An automatic testing method and system for substation relay protection devices

Abstract

This application relates to the field of relay protection testing technology, specifically to an automatic testing method and system for substation relay protection devices. The method includes: pre-setting different test currents for the relay protection device, substituting them into the standard inverse time characteristic equation to obtain the theoretical operating time under different test currents; intermittently inputting each test current into the relay protection device multiple times, recording the measured operating time corresponding to each input, and forming sample points with the theoretical operating time corresponding to the test current; using a random sampling consensus algorithm with cross-operating condition constraints to extract sample points corresponding to different test currents to construct temporary straight lines; separating the slope parameter characterizing the operational state of the logic circuit and the intercept parameter characterizing the mechanical state of the output circuit; identifying the complementary performance deviation phenomenon between the logic circuit and the execution circuit of the relay protection device, and testing the relay protection device. This improves the testing accuracy of substation relay protection devices.

Description

Technical Field

[0001] This application relates to the field of relay protection testing technology, specifically to an automatic testing method and system for substation relay protection devices. Background Technology

[0002] Substation relay protection devices are core equipment ensuring the safe operation of the power grid, and they are classified into various types according to the different protected objects. Among them, inverse-time overcurrent protection devices, as an important form of relay protection device, are widely used in distribution networks and transformer protection. The operating characteristics of this type of device are: the larger the input fault current, the shorter the protection operation time. In actual engineering, the total operating time of the device consists of two parts: one part is the time consumed by the logic circuit to perform inverse-time integral calculation, which varies with the current multiple; the other part is the fixed time consumed by the output relay to drive the mechanical contacts to close.

[0003] During long-term operation, different components inside the device will age in different ways. For example, drift in the parameters of the filter capacitor in the logic circuit may lead to a decrease in the efficiency of integral operation, which manifests as an increase in operation delay (slowing down); while fatigue or break-in of the mechanical spring of the output relay may lead to a decrease in the resistance to action, which manifests as a decrease in the mechanical action delay (speeding up).

[0004] Currently, routine maintenance of relay protection typically employs a whole-machine functional test method. This involves injecting simulated fault current into the device and measuring and verifying whether its total operating time is within the standard allowable error range. However, this testing method cannot identify compensatory hidden dangers. Because the test yields the total time, when a slowdown in the logic circuit (positive deviation) and a speedup in the mechanical circuit (negative deviation) occur simultaneously, the two deviations may cancel each other out, making the total operating time still within the acceptable range. This compensatory phenomenon masks severe aging of individual components, preventing maintenance personnel from detecting early failure risks in logic modules or mechanical parts. Summary of the Invention

[0005] To address the aforementioned technical problems, the purpose of this application is to provide an automatic testing method and system for substation relay protection devices. The specific technical solution adopted is as follows: In a first aspect, embodiments of this application provide an automatic testing method for substation relay protection devices, the method comprising the following steps: By using the starting current of the relay protection device to preset different test currents, and substituting them into the standard inverse time characteristic equation, the theoretical operating time under different test currents can be obtained. Under various test currents, the relay protection device is intermittently input multiple times, and the actual action time corresponding to each input is recorded. The sample points are formed by combining the actual action time corresponding to the test current with the theoretical action time. A random sampling consensus algorithm with cross-operating condition constraints is adopted to extract sample points corresponding to different test currents to construct temporary straight lines. Through iterative calculation, the slope parameter representing the operation state of the logic circuit and the intercept parameter representing the mechanical state of the output circuit are separated, and a linear model of the measured action time with respect to the theoretical action time is fitted. Based on the slope parameter, the logic loop state index is calculated; based on the intercept parameter, the output loop state index is calculated; based on the number of interior points of the fitted linear model in all sample points, the action consistency percentage is calculated; thereby identifying the complementary performance deviation phenomenon between the logic loop and the execution loop of the relay protection device, and testing the relay protection device.

[0006] In one embodiment, the preset of its different test currents includes: The starting current setting value of the relay protection device is amplified by a first factor to obtain the first test current, and the starting current setting value of the relay protection device is amplified by a second factor to obtain the second test current, wherein the first factor is greater than the second factor.

