A verification and experimental test method for contactor and relay fault prediction and health management technology

Abstract

The present invention relates to a verification and experimental test method for contactor and relay fault prediction and health management technology. The method is divided into a PHM algorithm, a technology verification method and a contactor relay device aging and life test test method. The method can be used to complete the experimental test of contactors and relays, operation parameter collection, health status monitoring and fault prediction and health management technology verification, evaluation, etc.

Description

Technical Field

[0001] The present invention relates to a technical verification and experimental testing method in the field of low-voltage electrical appliances, and in particular to a verification and experimental testing method for fault prediction and health management technology of a contactor relay. Background Art

[0002] Contactors and relays are electrical and signal-controlled electrical equipment that are widely used in a variety of power transmission, distribution, power consumption and control situations. Contactors and relays are typical electromagnetic switching devices. Such devices close or disconnect the main contacts according to the control signal on the control coil to achieve the on-off or signal control of the load. During the working process, the frequent on and off of the main contacts can easily cause the main contacts to arc. The instantaneous high-temperature arc will damage the main contacts or produce oxidized impurities on the surface of the main contacts, thereby reducing the electrical performance of the device. The health status of the device and whether it is faulty are closely related and have an important impact on whether the entire power transmission, distribution, power consumption system and signal control system can operate stably and normally.

[0003] Fault prediction and health management technology collects and monitors the operating status and characteristic parameter changes of the managed equipment. The characteristic data is processed by a specific algorithm to predict and infer potential faults, calculate the health status of the equipment and other information, and guide the repair and maintenance of the equipment in a targeted manner. The use of fault prediction and health management technology can effectively reduce the failure rate of equipment and improve operational stability and reliability.

[0004] Developing, verifying and testing fault prediction and health management technologies for contactors and relays, and then applying fault prediction and health management technologies to these devices can timely grasp the health level of the devices, so that they can be repaired and maintained in time, which is beneficial to improving the operating reliability of contactors, relays and electrical systems.

[0005] At present, the research and development of fault prediction and health management technologies specifically for contactors and relays is not mature enough. In addition, there is no system and method specifically for verification and experimental testing of this technology. The existing technology cannot effectively support the research and development of fault diagnosis and health management technologies for contactors and relays. Summary of the invention

[0006] In order to solve the shortcomings of the prior art, the present invention proposes a verification and experimental test method for contactor relay fault prediction and health management technology. The verification and experimental test method proposed by the present invention can complete a series of PHM method verification and experimental test work for electromagnetic switching devices such as contactors and relays.

[0007] The present invention proposes a verification and experimental test method for contactor relay fault prediction and health management technology, which is characterized by: the method is divided into a PHM algorithm verification method and a device aging and life test method.

[0008] The PHM algorithm verification method includes the following steps:

[0009] Step 1: Configure and load the PHM algorithm to be verified and the algorithm verification mode control process parameters to the main control computer;

[0010] Step 2: Configure and load algorithm verification mode control process parameters to the signal acquisition and processing computer and the state control computer;

[0011] Step 3: Run the algorithm verification process;

[0012] Step 4: Collect the operating data and parameters of the contactor in real time. The signal acquisition and processing computer compresses, processes, stores and uploads the operating data, and performs regular verification of the accuracy of signal acquisition and transmission;

[0013] Step 5: The main control computer runs the PHM algorithm to be verified;

[0014] Step 6: Determine whether the test is completed. If not, repeat step 3. If completed, proceed to step 7:

[0015] Step 7: Calculate the algorithm evaluation result score;

[0016] Step 8: Evaluate the effectiveness of the fault prediction and health management algorithms.

[0017] Step 8 of the PHM algorithm verification method of the present invention evaluates the effectiveness of the fault prediction and health management algorithm by using a composite indicator-based evaluation method. The composite indicator-based evaluation method uses a variety of algorithm performance evaluation indicators, such as statistical analysis performance, accuracy, stability, prediction performance and other indicators, to comprehensively evaluate the PHM algorithm from multiple angles and in all directions.

[0018] The PHM algorithm evaluation method based on composite indicators described in the present invention mainly selects four evaluation indicators: recall rate, detection accuracy, prediction stability coefficient and prediction range. First, the above four indicators are independently calculated for the algorithm to be evaluated, and the four indicators are normalized. Then, the hierarchical analysis evaluation method is used to calculate the weight coefficients of the four indicators. The four indicators are weighted according to the weight coefficients, and finally the evaluation effect score of the corresponding PHM algorithm is obtained.

[0019] The recall rate TPR mentioned in the present invention refers to the ratio of correctly classified true value cases to the number of all true value cases. This indicator is used to evaluate the detection ability of the PHM algorithm for true value cases. The expression of the recall rate TPR is:

[0020]

[0021] Among them, TP is the number of true positive cases, that is, the number of cases where the algorithm makes correct positive judgments; FN is the number of false negative cases, that is, the number of cases where the algorithm makes incorrect negative judgments. The recall rate TPR does not need to be normalized.

