Method and device for determining residual life of component, equipment and storage medium

Abstract

The invention relates to a method, device and equipment for determining the residual life of a component and a storage medium, and relates to the technical field of data processing, and the method comprises the steps: obtaining a test data set corresponding to a target component, carrying out the feature extraction of the test data set, and obtaining a feature index set, and determining the residual life of the target component according to the test data set under the condition that the characteristic indexes in the characteristic index set meet the alarm condition. And feature extraction is carried out, so that interference of single test data on alarm prediction can be avoided, and if the feature indexes in the feature index set meet the alarm condition, the residual life of the target component is determined according to the test data set, so that early warning can be carried out on the residual life of the target component in time before the target component fails. Therefore, the target component is replaced before failure, the failure of the target component is avoided, and the target component can be prevented from being replaced too early.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, device, equipment and storage medium for determining the remaining life of a component. Background Art

[0002] For some components, performance degradation may occur during long-term use. For example, wear and tear may cause performance degradation, which may lead to failure of the component.

[0003] In the related technology, two methods can be used: post-maintenance and preventive maintenance. Among them, post-maintenance will inevitably cause production stagnation and other damages. Preventive maintenance needs to rely on the life statistics of a large number of failed components to formulate the service life of the component. For the same component, the service life is different under different operating conditions, so it is difficult to formulate an accurate service life for a single component. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides a method, device, equipment and storage medium for determining the remaining life of a component. For a certain target component, a test data set for the target component can be obtained, and features can be extracted from the test data set to obtain feature indicators corresponding to multiple monitoring moments, thereby avoiding interference of a single test data with alarm prediction. The feature indicators are then processed. If the feature indicators in the feature indicator set meet the alarm conditions, the remaining life of the target component is determined based on the test data set, so that the remaining life of the target component can be warned in time before the target component fails, and then replaced before it fails, thereby avoiding failure of the target component and avoiding premature replacement of the target component.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for determining the remaining life of a component is provided, comprising:

[0006] Obtain the test data set corresponding to the target component;

[0007] Extracting features from the test data set to obtain a feature index set, wherein the feature index set includes feature indexes corresponding to multiple monitoring moments;

[0008] When the characteristic indicators in the characteristic indicator set meet the alarm condition, the remaining life of the target component is determined according to the test data set.

[0009] Optionally, the test data set includes test data subsets corresponding to multiple target test items;

[0010] The feature extraction of the test data set to obtain a feature index set includes:

[0011] Perform feature extraction on the test data subset corresponding to each target test item to obtain a feature indicator set corresponding to each target test item;

[0012] When the characteristic indicator in the characteristic indicator set meets the alarm condition, determining the remaining life of the target component according to the test data set includes:

[0013] For a characteristic indicator set corresponding to any target test item, when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm condition, determine the remaining sub-life of the target component corresponding to the any target test item according to the test data subset corresponding to the any target test item;

[0014] The remaining life of the target component is determined according to the remaining sub-life corresponding to at least one target test item.

[0015] Optionally, the extracting features of the test data subsets corresponding to each target test item to obtain a feature index set corresponding to each target test item includes:

[0016] For a test data subset corresponding to any target test item, the test data subset is extracted through a sliding window to obtain multiple groups of candidate test data, and a group of candidate test data is extracted at each monitoring moment;

[0017] Feature extraction is performed on each set of candidate test data to obtain feature indicators corresponding to multiple monitoring moments.

[0018] Optionally, when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm condition, before determining the remaining sub-life of the target component corresponding to any target test item according to the test data subset corresponding to the any target test item, the method further includes:

[0019] Determine the first characteristic indicator and the alarm threshold value corresponding to the current monitoring moment according to the characteristic indicator set corresponding to any target test item;

[0020] When the first characteristic indicator is less than the alarm threshold, it is determined that the first characteristic indicator corresponding to the current monitoring time satisfies the alarm condition.

[0021] Optionally, determining the first characteristic indicator and the alarm threshold corresponding to the current monitoring moment according to the characteristic indicator set corresponding to any target test item includes:

[0022] Acquire a first characteristic indicator corresponding to the current monitoring moment from the characteristic indicator set corresponding to any target test item;

[0023] Obtaining second characteristic indicators corresponding to multiple consecutive historical monitoring moments before the current monitoring moment;

[0024] The alarm threshold is determined according to the mean and standard deviation of the second characteristic indicator.

