as a continuous bias towards the same power supply polarity.

[0025] Polarity amplitude imbalance quantum satisfy Polarity Continuous Bias Subquant Pick and The ratio, where, for Inside Exceeding the bias determination threshold The longest consecutive period with the same sign. Polarity trend bias subquantity. satisfy in, for Numbered according to control cycle The changing fitted slope, The trend-normalized boundary. Arc initiation phase angle concentration. satisfy in, The imaginary unit satisfies , For the first The power supply phase angle corresponding to the maximum sampling point of arc energy in each control cycle. Polarity phase angle coupling subquantity. Take the current control cycle Concentration of phase angle with arcing The product of.

[0026] The current control cycle's ignition arc bias index satisfy in, For polarity data, confidence weights, , , , The weights are fused, and the sum of the fused weights is... . , , , , and Calibration is performed based on the arc energy distribution data of the same type of low-voltage switch under different load categories, different switching frequencies, and different contact degradation stages. , , and Before entering fused computing, all were limited to Inside; among them, Greater than Press at time count, Greater than Press at time count. Normalized restrictions Inside.

[0027] Polarity data confidence weights The product of the phase position confidence component, current confidence component, and action confidence component is calculated. The phase position confidence component is obtained by normalizing the zero-crossing interval deviation of the power supply phase reference signal; the current confidence component is obtained by normalizing the proportion of effective current sampling points to arc sampling points within the arcing interval; and the action confidence component is obtained by normalizing the time deviation between the contact action state transition edge and the contact voltage or contact current abrupt change edge. The values ​​of the above three confidence components are all within a certain range. , The range of values ​​is .

[0028] Let the confidence threshold be... . Below At that time, the polarity zone arc energy data of this control cycle is included in the historical record, but does not trigger an increase in the contact fault risk category on its own. Not less than At that time, the contact ignition and arcing bias index of the control cycle is used in the calculation of the contact failure risk prediction result. Calibrate based on power phase sampling accuracy, contact current sampling integrity, and contact action status detection accuracy.

[0029] Step 3: Generate contact failure risk prediction results based on the contact degradation state vector and contact arcing bias index in multiple consecutive control cycles; establish a candidate control action set based on the contact failure risk prediction results; select a candidate control action from the candidate control action set as the predicted control action and execute the predicted control action; incorporate contact state changes and polarity arcing bias into the failure risk judgment, and form an executable control strategy range based on the judgment results. The contact degradation state vector over multiple consecutive control cycles reflects the changing state of the contact, while the contact ignition bias index reflects the non-equilibrium of ignition energy in the polarity direction. Combining these two types of information generates a contact failure risk prediction result. This prediction result serves as the basis for establishing a set of candidate control actions, ensuring the set matches the current contact risk state. The predicted control actions are derived from the candidate control action set, and their execution results reflect the actual impact of the selected control action on the contact ignition bias index and the contact degradation state vector.

[0030] Contact failure risk prediction results include risk values. Risk Category And abnormal component records. Record the contact degradation state vector. The Middle The normalized value of each component is The set of component numbers involved in risk calculation is . According to the The degree of deviation of each degradation component from the healthy contact reference range and the life test fault boundary is determined. Abnormal component records include degradation components exceeding the healthy contact reference range, abnormal contact ignition bias index records, bias direction records, and corresponding control cycle numbers. Risk Value satisfy in, For the first Risk weights of each degradation component The weight of the arc ignition bias index is determined by the ignition bias. Weighting of the change in the arc ignition bias index. The change in the ignition and arc-breaking bias index between adjacent control cycles is taken as the value of the current control cycle. Compared with the previous control cycle The difference. Normalized amplitude limiting function. Less than Input values ​​by Processing will be greater than Input values ​​by Process the inputs while keeping other input values ​​unchanged. , , , The risk category determination boundary is calibrated based on the health contact benchmark data, life test data, and fault sample data of the same model of low-voltage switch. , and All weights are non-negative; medium-risk boundary Below the high-risk boundary .

