Full-performance test fault processing method

Abstract

The invention provides a full-performance test fault processing method. The method comprises the following steps: S1, acquiring operation state parameter data of test equipment in real time; s2, introducing a dynamic adjustment mechanism, considering a synergistic effect among state parameters of a plurality of test devices, performing anomaly detection processing on the parameter data based on a multi-dimensional parameter intelligent anomaly detection algorithm, and judging whether the test devices are abnormal or not; and S3, when the test equipment is judged to be abnormal, starting a fault diagnosis algorithm, performing fault diagnosis on the test equipment, determining a fault type and fault location by combining historical state parameter data of the test equipment and the fault mode library, and performing intelligent recovery according to the fault type. According to the invention, the operation state of the test equipment can be obtained in real time, the non-linear change and the potential fault mode of the operation state of the equipment can be accurately identified, the accurate identification of complex faults can be realized, and the identification precision is high.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of fault processing, and particularly relates to a full-performance test fault processing method. BACKGROUND

[0002] In the modern industrial equipment test process, full-performance test, as a key means to evaluate the comprehensive performance of equipment under various working conditions, has been widely applied in technology-intensive fields such as aerospace, rail transportation and energy power. Such tests usually involve the running behavior of equipment under extreme conditions, alternating loads and multi-working condition coordination. Through collecting the state parameter data of equipment during the test process, the safety, stability, responsiveness and reliability of the equipment are evaluated. However, in the actual test process, due to the complex structure of the equipment, the variety of operating parameters, the changeable test environment and other factors, the equipment is prone to various fault states, which further affects the stability of the test and the accuracy of the data.

[0003] The traditional equipment monitoring and fault processing method usually relies on single-parameter abnormal analysis or simple threshold judgment method. This method is easily affected by environmental fluctuations and equipment individual differences, leading to misjudgment or missed judgment. In addition, the fault of the test equipment is often not caused by the abnormality of a single parameter, but by the complex interaction of multiple state parameters. Therefore, relying only on single-parameter analysis cannot fully reflect the real health status of the equipment.

[0004] The patent with publication number CN 119274284A discloses a test fault early warning method, system, device and medium. The test safety monitoring system is divided into multiple subsystems, each subsystem includes multiple monitoring indicators, and each monitoring indicator is collected through multiple channels. The alarm value assignment rule of each monitoring indicator is formulated according to whether the channel of each monitoring indicator alarms. The score of each subsystem is calculated according to the alarm value assignment rule, the weighted coefficient of each subsystem is obtained, and the test safety score interval is obtained by weighting and superimposing the score of each subsystem according to the weighted coefficient of each subsystem. The fault type is judged according to the test safety score interval, and early warning is performed according to the fault type. However, it still has the technical problem of inaccurate fault judgment caused by poor accuracy of test parameter processing. SUMMARY

[0005] In order to solve the problems existing in the prior art, a full-performance test fault processing method is provided.

[0006] The technical scheme adopted by the application to solve its technical problems is:

[0007] The technical scheme provides a full-performance test fault processing method, which includes the following steps:

[0008] S1: During the full performance test, the running state parameter data of the test equipment is obtained in real time based on multiple sensors as the input of the multi-dimensional parameter intelligent anomaly detection algorithm;

[0009] S2: A dynamic adjustment mechanism is introduced, the synergistic effect between multiple test equipment state parameters is considered, the parameter data is processed by the multi-dimensional parameter intelligent anomaly detection algorithm for anomaly detection, the anomaly detection result is obtained, and it is judged whether the test equipment is abnormal;

[0010] S3: When the test equipment is judged to be abnormal, the fault diagnosis algorithm is started to diagnose the fault of the test equipment, the historical state parameter data of the test equipment and the fault mode library are combined to determine the fault type and fault location, and intelligent recovery is performed according to the fault type.

[0011] Preferably, in S1, the running state parameter data includes environmental parameters, electrical parameters, mechanical parameters and performance parameters;

[0012] The environmental parameters include temperature, humidity and pressure;

[0013] The electrical parameters include current, voltage and power;

[0014] The mechanical parameters include vibration frequency, rotating speed and machine load;

[0015] The performance parameters include output power and efficiency.