[0007] In one embodiment, separating the slope parameter characterizing the operational state of the logic loop and the intercept parameter characterizing the mechanical state of the output loop includes: Based on the longitudinal residual distance from each sample point to the temporary line, the number of interior points corresponding to each temporary line is counted. The temporary line with the most interior points in all iterations is taken as the final linear model. The slope parameter is the slope of the final linear model, and the intercept parameter is the intercept of the final linear model.

[0008] In one embodiment, determining the logic loop state index includes: Calculate the difference between the slope parameter and the natural number 1, and the logic loop state index is the maximum value between the difference and 0.

[0009] In one embodiment, the value of the exit loop state index is the intercept parameter.

[0010] In one embodiment, the action consistency percentage is the percentage of the number of interior points of the linear model out of all sample points.

[0011] In one embodiment, the testing of the relay protection device includes: If the percentage of consistent action is less than the preset consistency threshold, it is directly determined that there is an abnormal contact in the relay protection device; otherwise, the relay protection device is tested according to the logic circuit state index and the output circuit state index.

[0012] In one embodiment, the test of the relay protection device based on the logic loop state index and the output loop state index includes: If the output circuit status index is greater than the preset upper limit threshold of mechanical status, it is determined that the relay protection device has a mechanical hysteresis fault. If the output circuit status index is less than or equal to the preset upper limit threshold of mechanical status, the relay protection device is tested again based on the logic circuit status index and the output circuit status index.

[0013] In one embodiment, if the logic loop state index is greater than a preset logic loop state abnormality threshold and the output loop state index is less than or equal to a preset output relay inherent delay nominal value, the relay protection device is determined to be in a state of complementary performance deviation between the logic loop and the execution loop; if none of the above conditions are met, the relay protection device is determined to be in a healthy state.

[0014] Secondly, embodiments of this application also provide an automatic testing system for substation relay protection devices, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0015] This application has at least the following beneficial effects: This application employs a random sampling consensus algorithm with cross-operating condition constraints to construct temporary straight lines by extracting sample points corresponding to different test currents. Through iterative calculation, a linear model of the measured action time with respect to the theoretical action time is fitted, successfully decomposing the total action time into a time-varying delay gain coefficient (associated logic loop) and a fixed delay bias coefficient (associated mechanical loop). Under the condition that the total time appears to be acceptable, it can keenly capture the compensatory state where the logic loop ages slowly but is masked by the mechanical action speeding up, providing early warning of logic plug-in failure risks and achieving accurate identification of implicit compensatory faults in relay protection devices. Furthermore, the cross-operating condition sampling constraint mechanism overcomes the algorithm instability caused by uneven data distribution (dumbbell-shaped distribution), ensuring that stable component state parameters can still be calculated even under electromagnetic interference environments, avoiding misjudgments caused by single-point sampling fluctuations. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1This is a flowchart illustrating the steps of an automatic testing method for a substation relay protection device, provided as an embodiment of this application. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an automatic testing method and system for substation relay protection devices proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the automatic testing method and system for substation relay protection devices provided in this application.

[0021] Please see Figure 1 The diagram illustrates a flowchart of an automatic testing method for a substation relay protection device according to an embodiment of this application. The method includes the following steps: S1, using the starting current of the relay protection device to preset different test currents, substitute them into the standard inverse time characteristic equation to obtain the theoretical operating time under different test currents.

[0022] To achieve decoupled analysis of the internal component states of the inverse time overcurrent protection device, it is first necessary to obtain raw data that reflects the device's operating characteristics at different integration depths.

[0023] Read the setting sheet of the inverse time overcurrent protection device to obtain the starting current setting value. The standard inverse-time characteristic equation used. Based on the preset strategy, two representative current operating points are selected, specifically: High current instantaneous trip condition: set current multiple It is 10 times, recorded as the first multiple, which is the test current. Under this operating condition, the device operates extremely quickly with a shallow integration depth, primarily affected by mechanical delay.

[0024] Critical current inverse time condition: set current multiple It is 1.2 times, recorded as the second multiple, which is the test current. Under this operating condition, the device operates slowly, the integration depth is deep, and the cumulative effect of logic operation delay is significant.

[0025] Substituting the two test current values ​​into the standard inverse time equation, the standard inverse time equation in this embodiment is: Where I is the test current, K is the theoretical motion reference value. , , These are all setting values ​​for the inverse-time overcurrent protection device, which can be directly read from inside the device to calculate two fixed theoretical operating reference values: Short benchmark :correspond The theoretical action time under input (usually in the tens of milliseconds).