[0031] when Not less than the high-risk boundary hour, Recorded as high risk; when Less than And not less than the medium risk boundary hour, Recorded as medium risk; when Less than hour, Recorded as low risk. The candidate control action includes several candidate control actions. ,in, Number the candidate control action. The parameters of the candidate control action are limited to the rated operating parameters of the low-voltage switch, safe breaking capacity, coil thermal stability range, and controller output capability. When It is medium or high risk, and When continuously biased towards the same power supply polarity, candidate control actions that change the opening and closing phase angle window are recorded in the candidate control actions, and candidate control actions that change the coil drive advance or release delay are also recorded.

[0032] Step four: After the predictive control action is executed, the control action evaluation result is obtained based on the contact degradation state vector and the contact ignition bias index before and after the execution of the predictive control action. The arc energy negative feedback control quantity is then generated based on the evaluation result. The contact degradation state vector before and after the execution of the predictive control action provides a comparison of contact state changes, and the contact ignition bias index before and after the execution of the predictive control action provides a comparison of arc bias changes. The control action evaluation result, obtained from the above-mentioned relationship, reflects whether the predictive control action leads to an increase in arc energy, aggravation of polarity bias, enhanced phase angle locking, or accelerated contact degradation. The arc energy negative feedback control quantity is generated based on the control action evaluation result, reflecting the direction of feedback correction for unfavorable arc changes and contact degradation trends.

[0033] The contact degradation state vector and contact ignition bias index before the predictive control action are used as the pre-execution reference. The pre-execution reference consists of a set of reference periods. Sure, This includes historical control cycles that match the current control cycle in terms of load type, circuit breaker type, power supply phase range, and contact operation status detection conditions. It also includes the reference contact degradation state vector before execution. satisfy Pre-execution reference ignition arc offset index satisfy in, The weights for each benchmark period are: . The predicted contact degradation state vector after the execution of the control action is determined based on the differences in load current, control action parameters, and time intervals between the historical control cycle and the current control cycle. The trigger ignition bias indicator after the predictive control action is executed is marked as If the number of valid reference cycles is lower than the lower limit of the number of reference cycles, the evaluation record of this control action is a low-confidence evaluation.

[0034] Control action response deviation vector include , , , , and .in, The excess change in total arc energy, satisfying For the excess change in polarity bias, take and Difference deduction The non-negative part after deduction; the result is less than Time is recorded as . To lock the arc phase angle, take and Difference deduction The non-negative part after deduction; the result is less than Time is recorded as . For the excess of the bouncing arc coupling, take and Difference deduction The non-negative part after deduction; the result is less than Time is recorded as . The phase angle drift of the coil contacts satisfies Let be the phase angle contact resistance drift, taken as and The non-negative part of the difference; the difference is less than Time is recorded as The phase angle interval is the control phase angle window where the target action phase angle is located; when there are multiple candidate control phase angle windows, the control phase angle window corresponding to the executed predicted control action is recorded. and All are based on a phase angle value not less than and less than Range normalization processing. , , , , and These are the arc energy after the predicted control action is executed, the arc ignition bias index, the arc occurrence phase angle concentration, the overlap ratio between the bounce interval and the arc interval, the actual action phase angle, and the contact resistance within the phase angle interval. , , , and These are the baselines before execution; The phase angle of the target action; , , and To allow for fluctuations, the calibration is based on the normal fluctuation range observed during repeated switching tests of the same type of low-voltage switch.

[0035] Negative contribution of control action is marked as ,satisfy in, To evaluate the confidence level of the data, The range of values ​​is The determination is based on the number of reference cycles before execution, the proportion of effective arc sampling points after the execution of predictive control actions, and the stability of the power supply phase reference. Number the components in the control action response deviation vector. This is the set of component numbers that participate in the calculation of negative contributions. To control the deviation vector of the action response The Middle A positive deviation component For the first A safety boundary for a positive deviation component. For the first The contribution weights of each positive deviation component are given, and the sum of the contribution weights is... Safety boundary The threshold for determining negative contribution is determined based on the rated breaking capacity of the low-voltage switch, the limit of contact temperature rise, the limit of coil thermal stability, the test boundary of contact material life, and the accuracy of controller operation. and preset polarity bias threshold Calibration is based on the operational evaluation data and life test data of the same model of low-voltage switch under healthy contact conditions. Greater than ,and Greater than When the predictive control action is recorded as a negative contribution action, it is determined that the predictive control action carries the risk of increased ignition arc bias and accelerated contact degradation.