[0016] Preferably, the obtained parameter data is preprocessed, and any preprocessed running state parameter x i of the test equipment is calculated based on the historical equipment state parameter data to calculate the mean and standard deviation of each parameter, and the abnormality degree of each parameter is defined, and the calculation formula is represented as:

[0017]

[0018] In the formula, f(x i ) is the abnormality degree of the i-th preprocessed state parameter of the test equipment, which is used to measure the abnormality degree of the parameter, x i is the i-th preprocessed state parameter of the test equipment, is the mean of the i-th preprocessed state parameter of the test equipment, σ i is the standard deviation of the i-th preprocessed state parameter of the test equipment, which represents the range or dispersion of the parameter fluctuation;

[0019] α i is the first adjustment factor for controlling the non-linear influence in the abnormality degree of the i-th preprocessed state parameter of the test equipment, and β iis the second adjustment factor, used to control the influence degree of the exponential part in the i th pre-processed state parameter anomaly of the test equipment on the anomaly.

[0020] Preferably, in the formula (1),

[0021] The first term reflects the difference between the pre-processed state parameter x i and its mean value , the greater the deviation, the higher the value of the anomaly;

[0022] The second term introduces a nonlinear term, which is used to intensify the anomaly detection of the test equipment parameters. When the test equipment has nonlinear fluctuations or sudden changes, adjustments are made based on the periodic fluctuations of the test equipment, which can capture more dimensions of equipment changes;

[0023] The third term is the regularization term, which is used to suppress parameter deviations caused by external noise;

[0024] Through formula (1), the anomaly f(x i ) of each test equipment parameter is obtained;

[0025] According to the anomaly f(x i ) for preliminary evaluation of the health status of the test equipment, the anomaly f(x i ) of each test equipment parameter is compared with the preset anomaly threshold of the test equipment parameter. When the set anomaly threshold is exceeded, it is considered that there is an anomaly, and is marked.

[0026] Preferably, in the S2, the synergistic effect includes the calculation of the synergistic anomaly, and the synergistic anomaly of any two parameters x i and x j of the test equipment The calculation formula is as follows:

[0027]

[0028] In the formula, x j is the j th pre-processed state parameter of the test equipment, is the mean value of the j th pre-processed state parameter of the test equipment, σ j is the standard deviation of the j th pre-processed state parameter of the test equipment, indicating the range or dispersion of the parameter fluctuations, γ ij is the interaction adjustment parameter, used to adjust the strength of the interaction effect between any two pre-processed test equipment parameters, δ ij is the exponential adjustment parameter, used to control the influence of the difference between any two pre-processed test equipment parameters.

[0029] Preferably, in the formula (2), x

[0030] The first term Reflects the deviation of x i and x j from their respective mean values, through the calculation of this term, the abnormality degree of the pre-processed test equipment state parameters is evaluated;

[0031] The second term Captures the interaction effect between x i and x j ;

[0032] The third term Used to suppress the abnormality degree calculation bias caused by too large difference between parameters;

[0033] Obtain the joint abnormality degree of each pair of pre-processed test equipment state parameters x i and x j ; After that, compare it with the preset joint abnormality degree threshold value, if the joint abnormality degree exceeds the preset joint abnormality degree threshold value, it is considered that there is an abnormality, and is marked as the joint abnormality judgment result.

[0034] Preferably, the k-dimensional joint joint abnormality degree of k test equipment state parameters is calculated, and the calculation formula is represented as:

[0035]

[0036] In the formula, J is the high-dimensional joint joint abnormality degree formed by the joint abnormality degree of multiple pre-processed test equipment state parameters, which measures the mutual joint influence between multiple equipment parameters, and considers the abnormal correlation between parameters and its influence on the overall equipment health state;

[0037] xi is the i m th pre-processed test equipment state parameter, is the mean value of the i m th pre-processed test equipment state parameter, is the standard deviation of the i m th pre-processed test equipment state parameter, xi is the i n th pre-processed test equipment state parameter, is the mean value of the i n th pre-processed test equipment state parameter, is the standard deviation of the i n th pre-processed test equipment state parameter;

[0038] ​ is the i m and i n cross-regulation parameter of the i is the i m and i n exponential regulation parameter of the i j pre-processed test equipment state parameter, used to control the influence of the difference between any two pre-processed test equipment parameters, 1≤m<n≤k is the combination of all possible parameter pairs, the index of m and n ranges from 1 to k.