[0026] Long benchmark :correspond The theoretical action time under input (usually in the range of several seconds).

[0027] These two benchmark values and This will be used as the x-axis for constructing a two-dimensional observation point set, to measure the degree of deviation of the measured response from the theoretical value.

[0028] S2, under various test currents, intermittently input multiple times to the relay protection device, record the actual action time corresponding to each input, and form a sample point with the theoretical action time corresponding to the test current.

[0029] After determining the test current and theoretical reference value, the control relay protection tester performs a discrete current injection acquisition task. To ensure that each set of action time data acquired is statistically independent and to avoid systematic errors caused by the accumulation of mechanical residual magnetism, this step strictly follows a cycle of current injection-recording-forced waiting.

[0030] The specific data collection process is divided into two independent stages: Phase 1: Acquisition of instantaneous overcurrent response sequence, specifically: controlling the relay protection tester to output current to the inverse-time overcurrent protection device. Simultaneously, the timing is started. When a closing signal is detected at the device's outlet contact, the current measured action time is recorded. And immediately cut off the current output. At this time, the forced entry time is The relay enters a zero-output standby state. During this standby period, the relay protection tester performs no operation to ensure that the relay's magnetic circuit is fully demagnetized and the contact mechanical structure has completely returned to rest. The above process is repeated. Second-rate( This generates a rapid-break response sequence. .in, To enforce the reset waiting time, this embodiment sets it to 5 seconds, defining the time interval between two adjacent injection tests. It should be noted that this duration should be set longer than the physical time required for the relay magnetic circuit to completely demagnetize and for the mechanical spring vibration to subside, in order to eliminate the influence of the previous action on the subsequent action; N represents the number of repeated samplings under a single operating condition, which is set to 20 in this embodiment. The larger the value, the more accurate the statistics on the distribution of contact jitter.

[0031] Phase Two: Acquisition of inverse-time condition response sequence, controlling the relay protection tester to output current to the inverse-time overcurrent protection device. Similarly, perform the above-described injection-record-force wait loop operation. Second-rate( Record the time of each measured action. This generates an inverse time-limited operating condition response sequence. .

[0032] Using the above acquisition strategy, two sets of uncorrelated raw time data were obtained. Each set of data contains random discrete information reflecting the characteristics of contact jitter.

[0033] To enable a unified analysis of the response characteristics of inverse-time overcurrent protection devices at different integration depths, the two sets of heterogeneous time series collected above are mapped into a unified two-dimensional feature space, constructing a multi-condition joint observation point set. The specific construction process is as follows: Create a file containing... A set of sample points Each sample point From coordinates composition: All sample points corresponding to the rapid-cut condition are denoted as the first point cloud cluster, corresponding to the multi-condition joint observation point set. The front of the middle sample points ( ), where the x-axis Unified assignment to short baseline y-axis Taken from the instantaneous trip condition response sequence The first in element This set of sample points is represented on a two-dimensional plane with its horizontal coordinate fixed at... At this location, the vertical point cloud exhibits a certain discrete distribution along the ordinate.

[0034] Similarly, all sample points corresponding to the inverse time-limited operating condition are denoted as the second point cloud cluster, corresponding to the multi-condition joint observation point set. The last N sample points ( ), where the x-axis Unified assignment to long base y-axis Taken from the inverse time-limited operating condition response sequence The first in element This set of sample points is represented on a two-dimensional plane with the horizontal coordinate fixed at... At this location, the vertical point cloud exhibits a certain discrete distribution along the ordinate.

[0035] At this point, the original time-series test data has been transformed into a point set in the feature space. Geometrically, this point set is represented as two dumbbell-shaped clusters of point clouds separated along the X-axis, providing the necessary geometric support structure for the subsequent steps of separating component parameters using a cross-condition regression algorithm.

[0036] S3 employs a random sampling consensus algorithm with cross-condition constraints to extract sample points corresponding to different test currents to construct temporary straight lines. Through iterative calculation, the slope parameter representing the operation state of the logic circuit and the intercept parameter representing the mechanical state of the output circuit are separated, and a linear model of the measured action time with respect to the theoretical action time is fitted.