[0022] The detection accuracy ACC mentioned in the present invention refers to the ratio of the number of cases in which the PHM algorithm correctly classifies the fault to the corresponding total number. This indicator is used to evaluate the overall diagnosis and prediction accuracy of the PHM algorithm. The expression of the detection accuracy ACC is:

[0023]

[0024] Among them, TP is the number of true positive cases, that is, the number of cases where the algorithm makes a correct positive judgment; TN is the number of true negative cases, that is, the number of cases where the algorithm makes a correct negative judgment; P is the number of positive cases, that is, the number of all positive cases; N is the number of negative cases, that is, the number of all negative cases. The detection accuracy ACC does not need to be normalized.

[0025] The predicted stability coefficient α of the present invention Rob It refers to the percentage of the duration during which the prediction result no longer changes to the total time elapsed since the first prediction signal. This indicator is used to evaluate the prediction stability of the PHM algorithm. The prediction stability coefficient α Rob The expression is:

[0026]

[0027] Where Δsta is the duration of time during which the prediction result does not change, in seconds; Δocc is the total time that has passed since the first prediction signal, in seconds. Rob No further normalization is required.

[0028] The prediction range T of the present invention pre It refers to the maximum advance warning time that the PHM algorithm can provide under the premise of meeting the set confidence level. This indicator mainly evaluates the prediction advance of the PHM algorithm. The larger the prediction range, the earlier the fault signs can be found and the fault can be predicted. pre The expression is:

[0029] ΔT pre =T fault -T pre (4)

[0030] Among them, T faultis the time when the fault occurs, T pre In order to meet the set confidence level, the PHM algorithm predicts the alarm time for the first time. The following formula is used to normalize it and obtain the normalized result of the prediction range:

[0031]

[0032] Among them, T0 is the initial time of the prediction algorithm.

[0033] The above four evaluation indicators are obtained by independent calculation. After normalizing the indicators that need to be normalized, the hierarchical analysis evaluation method is used to integrate several rating indicators. The hierarchical analysis evaluation method requires a pairwise comparison of the importance levels between two evaluation indicators, assigning different judgment scales to form a judgment matrix. The comprehensive evaluation method described in the present invention uses four evaluation indicators, and the four evaluation indicators are compared with each other to form a 4*4 judgment matrix. The judgment scale and its meaning are shown in the table:

[0034]

[0035]

[0036] The judgment matrix is ​​constructed as follows:

[0037]

[0038] Then the consistency of the judgment matrix is ​​tested and the CI value is introduced:

[0039]

[0040] Among them, λ max is the maximum eigenvalue of the judgment matrix, and n is the order of the judgment matrix.

[0041] Calculate the random consistency ratio, denoted as CR:

[0042]

[0043] Among them, RI is the average random consistency index. The value of RI is determined by the matrix order. The RI values ​​of matrices from order 1 to order 9 are 0.00, 0.00, 0.58, 0.90, 1.12, 1.24, 1.32, 1.41 and 1.45 respectively.

[0044] Then the weights of the above evaluation indicators are standardized:

[0045]

[0046] The weight coefficients of the four indicators are w1, w2, w3, and w4 respectively.

[0047] Assign corresponding weight coefficients to the four indicators respectively, and calculate the evaluation effect scores of different PHM algorithms:

[0048] Q=w1·ACC+w2·TPR+w3·ΔT pre ′+w4·α Rob (9)

[0049] The advantages and disadvantages of different PHM algorithms can be compared by comparing the evaluation effect scores of different algorithms. The evaluation effect of the algorithm is positively correlated with the evaluation effect score.

[0050] Furthermore, a verification and test method for contactor relay fault prediction and health management technology also includes a device aging and life test method, comprising the following steps:

[0051] Step A: configure and load the "device aging and life test" control program and corresponding parameters to the main control computer, signal acquisition and processing computer and state control computer;

[0052] Step B: Run the aging and life test control program to perform the aging and life test process;

[0053] Step C: real-time acquisition and transmission of the operating data and parameter changes of the device under test, and periodic data verification;

[0054] Step D: The signal acquisition processing computer compresses, processes, and stores the operating data, and transmits the operating data to the main control computer;

[0055] Step E: The main control computer stores the operation data;

[0056] Step F: Determine whether the test is completed. If not, repeat step B. If completed, execute step G.

[0057] Step G: Export data and generate device test report.

[0058] Step 4 of the PHM algorithm verification method and step C of the device aging and life test method described in the present invention both include real-time acquisition of the operating data and parameters of the contactor. Since the computer system can only process and store discrete digital signals, and the parameter change during the operation is a continuous process, it needs to go through a signal acquisition process. The signal acquisition process is a discretization process, that is, converting a continuous-time signal into a discrete-time signal. The signal processing process includes two processes: signal acquisition and signal reconstruction. The reconstructed signal is to restore the discrete-time signal to a continuous-time signal by interpolation. In order to completely reconstruct a signal without causing signal distortion, according to Shannon's sampling theorem, the sampling frequency f s Need to be greater than or equal to the highest frequency f in the analog signal spectrum max2 times, that is:

[0059] f s ≥2f max (10)

[0060] The signal reconstruction process can be completed if the sampling frequency satisfies Shannon's sampling theorem, but too high a sampling frequency will impose a heavy burden on the sampling equipment and signal processing equipment, and in severe cases may cause the signal acquisition and processing functions to be paralyzed. Therefore, it is necessary to adjust and set the system sampling frequency in combination with the actual capabilities and parameters of the system hardware in the test and verification.