[0039] Preferably, the k-dimensional joint synergy abnormality degree is compared with a preset k-dimensional joint synergy abnormality degree threshold value, when the k-dimensional joint synergy abnormality degree exceeds the k-dimensional joint synergy abnormality degree threshold value, it is considered that there is an abnormality, and is marked as a k-dimensional abnormality judgment result;

[0040] The preliminary evaluation abnormality judgment result, the synergy abnormality judgment result, and the k-dimensional abnormality judgment result are combined to obtain an abnormality detection result, and the abnormality state of the test equipment is determined according to the abnormality detection result.

[0041] Preferably, in S3, a nonlinear interdependence measurement calculation formula is introduced by matching the current pre-processed test equipment state parameter with the fault mode in the fault mode library, and the calculation formula is represented as:

[0042]

[0043] In the formula, D(x, y j ) is the nonlinear interdependence measurement of the current equipment state and the fault mode, x is the pre-processed test equipment state parameter data set, is the i known fault mode state data in the fault mode library, is the i state data of the i is a nonlinear weighting factor, representing nonlinear weighted adjustment of the difference of each parameter;

[0044] σ i is the standard deviation of the i is an exponential regulation factor, controlling the sensitivity of the overall similarity measurement to the parameter difference, is an interaction regulation factor, used to regulate the interaction between each parameter, θ is a weight factor, determining the influence degree of the interaction on the overall measurement, controlling the relative weight of the complex interaction between the parameters in the measurement, is a balance factor, controlling the weighted balance of the interaction.

[0045] Preferably, the nonlinear interdependence measure of the current device state and all fault modes is calculated, and the maximum one is taken as the best match to determine the current device fault type.

[0046] Compared with the prior art, the present application has the following advantages:

[0047] 1. The present application can comprehensively and meticulously master the running state of the equipment by acquiring real-time running state parameters in multiple dimensions such as environment, electricity, machinery and performance, avoiding missed detection and misjudgment, and at the same time, using a multi-dimensional parameter intelligent anomaly detection algorithm combined with a dynamic adjustment mechanism, so that the anomaly detection is more flexible and can identify nonlinear changes and potential fault modes of the equipment running state.

[0048] 2. The present application effectively captures potential fault modes caused by multi-parameter linkage by fully considering the correlation between parameters and calculating the cooperative anomaly degree and k-dimensional joint cooperative anomaly degree, greatly improves the detection rate of complex and multi-source faults, and realizes efficient similarity matching and discrimination of the current equipment state and existing fault types by constructing a fusion analysis mechanism of the fault mode library and historical state data and introducing a matching algorithm based on nonlinear interdependence measure.

[0049] 3. The present application uses a combination of indicators such as parameter standard deviation, interaction adjustment factor and nonlinear weighting to evaluate the difference between the current state and historical modes in multiple dimensions, thereby enhancing the ability to identify complex faults. The above fault identification method is not only effective for single parameter anomaly, but also suitable for multi-variable interactive driven composite faults, effectively solving the problem of insufficient multi-source coupling fault identification ability of existing methods. BRIEF DESCRIPTION OF DRAWINGS

[0050] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

[0051] Figure 1 is the overall flowchart of the present application. DETAILED DESCRIPTION

[0052] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, in which the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0053] Example 1

[0054] As Figure 1As shown, the embodiment proposes a full performance test fault handling method, including the following steps:

[0055] S1: During the full performance test, based on multiple sensors, real-time acquisition of the running state parameter data of the test equipment as the input of the multi-dimensional parameter intelligent anomaly detection algorithm;

[0056] S2: Introduce a dynamic adjustment mechanism, considering the synergistic effect between multiple test equipment state parameters, based on the multi-dimensional parameter intelligent anomaly detection algorithm, perform anomaly detection processing on the parameter data, obtain the anomaly detection result, and judge whether the test equipment is abnormal;

[0057] S3: When the test equipment is judged to be abnormal, start the fault diagnosis algorithm, perform fault diagnosis on the test equipment, combine the historical state parameter data of the test equipment and the fault mode library, determine the fault type and fault location, and perform intelligent recovery according to the fault type.