[0037] To construct a multi-condition joint observation point set In this embodiment, the characteristic parameters that can independently characterize the operation state of the logic circuit and the mechanical state of the output circuit are calculated to overcome the outlier noise caused by contact jitter and the algorithm instability that may be caused by the dual-cluster data distribution. The Random Sample Consensus (RANSAC) algorithm with specific constraints is used for parameter estimation.

[0038] Before performing parameter calculations, it is essential to clarify the data analysis model upon which this embodiment is based. Although the inverse-time action equation of the relay protection device is nonlinear, the theoretical action reference value is used as the independent variable. In the feature space, the actual action time of the device and The relationship can be decoupled by engineering linearization.

[0039] Based on this, the following feature response model is constructed in this embodiment: in, (Independent variable) represents the theoretical operating reference value, that is, the operating time that the relay protection device should have under ideal conditions. (Dependent variable) represents the measured action response time, including various delay deviations. The slope parameter is defined as the time-varying delay gain coefficient. Statistically, the time-varying delay gain coefficient reflects the tendency of the deviation in the total device delay to be amplified proportionally with the integration depth (reference time). Numerically, This indicates that the actual computation time is greater than the theoretical value, which is related to the reduction in the integration efficiency of the logic circuit; The greater the deviation from 1, the more severe the aging of the logic circuit. The intercept parameter is defined as the fixed delay bias coefficient. Statistically, the fixed delay bias coefficient reflects the fixed deviation component in the total device delay that is independent of the integration depth. Numerically, It directly corresponds to the inherent time required for the mechanical contacts of the output relay to close, and is related to the inertial state of the mechanical circuit.

[0040] Due to oxidation or carbon buildup on the relay contacts, point collection In two-dimensional space, this is represented by two point cloud clusters with a certain degree of dispersion, often containing random noise points off-center. To extract the linear relationship that best represents the mainstream operating behavior of the device from these two separate point cloud clusters, while avoiding the extreme instability of the straight line slope (leverage effect) caused by random sampling within a single cluster, this embodiment implements the following improved iterative regression process: Step 1: Initialize iteration parameters and set the maximum number of iterations. In this embodiment, M=100 times, and the current optimal number of interior points is recorded. .

[0041] Step 2: Perform cross-condition constraint sampling, in the... In the next iteration ( Two sample points need to be randomly selected to construct a temporary straight line. To ensure that the constructed straight line has the maximum geometric support span, the following cross-condition sampling constraint is enforced: The selection of point A: must be from the point set The first cloud cluster in the middle (corresponding to the rapid overcurrent condition, i.e., the previous one) A sample point is randomly selected from the data. .

[0042] Selection of point B: must be from the point set The second point, cloud cluster (corresponding to the inverse time-limited operating condition, i.e., the later...) A sample point is randomly selected from the data. .

[0043] If the randomly selected points do not meet the above distribution conditions, for example, if both points come from the same working condition, the selection will be discarded and resampled. This constraint ensures that the two points A and B involved in the calculation have the maximum distance on the X-axis, thus mathematically guaranteeing the stability of the slope calculation.

[0044] Step 3: Constructing a temporary model Using constraints , Given two coordinates, calculate the current temporary slope. and temporary intercept : Step 4: Statistical Internal Point Consistency Traversing point sets All of them Each sample point. Calculate its distance to the current temporary line. Longitudinal residual distance : like Less than the preset jitter tolerance If the sample point is found to be an interior point that conforms to the current linear law, then the total number of interior points in this iteration is counted. In this embodiment, the permissible jitter deviation is... The value is set to 3ms, but implementers can set it according to their actual situation. This embodiment does not impose any restrictions on this.

[0045] Step 5: Update the optimal solution If the number of interior points of the current temporary line This means that the current line explains more data samples, and the update... And temporarily store the current parameters. For optimal model parameters .

[0046] Step 6: Parameter validity verification After completing all iterations, to ensure that the output conforms to basic engineering physics principles, the optimal model parameters are optimized. Perform validity verification. Specific verification conditions are as follows: It must be located within a reasonable physical range. In this embodiment, the physical range is... ,and It must be a positive value ( ).

[0047] If the optimal model parameters satisfy the above conditions, then the output is... and the corresponding maximum number of interior points .