[0061] The adjustment and setting of the sampling frequency described in the present invention follows the following rules: 1) Shannon sampling theorem; 2) maximum sampling processing rate allowable principle; 3) maximum storage speed allowable principle; 4) maximum storage capacity allowable principle. The Shannon sampling theorem has been explained above, and the other three rules and principles are now explained in detail.

[0062] The maximum sampling processing rate tolerance principle is a rule and principle for setting the maximum sampling frequency for testing and verifying the sampling and processing capabilities of the hardware system. The basis for this principle is mainly based on the sampling frequency f of the hardware. cs and signal processing rate f cc , the sampling frequency under the maximum sampling processing rate allowed principle is subject to the following constraints:

[0063] f s ≤80%f cs (11)

[0064] f s ≤50%f cc (12)

[0065] Combined with the sampling frequency of the hardware and the signal processing frequency limit, the sampling rate f under the maximum sampling processing allowable principle maxsp It is expressed as:

[0066] f maxsp =min{80%f cs ,50%f cc} (13)

[0067] Among them, min{} is the minimum value function.

[0068] The maximum storage speed tolerance principle is a rule and principle for setting the maximum sampling frequency for the storage speed of storage devices in the test and verification hardware system. The basis for this principle is mainly based on the maximum storage rate R of the storage device. st And the unit storage data size Size data , the sampling frequency under the maximum storage speed allowable principle is subject to the following constraints:

[0069]

[0070] Sampling frequency f under the principle of maximum storage speed allowance maxcst for:

[0071]

[0072] Among them, R st It is the maximum storage rate of the storage device, indicating the ability of the storage device to store data per second, and the unit is kilobits per second (kbit / S) or kilobit rate kbpS; S data The system-set unit for storing data size in kilobits (kbit).

[0073] The maximum storage capacity allowable principle is a rule and principle for setting the maximum sampling frequency for the storage space size of the storage device in the test and verification hardware system. The basis for this principle is mainly based on the maximum storage capacity C of the storage device. sto , Unit storage data size Size data and the test duration T du The sampling frequency under the principle of maximum storage capacity is subject to the following constraints:

[0074]

[0075] Among them, C sto The maximum storage capacity of the storage device, in kilobits (kbit); Size data Unit storage data size, the unit is kilobit (kbit); T du is the test duration in seconds (S).

[0076] Sampling frequency f under the principle of maximum storage speed allowance maxsst for:

[0077]

[0078] The adjustment and setting of the sampling frequency described in the present invention is combined with the above four rules and rules to determine the final sampling frequency f s :

[0079]

[0080] Among them, min{f maxsp ,f maxcst ,f maxsst} represents the sampling rate f under the maximum sampling processing allowable principle maxsp , the sampling frequency f under the principle of maximum storage speed allowance maxcstand the sampling frequency f under the principle of maximum storage speed allowance maxsst The minimum of the three.

[0081] In order to prevent data distortion and errors caused by hardware and software errors in the processes of signal acquisition and signal transmission, it is necessary to periodically interrupt the acquisition during the signal acquisition process and use an external verification signal injection device to inject a standard verification signal into the acquisition system. The system verifies the signal that has undergone the acquisition and transmission process to determine whether there are any problems in the acquisition and transmission process.

[0082] The PHM algorithm verification method and the device aging and life test method described in the present invention both include verification operations for data acquisition and transmission accuracy.

[0083] The verification process described in the present invention exists as a sub-process of the main process, and the verification process includes the following 8 sub-steps:

[0084] Step Ⅰ: Add 1 to the test number flag;

[0085] Step II: Determine whether the test number flag is equal to the specified value. If it is equal to the specified value, proceed to step III; if it is not equal to the specified value, end the verification and continue the test;

[0086] Step III: Set the check-allowed flag to 1 and the test number flag to 0;

[0087] Step IV: inject standard calibration signal;

[0088] Step V: The standard calibration signal is sampled, processed, and transmitted to the signal acquisition and processing computer;

[0089] Step VI: Determine whether the standard calibration signal is distorted. If it is distorted, proceed to step VII; if it is not distorted, proceed to step VIII;

[0090] Step VII: Output the error status and the number of the error device and other information;

[0091] Step VIII: Clear the check flag to 0 and continue the main test procedure.

[0092] The standard calibration signal distortion rate during calibration described in the present invention must be within 0.5%, and a distortion rate exceeding 0.5% is considered as signal distortion.

[0093] The standard verification signal described in the present invention is a specific signal formed by a combination of one or more signals of a preset time length. The specific form of the standard verification signal can be set according to actual conditions such as the characteristics of the acquisition system.

[0094] The main purpose of the data compression processing described in the present invention is to delete duplicate homogeneous data, compress the overall data volume, reduce the burden of data transmission and storage, delete periodic and repetitive data with no obvious changes, reduce the proportion of duplicate data, and increase the proportion of valid data containing feature information.