[0058] In S1, during the test, the test room staff selects sensors such as temperature sensors, pressure sensors, current and voltage sensors according to the characteristics of the test equipment to monitor the test equipment in real time, and acquires the running state parameter data of the test equipment. The running state parameter data includes environmental parameters, electrical parameters, mechanical parameters and performance parameters.

[0059] The environmental parameters include temperature, humidity and pressure, the electrical parameters include current, voltage and power, the mechanical parameters include vibration frequency, rotating speed and machine load, and the performance parameters include output power and efficiency.

[0060] Starting from the multi-dimensional state data of the test equipment, the acquired parameter data is preprocessed, including denoising, standardization, dimension processing and other operations on the parameter data, to obtain the preprocessed state parameters. For any preprocessed running state parameter x i (such as temperature, pressure, rotating speed, etc.), based on the historical equipment state parameter data, the mean and standard deviation of each parameter are calculated, and the abnormality degree of each parameter is defined, and the calculation formula is represented as:

[0061]

[0062] In the formula, f(x i ) is the abnormality degree of the i-th preprocessed state parameter of the test equipment, which is used to measure the abnormality degree of the parameter, x i is the i-th preprocessed state parameter of the test equipment, is the mean value of the i-th preprocessed state parameter of the test equipment, that is, the average value of the parameter in a period of time, and iis the standard deviation of the ith pre-processed state parameter of the test equipment, indicating the range or dispersion of the parameter fluctuation;

[0063] α i is the first adjustment factor to control the non-linear effect in the ith pre-processed state parameter of the test equipment, which determines the sensitivity of the abnormality degree to the deviation of the equipment state parameter, β i is the second adjustment factor to control the influence of the exponential part in the ith pre-processed state parameter of the test equipment on the abnormality degree.

[0064] In formula (1):

[0065] The first term reflects the difference between the pre-processed state parameter x i and its mean value , the greater the deviation, the higher the value of the abnormality degree, which reflects the rapid increase of the abnormality degree of the equipment parameter with the increase of the deviation;

[0066] The second term introduces a non-linear term to intensify the abnormality detection of the test equipment parameters, which can capture more dimensions of the equipment changes based on the periodic fluctuations of the test equipment when nonlinear fluctuations or sudden changes occur in the test equipment;

[0067] The third term is a regularization term to suppress the parameter deviation caused by external noise, which ensures that the abnormality degree will not fluctuate excessively due to noise or small errors by introducing this term;

[0068] Through formula (1), the abnormality degree f(x i ) of each test equipment parameter is obtained;

[0069] According to the abnormality degree f(x i ) for preliminary evaluation of the health status of the test equipment, the abnormality degree f(x i ) of each test equipment parameter is compared with the preset abnormality threshold of the test equipment parameter, and when it exceeds the set abnormality threshold, it is considered to be abnormal and is marked.

[0070] In S2, in order to improve the detection accuracy, the synergistic effect between multiple test equipment state parameters is considered, for example, temperature and pressure, pressure and vibration, and other parameters form a certain association, only through the joint abnormality detection of multiple parameters, the deep-seated abnormality can be effectively captured.