[0025] Optionally, determining the remaining sub-lifetime of the target component corresponding to any target test item according to the test data subset corresponding to the any target test item includes:

[0026] Determining a degradation model corresponding to the target component, and obtaining a life prediction model based on the degradation model;

[0027] Acquire multiple test data acquired after the at least two consecutive monitoring moments in the test data subset corresponding to any target test item to obtain a target data set;

[0028] The remaining sub-lifetime of the target component corresponding to any target test item is determined according to the target data set and the life prediction model.

[0029] Optionally, determining the remaining sub-lifetime of the target component corresponding to any target test item according to the target data set and the life prediction model includes:

[0030] Performing parameter estimation on the target data set to obtain a degradation rate and a degradation volatility intensity;

[0031] Determining degradation results corresponding to different test moments according to the degradation rate and the degradation volatility intensity;

[0032] The degradation rate, the degradation volatility intensity and the degradation result are processed by the life prediction model to obtain the remaining sub-life of the target component corresponding to any target test item.

[0033] Optionally, obtaining a test data set corresponding to the target component includes:

[0034] Determining at least one target test item related to the target component;

[0035] A test data set corresponding to each target test item is obtained respectively. The test data set corresponding to any target test item is obtained by periodically testing the target component using the test method corresponding to the target test item.

[0036] Optionally, determining the remaining life of the target component according to the remaining sub-life corresponding to at least one target test item includes:

[0037] The minimum value among the remaining sub-lifetimes corresponding to the at least one target test item is determined as the remaining lifetime of the target component.

[0038] According to a second aspect of an embodiment of the present disclosure, there is provided a device for determining the remaining life of a component, comprising:

[0039] An acquisition module is configured to acquire a test data set corresponding to a target component;

[0040] A feature extraction module is configured to extract features from the test data set to obtain a feature index set, wherein the feature index set includes feature indexes corresponding to multiple monitoring moments;

[0041] The first determination module is configured to determine the remaining life of the target component according to the test data set when the characteristic indicator in the characteristic indicator set meets the alarm condition.

[0042] According to a third aspect of an embodiment of the present disclosure, a device for determining the remaining life of a component is provided, comprising:

[0043] processor;

[0044] a memory for storing processor-executable instructions;

[0045] The processor is configured to implement the steps of the method for determining the remaining life of a component provided in the first aspect of the present disclosure when executing.

[0046] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the method for determining the remaining life of a component provided by the first aspect of the present disclosure are implemented.

[0047] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0048] For a certain target component, a test data set for the target component can be obtained, and features can be extracted from the test data set to obtain feature indicators corresponding to multiple monitoring moments, avoiding interference of a single test data on alarm prediction, and then processing the feature indicators. If the feature indicators in the feature indicator set meet the alarm conditions, the remaining life of the target component is determined based on the test data set, so that the remaining life of the target component can be warned in time before the target component fails, and then replaced before it fails, avoiding failure of the target component and avoiding premature replacement of the target component.

[0049] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0051] Figure 1 The present invention is a flow chart of a method for determining the remaining life of a component according to an exemplary embodiment.

[0052] Figure 2 The present invention is a flow chart showing a method for determining a remaining life according to an exemplary embodiment.

[0053] Figure 3 The present invention is a flowchart of a method for determining a feature indicator set according to an exemplary embodiment.

[0054] Figure 4 The present invention is a flowchart showing a method for determining a remaining sub-lifetime according to an exemplary embodiment.

[0055] Figure 5 The flowchart of another method for determining the remaining sub-lifetime is shown according to an exemplary embodiment.

[0056] Figure 6 The present invention is a flow chart showing another method for determining the remaining life of a component according to an exemplary embodiment.

[0057] Figure 7 It is a schematic diagram showing a prediction effect according to an exemplary embodiment.

[0058] Figure 8 is a schematic diagram showing another prediction effect according to an exemplary embodiment.

[0059] Fig. 9 The present invention is a block diagram of a device for determining the remaining life of a component according to an exemplary embodiment.