[0095] The PHM algorithm verification method and the device aging and life test method described in the present invention both include data compression processing.

[0096] The main objects of the data compression processing described in the present invention are homogenized periodic data and non-valid phase data of phase-change data. Among them, periodic data mainly refers to data changes with a periodic trend, such as AC voltage and AC current. Phase-change data mainly refers to data changes that do not have a periodic trend, but have obvious stages. The entire cycle of this type of data can be divided into an effective stage and an ineffective stage, such as the vibration signal generated by the contactor of the phase-action due to the attraction or disconnection. The data of the phase-change data in the ineffective stage has no change or no obvious change, and the ineffective stage data is invalid for the system. Therefore, in order to reduce the overall data volume, this part of the data needs to be discarded.

[0097] The specific steps of the periodic data compression processing method of the present invention are as follows: firstly, the periodic data is divided into units of length T, and then the data of one period is divided into equal parts of length T / 1000, which are recorded as x Δk , (k=1,2,…1000), calculate the mean of all Δk data respectively Compare the corresponding x of two adjacent periods i and i+1 Δk The mean and If all corresponding x in two adjacent periods Δk If the relative error of the mean is within 0.5%, it is considered that the periodic data has not changed significantly, and the i-th cycle data is discarded, and only the i+1 cycle data is transmitted and stored; if there is a waveform with a relative error greater than 0.5% in two adjacent cycles, it is considered that the periodic data has changed, and both cycle data are transmitted and stored. The above rules are expressed using the formula:

[0098]

[0099] Among them, i=1,2,…; k=1,2,…1000.

[0100] The specific steps of the compression processing method of the stage-by-stage change data described in the present invention are: first, the change edge of the stage-by-stage change data is identified, and the change edge is used as the center moment t0 of the acquisition cycle, and the acquisition time of one unit cycle ΔT is added before and after the center moment t0 to determine whether the acquisition signal data on the right side of the change edge is stable. If it is stable, the data is transmitted and saved; if it is unstable, the acquisition time of one unit cycle ΔT is added on the right side until the data is stable and no longer changes. The acquisition cycle of the stage-by-stage change data is expressed as [t0-ΔT, t0+nΔT], where (n=1, 2, ...).

[0101] The present invention can achieve the following beneficial effects:

[0102] The contactor and relay fault prediction and health management technology verification method designed by the present invention can programmatically and automatically perform fault prediction and health management technology verification operations on electromagnetic switch devices such as contactors and relays, automatically test and verify the algorithms and technologies to be verified, and evaluate and verify the effectiveness, accuracy and other performance of the algorithms and technologies. Compared with manual test verification and semi-automatic test verification systems and methods, the technology verification method designed by the present invention can improve the efficiency of technology verification and shorten the technology verification cycle.

[0103] By using the contactor fault prediction and health management test method designed by the present invention, it is possible to perform full life cycle performance tests, aging tests, accelerated aging tests and other tests on electromagnetic switching devices such as contactors and relays in a programmed and automated manner. During the test process, characteristic signals such as voltage, current, vibration, sound print, temperature, infrared, etc. of electromagnetic switching devices such as contactors and relays can be collected, processed, analyzed and stored. In combination with fault diagnosis and prediction algorithms, faults of electromagnetic switching devices such as contactors and relays can be diagnosed and predicted; in combination with health status assessment algorithms, the health status of electromagnetic switching devices such as contactors and relays can be assessed.

[0104] The sampling frequency setting method designed by the present invention can more fully utilize the system hardware performance, increase the sampling frequency, and improve the data resolution under the premise of satisfying Shannon's sampling theorem.

[0105] By using the data verification method designed in the present invention, the data can be verified periodically using a standard verification signal, system failures can be discovered in time, and the credibility and accuracy of the data can be guaranteed.

[0106] The data compression processing method designed by the present invention can effectively reduce the proportion of invalid and homogeneous data, increase the proportion of valid data, and reduce the pressure of data storage, transmission, and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] One or more embodiments are exemplarily described by corresponding drawings, which do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements, and the drawings do not constitute a scale limitation, and wherein:

[0108] Figure 1 It is a structural block diagram of a verification and experimental test system for contactor relay fault prediction and health management technology provided by an embodiment of the present disclosure.

[0109] Figure 2 It is a technical verification flow chart for contactor relay fault prediction and health management technology provided by an embodiment of the present disclosure.

[0110] Figure 3 It is an experimental test flow chart for contactor relay fault prediction and health management technology provided by an embodiment of the present disclosure.

[0111] Figure 4 It is a data verification flow chart of a signal acquisition process for contactor relay fault prediction and health management technology provided by an embodiment of the present disclosure.

[0112] Figure 5 It is a waveform diagram of a standard calibration signal provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0113] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0114] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as the content disclosed in this application is insufficient.

[0115] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0116] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a kind of", "the" and the like involved in this application do not indicate a quantitative limitation and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices.

