A relay array health state prediction method, device, equipment and medium

Abstract

The application belongs to the technical field of relay state monitoring and prediction, and particularly relates to a relay array health state prediction method, device, equipment and medium. The method comprises the following steps: generating detection result record information according to the first time interval, the second time interval, the conduction parameter value and the insulation parameter value, recording the detection result record information according to switch times or time sequences to obtain a historical detection sequence, establishing a health state model based on the historical detection sequence, generating prediction result information by using the health state model, and generating early warning information when the prediction result information meets a preset failure threshold condition. The application has the effect of realizing a closed loop from single determination to sequence prediction of state monitoring.

Description

Technical Field

[0001] This invention relates to the technical field of relay condition monitoring and prediction, and in particular to a method, apparatus, device and medium for predicting the health status of a relay array. Background Technology

[0002] Relays, as commonly used switching devices in power distribution, signal switching, and load control, are widely used in power control cabinets, automated production lines, rail transportation, automotive electronic control, and testing equipment. In practical applications, they are often arranged in arrays to meet the needs of multi-channel on / off control and multi-channel switching. When relays operate under long-term frequent operation and complex environmental conditions, factors such as contact wear, changes in contact state, coil performance drift, and changes in insulation state may lead to abnormal switching behavior, which in turn affects the reliable execution of control logic and the stability of equipment operation. Therefore, continuous monitoring of relay operating status and failure risk identification are common requirements in related fields.

[0003] Existing relay testing typically employs manual inspections or periodic testing methods. The testing content is mostly concentrated on single or a small number of electrical parameters. Testing schedules often adopt fixed cycles or fixed batch strategies, making it difficult to dynamically adjust based on the actual operating frequency of the relays, usage time, and environmental factors. At the same time, the granularity of the test data recording in the time dimension and the number of operations dimension is inconsistent, resulting in insufficient comparative correlation between different test batches and making it difficult to form a continuous and traceable historical data sequence. In array scenarios where multiple relays operate in parallel, there are also issues such as dispersed testing task scheduling, inconsistent testing result recording standards, and a lack of comparable data between different relays.

[0004] Furthermore, existing technologies for assessing relay failure risks typically rely on empirical thresholds or single-test conclusions, making it difficult to analyze parameter change trends based on historical test data and provide future risk assessments. This results in fault warnings remaining largely at the stage of "discovery after the fact" or "discovery only when failure is imminent," failing to meet the needs for continuous status management and early risk identification of relay arrays. Summary of the Invention

[0005] To achieve a closed loop of state monitoring from single judgment to sequence prediction, this invention provides a method, apparatus, device, and medium for predicting the health status of a relay array.

[0006] The above-mentioned objective of this invention is achieved through the following technical solution:

[0007] A method for predicting the health status of a relay array, the method comprising:

[0008] Signal relays are set on the two contact sides of the relay to be tested to construct the detection path corresponding to the relay to be tested.

[0009] The system acquires historical usage frequency information, test environment information, and historical test data information of the relay under test. Based on these information, it determines the combination information of the test modules and the test frequency information, and generates a multi-dimensional online test sequence.

[0010] According to the multidimensional online detection sequence, the target relay is subjected to a path state response detection operation through the detection path to obtain the first time interval and the second time interval.

[0011] According to the multidimensional online detection sequence, the conduction parameter measurement operation is performed on the target relay through the detection path to obtain the conduction parameter value;

[0012] According to the multidimensional online detection sequence, the insulation parameter measurement operation is performed on the target relay through the detection path to obtain the insulation parameter value;

[0013] The detection result record information is generated based on the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value. The detection result record information is recorded according to the number of switches or the time sequence to obtain the historical detection sequence.

[0014] A health status model is established based on historical detection sequences, and prediction results are generated using the health status model.

[0015] When the prediction results meet the preset failure threshold conditions, an early warning message is generated.

[0016] By adopting the above technical solution, a reusable detection path for a single relay under test can be formed in a relay array scenario. This allows subsequent detection actions to be executed in a predetermined order around the same detection path, reducing the uncertainty caused by repeated construction of detection paths. By introducing historical usage frequency information, test environment information, and historical detection data information, and determining the detection module combination information and detection frequency information accordingly, the arrangement of detection tasks is matched with the historical state of the relay under test. Furthermore, by associating detection time points with detection operations through a multi-dimensional online detection sequence, a schedulable and traceable online detection plan is formed. By obtaining the first time interval under the same detection sequence, The second time interval, conduction parameter value, and insulation parameter value are recorded to generate detection result information. Then, a historical detection sequence is formed according to the number of switching or time sequence, so that multiple detection results can be continuously compared in the same dimension, which is convenient for trend fitting and parameter evolution analysis. By establishing a health status model based on the historical detection sequence and outputting prediction result information, subsequent judgments no longer rely solely on single detection values, but are based on a set of predicted values ​​based on historical evolution relationships. By comparing the prediction result information with preset failure threshold conditions and generating early warning information, the early warning trigger has a clear judgment basis and is associated with the target relay, which facilitates the support of operation and maintenance decisions for the relay array.

[0017] Preferably, the step of generating a multidimensional online detection sequence based on the detection module combination information and the detection frequency information includes:

[0018] The detection time point sequence corresponding to each relay to be tested is determined based on the detection frequency information.

[0019] Based on the detection module combination information, the detection operation set corresponding to each relay to be tested is determined. The detection operation set includes at least one of the following: circuit status response detection operation, continuity parameter measurement operation, and insulation parameter measurement operation.

[0020] The detection time point sequence is associated with the detection operation set to generate a detection task item corresponding to each relay to be detected. Multiple detection task items are arranged in chronological order to obtain a multidimensional online detection sequence.

[0021] By adopting the above technical solutions, a clear detection time point sequence can be formed for each relay to be tested based on the detection frequency information, so that the detection trigger time has a traceable time reference; a corresponding set of detection operations can be determined for each relay to be tested based on the detection module combination information, so that the detection content remains consistent and controllable as the configuration changes; the detection time point sequence and the set of detection operations can be associated to generate detection task items, and multiple detection task items can be arranged in chronological order to obtain a multi-dimensional online detection sequence, so that multiple relays and multiple detection operations form a unified scheduling sequence on the same time axis, which is convenient for sequential execution, recording and backtracking.

[0022] Preferably, the step of performing a path state response detection operation on the target relay through a detection path according to a multidimensional online detection sequence to obtain a first time interval and a second time interval includes:

[0023] Determine the corresponding path state response detection operation of the target relay according to the multidimensional online detection sequence;

[0024] Based on the path status response detection operation control, the detection path is switched to the path status response detection state corresponding to the target relay;

[0025] A first state indication signal is applied to one end of the target relay by the detection path in the path state response detection state, and a state detection operation is performed at the other end of the target relay.

[0026] Send a closing control signal to the target relay, obtain the first time point when the closing control signal is sent, obtain the second time point when the state detection operation detects the first state indication signal, and obtain the first time interval based on the time difference between the first time point and the second time point;

[0027] A disconnection control signal is sent to the target relay, the third time point at which the disconnection control signal is sent is obtained, and the fourth time point is obtained when the second state indication signal is detected during the state detection operation. The second time interval is obtained based on the time difference between the third and fourth time points.

[0028] By adopting the above technical solution, the path state response detection operation corresponding to the target relay can be clearly identified under the constraints of multi-dimensional online detection sequence, and the detection path can be switched to the path state response detection state that matches the target relay, so that the detection action corresponds one-to-one with the target relay. By applying a first state indication signal at one end of the target relay and performing a state detection operation at the other end, a quantifiable time correspondence can be established between the closing control signal and the state detection result. Thus, the first time interval is obtained by the time difference between the first time point and the second time point, and the second time interval is obtained by the time difference between the third time point and the fourth time point. This allows the path state response process to be recorded on a unified time scale and used for subsequent serialization analysis and comparison.

[0029] Preferably, the step of performing a conduction parameter measurement operation on the target relay through a detection path according to a multidimensional online detection sequence to obtain conduction parameter values ​​includes:

[0030] The measurement operation for determining the conduction parameters corresponding to the target relay is performed according to the multidimensional online detection sequence.

[0031] Based on the continuity parameter measurement operation, the detection path is switched to the continuity parameter measurement state corresponding to the target relay;

[0032] The detection path, which is in the state of conducting parameter measurement, sends a closing control signal to the target relay and controls the electrical parameter measurement unit to apply a test current to the target relay.

[0033] The voltage between the contacts of the target relay under the test current is obtained by the electrical parameter measurement unit, and the conduction parameter value is determined based on the test current and the voltage between the contacts.

[0034] By adopting the above technical solution, the conduction parameter measurement operation corresponding to the target relay can be determined under the constraint of multi-dimensional online detection sequence. Based on the conduction parameter measurement operation, the detection path is switched to the conduction parameter measurement state that matches the target relay, so that the measurement path is consistent with the target relay. By sending a closing control signal and applying a test current to the target relay in the conduction parameter measurement state, the conduction parameter measurement is established under a unified excitation condition. By collecting the contact voltage under the test current and determining the conduction parameter value based on the test current and the contact voltage, the quantitative result of the conduction state has a clear calculation basis, which is convenient for comparison and tracking with historical records.

[0035] Preferably, according to the multidimensional online detection sequence, the insulation parameter measurement operation is performed on the target relay through the detection path to obtain the insulation parameter value, including:

[0036] Determine the insulation parameter measurement operation corresponding to the target relay according to the multidimensional online detection sequence;

[0037] Based on the insulation parameter measurement operation control, the detection path is switched to the insulation parameter measurement state corresponding to the target relay;

[0038] A disconnection control signal is sent to the target relay through the detection path that is in the state of measuring insulation parameters;

[0039] An electrical parameter measurement unit applies a test voltage between the two contacts of the target relay and acquires the response current, then determines the insulation parameter value based on the response current.