[0071] The synergistic effect includes the calculation of the synergistic abnormality degree, which describes the interaction between two parameters, and the synergistic abnormality degree of any two parameters x i and x j of the test equipment The calculation formula is represented as:

[0072]

[0073] In the formula, x j is the jth pre-processed state parameter of the test equipment, is the mean value of the jth pre-processed state parameter of the test equipment, i.e., the average value of the parameter in a period of time, σ j is the standard deviation of the jth pre-processed state parameter of the test equipment, indicating the range or dispersion of the parameter, γ ij is an interaction adjustment parameter, used to adjust the strength of the interaction effect between any two pre-processed test equipment parameters, δ ij is an exponential adjustment parameter, used to control the influence of the difference between any two pre-processed test equipment parameters, especially to suppress excessive calculation of abnormality in the case of excessively large parameter difference.

[0074] In formula (2):

[0075] The first term reflects the deviation of x i and x j from their respective mean values, and through the calculation of this term, the abnormality degree of the pre-processed test equipment state parameter is evaluated;

[0076] The second term captures the interaction effect between x i and x j , especially the change trend, and if the two parameters jointly deviate, this term will exacerbate their synergistic abnormality;

[0077] The third term is used to suppress the abnormality degree calculation bias caused by excessively large difference between parameters, and when the difference between two pre-processed test equipment state parameters is too large, the third term is introduced to suppress the excessive influence of their abnormality degree;

[0078] The synergistic abnormality degree i of each pair of pre-processed test equipment state parameters x j is obtained After that, the preset synergistic abnormality threshold is compared, and if the synergistic abnormality degree exceeds the preset synergistic abnormality threshold, it is considered to be abnormal and is marked as a synergistic abnormality judgment result.

[0079] The k-dimensional joint synergistic abnormality degree of k test equipment state parameters is calculated, and the calculation formula is represented as:

[0080]

[0081] wherein, is a high-dimensional joint synergistic anomaly degree formed by the joint anomaly degrees of a plurality of pre-processed test equipment state parameters , measures the mutual synergistic influence between a plurality of equipment parameters, and considers the abnormal correlation between parameters and its influence on the overall equipment health state;

[0082] is the i m th pre-processed test equipment state parameter, is the mean of the i m th pre-processed test equipment state parameter, is the standard deviation of the i m th pre-processed test equipment state parameter, is the i n th pre-processed test equipment state parameter, is the mean of the i n th pre-processed test equipment state parameter, is the standard deviation of the i n th pre-processed test equipment state parameter;

[0083] is the interaction adjustment parameter of the i m th and i n th pre-processed test equipment state parameters, representing the strength of the synergistic effect between two test equipment, is the exponential adjustment parameter of the i m th and i n th pre-processed test equipment state parameters, used to control the influence of the difference between any two pre-processed test equipment parameters, especially to suppress excessive calculation of anomaly degree in the case of excessive difference between parameters when calculating synergistic anomaly degree, 1≤m<n≤k is the combination of all possible parameter pairs, and the index values of m and n range from 1 to k.

[0084] The k-dimensional joint synergistic anomaly degree is compared with a preset k-dimensional joint synergistic anomaly degree threshold value, and when the k-dimensional joint synergistic anomaly degree exceeds the k-dimensional joint synergistic anomaly degree threshold value, it is considered to be abnormal, and is marked as the k-dimensional anomaly judgment result;

[0085] The preliminary evaluation anomaly judgment result, the synergistic anomaly judgment result, and the k-dimensional anomaly judgment result are merged to obtain an anomaly detection result, and the anomaly state of the test equipment is determined according to the anomaly detection result.

[0086] In S3, when the test equipment is judged to be abnormal, that is, the abnormality of a certain test equipment state parameter or parameter combination exceeds the preset threshold, it will be judged that the test equipment has a potential fault, and then the fault diagnosis mechanism will be triggered. When the fault diagnosis mechanism is triggered, the historical operating data of the test equipment (temperature, pressure, current, voltage, mechanical parameters, output power, efficiency, etc.) and the fault mode library (the equipment's past common fault cases and their corresponding characteristic parameters (such as abnormal fluctuations, trend changes, etc.)) are obtained from the existing database.