[0036] Arc energy negative feedback control quantity include , , and . For the negative feedback component of the total arc energy, satisfying For polarity-balanced negative feedback components, satisfying To lock the negative feedback component for phase angle, satisfy For coil timing correction components, satisfy in, to The negative feedback fusion coefficient is calibrated based on control action evaluation data, contact life test data, and coil drive response. It is compatible with the frequency constraints of opening and closing, the allowable conditions for reclosing, and the width of the phase angle window for opening and closing; Matching the center position of the phase angle window for opening and closing; Matching the phase angle window that avoids the phase angle locking interval of the combustion arc; It matches the coil drive advance and release delay.

[0037] Step 5: Update the action control priority of each candidate control action in the candidate control action set according to the arc energy negative feedback control amount, and select the target subsequent control action from the updated candidate control action set to reduce the contact arc ignition bias index in the subsequent control cycle and suppress contact degradation. Adjust the sorting relationship of each candidate control action in the candidate control action set according to the arc energy negative feedback control amount. The updating of action control priorities reflects the value of different candidate control actions in correcting arc energy, polarity bias, action phase angle, and coil drive timing. The target subsequent control action is determined from the updated set of candidate control actions. The selection result corresponds to the currently obtained arc energy negative feedback regulation amount, which can continuously constrain the contact arc bias index and contact degradation trend in subsequent control cycles, forming a control closed loop from acquisition, prediction, execution, evaluation to feedback regulation.

[0038] Candidate control actions Action control priority is denoted as The range of values ​​is Updated action control priority satisfy in, For priority limiting functions, it will be less than Input values ​​by Processing will be greater than Input values ​​by Process the inputs while keeping other input values ​​unchanged; Candidate control actions With arc energy negative feedback control amount Directional consistency score; Candidate control actions The corresponding negative contribution index of control actions; Increase the step size to prioritize Reduce the step size for priority. The range of values ​​is A positive value indicates a candidate control action. With arc energy negative feedback control amount The adjustment direction is the same, and a negative value indicates that the adjustment direction is opposite. Select candidate control actions Historical performance evaluation values; when historical performance evaluation values ​​have not yet been formed. Take the initial value from the operation evaluation data of the same model of low-voltage switch. The target control action is determined based on the matching degree of four categories of candidate control action parameters: reduction of total arcing energy, reduction of polarity bias, dispersion of arcing phase angle, and correction of coil timing. The target control action is selected from candidate control actions that satisfy the reclosing allowable conditions, opening and closing frequency constraints, coil thermal stability boundaries, and safe breaking capacity boundaries, according to... The size is determined.

[0039] 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 scope of the technology 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 predictive control of low-voltage switch faults based on contact electrical parameters, characterized in that, Perform the following steps according to the control cycle: Step 1: Obtain the opening and closing control command of the low-voltage switch at the time of the current control cycle, collect the contact electrical parameters, coil drive parameters, power supply phase reference signal and contact operation status of the low-voltage switch, and form the current control cycle data. Step 2: Obtain the contact degradation state vector of the current control cycle based on the current control cycle data, and construct the polarity partition arc energy sequence based on the control cycle data corresponding to multiple consecutive control cycles. The ignition bias index for the current control cycle is calculated based on the polarity zone arc energy sequence. Step 3: Generate contact failure risk prediction results based on the contact degradation state vector and contact arc ignition bias index in multiple consecutive control cycles; A set of candidate control actions is established based on the contact failure risk prediction results; one candidate control action is selected from the set of candidate control actions as the predicted control action, and the predicted control action is executed. Step 4: After the predictive control action is executed, obtain the control action evaluation result based on the contact degradation state vector and the contact arc ignition bias index before and after the predictive control action is executed, and generate the arc energy negative feedback control quantity based on the control action evaluation result. Step 5: Update the action control priority of each candidate control action in the candidate control action set according to the negative feedback regulation of arc energy, and select the target subsequent control action from the updated candidate control action set to reduce the contact arc ignition bias index in the subsequent control cycle and suppress contact degradation.

2. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 1, characterized in that, The current control cycle is any of the opening and closing control cycles of the low-voltage switch to be evaluated. The contact electrical parameters include contact voltage and contact current, and the coil drive parameters include coil voltage and coil current.

3. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 2, characterized in that, The polarity-partitioned arc energy sequence includes the positive half-cycle arc energy and negative half-cycle arc energy divided according to the power supply polarity within multiple consecutive control cycles; the contact degradation state vector of the current control cycle includes contact resistance, contact voltage drop, arc time, arc energy, bounce time, actual pull-in phase angle, and actual break-out phase angle; wherein, the arc energy includes at least one of the current cycle arc energy and the cycle cumulative arc energy.

4. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 3, characterized in that, The contact ignition bias index is obtained by fusing the polarity amplitude imbalance subquantity, the polarity continuous bias subquantity, the polarity trend bias subquantity, and the polarity phase angle coupling subquantity according to the polarity data confidence weight. Among them, the polarity phase angle coupling subquantity is determined based on the contact action phase angle, the positive half-cycle arc energy, and the negative half-cycle arc energy; the contact action phase angle is determined based on at least one of the actual engagement phase angle and the actual disengagement phase angle; and the polarity data confidence weight is determined based on the power supply phase reference signal, the contact current, and the contact action state.

5. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 4, characterized in that, The polarity amplitude imbalance subquantitative is determined based on the normalized energy difference between the arc energy in the positive half-cycle and the arc energy in the negative half-cycle within multiple consecutive control cycles. The polarity continuous bias subquantitative is determined based on the degree to which the arc energy continuously biases towards the same power source polarity within multiple consecutive control cycles. The polarity trend bias subquantitative is determined based on the increasing trend of the normalized energy difference as the control cycle progresses.

6. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 1, characterized in that, The candidate control action set includes multiple candidate control actions, and each candidate control action includes at least one control action parameter among the following: opening and closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowable conditions, and opening and closing frequency.

7. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 6, characterized in that, The control action evaluation results include the control action response deviation vector, polarity bias change, and control action negative contribution index. The control action response deviation vector consists of the total arc energy excess change, polarity bias excess change, arc phase angle locking amplification, bouncing arc coupling excess, coil contact phase angle drift, and phase angle contact resistance drift. The polarity bias excess change is used as the polarity bias change.

8. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 7, characterized in that, The negative contribution index of control action is obtained by fusing the positive deviation component in the control action response deviation vector pointing to the direction of contact degradation acceleration after safety boundary normalization, contribution weight allocation and data confidence correction; When the negative contribution index of the control action exceeds the preset negative contribution judgment threshold and the change in polarity bias exceeds the preset polarity bias threshold, it is determined that the predicted control action has the risk of increased ignition and arcing bias index and accelerated contact degradation.

9. The low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 8, characterized in that, The arc energy negative feedback control quantity is obtained by fusing the total arc energy negative feedback component, the polarity balance negative feedback component, the phase angle delocking negative feedback component, and the coil timing correction component. Among them, the total arc energy negative feedback component is determined based on the negative contribution index of the control action and the excess change of the total arc energy; the polarity balance negative feedback component is determined based on the contact arc ignition bias index and the change of the polarity bias; the phase angle delocking negative feedback component is determined based on the arc phase angle locking amplification; and the coil timing correction component is determined based on the coil contact phase angle drift.

10. A low-voltage switch fault prediction and control method based on contact electrical parameters according to claim 9, characterized in that, The action control priority is updated based on the negative feedback regulation of the arc energy, and the target subsequent control action is determined based on the updated action control priority; the target subsequent control action includes at least one of the following control action parameters: opening and closing phase angle window, coil drive advance, coil holding current, release delay, reclosing allowable conditions, and opening and closing frequency.