[0040] By adopting the above technical solution, the insulation parameter measurement operation corresponding to the target relay can be determined under the constraint of multi-dimensional online detection sequence. Based on the insulation parameter measurement operation, the detection path is switched to the insulation parameter measurement state that matches the target relay, so that the measurement path is consistent with the target relay. By sending a disconnection control signal to the target relay in the insulation parameter measurement state, the insulation parameter measurement is established on the premise that the target relay is in the disconnected state. By applying a test voltage between two contacts and obtaining the response current, and then determining the insulation parameter value based on the response current, the quantitative result of the insulation state has a clear basis for acquisition and calculation, which is convenient for subsequent recording, comparison and trend analysis.

[0041] Preferably, a health status model is established based on historical detection sequences, and prediction result information is generated using the health status model, including:

[0042] Based on historical detection sequences, extract the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence corresponding to time or number of switching operations;

[0043] The first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence are aligned by time or switching count to obtain an aligned sequence.

[0044] Based on the aligned sequence, a regression algorithm is used to perform parameter fitting to obtain the parameters of the health status model, and a health status model is generated based on the parameters of the health status model.

[0045] Based on the health status model, the future time point or the future number of switching is determined, and the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value and the insulation parameter prediction value corresponding to the future time point or the future number of switching are generated.

[0046] Prediction results are generated based on the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter.

[0047] By adopting the above technical solution, it is possible to extract the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence corresponding to time or switching frequency based on historical detection sequences, so that various detection data can form a serialized input that can be used for modeling; by performing time or switching frequency alignment processing on each sequence, an aligned sequence is obtained, so that different parameters can establish a correspondence under the same index dimension; by performing parameter fitting processing on the aligned sequence using a regression algorithm, the parameters of the health state model are obtained and a health state model is generated, so that the historical sequence and the parameter evolution relationship are mapped; by determining the future time point or the future switching frequency based on the health state model, the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter are output, so that the prediction output has a unified index; by aggregating the prediction values ​​to generate prediction result information, the subsequent threshold determination and early warning generation have a data foundation.

[0048] Preferably, when the prediction result information meets the preset failure threshold condition, a warning message is generated, including:

[0049] The preset failure threshold conditions include a first time interval threshold condition, a second time interval threshold condition, a conduction parameter threshold condition, and an insulation parameter threshold condition;

[0050] The predicted value of the first time interval is compared with the first time interval threshold condition, the predicted value of the second time interval is compared with the second time interval threshold condition, the predicted value of the conduction parameter is compared with the conduction parameter threshold condition, and the predicted value of the insulation parameter is compared with the insulation parameter threshold condition to obtain threshold determination result information;

[0051] When the threshold determination result information indicates that at least one of the corresponding preset failure threshold conditions is met, an early warning information is generated.

[0052] By adopting the above technical solution, the preset failure threshold conditions can be refined into a first time interval threshold condition, a second time interval threshold condition, a conduction parameter threshold condition, and an insulation parameter threshold condition, so that different predicted quantities correspond to clear judgment criteria. By comparing the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter with the corresponding threshold conditions and generating threshold judgment result information, the comparison process and conclusion of each predicted quantity are recorded in a structured manner. When the threshold judgment result information indicates that at least one of the corresponding preset failure threshold conditions is met, an early warning information is generated, so that the early warning trigger and the threshold judgment result are correlated, which facilitates the recording and traceability of risk warnings for the target relay.

[0053] The second objective of this invention is achieved through the following technical solution:

[0054] A relay array health status prediction device, the relay array health status prediction device comprising:

[0055] The detection path construction module is used to set up signal relays on the two contact sides of the relay to be tested, and construct the detection path corresponding to the relay to be tested.

[0056] The detection sequence generation module is used to acquire historical usage frequency information, test environment information, and historical test data information of the relay to be tested, and to determine the combination information of detection modules and detection frequency information based on the historical usage frequency information, test environment information, and historical test data information, and to generate a multi-dimensional online detection sequence based on the combination information of detection modules and detection frequency information.

[0057] The path status response detection module is used to perform path status response detection operation on the target relay through the detection path according to the multidimensional online detection sequence, and obtain the first time interval and the second time interval.

[0058] The continuity parameter measurement module is used to perform continuity parameter measurement operations on the target relay through the detection path according to the multi-dimensional online detection sequence, and obtain the continuity parameter value;

[0059] The insulation parameter measurement module is used to perform insulation parameter measurement operations on the target relay according to a multi-dimensional online detection sequence and through the detection path to obtain insulation parameter values.

[0060] The detection result recording module is used to generate detection result recording information based on the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value, and to record the detection result recording information according to the number of switches or the time sequence to obtain the historical detection sequence;

[0061] The health status model construction and prediction module is used to build a health status model based on historical detection sequences and generate prediction results information using the health status model;

[0062] The early warning generation module is used to generate early warning information when the prediction result information meets the preset failure threshold conditions.

[0063] By adopting the above technical solution, a reusable detection path for a single relay under test can be formed in a relay array scenario. This allows subsequent detection actions to be executed in a predetermined order around the same detection path, reducing the uncertainty caused by repeated construction of detection paths. By introducing historical usage frequency information, test environment information, and historical detection data information, and determining the detection module combination information and detection frequency information accordingly, the arrangement of detection tasks is matched with the historical state of the relay under test. Furthermore, by associating detection time points with detection operations through a multi-dimensional online detection sequence, a schedulable and traceable online detection plan is formed. By obtaining the first time interval under the same detection sequence, The second time interval, conduction parameter value, and insulation parameter value are recorded to generate detection result information. Then, a historical detection sequence is formed according to the number of switching or time sequence, so that multiple detection results can be continuously compared in the same dimension, which is convenient for trend fitting and parameter evolution analysis. By establishing a health status model based on the historical detection sequence and outputting prediction result information, subsequent judgments no longer rely solely on single detection values, but are based on a set of predicted values ​​based on historical evolution relationships. By comparing the prediction result information with preset failure threshold conditions and generating early warning information, the early warning trigger has a clear judgment basis and is associated with the target relay, which facilitates the support of operation and maintenance decisions for the relay array.

[0064] The above-mentioned objective three of the present invention is achieved through the following technical solution:

[0065] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described relay array health status prediction method.

[0066] The above-mentioned objective four of the present invention is achieved through the following technical solution:

[0067] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described relay array health status prediction method.

[0068] In summary, the present invention has at least one of the following beneficial technical effects:

[0069] 1. By adopting the above technical solution, a reusable detection path for a single relay under test can be formed in a relay array scenario. This allows subsequent detection actions to be executed in a predetermined order around the same detection path, reducing the uncertainty caused by repeated construction of detection paths. By introducing historical usage frequency information, test environment information, and historical detection data information, and determining the detection module combination information and detection frequency information accordingly, the arrangement of detection tasks is matched with the historical state of the relay under test. Furthermore, by associating detection time points with detection operations through a multi-dimensional online detection sequence, a schedulable and traceable online detection plan is formed. The first time interval is obtained within the same detection sequence. The system records the second time interval, conduction parameter values, and insulation parameter values, generating detection result records. These records are then arranged according to the number of switches or time sequence to form a historical detection sequence. This allows for continuous comparison of multiple detection results within the same dimension, facilitating trend fitting and parameter evolution analysis. By establishing a health status model based on the historical detection sequence and outputting prediction results, subsequent judgments no longer rely solely on single detection values ​​but are based on a set of predicted values ​​rooted in historical evolutionary relationships. Furthermore, by comparing the prediction results with preset failure threshold conditions and generating early warning information, the early warning trigger has a clear basis for judgment and is associated with the target relay, supporting decision-making regarding the operation and maintenance of the relay array. Attached Figure Description

[0070] Figure 1 This is an implementation diagram of a relay array health status prediction method in one embodiment of the present invention.

[0071] Figure 2 This is a circuit diagram of a relay array health status prediction method in one embodiment of the present invention.

[0072] Figure 3 This is a circuit diagram of a relay array health status prediction method in one embodiment of the present invention;

[0073] Figure 4 This is another circuit diagram in a relay array health status prediction method according to an embodiment of the present invention. Detailed Implementation

[0074] The present invention will be further described in detail below with reference to the accompanying drawings.

[0075] In one embodiment, such as Figure 1 , Figure 2 and Figure 3 As shown, this invention discloses a method for predicting the health status of a relay array, which specifically includes the following steps:

[0076] S10: Set a signal relay on the two contact sides of the relay to be tested to construct the detection path corresponding to the relay to be tested.

[0077] In this embodiment, SUM1, SUM2, and SUM3 refer to the detection interface identifiers on the static test equipment side; K1 to K12 refer to the identifiers of the relays to be tested in the relay array; T1 to T8 refer to the detection channel identifiers corresponding to the detection function; m11, m12, m21, m22, m31, m32, m41, and m42 refer to the measurement connection point identifiers electrically connected to the contact side of the relays to be tested; and s11, s12, s21, s22, s31, s32, s41, and s42 refer to the switching connection point identifiers electrically connected to the switching terminal of the signal relay. The tens digit of the number is used to represent the relay number to be tested, and the units digit is used to represent the contact side number.

[0078] The dynamic path performance verification module refers to the detection module corresponding to the path state response detection operation. The detection channels for the dynamic path performance verification module are labeled T1 and T2. T1 is used to output a high-level signal and form an electrical connection with one end of the target relay, while T2 is used to perform level detection and form an electrical connection with the other end of the target relay. The contact resistance measurement module refers to the detection module corresponding to the conduction parameter measurement operation. The detection channels for the contact resistance measurement module are labeled T3, T4, T5, and T6. T3 and T4 are used to apply a test current to the target relay and form an electrical connection, while T5 and T6 are used to collect the voltage between the contacts and form an electrical connection. The insulation leakage current measurement module refers to the detection module corresponding to the insulation parameter measurement operation. The detection channels for the insulation leakage current measurement module are labeled T7 and T8. T7 is used to apply a test voltage between the two contacts of the target relay and form an electrical connection. T8 is used to acquire the response current and form an electrical connection. K1, K2, K3, and K4 refer to the relay identifiers of the relay array to be tested. m11, m12, m21, m22, m31, m32, m41, and m42 refer to the measurement connection point identifiers that are electrically connected to the contact side of the relay to be tested. s11, s12, s21, s22, s31, s32, s41, and s42 refer to the switching connection point identifiers that are electrically connected to the switching terminal of the signal relay. The tens digit is used to represent the relay number to be tested, and the units digit is used to represent the contact side number.