[0087] By matching the pre-processed test equipment state parameters with the failure modes in the failure mode library, a nonlinear interdependence metric calculation formula is introduced, which is expressed as:

[0088]

[0089] In the formula, D(x,y) j ) is a measure of the nonlinear interdependence between the current equipment state and the failure mode, and x is the preprocessed dataset of test equipment state parameters. It is the first in the fault mode library Known fault mode status data, It is the first in the fault mode library The i-th state data for a known fault mode, where N is the number of dimensions of the device state parameters. It is a non-linear weighting factor, representing a non-linear weighted adjustment for the difference in each parameter;

[0090] σ i It is the standard deviation of the state parameters of the i-th pretreated test equipment. It is an exponential adjustment factor that controls the sensitivity of the overall similarity measure to parameter differences. θ is the interaction modulator, used to adjust the interaction between each parameter. θ is the weighting factor, which determines the degree of influence of the interaction on the overall metric, controlling the relative weights of complex interactions between parameters in the metric. It is a balancing factor that controls the weighted balance of interactions.

[0091] Calculate the nonlinear interdependence measure between the current equipment state and all fault modes, take the largest one as the best match, and determine the current equipment fault type.

[0092] Example 2

[0093] like Figure 1 As shown, this embodiment proposes a fault handling method for full-performance testing, including the following steps:

[0094] Test equipment state parameter influence analysis: By analyzing the abnormality of each test equipment state parameter, identify which parameter changes have a significant impact on the equipment state;

[0095] For example, if temperature and pressure simultaneously appear abnormal fluctuations, it can be inferred that the cooling system of the equipment may have failed, multi-parameter correlation analysis: through the calculation of multi-parameter cooperative abnormality, analyze the mutual influence between multiple parameters, further narrow down the area of fault occurrence.

[0096] For example, if the vibration parameter is abnormal and cooperates with the abnormality of temperature, pressure and other parameters, it may be located to mechanical structure or driving system failure, fault source tracing: combined with the working principle of the equipment and the fault mode library, according to the abnormal parameter value and its change trend, use the existing rule-based reasoning engine to trace the possible fault source.

[0097] For example, if the current and voltage fluctuate abnormally and frequently, it may be that a component in the electrical system (such as a power module) has failed.

[0098] Finally, according to the determined fault type and fault location result, intelligent recovery is carried out, including recovery strategy selection; for different fault types, automatically select the corresponding recovery strategy. For example:

[0099] Electrical fault: When an overload or short circuit fault of the electrical system is detected, automatic power-off protection can be performed to avoid damaging the equipment, or load distribution can be adjusted to allow the equipment to continue working within a safe range.

[0100] Mechanical failure: When a mechanical failure occurs, the working parameters of the equipment (such as speed, load, etc.) can be automatically adjusted, or standby equipment can be enabled or the working mode can be adjusted to ensure the continuation of the test.

[0101] Temperature anomaly: For equipment overheating problems, the cooling system can be started or the working load can be reduced to bring the equipment temperature back to a safe range.

[0102] Adjusting parameters and equipment control: In the process of intelligent recovery, the equipment parameters (such as reducing output power, adjusting speed, etc.) are adjusted to avoid the equipment being in a fault state. By adjusting the running parameters of the equipment in real time, the equipment can be brought back to normal operation state from abnormal state.

[0103] By obtaining the running state parameters of multiple dimensions such as environment, electricity, machinery and performance in real time, the running state of the equipment can be comprehensively and carefully mastered, avoiding missed detection and misjudgment. At the same time, multi-dimensional parameter intelligent anomaly detection algorithm is adopted, combined with dynamic adjustment mechanism, making the anomaly detection more flexible, which can identify the nonlinear change of equipment running state and potential fault mode.

[0104] By fully considering the correlation between parameters, through the calculation of collaborative abnormality and k-dimensional joint collaborative abnormality, the potential failure mode caused by multi-parameter linkage is effectively captured, the detection rate of complex and multi-source faults is greatly improved, and through the fusion analysis mechanism of the construction of the failure mode library and the historical state data, and the introduction of the matching algorithm based on the nonlinear interdependence measure, the efficient similarity matching and discrimination of the current device state and the existing failure type are realized.