[0117] The present invention is further described with reference to the embodiments and the accompanying drawings:

[0118] The core of the present invention is to provide a verification and experimental testing method for contactor relay fault prediction and health management technology. By using the method designed by the present invention, verification and experimental testing of fault prediction and health management technology can be performed on contactor relays.

[0119] Embodiment 1:

[0120] like Figure 1 As shown in the figure, the whole system consists of a main control computer, a signal acquisition and processing computer, a state control computer, a signal conditioning panel, a control drive panel, a contactor panel under test, a power supply, and a multi-channel electronic load. Among them, the signal conditioning panel contains a signal conditioning circuit, a voltage sensor, a current sensor, a temperature sensor, a vibration sensor, a noise sensor, and a calibration signal injection circuit; the control drive panel contains a drive control circuit and a multi-channel drive circuit; and the contactor panel under test contains multiple contactors under test, relays, and other devices under test.

[0121] Specifically, the present invention can be used to complete the verification and evaluation of fault prediction and health management algorithms and technologies for contactors and relays. The present invention refers to this function, state and mode as the PHM algorithm verification mode. In the PHM algorithm verification mode, the program run by the signal acquisition computer and the state control computer is the algorithm verification mode control program, and the main control computer runs the PHM algorithm verification mode main program and the PHM algorithm to be verified. In the PHM algorithm verification mode, the research goal is to verify, evaluate and assess fault prediction and health management technologies. The research objects of the PHM algorithm verification mode are methods, computer programs and algorithms involved in fault prediction and health management technologies.

[0122] In order for the system to complete the PHM algorithm verification work, the entire system needs to be adjusted to the PHM algorithm verification state. The specific operations and configurations are as follows:

[0123] The method to be verified, the computer program and the algorithm verification mode control program configuration are loaded into the 01 main control computer, and the algorithm verification mode control program configuration is loaded into the 02 signal acquisition and processing computer and the 03 state control computer. The 03 state control computer receives the instructions of the 01 main control computer or performs state control on the 051 drive control circuit, the 07 power supply and the 08 multi-channel electronic load according to its internal algorithm verification mode control program, so that these controlled devices operate according to the settings of the algorithm verification mode control program. The 051 drive control circuit receives the control instructions and sends a drive signal to the 052 multi-channel drive circuit, and then the 052 multi-channel drive circuit controls multiple contactors under test such as KM1 to KMn to control the connection and disconnection of the contactor contacts according to the program settings. At the same time, the 03 state control computer controls the 07 power supply and the 08 multi-channel electronic load, so that the 07 power supply and the 08 multi-channel electronic load change the power supply state and load state for the test circuit according to the algorithm verification mode control program. 08 Multi-channel electronic load can receive the control of the host computer to realize operations such as load properties, load parameters, and load switching to change the working state of the contactor under test and simulate the specific working conditions of the contactor under test. 042 Voltage sensor, 043 Current sensor, 044 Temperature sensor, 045 Vibration sensor, 046 Noise sensor detect and collect various parameters of the contactor under test in real time, and upload the data to 02 Signal acquisition and processing computer through 041 signal conditioning circuit. 02 Signal acquisition and processing computer compresses, processes, stores and transmits the received data. The data transmitted to 01 main control computer in this working mode is compressed data; in addition, 02 Signal acquisition and processing computer can also upload all raw data to 01 main control computer through the communication interface. Data processing and storage, contactor fault prediction and health management operations are all completed by 01 main control computer, and the prediction results and health management evaluation results are stored. After completing the entire or a certain stage of the verification test, the actual fault occurrence record of the contactor under test is compared with the predicted result, the actual operation record of the contactor under test is compared with the health assessment result, and the methods, computer programs and algorithm effects involved in the verification of the PHM method are evaluated according to the evaluation method based on composite indicators. The effectiveness of the PHM algorithms, procedures and technologies is verified and evaluated to realize the verification and evaluation of the PHM algorithms and technologies.

[0124] Figure 2It is a technical verification flow chart for contactor and relay fault prediction and health management technology. To start the technical verification process, first perform step 1: configure and load the PHM algorithm to be verified and the algorithm verification mode control program on the main control computer. This step mainly completes the configuration of the main control computer; step 2: configure and load the algorithm verification mode control program on the signal acquisition and processing computer and the state control computer. This step completes the configuration of the signal acquisition and processing computer and the state control computer; after completing the above two configuration steps, perform step 3: start running the algorithm verification test. After step 3 starts, the contactor starts to control the on and off according to the verification mode program to simulate different working conditions and perform a standard test process. ; Step 4: Verify that all types of sensors in the system collect operating data in real time, and the signal acquisition and processing computer processes, analyzes, and stores the operating data, and uploads the operating data to the main control computer according to the system configuration; Step 5: The main control computer runs the PHM algorithm to be verified to process the data, thereby performing fault prediction and health status assessment and management; Step 6: The program determines whether the test is completed. If completed, proceed to step 7. If not completed, repeat step 3. The judgment criteria and basis for determining whether the test is completed can be determined in the program configuration and settings. Commonly used judgment criteria include but are not limited to the contactor completing a fixed number of cyclic operating condition tests, the contactor parameters exceeding the limit or completely failing, etc. Step 7: Compare the actual operation records of the device with the prediction results and health assessment results; Step 8: Evaluate and assess the effectiveness of the fault prediction and health assessment algorithms based on the comparison results obtained in step 7, and end the entire technical verification process after completing the evaluation and assessment.