[0079] Specifically, firstly, the target relay is identified based on the relay identifier in the relay array and its number X is obtained. Then, signal relays are set on both contact sides of the target relay, with the measurement connection point corresponding to the first contact side of the target relay identified as mX1, the measurement connection point corresponding to the second contact side of the target relay identified as mX2, the switching connection point corresponding to the first contact side of the target relay identified as sX1, and the switching connection point corresponding to the second contact side of the target relay identified as sX2, where X is the target relay number. Next, the detection interface identifiers SUM1, SUM2, and SUM3 on the static test equipment side are electrically connected to the input side connection line of the relay array, and the output side connection line of the relay array is electrically connected to the device under test (DUT), so that the on / off path of the DUT forms a switchable electrical connection via the relay array. Then, the detection operation corresponding to the target relay is determined based on the multi-dimensional online detection sequence, and the detection module and detection channel corresponding to the detection operation are selected. When the detection operation is a path state response detection operation... Select the dynamic path performance verification module and select detection channel identifiers T1 and T2. Establish an electrical connection between T1 and mX1, and between T2 and mX2. When the detection operation is a continuity parameter measurement operation, select the contact resistance measurement module and select detection channel identifiers T3, T4, T5, and T6. Establish an electrical connection between T3 and mX1, T4 and mX2, T5 and mX1, and T6 and mX2. When the detection operation is an insulation parameter measurement operation, select the insulation leakage current measurement module and select detection channel identifiers T7 and T8. Establish an electrical connection between T7 and mX1, and between T8 and mX2. Then, control the switching terminal of the signal relay electrically connected to sX1 to be in the conducting state, and control the switching terminal of the signal relay electrically connected to sX2 to be in the conducting state. This connects the selected detection channel to the first contact side of the target relay via mX1 and sX1, and to the second contact side of the target relay via mX2 and sX2, thereby constructing the detection path corresponding to the target relay.

[0080] S20: Obtain the historical usage frequency information, test environment information, and historical test data information of the relay to be tested, and determine the detection module combination information and detection frequency information based on the historical usage frequency information, the test environment information, and the historical test data information, and generate a multi-dimensional online detection sequence according to the detection module combination information and the detection frequency information.

[0081] In this embodiment, historical usage frequency information refers to the number of times the relay under test is closed and opened, statistically analyzed by time or by the number of switches. Test environment information refers to environmental parameter information related to the operation of the relay under test. Historical test data information refers to the record information of the first time interval, second time interval, conduction parameter value, and insulation parameter value of the relay under test in previous tests. Test module combination information refers to the selection and combination information of the circuit status response test operation, conduction parameter measurement operation, and insulation parameter measurement operation. Test frequency information refers to the frequency information of the test time points arranged for the relay under test. Multidimensional online test sequence refers to the test task sequence information obtained by associating the test time point sequence with the test operation set and arranging them in chronological order.

[0082] Specifically, firstly, the number of closing and opening cycles within a preset statistical period is read from the operation record of the relay under test to form historical usage frequency information. Then, environmental parameters within the same operating cycle as the relay under test are read from the environmental record to form test environment information. Next, the first time interval record value, second time interval record value, conduction parameter record value, and insulation parameter record value corresponding to the relay under test are extracted from the test result record information to form historical test data information. Then, the historical usage frequency information is matched with a preset frequency range to determine the basic test frequency. The test environment information is matched with a preset environmental range to perform frequency adjustment processing on the basic test frequency. Simultaneously, the historical test data information is matched with a preset data range to perform correction processing on the adjusted test frequency to obtain the test frequency information. Finally, based on the historical test... The relative positional relationship between the first time interval record value, the second time interval record value, the continuity parameter record value, and the insulation parameter record value and the corresponding preset threshold interval in the test data information is used to select at least one of the following operations: path state response detection operation, continuity parameter measurement operation, and insulation parameter measurement operation, and form detection module combination information. Then, based on the detection frequency information, a corresponding detection time point sequence is generated, and the detection time point sequence is associated with the detection module combination information to generate detection task items. Finally, multiple detection task items are arranged in chronological order to obtain a multidimensional online detection sequence. For example, when the historical usage frequency information indicates that the number of closing and opening times is in a high range and the continuity parameter record value in the historical test data information is close to the upper boundary of the preset threshold interval, the detection module combination information includes the continuity parameter measurement operation and the detection frequency information corresponds to a denser detection time point sequence.

[0083] S30: According to the multidimensional online detection sequence, the target relay is subjected to a path state response detection operation through the detection path to obtain a first time interval and a second time interval.

[0084] In this embodiment, Figure 4The relay control signal in this context refers to the control signal used to control the switching of the target relay contact state. Figure 4 The high-level signal in the detection path refers to the status indication signal applied in the detection path and used to characterize the path state. Figure 4 Level detection in this context refers to the process of sampling the signal level at the other end of the target relay and determining the circuit status accordingly. Figure 4 The activation interval ΔT1 corresponds to the first time interval. Figure 4 The closing interval ΔT2 corresponds to the second time interval.

[0085] Specifically, firstly, the path state response detection operation associated with the target relay is obtained by parsing the multidimensional online detection sequence, and the detection path is controlled to be in the path state response detection state corresponding to the target relay. Then, a high-level signal is applied to one end of the target relay through the detection path, and level detection is started at the other end of the target relay to obtain a level detection sampling value sequence. The level detection sampling value sequence refers to the set of level values ​​obtained by continuous sampling according to a preset sampling period. Then, a closing control signal is sent to the target relay, and the first time point of issuing the closing control signal is obtained. The first time point is determined by the output trigger time of the closing control signal. Then, based on the level detection sampling value sequence, the time when the level changes from low level to high level is identified, and a second time point is obtained. The first time point is determined by the sampling time when the high-level signal is first detected. Then, the time difference between the first and second time points is calculated to obtain the first time interval. The time difference calculation is done by subtracting the first time point from the second time point. Next, a disconnection control signal is sent to the target relay and the third time point of issuing the disconnection control signal is obtained. The third time point is determined by the output trigger time of the disconnection control signal. Then, the time point when the level changes from high to low based on the level detection sampling value sequence is identified and the fourth time point is obtained. The fourth time point is determined by the sampling time when the high-level signal is first detected as invalid. Then, the time difference between the third and fourth time points is calculated to obtain the second time interval. The time difference calculation is done by subtracting the third time point from the fourth time point.

[0086] S40: According to the multidimensional online detection sequence, the conduction parameter measurement operation is performed on the target relay through the detection path to obtain the conduction parameter value.

[0087] In this embodiment, the continuity parameter measurement operation refers to the process of applying a preset test current and collecting the voltage between the contacts through the detection path when the target relay is in the closed state. The continuity parameter value refers to the parameter value used to characterize the continuity state of the target relay, calculated based on the test current and the voltage between the contacts. The test current refers to the preset current value used for continuity parameter measurement, and the voltage between the contacts refers to the voltage value formed between the two contacts of the target relay under the action of the test current.

[0088] Specifically, first, the next detection task item in the multidimensional online detection sequence is read and parsed to obtain the target relay identification information and the conduction parameter measurement operation identification information. Then, based on the target relay identification information, the detection path is controlled to complete the electrical connection switching with the target relay so that the detection path passes through the two contact sides of the target relay. Next, a closing control signal is sent to the target relay, and the electrical connection between the two ends of the detection path and the two contacts of the target relay is confirmed based on the closing state. Then, the test current is applied through the detection path, and the voltage between the contacts is collected during the period when the test current remains stable. Then, the ratio between the voltage between the contacts and the test current is calculated to obtain the conduction parameter value. The ratio calculation is completed by dividing the value of the voltage between the contacts by the value of the test current. For example, when the test current is 10mA and the voltage between the contacts is 2mV, 2mV is divided by 10mA to obtain 0.2Ω as the conduction parameter value.

[0089] S50: According to the multidimensional online detection sequence, the insulation parameter measurement operation is performed on the target relay through the detection path to obtain the insulation parameter value.

[0090] In this embodiment, the insulation parameter measurement operation refers to the process of applying a test voltage and collecting the response current between the two contacts through the detection path when the target relay is in the open state. The insulation parameter value refers to the parameter value determined based on the response current to characterize the open state of the target relay. The test voltage refers to the preset voltage value used for insulation parameter measurement. The response current refers to the current value flowing between the two contacts of the target relay under the action of the test voltage.

[0091] Specifically, first, the next detection task item in the multidimensional online detection sequence is read and parsed to obtain the target relay identification information and the insulation parameter measurement operation identification information. Then, according to the target relay identification information, the detection path is controlled to complete the electrical connection switching with the target relay so that the detection path passes through the two contact sides of the target relay. Then, a disconnection control signal is sent to the target relay, and the disconnection state is confirmed to be in a disconnected electrical connection state between the two contacts of the target relay. Then, the test voltage is applied between the two contacts of the target relay through the detection path, and the response current is collected during the time period when the test voltage is stable. Then, the response current is recorded as the insulation parameter value. Alternatively, the response current is sampled multiple times within a preset sampling window, and the average calculation is performed on the multiple sampled values ​​to obtain the insulation parameter value. The average calculation is performed by adding the multiple sampled values ​​and dividing by the number of samplings. For example, when the test voltage is 100V and the collected response current is 50nA, or when 48nA, 52nA, and 50nA are collected within the sampling window and the average value is calculated to be 50nA, 50nA is used as the insulation parameter value.