[0105] By using parameter standard deviation, interaction adjustment factor, nonlinear weighting and other combined indicators, the difference between the current state and the historical mode is evaluated in multiple dimensions, thereby enhancing the recognition ability of complex faults. The above fault recognition method is not only effective for single parameter abnormality, but also suitable for multi-variable interactive driven composite faults, effectively solving the problem of insufficient recognition ability of existing methods for multi-source coupled faults.

[0106] Although the embodiments of the present application have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and purposes of the present application, and the scope of the present application is defined by the claims and their equivalents.

Claims

1. A method for troubleshooting faults in a full-performance test, characterized in that, Includes the following steps: S1: During the full-performance test, the operating status parameter data of the test equipment is acquired in real time based on multiple sensors, which serves as the input for the multi-dimensional parameter intelligent anomaly detection algorithm; S2: Introduce a dynamic adjustment mechanism, and consider the synergistic effect between the state parameters of multiple test equipment. Based on the multi-dimensional parameter intelligent anomaly detection algorithm, perform anomaly detection processing on the parameter data, obtain the anomaly detection result, and determine whether there is an anomaly in the test equipment. S3: When the test equipment is determined to be abnormal, the fault diagnosis algorithm is activated to diagnose the fault of the test equipment. Combining the historical status parameter data of the test equipment and the fault mode library, the fault type and fault location are determined, and intelligent recovery is performed according to the fault type.

2. The method for handling faults in a full-performance test according to claim 1, characterized in that, In S1, the operating status parameter data includes environmental parameters, electrical parameters, mechanical parameters, and performance parameters; Environmental parameters include temperature, humidity, and pressure; Electrical parameters include current, voltage, and power; Mechanical parameters include vibration frequency, rotational speed, and machine load; Performance parameters include output power and efficiency.

3. The method for handling faults in a full-performance test according to claim 1, characterized in that, The acquired parameter data is preprocessed, and any preprocessed operating status parameter x of the test equipment is processed. i Based on historical equipment status parameter data, the mean and standard deviation of each parameter are calculated, and the outlier of each parameter is defined. The calculation formula is expressed as follows: In the formula, f(x) i ) represents the anomaly degree of the i-th preprocessed state parameter of the test equipment, used to measure the degree of anomaly of that parameter, x i It is the i-th pre-processed state parameter of the test equipment. σ is the mean of the i-th pre-processed state parameters of the experimental equipment. i It is the standard deviation of the i-th pre-processed state parameter of the test equipment, representing the range or dispersion of the parameter's fluctuation. α i It is the first adjustment factor, used to control the nonlinear influence of the anomaly degree of the i-th pre-treated state parameter of the experimental equipment, β. i It is the second adjustment factor, used to control the influence of the exponential part of the anomaly degree of the i-th pre-processed state parameter of the test equipment on the anomaly degree.

4. The method for handling faults in a full-performance test according to claim 3, characterized in that, In the above formula (1): First item This reflects the preprocessed state parameter x i Its mean The greater the difference between them, the higher the anomaly value; Second item By introducing nonlinear terms, the detection of anomalies in the test equipment parameters is enhanced. When nonlinear fluctuations or sudden changes occur in the test equipment, adjustments are made based on the periodic fluctuations of the test equipment, which can capture more dimensions of equipment changes. Third item It is a regularization term used to suppress parameter deviations caused by external noise; The anomaly degree f(x) of each test equipment parameter is obtained using equation (1). i ); Based on the anomaly degree f(x) i To conduct a preliminary assessment of the health status of the test equipment, the abnormality degree f(x) of each test equipment parameter is calculated. i The abnormal threshold is compared with the preset abnormal threshold of the test equipment parameters. When the abnormal threshold is exceeded, it is considered to be abnormal and marked.

5. The method for handling faults in a full-performance test according to claim 1, characterized in that, In S2, the synergistic effect includes the calculation of the synergistic anomaly degree, and any two parameters x of the experimental equipment. i and x j Cooperative anomaly The calculation formula is expressed as follows: In the formula, x j It is the j-th pre-processed state parameter of the test equipment. σ is the mean of the j-th pre-processed state parameters of the experimental equipment. j γ is the standard deviation of the j-th pretreated state parameter of the test equipment, representing the range or dispersion of the parameter's fluctuation. ij It is an interactive adjustment parameter used to adjust the intensity of the interaction effect between any two pretreated experimental equipment parameters, δ. ij It is an exponential adjustment parameter used to control the influence of differences between any two pretreated test equipment parameters.