[0125] According to the program setting, the data verification operation is performed periodically at the test interval. The 047 verification signal injection circuit injects the standard verification signal into the 041 signal conditioning circuit. After the signal acquisition and transmission process, it is transmitted to the 02 signal acquisition and processing computer. The 02 signal acquisition and processing computer determines whether the signal is distorted to determine whether there is a fault in the acquisition and transmission process. The 047 verification signal injection circuit is essentially a signal generating circuit that can inject a verification signal into the system according to the setting.

[0126] Embodiment 2:

[0127] Specifically, the present invention can also be used to perform device aging and life test tests. The present invention refers to this function, state and mode as a device aging and life test test mode. In the device aging and life test test mode, the program run by the main control computer, the signal acquisition and processing computer and the state control computer is an aging and life test test program. The research goal is to perform device aging and life tests on the tested device according to predetermined test standards and procedures, and to detect, collect and store the state information and parameter changes of the tested device in real time by loading different working conditions on the tested device, changing the test conditions, etc., to obtain the aging test and life test data of the tested device, and then accumulate test data and provide data support for the subsequent prediction of faults and evaluation of the health status of the tested device. The research object of the system in this mode is the contactor and relay under test.

[0128] In order to enable the system to realize the device aging and life test functions, the system needs to be configured into the device aging and life test mode. The specific process and operations are as follows:

[0129] The aging and life test program is loaded into the 01 main control computer, 02 signal acquisition and processing computer and 03 state control computer. According to the setting of the aging and life test program, the 03 state control computer sends control signals to the 07 power supply, 08 multi-channel electronic load and 051 drive control circuit to adjust the power supply state, load state and on-off state of the contactor under test. The three types of states act alone or in combination to provide a variety of different working states and working condition excitations for the contactor under test. The 03 state control computer receives the instructions of the 01 main control computer or controls the state of the 051 drive control circuit, 07 power supply and 08 multi-channel electronic load according to its internal aging and life test program, so that these controlled devices operate according to the setting of the aging and life test program. The 051 drive control circuit receives the control instruction and sends a drive signal to the 052 multi-channel drive circuit, and then the 052 multi-channel drive circuit controls multiple contactors under test such as KM1 to KMn to control the connection and disconnection of the contactor contacts according to the program setting. At the same time, the 03 state control computer controls the 07 power supply and the 08 multi-channel electronic load, so that the 07 power supply and the 08 multi-channel electronic load change the power supply state and load state for the test circuit according to the aging and life test test procedures. The 08 multi-channel electronic load can receive the control of the host computer, realize the load properties, load parameters, load switching and other operations, so as to change the working state of the contactor under test, so that the contactor under test completes the predetermined aging and life test test process. The 042 voltage sensor, 043 current sensor, 044 temperature sensor, 045 vibration sensor, and 046 noise sensor detect and collect various parameter data of the tested device in real time, and upload the data to the 02 signal acquisition and processing computer through the 041 signal conditioning circuit. In the device aging and life test test mode, the 02 signal acquisition and processing computer processes, stores and transmits the received data. The 02 signal acquisition and processing computer can upload all data to the 01 main control computer through the communication interface, and the data processing, analysis and storage operations can also be completed by the 01 main control computer. After completing the entire or a certain stage of the aging and life test, the data during the aging and life test of the contactor under test is stored and exported to obtain the aging and life test data of the contactor under test, and complete the aging and life test of the contactor under test.

[0130] Figure 3It is a test flow chart for contactor fault prediction and health management technology. The test flow includes steps A to G, a total of 7 steps. Step A: configure the main control computer, signal acquisition and processing computer, and state control computer, and load the aging and life test control program; Step B: run the aging and life test control program, and the device under test performs aging and life tests according to the program control; Step C: each sensor in the system collects operating data in real time; Step D: The signal acquisition and processing computer processes, analyzes, stores the operating data and uploads the data to the main control computer; Step E: The main control computer processes, analyzes, and stores the data; Step F: Determine whether the test is completed. If not, repeat step B. If completed, proceed to step G. The judgment criteria and basis for determining whether the test is completed can be determined in the program configuration and settings. Commonly used judgment criteria include but are not limited to the contactor completing a fixed number of cyclic working condition tests, the contactor parameters exceeding the limit or completely failing, etc.; Step G: Data packaging, exporting, and generating device test reports. End the entire aging and test process.

[0131] According to the program settings, periodic data verification operations are performed at test intervals. The 047 verification signal injection circuit injects a standard verification signal into the 041 signal conditioning circuit. After the signal acquisition and transmission processes, it is transmitted to the 02 signal acquisition and processing computer. The 02 signal acquisition and processing computer determines whether the signal is distorted to determine whether there is a fault in the acquisition and transmission process.

[0132] Embodiment 3:

[0133] This embodiment takes the setting of the signal sampling frequency as an example.