[0092] S60: Generate detection result record information based on the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value, and record the detection result record information according to the number of switching or time sequence to obtain a historical detection sequence.

[0093] In this embodiment, the detection result record information refers to the record content information corresponding to a single detection task, and the historical detection sequence refers to the sequence information obtained by arranging the detection result record information corresponding to multiple detection tasks in order of the number of switches or time.

[0094] Specifically, after completing the circuit status response detection operation, conduction parameter measurement operation, and insulation parameter measurement operation, the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value are summarized. The target relay identification information and the detection time point information or switching count information corresponding to the detection task item are then read. Finally, the target relay identification information, the detection time point information or switching count information, the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value are written into the same detection result record information in a preset field order. The preset field order refers to the agreed-upon order of the fields in the record content, such as by the target relay identification... Information, detection time point information, number of switches information, first time interval, second time interval, conduction parameter value, and insulation parameter value are written in sequence. Then, the detection result record information generated each time is appended to the record set and sequential insertion processing is performed according to the detection time point information or the number of switches information. Sequential insertion processing means finding a record position in the record set that is larger than the current detection time point information or the current number of switches information, and inserting the current detection result record information before that record position, so that the record set is always arranged in ascending order according to the detection time point information or the number of switches information. Finally, the record set obtained by ascending arrangement is determined as the historical detection sequence.

[0095] S70: Establish a health status model based on the historical detection sequence, and use the health status model to generate prediction result information.

[0096] In this embodiment, the health status model refers to a model obtained by regression algorithm and used to characterize the relationship between the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value and the change over time or the number of switching operations. The prediction result information refers to the set of predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter corresponding to future time points or future switching operations based on the health status model.

[0097] Specifically, first, the system reads each detection result record from the historical detection sequence and sequentially extracts the corresponding switching frequency information or detection time point information, first time interval, second time interval, conduction parameter value, and insulation parameter value. Then, using the switching frequency information or detection time point information as independent variables and the first time interval, second time interval, conduction parameter value, and insulation parameter value as dependent variables, it constructs a training data set for the regression algorithm. The training data set for the regression algorithm refers to a data set composed of multiple sets of independent variables and multiple sets of dependent variables according to their corresponding relationships. For example, when the switching frequency information is 1000, 2000, and 3000, the corresponding conduction parameter values ​​are 0.2Ω, 0.23Ω, and 0.27Ω, respectively, and the above correspondence is written into the training data set. Then, a parameter fitting operation is performed on the training data set to determine the parameters of the health state model. The parameter fitting operation is... This refers to the process of solving for model parameters by minimizing regression error, where regression error is the difference between the output value of the health status model and the recorded values ​​in the historical detection sequence. The health status model parameters are then bound to the model form of the regression algorithm to generate the health status model. Subsequently, based on the maximum number of switching events or the maximum detection time point information in the historical detection sequence, the future number of switching events or the future time point information is determined and substituted into the health status model as input to obtain the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter. Finally, the future number of switching events or the future time point information, along with the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter, are collected in the prediction result information according to the preset field order.

[0098] S80: When the prediction result information meets the preset failure threshold condition, an early warning information is generated.

[0099] In this embodiment, the preset failure threshold condition refers to the set of conditions used to perform threshold determination on the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter, and the warning information refers to the prompt content information used to characterize the threshold determination result and associated with the target relay.

[0100] Specifically, the prediction results are first read and the future time point information or future switching frequency information, the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value are extracted. Then, the preset failure threshold conditions are read and the first time interval threshold conditions, the second time interval threshold conditions, the conduction parameter threshold conditions, and the insulation parameter threshold conditions are obtained respectively. The threshold conditions refer to the judgment rules composed of threshold values ​​and comparison relationships. For example, the first time interval threshold condition can use the comparison relationship that the first time interval prediction value is greater than the preset first time interval threshold; the conduction parameter threshold condition can use the comparison relationship that the conduction parameter prediction value is greater than the preset conduction parameter threshold; and the insulation parameter threshold condition can use the comparison relationship that the insulation parameter prediction value is greater than the preset insulation parameter threshold or the comparison relationship that the insulation parameter prediction value is less than the preset insulation parameter threshold. Finally, the first time interval prediction value is compared with the first time interval threshold conditions. The system compares and outputs the first time interval threshold determination result, compares the second time interval predicted value with the second time interval threshold condition and outputs the second time interval threshold determination result, compares the conduction parameter predicted value with the conduction parameter threshold condition and outputs the conduction parameter threshold determination result, compares the insulation parameter predicted value with the insulation parameter threshold condition and outputs the insulation parameter threshold determination result, and then performs a summary determination on the first time interval threshold determination result, the second time interval threshold determination result, the conduction parameter threshold determination result, and the insulation parameter threshold determination result to obtain threshold determination result information. The summary determination means that when at least one threshold determination result indicates that the corresponding threshold condition is met, the threshold determination result information is marked as meeting the preset failure threshold condition. Finally, when the threshold determination result information is marked as meeting the preset failure threshold condition, an early warning information is generated and written into the target relay identification information and future time point information or future switching number information.

[0101] In one embodiment, such as Figure 2 As shown, in step S20, namely generating a multidimensional online detection sequence based on the detection module combination information and the detection frequency information, the following steps are included:

[0102] S201: Determine the detection time point sequence corresponding to each relay to be detected based on the detection frequency information.

[0103] In this embodiment, the detection time point sequence refers to a set of multiple detection time point information arranged sequentially on the time axis, and each detection time point information is used to indicate the time position of the detection task.

[0104] Specifically, the detection frequency information is first read and parsed to obtain the frequency parameter, which is used to characterize the time interval between two adjacent detection time points. Then, the starting time point information corresponding to each relay to be tested is read and used as the first item of the detection time point sequence. Then, the subsequent detection time point information is generated in a recursive manner based on the frequency parameter. The recursive manner means adding the previous detection time point information to the frequency parameter to obtain the next detection time point information and repeating the process to form a sequence. For example, when the detection frequency information indicates that the time interval between adjacent detection time points is 10 minutes and the starting time point information is 08:00, 08:10, 08:20, and 08:30 are generated sequentially as the subsequent detection time point information. Then, the recursive generation stops according to the preset detection cycle termination condition. The detection cycle termination condition is used to limit the length of the detection time point sequence. The detection cycle termination condition can be achieved by reaching a preset end time point information or reaching a preset number of detections. Finally, the multiple detection time point information obtained in chronological order are collected to determine the detection time point sequence.

[0105] S202: Based on the combined information of the detection modules, determine the set of detection operations corresponding to each relay to be tested, wherein the set of detection operations includes at least one of the following: a circuit status response detection operation, a continuity parameter measurement operation, and an insulation parameter measurement operation.

[0106] In this embodiment, the detection operation set refers to a set of one or more detection operation identification information associated with the same relay to be tested. The detection operation identification information is used to indicate the specific selection result among the path status response detection operation, the continuity parameter measurement operation, and the insulation parameter measurement operation.

[0107] Specifically, the detection module combination information is first read and parsed to obtain the combination content. The combination content is used to characterize the selection relationship of the circuit state response detection operation, the continuity parameter measurement operation, and the insulation parameter measurement operation. Then, for each relay to be tested, the combination content associated with the relay to be tested is determined and the combination content is converted into a set of detection operation identification information. The conversion means mapping the selected detection operation in the combination content to the corresponding detection operation identification information and aggregating multiple detection operation identification information to form a set. For example, when the detection module combination information indicates that the circuit state response detection operation and the continuity parameter measurement operation are selected for a certain relay to be tested, but the insulation parameter measurement operation is not selected, the circuit state response detection operation identification information and the continuity parameter measurement operation identification information are aggregated as the detection operation set. As another example, when the detection module combination information indicates that only the insulation parameter measurement operation is selected for another relay to be tested, the insulation parameter measurement operation identification information is used as the detection operation set. Finally, the detection operation identification information set obtained for each relay to be tested is determined as the corresponding detection operation set.

[0108] S203: Associate the detection time point sequence with the detection operation set to generate a detection task item corresponding to each relay to be detected, and arrange the multiple detection task items in time order to obtain the multidimensional online detection sequence.

[0109] In this embodiment, a detection task item refers to a recording unit used to describe the content of a single detection execution, and a multidimensional online detection sequence refers to a sequence of information formed by arranging multiple detection task items in the order of their detection time points.

[0110] Specifically, for each relay to be tested, the detection time point information in the detection time point sequence is read and the detection operation identifier information in the detection operation set is read. Then, a detection task item is generated based on the combination of the detection time point information and the detection operation identifier information. Generation means writing the relay identifier information, detection time point information and detection operation identifier information into the same record in the order of preset fields to form a detection task item. For example, the content of the relay identifier information being R1, the detection time point information being 08:10, and the detection operation identifier information being the conduction parameter measurement operation identifier information is written into a detection task item. Then, multiple detection task items for the same relay to be tested are arranged in ascending order of detection time point information to form a local sequence. Subsequently, the local sequences corresponding to multiple relays to be tested are summarized and global sorting is performed based on the detection time point information. Global sorting means comparing the detection time point information of different detection task items and rearranging them in chronological order to obtain a unified sequential queue. Finally, the set of detection task items after global sorting is determined as a multi-dimensional online detection sequence.

[0111] In one embodiment, such as Figure 3 As shown, in step S30, i.e., according to the multidimensional online detection sequence, the target relay is subjected to a path state response detection operation through the detection path to obtain a first time interval and a second time interval, including:

[0112] S301: Determine the path status response detection operation corresponding to the target relay according to the multidimensional online detection sequence.

[0113] In this embodiment, the target relay refers to the relay to be detected determined by the detection task item in the multidimensional online detection sequence, and the path status response detection operation refers to the detection execution content corresponding to the detection operation type identified in the detection task item being the path status response detection operation.