6. The method for handling faults in a full-performance test according to claim 5, characterized in that, In the above formula (2): First item Reflects x i and x j The deviation from their respective means is used to assess the degree of abnormality of the pretreated test equipment state parameters through the calculation of this item; Second item Capture x i and x j Interaction effects between them; Third item Used to suppress the deviation in anomaly calculation caused by excessive differences between parameters; Obtain the state parameter x of each pair of pretreated test equipment. i and x j Cooperative anomaly Then, it is compared with the preset collaboration anomaly threshold. If the collaboration anomaly exceeds the preset collaboration anomaly threshold, it is considered to be an anomaly and is marked as the collaboration anomaly judgment result.

7. The method for handling faults in a full-performance test according to claim 6, characterized in that, The k-dimensional joint cooperative anomaly degree of k test equipment state parameters is calculated using the following formula: In the formula, It is a high-dimensional joint synergistic anomaly formed by the joint anomaly of multiple pre-processed test equipment state parameters. It measures the mutual synergistic influence between multiple equipment parameters and considers the abnormal correlation between parameters and their impact on the overall equipment health status. is the i-th m pre-processed test equipment status parameter, is the mean value of the i-th m pre-processed test equipment status parameters, is the standard deviation of the i-th m pre-processed test equipment status parameters, is the i-th n pre-processed test equipment status parameter, is the mean value of the i-th n pre-processed test equipment status parameters, is the standard deviation of the i-th n pre-processed test equipment status parameters; γi min is the interaction adjustment parameter of the i-th m and the i-th n pre-processed test equipment status parameters, representing the synergy strength between two test equipment, is the exponential adjustment parameter of the i-th m and the i-th n pre-processed test equipment status parameters, used to control the influence of the difference between any two pre-processed test equipment parameters, 1 ≤ m < n ≤ k is to combine all possible parameter pairs, and the index value ranges of m and n are from 1 to k.

8. The method for handling faults in a full-performance test according to claim 7, characterized in that, k-dimensional joint cooperative anomaly The result is compared with a preset k-dimensional joint collaborative anomaly threshold. If the result exceeds the k-dimensional joint collaborative anomaly threshold, an anomaly is considered to exist and is marked as the k-dimensional anomaly judgment result. The preliminary anomaly judgment results, the collaborative anomaly judgment results, and the k-dimensional anomaly judgment results are merged to obtain the anomaly detection results. Based on the anomaly detection results, the abnormal state of the test equipment is determined.

9. The method for handling faults in a full-performance test according to claim 1, characterized in that, In step S3, by matching the pre-processed test equipment state parameters with the fault modes in the fault mode library, a nonlinear interdependence metric calculation formula is introduced, which is expressed as: In the formula, D(x,y) j ) is a measure of the nonlinear interdependence between the current equipment state and the failure mode, and x is the preprocessed dataset of test equipment state parameters. It is the first in the fault mode library. Known fault mode status data, It is the first in the fault mode library. The i-th state data for a known fault mode, where N is the number of dimensions of the device state parameters. It is a non-linear weighting factor, representing a non-linear weighted adjustment for the difference in each parameter; σ i It is the standard deviation of the state parameters of the i-th pretreated test equipment. It is an exponential adjustment factor that controls the sensitivity of the overall similarity measure to parameter differences. θ is the interaction modulator, used to adjust the interaction between each parameter. θ is the weighting factor, which determines the degree of influence of the interaction on the overall metric, controlling the relative weights of complex interactions between parameters in the metric. It is a balancing factor that controls the weighted balance of interactions.

10. A method for handling faults in a full-performance test according to claim 9, characterized in that, Calculate the nonlinear interdependence measure between the current equipment state and all fault modes, take the largest one as the best match, and determine the current equipment fault type.

Images