[0134] In this embodiment, the highest frequency f in the analog signal spectrum max The sampling frequency of the hardware sampling system is 10kHz; cs is 100kHz and the signal processing rate f cc The maximum storage rate of the storage device is 800kHz. st 3×10 6 Kilobit per second (kbit / S), the system sets the unit to store data size S data 0.1 kilobit (kbit); the maximum storage capacity of the storage device C sto 5×10 9 Kilobit (kbit), unit storage data size Size data 0.1 kilobit (kbit); test duration T du For 10 6 Seconds(S).

[0135] f maxsp =min{80%fcs ,50%f cc}=min{80%×100kHz,50%×800kHz}=80kHz (20)

[0136]

[0137]

[0138] The adjustment and setting of the sampling frequency according to the present invention is combined with the above four rules and rules to determine the final sampling frequency f s :

[0139] Specific embodiment 4:

[0141] This embodiment uses an example to illustrate collection verification.

[0142] The verification process described in the present invention exists as a sub-process of the main process, and the verification process includes the following 8 sub-steps:

[0143] Step Ⅰ: Add 1 to the test number flag;

[0144] Step II: Determine whether the test number flag is equal to the specified value. The specified value = 10, that is, the test is performed once every 10 times. If the test number flag is equal to the specified value, proceed to step III.

[0145] Step III: Set the check-allowed flag to 1 and the test number flag to 0;

[0146] Step IV: injecting a standard calibration signal, wherein the standard calibration signal is selected from the signal in specific embodiment 5;

[0147] Step V: The standard calibration signal is sampled, processed, and transmitted to the signal acquisition and processing computer;

[0148] Step VI: Determine whether the standard calibration signal is distorted. The distortion rate of the standard calibration signal during the calibration of the present invention must be within 0.5%. If the distortion rate exceeds 0.5%, it is considered as signal distortion. If no distortion occurs, proceed to step VIII;

[0149] Step VII: Output the error status and the number of the error device and other information (this embodiment skips this step);

[0150] Step VIII: Clear the check flag to 0 and continue the main test procedure.

[0151] Embodiment 5:

[0152] This embodiment is described with reference to a standard calibration signal as an example.

[0153] This embodiment uses a standard sine wave signal and a triangular wave signal to form a standard calibration signal. Figure 5 The period of the standard calibration signal is set to 0.5 seconds, the first 0.3 seconds is a standard sine signal with a frequency of 20Hz and an amplitude range of [-1V, +1V]; 0.3-0.5 seconds is a triangle wave signal with a frequency of 10Hz and an amplitude range of [-1V, +1V].

[0154] Embodiment 6:

[0155] This embodiment uses examples to illustrate data compression processing. This embodiment illustrates data compression processing of AC contactor working current and contactor vibration signal respectively.

[0156] When an AC contactor is in a healthy state, the load it controls is purely resistive, and the load does not change, the working current after it is turned on is a standard sine waveform. Select the data of two adjacent cycles, divide the two data into one thousandth of a cycle, and calculate For i=1,2; k=1,2,…1000, the index of less than 0.5% is met, which proves that there is no obvious difference in the current waveforms of two adjacent cycles. The data of the two cycles are considered to be homogeneous within the allowable range. Only the data of one cycle needs to be stored, that is, only the data of the next cycle is transmitted and stored. After data compression processing, the data transmission and storage amount are significantly reduced.

[0157] The first changing edge of the contactor vibration signal appears at t0. The unit time length is set to 0.05 seconds. It is determined whether the vibration signal is stable at t0+0.05. If it is stable and no longer changes, the data acquisition time interval is set to [t0-0.05, t0+0.05]. If the vibration signal is not stable at t0+0.05, and the data no longer changes after being extended to the right by 3 unit time lengths, the data acquisition time interval is set to [t0-0.05, t0+0.20].

[0158] Embodiment 7:

[0159] This embodiment uses an example to illustrate the effectiveness evaluation method of evaluating the fault prediction and health management algorithm in step 8 of the PHM algorithm verification method.

[0160] The four evaluation indicators of PHM algorithm No. 1 to be evaluated are: 0.95 / 0.8 / 0.6 / 0.75, and the four evaluation indicators of PHM algorithm No. 2 to be evaluated are: 0.98 / 0.6 / 0.69 / 0.8.

[0161] The judgment matrix of this embodiment is as follows:

[0162]

[0163] Then the consistency of the judgment matrix is ​​tested and the CI value is introduced:

[0164]

[0165] Calculate the random consistency ratio, denoted as CR:

[0166]

[0167] This shows that the judgment matrix is ​​consistent and can be used for subsequent evaluation.

[0168] Then the weights of the above evaluation indicators are standardized:

[0169]

[0170] The weight coefficients of the four indicators are 0.537, 0.279, 0.135, and 0.049 respectively.

[0171] Assign corresponding weight coefficients to the four indicators respectively, and calculate the evaluation effect scores of different PHM algorithms:

[0172] Q=w1·ACC+w2xxx+w3·ΔT pre ′+w4·α Rob (27)

[0173] The evaluation effect scores of the two PHM algorithms to be evaluated are calculated to be 0.8511 and 0.8260 respectively, which proves that the evaluation effect of PHM algorithm No. 1 to be evaluated is better.