[0114] Specifically, the next detection task item in the multidimensional online detection sequence is first read and parsed to obtain the identification information of the relay to be detected, the detection time point information, and the detection operation identification information. Then, the identification information of the relay to be detected is determined as the target relay identification information, and the target relay is determined accordingly. Next, the detection operation identification information is matched with the path status response detection operation identification information to determine whether the detection task item indicates a path status response detection operation. When the matching result indicates that they are consistent, the detection task item is determined as the path status response detection operation corresponding to the target relay, and the execution flow of the path status response detection operation is entered. When the matching result indicates that they are inconsistent, the next detection task item in the multidimensional online detection sequence is read again, and the parsing and matching process is repeated until a detection task item whose detection operation identification information is the path status response detection operation identification information is matched is found.

[0115] S302: Based on the path status response detection operation, control the detection path to switch to the path status response detection state corresponding to the target relay.

[0116] In this embodiment, the path status response detection state refers to the path connection state in which the detection path and the two contact sides of the target relay form an electrical connection and meet the requirements for performing the path status response detection operation.

[0117] Specifically, the target relay is first located based on the target relay identification information corresponding to the path status response detection operation, and the detection path connection relationship associated with the target relay identification information is read. The detection path connection relationship refers to the connection correspondence used to indicate that the two ends of the detection path are respectively connected to the two contact sides of the target relay. Then, the switching state of the control signal relay is adjusted so that the detection path passes through the two contact sides of the target relay according to the detection path connection relationship. The switching state of the control signal relay includes controlling the first contact side signal relay to switch to the conducting state and controlling the second contact side signal relay to switch to the conducting state, so that the first wire of the detection path forms an electrical connection with the first contact side of the target relay and the second wire of the detection path forms an electrical connection with the second contact side of the target relay. Subsequently, a connectivity confirmation process is performed on the electrical connection state of the detection path. The connectivity confirmation process refers to applying an electrical signal for connectivity confirmation at both ends of the detection path and determining whether the electrical signal can be transmitted along the detection path to the expected position to confirm that the detection path and the two contact sides of the target relay have formed an electrical connection. Finally, when the connectivity confirmation process indicates that the detection path meets the execution conditions of the path status response detection operation, the current electrical connection state of the detection path is determined as the path status response detection state corresponding to the target relay.

[0118] S303: The detection path, which is in the state of the path state response detection state, applies a first state indication signal to one end of the target relay and performs a state detection operation at the other end of the target relay.

[0119] In this embodiment, the first state indication signal is an electrical signal used to indicate the conduction state of the detection path, and the state detection operation is the operation process of sampling the electrical signal at the other end of the target relay and outputting the state detection result.

[0120] Specifically, the application end and detection end of the detection path are first determined. The application end refers to the detection path port connected to one end of the target relay and used to output the first state indication signal. The detection end refers to the detection path port connected to the other end of the target relay and used to perform the state detection operation. Then, the first state indication signal is output through the application end and kept in a stable output state for a preset holding time. The stable output state means that the level or amplitude of the first state indication signal remains unchanged within a preset allowable fluctuation range. Subsequently, the state detection operation is performed at the detection end to sample the level or amplitude acquired by the detection end and compare the sampled value with a preset judgment threshold to obtain the state detection result. The sampling is completed by repeatedly reading the level of the detection end according to a preset sampling period. The comparison is completed by judging whether the sampled value is higher or lower than the preset judgment threshold. For example, when the first state indication signal is a high-level signal and the preset judgment threshold is 2.0V, if the sampled value of the detection end is higher than 2.0V, the state detection result is determined to be that the first state indication signal is detected. If the sampled value of the detection end is lower than 2.0V, the state detection result is determined to be that the first state indication signal is not detected.

[0121] S304: Send a closing control signal to the target relay, obtain the first time point at which the closing control signal is sent, obtain the second time point when the state detection operation detects the first state indication signal, and obtain the first time interval based on the time difference between the first time point and the second time point.

[0122] In this embodiment, the closing control signal refers to the control signal used to drive the target relay to switch from the open state to the closed state. The first time point refers to the time marker when the closing control signal is issued. The second time point refers to the time marker when the state detection operation first detects the first state indication signal. The first time interval refers to the time difference between the second time point and the first time point.

[0123] Specifically, a closing control signal is first output to the target relay under the condition that the target relay is in the open state and the first state indication signal is maintained. At the start of the closing control signal output, a timing reference is read to record the first time point. The timing reference is a timing source used to output time point information. Then, in the sampling loop of the state detection operation, the sampled value at the detection end is continuously read and a threshold comparison process is performed to determine whether the first state indication signal is detected. When the threshold comparison process first indicates that the first state indication signal is detected, the timing reference is read to record the second time point. Then, the time difference between the first time point and the second time point is calculated to obtain the first time interval. The time difference calculation is completed by subtracting the timing value corresponding to the first time point from the timing value corresponding to the second time point. For example, if the timing value corresponding to the first time point is 100.000ms and the timing value corresponding to the second time point is 103.500ms, 103.500ms is subtracted from 100.000ms to obtain 3.500ms as the first time interval.

[0124] S305: Send a disconnection control signal to the target relay, obtain a third time point at which the disconnection control signal is sent, obtain a fourth time point when the state detection operation detects the second state indication signal, and obtain a second time interval based on the time difference between the third time point and the fourth time point.

[0125] In this embodiment, the disconnect control signal refers to the control signal used to drive the target relay to switch from the closed state to the open state, the third time point refers to the time marker when the disconnect control signal is issued, the fourth time point refers to the time marker when the state detection operation first detects the second state indication signal, the second time interval refers to the time difference between the fourth time point and the third time point, and the second state indication signal refers to the electrical signal used to indicate the disconnected state of the detection path.

[0126] Specifically, firstly, under the condition that the target relay is in the closed state and the state detection operation is continuously sampling, a disconnection control signal is output to the target relay. At the start of the disconnection control signal output, a timing reference is read to record a third time point. Subsequently, during the sampling loop of the state detection operation, the sampled value at the detection end is continuously read, and a threshold comparison is performed to determine whether a second state indication signal has been detected. The determination of the second state indication signal is accomplished by comparing the sampled value at the detection end with a preset determination threshold and outputting a determination result corresponding to the disconnection state. For example, when the first state indication signal is a high-level signal and the preset determination threshold is 2.0V, the sampled value at the detection end... A sample value below 2.0V is considered a detected second state indication signal. When the threshold comparison process first indicates that the second state indication signal has been detected, the timing reference is read to record the fourth time point. Then, the time difference between the third and fourth time points is calculated to obtain the second time interval. The time difference calculation is done by subtracting the timing value corresponding to the third time point from the timing value corresponding to the fourth time point. For example, if the timing value corresponding to the third time point is 200.000ms and the timing value corresponding to the fourth time point is 204.200ms, then 200.000ms is subtracted from 204.200ms to obtain 4.200ms as the second time interval.

[0127] In one embodiment, such as Figure 4 As shown, in step S40, i.e., according to the multidimensional online detection sequence, the conduction parameter measurement operation is performed on the target relay through the detection path to obtain the conduction parameter value, including:

[0128] S401: Determine the conduction parameter measurement operation corresponding to the target relay according to the multidimensional online detection sequence.

[0129] In this embodiment, the conduction parameter measurement operation refers to the detection execution content corresponding to the detection operation indicated by the detection task item in the multidimensional online detection sequence when the conduction parameter measurement operation is performed.

[0130] Specifically, the next detection task item in the multidimensional online detection sequence is first read, and parsing processing is performed on the detection task item. Parsing processing refers to extracting the identification information of the relay to be detected, the detection time point information, and the detection operation identification information from the detection task item. Then, the identification information of the relay to be detected is determined as the identification information of the target relay, and the target relay is determined accordingly. Next, matching processing is performed on the detection operation identification information and the conduction parameter measurement operation identification information. Matching processing refers to performing consistency judgment on the detection operation identification information and the conduction parameter measurement operation identification information. When the consistency judgment indication is consistent, the detection task item is determined as the conduction parameter measurement operation corresponding to the target relay. When the consistency judgment indication is inconsistent, the next detection task item in the multidimensional online detection sequence is read again, and the parsing processing and the matching processing are repeated until a detection task item whose detection operation identification information is the conduction parameter measurement operation identification information is matched.

[0131] S402: Based on the conduction parameter measurement operation, control the detection path to switch to the conduction parameter measurement state corresponding to the target relay.

[0132] In this embodiment, the continuity parameter measurement state refers to the connection state of the detection path and the two contact sides of the target relay forming an electrical connection, and satisfying the requirement to apply a test current and collect the voltage between the contacts.

[0133] Specifically, the target relay is located based on the target relay identification information corresponding to the continuity parameter measurement operation. The connection relationship of the detection path associated with the target relay identification information is read. Then, the switching state of the control signal relay is adjusted so that the detection path is connected to both contact sides of the target relay according to the detection path connection relationship. The switching state of the control signal relay includes switching the signal relay on the first contact side to the on state and switching the signal relay on the second contact side to the on state, so that the first wire of the detection path is electrically connected to the first contact side of the target relay and the second wire of the detection path is electrically connected to the second contact side of the target relay. Subsequently, a closing control signal is sent to the target relay to put the target relay in the closed state. The continuity confirmation process is performed on the electrical connection state of the detection path to confirm that both ends of the detection path are electrically connected to the two contact sides of the target relay. Finally, when the continuity confirmation process indicates that the execution conditions of applying test current and collecting voltage between contacts are met, the current electrical connection state of the detection path is determined as the continuity parameter measurement state corresponding to the target relay.

[0134] S403: Send a closing control signal to the target relay through the detection path in the conduction parameter measurement state, and control the electrical parameter measurement unit to apply a test current to the target relay.