[0060] Fig.10 The present invention is a block diagram of a device for determining the remaining life of a component according to an exemplary embodiment. DETAILED DESCRIPTION

[0061] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0062] It should be noted that all actions of acquiring signals, information or data in the present disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the device is located and with the authorization given by the owner of the corresponding device.

[0063] For some components, performance degradation may occur during long-term use, which may lead to failure of the component. For example, motherboard testing is a key process in the manufacturing process of mobile phones, which can test the radio frequency performance of mobile phones. In the board testing process, the test equipment contacts the motherboard under test through probes for testing, and evaluates the performance, function, stability and safety of the motherboard based on the test results to prevent defective products from flowing into subsequent processes. However, the testing of the board testing equipment relies on the probe to be plugged in and out of the motherboard under test for a long time and many times, which inevitably causes the probe to wear. As the test equipment continues to run, the wear of the probe accumulates over a long period of time, causing the probe to eventually fail, resulting in the failure of multiple consecutive motherboard tests in production.

[0064] In the related technology, two methods can be used: post-maintenance and preventive maintenance. Among them, post-maintenance will inevitably cause production stagnation and other damages. Preventive maintenance needs to rely on the life statistics of a large number of failed components to formulate the service life of the component. If the current component has exceeded its service life, the component should be replaced. However, for the same component, the service life is different due to different operating conditions, making it difficult to accurately formulate the service life of a single component. Inaccurate service life formulation will cause the component to be replaced too early or too late, resulting in waste of cost or loss of production capacity.

[0065] In response to the above technical problems, the present invention provides a method, device, equipment and storage medium for determining the remaining life of a component. For a certain target component, a test data set for the target component can be obtained, and features can be extracted from the test data set to obtain feature indicators corresponding to multiple monitoring moments, thereby avoiding interference of a single test data with alarm prediction. The feature indicators are then processed. If the feature indicators in the feature indicator set meet the alarm conditions, the remaining life of the target component is determined based on the test data set, so that the remaining life of the target component can be warned in time before the target component fails, and then replaced before it fails, thereby avoiding failure of the target component and avoiding premature replacement of the target component.

[0066] Figure 1 is a flow chart of a method for determining the remaining life of a component according to an exemplary embodiment. Figure 1 As shown, the method can be applied to a server and may include the following steps.

[0067] In step S101, a test data set corresponding to a target component is obtained.

[0068] In this embodiment, the target component can be tested periodically or in real time to obtain its corresponding test data set, which can include multiple test data, and the test data can be data obtained by testing the target component at different test times. The target component can be tested continuously, so that the corresponding test data can be continuously obtained, so that the operating status of the target component can be obtained in time. Among them, the target component can be any component that will degrade in performance during use, for example, it can be a probe for testing radio frequency performance or a probe for testing current, and it can also be a nozzle, a screwdriver, a motor, a bearing, etc.

[0069] In step S102, feature extraction is performed on the test data set to obtain a feature index set, which includes feature indexes corresponding to multiple monitoring moments.

[0070] In this embodiment, multiple test data corresponding to the test data set can be obtained at different monitoring moments, and feature extraction can be performed on the multiple test data to obtain a feature index, that is, one monitoring moment corresponds to one feature index, so that multiple feature indexes can be obtained at multiple monitoring moments, and the multiple feature indexes can constitute a feature index set. That is, each feature index is obtained from multiple continuous test data. Based on the continuously acquired test data, feature extraction can be continuously performed on the test data set to obtain the corresponding feature index. Feature extraction of the test data set can obtain the changing trend of the test data, and can avoid directly using the test data to determine whether to calculate the remaining life. Among them, errors are inevitable in the testing process, and a single test data may have a large error, resulting in inaccurate judgment results obtained using it. After feature extraction of the test data set to obtain a feature index set, and then determine whether to calculate the remaining life, the judgment result can be more accurate.

[0071] In step S103, when the characteristic indicators in the characteristic indicator set meet the alarm condition, the remaining life of the target component is determined according to the test data set.