[0048] If the optimal model parameters do not meet the conditions, the test data is determined to be abnormal, an error code indicating invalid sampled data is generated, and the steps of collecting the instantaneous overcurrent condition response sequence and the inverse time-limit condition response sequence are re-executed to prevent the output of incorrect diagnostic conclusions.

[0049] Through the above steps, the RANSAC algorithm is used to remove outlier noise (corresponding to contact jitter) in the point clusters, and cross-condition constraints are used to lock the connection between the high-density cores of two point cloud clusters, thereby obtaining the most stable and realistic component state parameters.

[0050] S4. Calculate the logic loop state index based on the slope parameter, calculate the output loop state index based on the intercept parameter, and calculate the action consistency percentage based on the number of interior points of the fitted linear model in all sample points. Based on this, identify the complementary performance deviation phenomenon between the logic loop and execution loop of the relay protection device, and test the relay protection device.

[0051] To transform the abstract parameters obtained from mathematical regression into intuitive health indicators of device components, and based on multidimensional threshold judgment logic, to identify various equipment states, including implicit compensatory faults.

[0052] First, obtain the optimal model parameters and maximum number of interior points This is transformed into three independent component state indices. Based on the slope parameter. Calculate the state index of the logic loop The specific calculation method is as follows: In the formula, max() is the maximum value function, and the logic loop state index is... This reflects the degree to which the efficiency of the integration operation of the logic circuit deviates from the ideal state. In the ideal state, ,at this time When the filter capacitors in the logic module age, causing the integration to slow down, It will be significantly greater than 1, resulting in a logic loop state exponent. The value increases. The larger the logic circuit state index, the more severe the aging of the logic circuit.

[0053] Based on intercept parameter Calculate the export loop state index The specific calculation method is as follows: Export loop status index The value is directly taken from the regression intercept, reflecting the pure mechanical action time of the output relay after deducting the calculation time. Output circuit state index. The larger the value, the more severe the hysteresis or jamming in the mechanical transmission part of the output relay; if the value is significantly less than the nominal value, it may indicate that the mechanical parts are abnormally sensitive or that there is a measurement error.

[0054] Based on the maximum number of interior points The percentage of consistent actions is calculated based on the proportion of all sample points. The specific calculation method is as follows: Among them, the percentage of actions consistent This value reflects the stability of the contact action. The closer the value is to 100%, the more likely the vast majority of test samples are closely following the mainstream linear law, indicating stable contact performance. If the value is low, it indicates the presence of a large number of outliers, suggesting that the contact resistance may fluctuate or mechanically vibrate due to oxidation or carbon buildup.

[0055] At this point, a three-dimensional state vector has been generated. .

[0056] Finally, a pre-defined diagnostic threshold set includes: logical state anomaly thresholds. Mechanical condition upper limit threshold The nominal value of the inherent delay of the export relay and consistency threshold In this embodiment, the logical state abnormality threshold is... Set to 0.05, upper limit threshold for mechanical state. The output relay's inherent delay is set to 35ms. Consistency threshold Set to 90%.

[0057] Based on the state vector The diagnostics are performed according to the following priority logic: Level 1: Contact stability screening, first check the percentage of consistent actions. .like This indicates an insufficient proportion of effective internal contacts. Regardless of other indicators, this directly indicates an abnormal contact in the relay protection device. This usually means that the surface condition of the output relay contacts has deteriorated, posing a risk of random vibration.

[0058] Second level: Detection of explicit mechanical faults. If the contact stability is qualified, check the status index of the output circuit. .like This means the inherent mechanical delay exceeds the allowable limit. The relay protection device is determined to have a mechanical hysteresis fault. This indicates that the mechanical components of the output relay may have a stuck shaft or spring fatigue, resulting in a significantly slower operation.

[0059] The third level: uncovering hidden compensatory faults. If the first two levels are within acceptable limits, further check for compensatory phenomena involving parameter complementarity. This is a hidden danger that traditional total time testing methods cannot detect. Compensatory faults are geometrically manifested as the fitted straight line falling within the acceptable error band (e.g., ...). Significant rotation occurs within the range. Specifically, this manifests as: the slope of the straight line... Deviation from ideal value ( (Representation logic slows down), while intercept Reverse compensation occurs ( By reducing the mechanical acceleration (i.e., ensuring the total time for both instantaneous and inverse-time operating points remains within the acceptable error band), the linear posture already indicates the component's aging trend. The specific judgment process is as follows: If the following combined conditions are simultaneously met: (This indicates that the logic loop has slowed down significantly); and (This indicates that the mechanical circuit's operating time is normal or slightly faster than normal, with no hysteresis.) The relay protection device is determined to be in a compensatory vulnerability state. At this point, although the total operating time may be within acceptable limits, the integral efficiency of the logic module has substantially decreased, only masked by the faster mechanical operation.