[0135] In this embodiment, the electrical parameter measurement unit refers to the measuring component used to output test current and collect voltage between contacts. The test current is a preset current value that guides the parameter measurement operation.

[0136] Specifically, under the condition that the detection path is already in the conduction parameter measurement state, a closing control signal is output to the target relay through the detection path. After the closing control signal is output, the target relay is judged to enter the closed state based on a preset waiting time or based on the path state confirmation result. The preset waiting time refers to the waiting period set to cover the relay switching time. The path state confirmation result refers to the path state response result obtained through the state detection operation. Subsequently, the electrical parameter measurement unit is controlled to enter the test current output mode. The test current output mode refers to the working mode of adjusting the output terminal to stabilize the output current by adjusting the preset current target value. Then, the electrical parameter measurement unit... The output terminal is connected to the two contact sides of the target relay through the detection path and outputs the test current. In order to ensure that the test current remains stable within the measurement window, the current deviation value is obtained by sampling the output current in real time and calculating the difference between the sampled value and the preset current target value. Then, the output terminal drive is adjusted according to the current deviation value to reduce the current deviation value, so that the test current remains stable within the preset allowable fluctuation range. For example, when the preset current target value is 10mA and the real-time sampled value is 9.8mA, the difference between 9.8mA and 10mA is 0.2mA as the current deviation value and the output terminal drive is increased to make the real-time sampled value approach 10mA.

[0137] S404: The voltage between the contacts of the target relay under the test current is obtained through the electrical parameter measurement unit, and the conduction parameter value is determined based on the test current and the voltage between the contacts.

[0138] In this embodiment, the inter-contact voltage refers to the voltage value formed between the two contacts of the target relay under the action of the test current, and the conduction parameter value refers to the parameter value calculated from the test current and the inter-contact voltage and used to characterize the conduction state of the target relay.

[0139] Specifically, under the condition that the electrical parameter measurement unit continuously outputs test current, the electrical parameter measurement unit is controlled to enter the contact voltage acquisition mode. The contact voltage acquisition mode refers to the working mode of sampling the voltage sampling terminals connected to the two contact sides of the target relay and outputting the voltage value. Then, the contact voltage is sampled multiple times according to the preset sampling period, and the voltage sampling value obtained from each sampling is recorded. In order to reduce the influence of sampling fluctuation on the calculation results, the contact voltage is obtained by averaging the multiple voltage sampling values. The averaging calculation is completed by adding the multiple voltage sampling values ​​and dividing by the number of samplings. Then, the current value corresponding to the test current is read, and the ratio between the contact voltage and the test current is calculated to determine the conduction parameter value. The ratio calculation is completed by dividing the value of the contact voltage by the value of the test current. For example, when the test current is 10mA and the voltage sampling values ​​obtained from multiple samplings are 2.0mV, 2.1mV, and 1.9mV, and the average value is 2.0mV, 2.0mV is divided by 10mA to get 0.2Ω, and 0.2Ω is used as the conduction parameter value.

[0140] In one embodiment, step S50, namely, performing an insulation parameter measurement operation on the target relay through the detection path according to the multidimensional online detection sequence to obtain insulation parameter values, includes:

[0141] S501: Determine the insulation parameter measurement operation corresponding to the target relay according to the multidimensional online detection sequence.

[0142] In this embodiment, the insulation parameter measurement operation refers to the detection execution content corresponding to the detection operation indicated by the detection task item in the multidimensional online detection sequence when the detection operation is the insulation parameter measurement operation.

[0143] Specifically, the next detection task item in the multidimensional online detection sequence is read, and parsing processing is performed on the detection task item to extract the identification information of the relay to be detected, the detection time point information, and the detection operation identification information. Then, the identification information of the relay to be detected is determined as the identification information of the target relay, and the target relay is determined accordingly. Next, the detection operation identification information and the insulation parameter measurement operation identification information are matched to perform consistency judgment. When the consistency judgment indication is consistent, the detection task item is determined as the insulation parameter measurement operation corresponding to the target relay. When the consistency judgment indication is inconsistent, the next detection task item in the multidimensional online detection sequence is read again, and the parsing processing and the matching processing are repeated until a detection task item whose detection operation identification information is the insulation parameter measurement operation identification information is matched.

[0144] S502: Based on the insulation parameter measurement operation, control the detection path to switch to the insulation parameter measurement state corresponding to the target relay.

[0145] In this embodiment, the insulation parameter measurement state refers to the path connection state in which the detection path forms an electrical connection with the two contact sides of the target relay and satisfies the requirement of applying a test voltage between the two contacts and collecting the response current.

[0146] Specifically, the target relay is located based on the target relay identification information corresponding to the insulation parameter measurement operation. The connection relationship of the detection path associated with the target relay identification information is read. Then, the switching state of the control signal relay is adjusted so that the detection path is connected to both contact sides of the target relay according to the detection path connection relationship. The switching state of the control signal relay includes switching the signal relay on the first contact side to the conducting state and switching the signal relay on the second contact side to the conducting state, so that the first wire of the detection path is electrically connected to the first contact side of the target relay and the second wire of the detection path is electrically connected to the second contact side of the target relay. Subsequently, a disconnection control signal is sent to the target relay to put the target relay in the disconnected state. The continuity confirmation process of the electrical connection state of the detection path is performed to confirm that both ends of the detection path are electrically connected to the two contact sides of the target relay. Finally, when the continuity confirmation process indicates that the execution conditions of applying test voltage and collecting response current are met, the current electrical connection state of the detection path is determined as the insulation parameter measurement state corresponding to the target relay.

[0147] S503: Send a disconnection control signal to the target relay through the detection path that is in the insulation parameter measurement state.

[0148] In this embodiment, the disconnect control signal refers to the control signal used to drive the target relay to switch from a closed state to an open state.

[0149] Specifically, under the condition that the detection path is already in the state of insulation parameter measurement, first confirm that the detection task item corresponding to the multi-dimensional online detection sequence indicates the insulation parameter measurement operation, then output a disconnection control signal to the target relay through the detection path and record the output start time of the disconnection control signal. Subsequently, the connection relationship of the detection path remains unchanged within the preset waiting time and waits for the target relay to complete the contact separation action. The preset waiting time is the waiting period set to cover the mechanical response time required for the relay contacts to go from closed to open. Then, based on the sampling results of the state detection operation, it is determined whether the two contacts of the target relay have switched from the conducting state to the disconnected state. The sampling results of the state detection operation are the judgment results obtained by comparing the sampled value at the detection end with the preset judgment threshold. When the judgment result indicates that the disconnected state is established, the disconnection control signal sending process ends and the subsequent process of applying test voltage and collecting response current begins.

[0150] S504: Apply a test voltage between the two contacts of the target relay using the electrical parameter measurement unit, obtain the response current, and determine the insulation parameter value based on the response current.

[0151] In this embodiment, the test voltage refers to the preset voltage value in the insulation parameter measurement operation, the response current refers to the current value flowing between the two contacts of the target relay under the action of the test voltage, and the insulation parameter value refers to the parameter value determined by the response current and used to characterize the open state of the target relay.

[0152] Specifically, with the target relay in the open state and the detection path maintaining continuity between the two contacts of the target relay, the control electrical parameter measurement unit enters the test voltage output mode. The test voltage output mode refers to a working mode where the output voltage is adjusted to a preset target voltage value to stabilize the output voltage. Then, the voltage output terminal of the electrical parameter measurement unit is connected between the two contacts of the target relay via the detection path to output the test voltage. To ensure the test voltage remains stable within the measurement window, the output voltage is sampled in real time, and the difference between the sampled value and the preset target voltage value is calculated to obtain the voltage deviation value. The output drive is then adjusted based on the voltage deviation value to reduce the voltage deviation, ensuring the test voltage remains stable within a preset allowable fluctuation range. Subsequently, the control... The electrical parameter measurement unit enters the response current acquisition mode. The response current acquisition mode refers to the working mode of sampling the current sampling terminal connected between the two contacts of the target relay and outputting the current value. Then, the response current is sampled multiple times according to the preset sampling period, and the current sampling value obtained from each sampling is recorded. In order to reduce the impact of sampling fluctuation on the consistency of judgment, the response current is calculated by averaging the multiple current sampling values. The averaging calculation is completed by adding the multiple current sampling values ​​and dividing by the number of samplings. Finally, the response current is recorded as the insulation parameter value. For example, if the test voltage is 100V and the current sampling values ​​obtained from multiple samplings are 48nA, 52nA, and 50nA, and the average value is 50nA, then 50nA is recorded as the insulation parameter value.

[0153] In one embodiment, step S70, namely, establishing a health status model based on the historical detection sequence and generating prediction result information using the health status model, includes:

[0154] S701: Based on the historical detection sequence, extract the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence corresponding to the time or the number of switching.

[0155] In this embodiment, the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence refer to numerical sequences formed by collecting the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value corresponding to each detection result record information in the historical detection sequence in order of time or number of switching.

[0156] Specifically, each detection result record in the historical detection sequence is read and field parsing processing is performed on the detection result record information. Field parsing processing refers to extracting the detection time point information or switching count information, the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value from the detection result record information. Then, using the detection time point information or switching count information as the index key, the first time interval is written to the corresponding position in the first time interval sequence, the second time interval is written to the corresponding position in the second time interval sequence, the conduction parameter value is written to the corresponding position in the conduction parameter value sequence, and the insulation parameter value is written to the corresponding position in the insulation parameter value sequence. The position corresponding to the index key refers to the sequence number position after being arranged in ascending order of detection time point information or switching count information. When there are multiple detection result records with the same detection time point information or the same switching count information in the historical detection sequence, a single value is formed by retaining the latest written record or by averaging multiple values ​​corresponding to the same index key. The averaging calculation is completed by adding multiple values ​​and then dividing by the number of values. Finally, the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence arranged in the order of the index key are output as the extraction results.

[0157] S702: Perform time or switching frequency alignment processing on the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence to obtain an aligned sequence.