[0072] In this embodiment, when the characteristic indicators in the characteristic indicator set meet the alarm conditions, it is determined that the target component is at risk of failure, an alarm can be issued, and the remaining life of the target component can be determined based on the test data set. The alarm can make the user pay attention to the target component, and the determination of the remaining life of the target component can more intuitively determine how long the target component can be used, so as to reasonably arrange the use of the target component. In addition, maintenance operations can be arranged in advance according to the remaining life of the target component. For example, the aging time can be determined according to the remaining life, and according to the production line inspection plan, the nearest inspection before the failure time is selected to perform maintenance on the target component. In this way, the use time of the target component can be guaranteed as much as possible, avoiding the waste of resources caused by premature replacement, and it can also be replaced before it fails to avoid the failure of the target component.

[0073] In one possible implementation, the test data set may include test data subsets corresponding to multiple target test items, and a method for performing feature extraction on the test data set to obtain a feature indicator set may be: performing feature extraction on the test data subset corresponding to each target test item respectively to obtain a feature indicator set corresponding to each target test item.

[0074] In this embodiment, a plurality of target test items may be tested on the target component, and a corresponding test data subset may be obtained for each target test item. Feature extraction may be performed on the test data subset corresponding to each target test item to obtain a feature index set corresponding to each target test item, that is, a test data subset corresponding to a target test item obtains a feature index set corresponding thereto.

[0075] In a possible implementation, a method for acquiring a test data set corresponding to a target component may be: determining at least one target test item related to the target component; respectively acquiring a test data set corresponding to each target test item, wherein the test data set corresponding to any target test item is obtained by periodically testing the target component using a test method corresponding to the target test item.

[0076] In this embodiment, at least one target test item related to the performance of the target component can be screened out, and based on the test method corresponding to the at least one target test item, the target component is periodically tested to obtain a test data set corresponding to each target test item.

[0077] Figure 2 is a flow chart showing a method for determining the remaining life according to an exemplary embodiment. Figure 2As shown, in a possible implementation manner, when a characteristic indicator in a characteristic indicator set meets an alarm condition, determining the remaining life of a target component according to a test data set may include the following steps:

[0078] In step S201, for a characteristic indicator set corresponding to any target test item, when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm conditions, the remaining sub-life of the target component corresponding to any target test item is determined based on the test data subset corresponding to the any target test item.

[0079] In step S202, the remaining life of the target component is determined according to the remaining sub-life corresponding to the at least one target test item.

[0080] In this embodiment, based on the alarm condition, for the characteristic indicator set corresponding to any target test item, each characteristic indicator in the acquired characteristic indicator set can be judged to determine whether the characteristic indicator meets the alarm condition. If the characteristic indicators corresponding to at least two consecutive monitoring moments meet the alarm condition, it is determined that the target component is at risk of failure, an alarm can be issued, and the remaining sub-life of the target component corresponding to any target test item can be determined based on the test data subset corresponding to the any target test item. Among them, there may be more than one characteristic indicator set corresponding to the target test item that triggers the alarm and calculates the corresponding remaining life, and for different target test items, the corresponding test data sets are different, and the corresponding characteristic indicator sets are also different, so the remaining life obtained is also different, and the remaining life of the target component can be determined based on the remaining sub-life corresponding to the at least one target test item.

[0081] In a possible implementation, when the characteristic indicators in the characteristic indicator set meet the alarm conditions, the method for determining the remaining life of the target component according to the test data set may be: for the characteristic indicator set corresponding to any target test item, when there is a characteristic indicator that is less than the alarm threshold, the total number of alarms is obtained, wherein if there is a characteristic indicator that is less than the alarm threshold, an alarm is issued, that is, the total number of alarms is increased by 1, and when the total number of alarms is greater than the alarm number threshold, the remaining sub-life of the target component corresponding to any target test item is determined according to the test data subset corresponding to the any target test item; the remaining life of the target component is determined according to the remaining sub-life corresponding to the at least one target test item.

[0082] Figure 3 is a flow chart showing a method for determining a feature index set according to an exemplary embodiment. Figure 3As shown, in a possible implementation manner, feature extraction is performed on the test data subset corresponding to each target test item to obtain a feature indicator set corresponding to each target test item, which may include the following steps:

[0083] In step S301, for a test data subset corresponding to any target test item, the test data subset is extracted through a sliding window to obtain multiple groups of candidate test data, and a group of candidate test data is extracted at each monitoring moment.