[0174] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A verification and experimental test method for contactor and relay fault prediction and health management technology, characterized in that: The method includes a PHM algorithm verification method and a contactor relay device aging and life test method; The PHM algorithm verification method includes the following steps: Step 1: Configure and load the PHM algorithm to be verified and the algorithm verification mode control process parameters to the main control computer; Step 2: Configure and load algorithm verification mode control process parameters to the signal acquisition and processing computer and the state control computer; Step 3: Run the algorithm verification process; Step 4: Collect the operating data and parameters of the contactor in real time. The signal acquisition and processing computer compresses, processes, stores and uploads the operating data, and performs regular verification of the accuracy of signal acquisition and transmission; Step 5: The main control computer runs the PHM algorithm to be verified; Step 6: Determine whether the test is completed. If not, repeat step 3. If completed, proceed to step 7: Step 7: Calculate the algorithm evaluation result score; Step 8: Evaluate the effectiveness of the fault prediction and health management algorithms; The device aging and life test test method includes the following steps: Step A: configure and load the "device aging and life test" control program and corresponding parameters to the main control computer, signal acquisition and processing computer and state control computer; Step B: Run the aging and life test control program to perform the aging and life test process; Step C: real-time acquisition and transmission of the operating data and parameter changes of the device under test, and periodic data verification; Step D: The signal acquisition processing computer compresses, processes, and stores the operating data, and transmits the operating data to the main control computer; Step E: The main control computer stores the operation data; Step F: Determine whether the test is completed. If not, repeat step B. If completed, execute step G. Step G: Export data and generate device test report; Data and signal verification includes the following steps: Step Ⅰ: Add 1 to the test number flag; Step II: Determine whether the test number flag is equal to the specified value. If it is equal to the specified value, proceed to step III; if it is not equal to the specified value, end the verification and continue the test; Step III: Set the check-allowed flag to 1 and the test number flag to 0; Step IV: inject standard calibration signal; Step V: The standard calibration signal is sampled, processed, and transmitted to the signal acquisition and processing computer; Step VI: Determine whether the standard calibration signal is distorted. If it is distorted, proceed to step VII; if it is not distorted, proceed to step VIII; Step VII: Output the error status and the number of the error device; Step VIII: Clear the check flag to 0 and continue the main test procedure; In evaluating the effectiveness of fault prediction and health management algorithms, a comprehensive evaluation method is implemented, which uses four evaluation indicators, namely, recall rate TPR, detection accuracy ACC, prediction stability coefficient α Rob , prediction range T pre ; The expression of the recall rate TPR is: Among them, TP is the number of true positive cases, that is, the number of cases where the algorithm makes correct positive judgments; FN is the number of false negative cases, that is, the number of cases where the algorithm makes incorrect negative judgments; The detection accuracy ACC expression is: Among them, TP is the number of true positive cases, that is, the number of cases where the algorithm makes a correct positive judgment; TN is the number of true negative cases, that is, the number of cases where the algorithm makes a correct negative judgment; P is the number of positive cases, that is, the number of all positive cases; N is the number of negative cases, that is, the number of all negative cases; The predicted stability coefficient α Rob It refers to the percentage of the duration during which the prediction result no longer changes to the total time elapsed since the first prediction signal. The prediction stability coefficient is used to evaluate the prediction stability of the PHM algorithm, and its expression is: Where Δsta is the duration of time during which the prediction result does not change, in seconds; Δocc is the total time that has passed since the first prediction signal, in seconds; The prediction range T pre It refers to the first prediction alarm time of the PHM algorithm under the premise of meeting the set confidence level. The prediction range evaluates the prediction advance degree of the PHM algorithm. The larger the prediction range, the earlier the fault signs can be found and the fault can be predicted early. Its expression is: ΔT pre =T fault -T pre Among them, T fault is the time when the fault occurs, T pre In order to meet the first prediction of the alarm time by the PHM algorithm under the premise of setting confidence, the following formula is used to normalize it and obtain the normalized result of the prediction range: Among them, T0 is the initial time of the prediction algorithm; The four evaluation indicators are obtained by independent calculation. After normalization of the indicators that need to be normalized, the importance levels of two evaluation indicators are compared pairwise to form a 4*4 judgment matrix consisting of 16 judgment scale elements. The weight coefficient of each indicator is obtained after standardization, and the evaluation effect score is finally obtained by weighted calculation. The verification and evaluation of different PHM algorithms are achieved by comparing the evaluation effect scores of different algorithms.

2. A verification and experimental test method for contactor and relay fault prediction and health management technology according to claim 1, characterized in that: The PHM algorithm verification method evaluates the effectiveness of the fault prediction and health management algorithms by using an evaluation method based on composite indicators.

3. The verification and experimental testing method for contactor and relay fault prediction and health management technology according to claim 1 is characterized in that: The adjustment and setting of sampling frequency follows the following rules: 1) Shannon sampling theorem; 2) maximum sampling processing rate allowable principle; 3) maximum storage speed allowable principle; 4) maximum storage capacity allowable principle.

Images