[0158] In this embodiment, the alignment sequence refers to the sequence data formed by combining the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value according to the corresponding relationship under the same time point information or the same number of switching information.

[0159] Specifically, an alignment benchmark is determined, which is either detection time point information or switching count information and remains consistent. An alignment index list is then generated based on this benchmark. The alignment index list refers to a set of multiple alignment index values ​​arranged in ascending order. Index matching processing is then performed on the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence. Index matching processing involves finding the position in each sequence corresponding to each alignment index value in the alignment index list and reading the sequence value at that position. When a sequence does not have a corresponding sequence value under a certain alignment index value, the sequence value is supplemented using a missing value marking method or a nearest-neighbor index value interpolation method. The missing value marking method involves marking the position as missing. The position is skipped during the construction of subsequent training datasets. The nearest neighbor index value interpolation method refers to taking the two existing sequence values ​​before and after the alignment index value and calculating the interpolation value according to the relative position of the alignment index value between them. The interpolation value calculation is completed using linear interpolation, which means that the previous sequence value and the next sequence value are weighted and summed proportionally to obtain the interpolation value. Then, under each alignment index value, the corresponding values ​​of the first time interval sequence, the corresponding values ​​of the second time interval sequence, the corresponding values ​​of the conduction parameter value sequence, and the corresponding values ​​of the insulation parameter value sequence are combined into an alignment record according to the preset field order. Multiple alignment records are arranged in ascending order of alignment index value. Finally, the sequence of alignment records obtained by the arrangement is determined as the alignment sequence.

[0160] S703: Based on the aligned sequence, a regression algorithm is used to perform parameter fitting to obtain health status model parameters, and the health status model is generated based on the health status model parameters.

[0161] In this embodiment, the health status model parameters refer to the set of parameters determined by the regression algorithm and used to characterize the relationship between each aligned record in the alignment sequence and the change of time or number of switching. The parameter fitting process refers to the process of determining the parameter set based on the deviation between the record values ​​in the alignment sequence and the output values ​​of the regression algorithm.

[0162] Specifically, each alignment record in the alignment sequence is read, and the corresponding alignment index value, first time interval, second time interval, conduction parameter value, and insulation parameter value are extracted. The alignment index value is then used as the input variable for the regression algorithm, and the first time interval, second time interval, conduction parameter value, and insulation parameter value are used as the output target value for the regression algorithm. Subsequently, the model expression for the regression algorithm is determined, and the parameter values ​​in the model expression are initialized. The model expression is a function that maps the alignment index value to the first time interval, second time interval, conduction parameter value, and insulation parameter value. Initializing the parameter values ​​in the model expression means assigning initial values ​​to each parameter in the model expression to facilitate fitting calculations. Then, based on the alignment sequence, the alignment index value is input line by line, and the regression algorithm output value is calculated. Simultaneously, the regression algorithm output value is compared with... The deviation values ​​between the recorded values ​​in the alignment record are calculated by subtracting the regression algorithm output value from the recorded value and taking the absolute or squared value of the difference. The deviation values ​​corresponding to multiple alignment records are then summarized as the fitting error, and the parameter values ​​in the model expression are updated accordingly. The parameter values ​​in the model expression are updated by repeatedly calculating the fitting error after each update and selecting the parameter values ​​that reduce the fitting error, until the fitting error meets the preset stopping condition or the number of updates reaches the preset number threshold. The parameter values ​​at the time of stopping are then determined as the health status model parameters. Finally, the health status model parameters are bound to the model expression of the regression algorithm and written into the model record. The model record refers to the record information containing the model expression identification information and the health status model parameters, thereby generating the health status model.

[0163] S704: Based on the health status model, determine future time points or future switching counts, and generate a first time interval prediction value, a second time interval prediction value, a conduction parameter prediction value, and an insulation parameter prediction value corresponding to the future time points or the future switching counts.

[0164] In this embodiment, future time points or future switching counts refer to time point information or switching count information located after the coverage of the historical detection sequence. The first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value refer to the predicted output values ​​obtained by substituting the future time point information or future switching count information into the health status model.

[0165] Specifically, the system reads the model expression identifier information and health status model parameters recorded in the health status model. Then, it reads the maximum detection time point or the maximum number of on / off cycles from the historical detection sequence as the baseline index value. Subsequently, it generates future time point information or future on / off cycle information based on a preset prediction step size. The preset prediction step size refers to the increment between two adjacent future index values. The increment uses either a time increment or an on / off cycle increment and maintains consistency. For example, if the preset prediction step size is 10 minutes, future time point information is generated sequentially after the maximum detection time point information, adding 10 minutes, 20 minutes, and 30 minutes. Similarly, if the preset prediction step size is 1000 cycles, future time point information is generated sequentially after the maximum on / off cycle information, adding 1000 cycles, 2000 cycles, and 30 minutes. The system first obtains information on the number of future switches (00). Then, it substitutes each future time point or future switch count information as an independent variable into the health state model and calculates the corresponding predicted values ​​for the first time interval, the second time interval, the conduction parameter, and the insulation parameter according to the model expression. Substituting and calculating refers to the process of substituting the independent variable input into the model expression and combining it with the health state model parameters to complete the numerical calculation and obtain the output result. Finally, it aggregates each future time point or future switch count information with the corresponding predicted values ​​for the first time interval, the second time interval, the conduction parameter, and the insulation parameter according to the preset field order to form a prediction record. The multiple prediction records are then arranged in ascending order of future time point information or future switch count information.

[0166] S705: Generate the prediction result information based on the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value.

[0167] In this embodiment, the prediction result information refers to a set of recorded information used to carry future time point information or future switching frequency information, as well as the first time interval prediction value, second time interval prediction value, conduction parameter prediction value, and insulation parameter prediction value corresponding to the future time point information or future switching frequency information.

[0168] Specifically, the system reads the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value output by the health status model. It also reads the future time point information or future switching frequency information corresponding to the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value. Then, it constructs a prediction record according to a preset field order. The preset field order refers to the agreed-upon order of the fields in the prediction record. Constructing a prediction record means writing the future time point information or future switching frequency information, the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value into the same record. For example, it might be based on the future time point information, the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value, and the insulation parameter prediction value. The measured values, the predicted values ​​of the second time interval, the predicted values ​​of the conduction parameters, and the predicted values ​​of the insulation parameters are written in sequence. Then, a consistency check is performed on multiple prediction records. The consistency check process checks whether each prediction record contains the predicted values ​​of the first time interval, the second time interval, the predicted values ​​of the conduction parameters, and the predicted values ​​of the insulation parameters, and checks whether the field values ​​meet the preset format requirements. When the consistency check process indicates that there are missing fields, the missing fields are marked using a missing field marking method and the prediction record is retained. When the consistency check process indicates that all fields are complete, the prediction record is appended to the prediction result information set. Finally, the prediction record set arranged in ascending order according to the future time point information or the future switching frequency information is determined as the prediction result information.

[0169] In one embodiment, step S80, namely generating early warning information when the prediction result information meets the preset failure threshold condition, includes:

[0170] S801: The preset failure threshold conditions include a first time interval threshold condition, a second time interval threshold condition, a conduction parameter threshold condition, and an insulation parameter threshold condition.

[0171] Specifically, the configuration records of preset failure threshold conditions are read and parsed to obtain four types of threshold condition entries. Configuration records are records that store threshold condition content in the form of fields. Parsing involves extracting fields from the configuration records and restoring them into threshold condition entries that can be compared. The four types of threshold condition entries are then mapped to first time interval threshold conditions, second time interval threshold conditions, conduction parameter threshold conditions, and insulation parameter threshold conditions, respectively. Each type of threshold condition entry includes a threshold value and a comparison relationship. The threshold value is a value used to define the failure judgment boundary, and the comparison relationship is a rule indicating whether the predicted value and the threshold value are judged using methods such as greater than, less than, or outside the interval. Then, the first time interval threshold condition, the second time interval threshold condition, and the second time interval threshold condition are further processed. The threshold conditions, conduction parameter threshold conditions, and insulation parameter threshold conditions undergo field integrity checks to confirm that all four types of threshold condition entries contain a threshold value field and a comparison relationship field. Field integrity checks refer to the process of checking whether a field is missing. Then, the four types of threshold condition entries that pass the field integrity check are written into the preset failure threshold condition set to complete the construction of the preset failure threshold conditions. For example, the first time interval threshold condition is written with a threshold value of 5ms and a comparison relationship of greater than, the second time interval threshold condition is written with a threshold value of 6ms and a comparison relationship of greater than, the conduction parameter threshold condition is written with a threshold value of 0.5Ω and a comparison relationship of greater than, and the insulation parameter threshold condition is written with a threshold value of 80nA and a comparison relationship of greater than.

[0172] S802: Compare the predicted value of the first time interval with the first time interval threshold condition, compare the predicted value of the second time interval with the second time interval threshold condition, compare the predicted value of the conduction parameter with the conduction parameter threshold condition, and compare the predicted value of the insulation parameter with the insulation parameter threshold condition to obtain threshold determination result information.

[0173] In this embodiment, the threshold determination result information refers to the result record information formed by recording and compiling the comparison results of the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter with the corresponding threshold conditions.