[0084] In this embodiment, for a test data subset corresponding to any target test item, the test data subset can be extracted through a sliding window. The length of the sliding window can be determined according to actual conditions. For example, the sliding window may include 10 test data, that is, each time the sliding window slides, it includes 10 test data. The sliding window may be a non-overlapping sliding window or a partially overlapping sliding window. For example, for a non-overlapping sliding window, if there are currently 20 test data, the sliding window may include the first 10 test data at one time, and the sliding window may include the last 10 test data at another time. For a partially overlapping sliding window, if the number of overlaps is 9, the sliding window only slides the distance of one test data at a time. For example, the sliding window includes the 1st to 10th test data at a time, and after the sliding window moves once, it includes the 2nd to 11th test data.

[0085] Through the above sliding window, a group of candidate test data is obtained by sliding once at each monitoring moment, until the test data in the current test data subset is completely slid, and multiple groups of candidate test data can be obtained.

[0086] In step S302, feature extraction is performed on each group of candidate test data to obtain feature indicators corresponding to multiple monitoring moments.

[0087] In this embodiment, feature extraction can be performed for each set of candidate test data, wherein one monitoring time corresponds to one set of candidate test data, and one set of candidate test data can obtain a feature index. For any set of candidate test data, the root mean square of the set of candidate test data can be extracted to obtain the feature index corresponding to the monitoring time of the set of candidate test data. Optionally, the mean or kurtosis of the set of candidate test data can also be extracted as the feature index.

[0088] In a possible implementation, when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm condition, before determining the remaining sub-life of the target component corresponding to any target test item according to the test data subset corresponding to the any target test item, it can be determined whether the characteristic indicator corresponding to each monitoring moment satisfies the alarm condition. Wherein, it can be determined whether the characteristic indicator satisfies the alarm condition according to the alarm threshold, for example, if any characteristic indicator is less than the alarm threshold, it is determined that the characteristic indicator satisfies the alarm condition.

[0089] Among them, the method for judging whether any characteristic indicator meets the alarm condition can be specifically as follows: according to the characteristic indicator set corresponding to any target test item, determine the first characteristic indicator and alarm threshold corresponding to the current monitoring moment; when the first characteristic indicator is less than the alarm threshold, determine that the first characteristic indicator corresponding to the current monitoring moment meets the alarm condition.

[0090] In this embodiment, each monitoring moment corresponds to a characteristic indicator, and the corresponding first characteristic indicator can be obtained from the characteristic indicator set according to the current monitoring moment. The alarm threshold can also be determined according to the characteristic indicator set corresponding to any target test item.

[0091] In a possible implementation, the method for determining the alarm threshold may be: obtaining a second characteristic indicator corresponding to a plurality of consecutive historical monitoring moments before the current monitoring moment; and determining the alarm threshold according to the mean and standard deviation of the second characteristic indicator.

[0092] In this embodiment, the current monitoring time can be recorded as t k , take the historical feature length as N, where N can be 50, that is, obtain the second feature index corresponding to the 50 consecutive historical monitoring moments before the current monitoring moment. At this time, the second feature index can be X k-50:k-1 , calculate its mean(X k-50:k-1 ) and standard deviation std(X k-50:k-1 ), and then the value of the alarm threshold can be set to: mean(X k-50:k-1 )-3×std(X k-50:k-1 ). Wherein, k is a positive integer.

[0093] The characteristic index corresponding to the current detection time can be recorded as x k , if x k If the value is less than the alarm threshold, it is determined that the first characteristic indicator corresponding to the current monitoring time meets the alarm condition.

[0094] In a possible implementation, the interval alarm counter and the continuous alarm counter can be used to determine whether there are at least two consecutive monitoring moments corresponding to characteristic indicators that meet the alarm condition. For example, if the characteristic indicator corresponding to the current detection moment is x k If the value is lower than the alarm threshold, the interval alarm counter increases its count value l=k; otherwise, k Incorporate the test data set to obtain a new second feature index Xk-50:k and update the alarm threshold. k+1 At this moment, a new characteristic index and alarm threshold are obtained, and it is determined whether the new characteristic index is lower than the alarm threshold. If it is lower, an alarm will be issued. If t k and t k+1 If an alarm occurs at all times, the value of the continuous alarm counter A is increased. l = k + 1. With the continuous acquisition of test data, new characteristic indicators will be continuously obtained. The count of the continuous alarm counter can also continue to increase. l -A l-1 <2, an alarm is output and the remaining life of the target component is predicted.