[0060] Level 4: Health status confirmation. If none of the above abnormal conditions are met, that is... , Furthermore, no compensation determination was triggered. The relay protection device is determined to be in a healthy state.

[0061] Through the above-mentioned hierarchical diagnostic logic, not only can explicit mechanical faults and contact abnormalities be detected, but more importantly, implicit compensatory states with complementary parameters can be identified, thus achieving a deep understanding of the equipment's health status.

[0062] Based on the same inventive concept as the above method, this application embodiment also provides an automatic testing system for substation relay protection devices, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described automatic testing methods for substation relay protection devices.

[0063] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0064] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0065] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. An automatic testing method for substation relay protection devices, characterized in that, The method includes the following steps: By using the starting current of the relay protection device to preset different test currents, and substituting them into the standard inverse time characteristic equation, the theoretical operating time under different test currents can be obtained. Under various test currents, the relay protection device is intermittently input multiple times, and the actual action time corresponding to each input is recorded. The sample points are formed by combining the actual action time corresponding to the test current with the theoretical action time. A random sampling consensus algorithm with cross-operating condition constraints is adopted to extract sample points corresponding to different test currents to construct temporary straight lines. Through iterative calculation, the slope parameter representing the operation state of the logic circuit and the intercept parameter representing the mechanical state of the output circuit are separated, and a linear model of the measured action time with respect to the theoretical action time is fitted. Based on the slope parameter, the logic loop state index is calculated; based on the intercept parameter, the output loop state index is calculated; based on the number of interior points of the fitted linear model in all sample points, the action consistency percentage is calculated; thereby identifying the complementary performance deviation phenomenon between the logic loop and the execution loop of the relay protection device, and testing the relay protection device.

2. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The preset different test currents include: The starting current setting value of the relay protection device is amplified by a first factor to obtain the first test current, and the starting current setting value of the relay protection device is amplified by a second factor to obtain the second test current, wherein the first factor is greater than the second factor.

3. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The separation of the slope parameter characterizing the operational state of the logic circuit and the intercept parameter characterizing the mechanical state of the output circuit includes: Based on the longitudinal residual distance from each sample point to the temporary line, the number of interior points corresponding to each temporary line is counted. The temporary line with the most interior points in all iterations is taken as the final linear model. The slope parameter is the slope of the final linear model, and the intercept parameter is the intercept of the final linear model.

4. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The determination of the logic loop state index includes: Calculate the difference between the slope parameter and the natural number 1, and the logic loop state index is the maximum value between the difference and 0.

5. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The value of the exit loop state index is the intercept parameter.

6. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The action consistency percentage is the proportion of the number of interior points in the linear model out of all sample points.

7. The automatic testing method for substation relay protection devices as described in claim 1, characterized in that, The testing of the relay protection device includes: If the percentage of consistent action is less than the preset consistency threshold, it is directly determined that there is an abnormal contact in the relay protection device; otherwise, the relay protection device is tested according to the logic circuit state index and the output circuit state index.

8. The automatic testing method for substation relay protection devices as described in claim 7, characterized in that, The test of the relay protection device based on the logic circuit state index and the output circuit state index includes: If the output circuit status index is greater than the preset upper limit threshold of mechanical status, it is determined that the relay protection device has a mechanical hysteresis fault. If the output circuit status index is less than or equal to the preset upper limit threshold of mechanical status, the relay protection device is tested again based on the logic circuit status index and the output circuit status index.

9. The automatic testing method for substation relay protection devices as described in claim 8, characterized in that, If the logic circuit state index is greater than the preset logic state abnormality threshold and the output circuit state index is less than or equal to the preset output relay inherent delay nominal value, the relay protection device is determined to be in a state of complementary performance deviation between the logic circuit and the execution circuit; if none of the above conditions are met, the relay protection device is determined to be in a healthy state.

10. An automatic testing system for substation relay protection devices, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-9.

Images