[0174] Specifically, a prediction record is read from the prediction result information, and the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter are extracted. Then, the threshold conditions of the first time interval, the second time interval, the conduction parameter, and the insulation parameter are read, and the threshold values ​​and comparison relationships are extracted respectively. Subsequently, a comparison operation is performed between the predicted value of the first time interval and the threshold value of the first time interval according to the comparison relationship to obtain the first comparison result. The comparison operation refers to the process of determining whether the predicted value of the first time interval meets the threshold value constraint based on the comparison relationship. For example, if the comparison relationship is greater than, it is determined whether the predicted value of the first time interval is greater than the threshold value of the first time interval. Then, the comparison operation is performed between the predicted value of the second time interval and the threshold value of the second time interval in the same way. To obtain the second comparison result, a comparison operation is performed between the predicted value of the conduction parameter and the threshold value of the conduction parameter to obtain the third comparison result. A comparison operation is performed between the predicted value of the insulation parameter and the threshold value of the insulation parameter to obtain the fourth comparison result. When the comparison relationship is outside the interval, the comparison result is obtained by simultaneously judging whether the predicted value is less than the lower boundary or greater than the upper boundary. Then, the first comparison result, the second comparison result, the third comparison result, and the fourth comparison result are written into the same threshold judgment result record in the order of preset fields. The future time point information or future switching number information corresponding to the prediction record is also written into the threshold judgment result record. Finally, the multiple threshold judgment result records obtained from multiple prediction records in the prediction result information are aggregated in ascending order of future time point information or future switching number information to form the threshold judgment result information.

[0175] S803: When the threshold determination result information indicates that at least one of the corresponding preset failure threshold conditions is met, an early warning information is generated.

[0176] In this embodiment, the warning information refers to the prompt content information associated with the target relay identification information and used to characterize the threshold determination result information.

[0177] Specifically, the system reads a threshold determination result record from the threshold determination result information and extracts future time point information or future switching frequency information, as well as the first comparison result, second comparison result, third comparison result, and fourth comparison result. Then, it performs a summary determination process on the first comparison result, second comparison result, third comparison result, and fourth comparison result. The summary determination process determines whether at least one of the first comparison result, second comparison result, third comparison result, and fourth comparison result indicates that the corresponding threshold condition is met and outputs a summary determination flag. When the summary determination flag indicates that at least one condition is met, the system reads the target relay identification information associated with that threshold determination result record and constructs an early warning record. Constructing an early warning record means... The relay identification information, future time point information or future switching count information, and comparison result item identification information that meets the threshold conditions are written into the same record. The comparison result item identification information that meets the threshold conditions is used to indicate the specific trigger item among the first time interval threshold conditions, the second time interval threshold conditions, the conduction parameter threshold conditions, and the insulation parameter threshold conditions. Then, the warning record is written into the warning information set and arranged in ascending order of future time point information or future switching count information. Finally, the set of warning records obtained by the arrangement is determined as the warning information. For example, when the third comparison result indicates that the predicted value of the conduction parameter meets the conduction parameter threshold condition, the conduction parameter threshold condition identification information is written into the warning record, and the corresponding future time point information is also written into the warning record.

[0178] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0179] In one embodiment, a relay array health status prediction device is provided, which corresponds one-to-one with the relay array health status prediction method in the above embodiments. This relay array health status prediction device includes a detection path construction module, a detection sequence generation module, a path status response detection module, a continuity parameter measurement module, an insulation parameter measurement module, a detection result recording module, a health status model construction and prediction module, and an early warning generation module.

[0180] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for predicting the health status of a relay array, characterized in that, The relay array health status prediction method includes: Signal relays are set on the two contact sides of the relay to be tested to construct the detection path corresponding to the relay to be tested. The system acquires historical usage frequency information, test environment information, and historical test data information of the relay under test. Based on these information, it determines the combination information of the test modules and the test frequency information, and generates a multi-dimensional online test sequence. According to the multidimensional online detection sequence, the target relay is subjected to a path state response detection operation through the detection path to obtain the first time interval and the second time interval. According to the multidimensional online detection sequence, the conduction parameter measurement operation is performed on the target relay through the detection path to obtain the conduction parameter value; According to the multidimensional online detection sequence, the insulation parameter measurement operation is performed on the target relay through the detection path to obtain the insulation parameter value; The detection result record information is generated based on the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value. The detection result record information is recorded according to the number of switches or the time sequence to obtain the historical detection sequence. A health status model is established based on historical detection sequences, and prediction results are generated using the health status model. When the prediction results meet the preset failure threshold conditions, an early warning message is generated.

2. The method for predicting the health status of a relay array according to claim 1, characterized in that, The step of generating a multidimensional online detection sequence based on the detection module combination information and detection frequency information includes: The detection time point sequence corresponding to each relay to be tested is determined based on the detection frequency information. Based on the detection module combination information, the detection operation set corresponding to each relay to be tested is determined. The detection operation set includes at least one of the following: circuit status response detection operation, continuity parameter measurement operation, and insulation parameter measurement operation. The detection time point sequence is associated with the detection operation set to generate a detection task item corresponding to each relay to be detected. Multiple detection task items are arranged in chronological order to obtain a multidimensional online detection sequence.

3. The method for predicting the health status of a relay array according to claim 1, characterized in that, The step of performing a path state response detection operation on the target relay through a detection path according to a multidimensional online detection sequence to obtain a first time interval and a second time interval includes: Determine the corresponding path state response detection operation of the target relay according to the multidimensional online detection sequence; Based on the path status response detection operation control, the detection path is switched to the path status response detection state corresponding to the target relay; A first state indication signal is applied to one end of the target relay by the detection path in the path state response detection state, and a state detection operation is performed at the other end of the target relay. Send a closing control signal to the target relay, obtain the first time point when the closing control signal is sent, obtain the second time point when the state detection operation detects the first state indication signal, and obtain the first time interval based on the time difference between the first time point and the second time point; A disconnection control signal is sent to the target relay, the third time point at which the disconnection control signal is sent is obtained, and the fourth time point is obtained when the second state indication signal is detected during the state detection operation. The second time interval is obtained based on the time difference between the third and fourth time points.

4. The method for predicting the health status of a relay array according to claim 1, characterized in that, The step of performing conduction parameter measurement on the target relay according to the multidimensional online detection sequence and obtaining conduction parameter values ​​includes: The measurement operation for determining the conduction parameters corresponding to the target relay is performed according to the multidimensional online detection sequence. Based on the continuity parameter measurement operation, the detection path is switched to the continuity parameter measurement state corresponding to the target relay; The detection path, which is in the state of conducting parameter measurement, sends a closing control signal to the target relay and controls the electrical parameter measurement unit to apply a test current to the target relay. The voltage between the contacts of the target relay under the test current is obtained by the electrical parameter measurement unit, and the conduction parameter value is determined based on the test current and the voltage between the contacts.

5. The method for predicting the health status of a relay array according to claim 1, characterized in that, The step of performing insulation parameter measurement on the target relay according to the multidimensional online detection sequence and obtaining insulation parameter values ​​includes: Determine the insulation parameter measurement operation corresponding to the target relay according to the multidimensional online detection sequence; Based on the insulation parameter measurement operation control, the detection path is switched to the insulation parameter measurement state corresponding to the target relay; A disconnection control signal is sent to the target relay through the detection path that is in the state of measuring insulation parameters; An electrical parameter measurement unit applies a test voltage between the two contacts of the target relay and acquires the response current, then determines the insulation parameter value based on the response current.

6. The method for predicting the health status of a relay array according to claim 1, characterized in that, The process of establishing a health status model based on historical detection sequences and generating prediction results using the health status model includes: Based on historical detection sequences, extract the first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence corresponding to time or number of switching operations; The first time interval sequence, the second time interval sequence, the conduction parameter value sequence, and the insulation parameter value sequence are aligned by time or switching count to obtain an aligned sequence. Based on the aligned sequence, a regression algorithm is used to perform parameter fitting to obtain the parameters of the health status model, and a health status model is generated based on the parameters of the health status model. Based on the health status model, the future time point or the future number of switching is determined, and the first time interval prediction value, the second time interval prediction value, the conduction parameter prediction value and the insulation parameter prediction value corresponding to the future time point or the future number of switching are generated. Prediction results are generated based on the predicted values ​​of the first time interval, the second time interval, the conduction parameter, and the insulation parameter.

7. The method for predicting the health status of a relay array according to claim 1, characterized in that, When the prediction result information meets the preset failure threshold condition, an early warning information is generated, including: The preset failure threshold conditions include a first time interval threshold condition, a second time interval threshold condition, a conduction parameter threshold condition, and an insulation parameter threshold condition; The predicted value of the first time interval is compared with the threshold condition of the first time interval, the predicted value of the second time interval is compared with the threshold condition of the second time interval, the predicted value of the conduction parameter is compared with the threshold condition of the conduction parameter, and the predicted value of the insulation parameter is compared with the threshold condition of the insulation parameter to obtain the threshold determination result information. When the threshold determination result indicates that at least one of the corresponding preset failure threshold conditions is met, an early warning message is generated.

8. A relay array health status prediction device, characterized in that, The relay array health status prediction device includes: The detection path construction module is used to set up signal relays on the two contact sides of the relay to be tested, and construct the detection path corresponding to the relay to be tested. The detection sequence generation module is used to acquire historical usage frequency information, test environment information, and historical test data information of the relay to be tested, and to determine the combination information of detection modules and detection frequency information based on the historical usage frequency information, test environment information, and historical test data information, and to generate a multi-dimensional online detection sequence based on the combination information of detection modules and detection frequency information. The path status response detection module is used to perform path status response detection operation on the target relay through the detection path according to the multidimensional online detection sequence, and obtain the first time interval and the second time interval. The continuity parameter measurement module is used to perform continuity parameter measurement operations on the target relay through the detection path according to the multi-dimensional online detection sequence, and obtain the continuity parameter value; The insulation parameter measurement module is used to perform insulation parameter measurement operations on the target relay according to a multi-dimensional online detection sequence and through the detection path to obtain insulation parameter values. The detection result recording module is used to generate detection result recording information based on the first time interval, the second time interval, the conduction parameter value, and the insulation parameter value, and to record the detection result recording information according to the number of switches or the time sequence to obtain the historical detection sequence; The health status model construction and prediction module is used to build a health status model based on historical detection sequences and generate prediction results information using the health status model. The early warning generation module is used to generate early warning information when the prediction result information meets the preset failure threshold conditions.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the relay array health status prediction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the relay array health status prediction method as described in any one of claims 1 to 7.

Images