[0095] Among them, the number of alarms of the interval alarm counter can also be counted to obtain the total number of alarms, and when the characteristic indicators corresponding to at least two consecutive monitoring moments meet the alarm conditions and the total number of alarms is greater than the alarm number threshold, the remaining sub-life of the target component corresponding to any target test item is determined based on the test data subset corresponding to any target test item.

[0096] Figure 4 is a flow chart showing a method for determining a remaining sub-lifetime according to an exemplary embodiment. Figure 4 As shown, in a possible implementation manner, determining the remaining sub-lifetime corresponding to any target test item of a target component according to the test data subset corresponding to the any target test item may include the following steps:

[0097] In step S401, a degradation model corresponding to a target component is determined, and a life prediction model is obtained based on the degradation model.

[0098] In this embodiment, a degradation model may be constructed according to the trend characteristics of the test data of the target component. For example, the degradation model may be:

[0099]

[0100] Among them, if the time when the remaining life calculation is triggered is t FPT , then t is t FPT A certain test time after that; s(t) represents the test data of the target component at the test time t; s0 is the initial test data of the target test item, which can be eliminated by difference; v is the degradation rate; σ is the degradation volatility intensity, which is a parameter describing the degradation volatility intensity of the target component; B(t) is the standard Brownian motion.

[0101] According to the degradation model, it can be seen that the remaining life of the target component at the test time t obeys the probability distribution of the following life prediction model:

[0102]

[0103] Among them, Γ is the test value threshold of the current target test item, π is the pi, is the estimated result of v, is the estimated result of σ, τ is any time, f RUL (τ) is the probability at any time. The time corresponding to the maximum probability is the remaining life.

[0104] In step S402, a plurality of test data obtained after at least two consecutive monitoring moments in a test data subset corresponding to any target test item is obtained to obtain a target data set.

[0105] In step S403, the remaining life of the target component is determined according to the remaining sub-life corresponding to at least one target test item.

[0106] In this embodiment, at the above-mentioned at least two consecutive monitoring moments, the corresponding characteristic indicators all meet the alarm conditions. At this time, an alarm can be issued and the remaining life can be calculated. By obtaining multiple test data obtained after at least two consecutive monitoring moments in the test data subset corresponding to any target test item, and using it as the target data set to predict the remaining life, the remaining life of the target component can be determined more accurately.

[0107] Figure 5 is a flow chart showing another method for determining the remaining sub-lifetime according to an exemplary embodiment. Figure 5 As shown, in a possible implementation manner, determining the remaining sub-lifetime corresponding to the target component for any target test item according to the target data set and the life prediction model may include the following steps:

[0108] In step S501, parameter estimation is performed on the target data set to obtain the degradation rate and degradation volatility intensity.

[0109] In this embodiment, a logarithmic maximum likelihood function may be constructed, and the maximum likelihood estimation of parameters of the target data set may be performed using the logarithmic maximum likelihood function to obtain estimation results of the degradation rate and the degradation volatility intensity.

[0110] Among them, the logarithmic maximum likelihood function can be:

[0111]

[0112] Among them, ΔS t For S t , Δt is the difference between the previous and next test times.

[0113] Through the logarithmic maximum likelihood function, the estimated results of the degradation rate and degradation volatility intensity can be obtained as follows:

[0114]

[0115] Where N is the length of the difference sequence.

[0116] In step S502, the degradation results corresponding to different test moments are determined according to the degradation rate and the degradation fluctuation intensity.

[0117] In this embodiment, based on the degradation rate and the intensity of degradation fluctuation, Kalman filtering can be used to estimate the degradation state of the target component to obtain the degradation results corresponding to different test times, which can be recorded as

[0118] In step S503, the degradation rate, the degradation volatility intensity and the degradation result are processed by the life prediction model to obtain the remaining sub-life of the target component corresponding to any target test item.

[0119] In this embodiment, the degradation rate, the degradation volatility intensity and the degradation result are input into the life prediction model, so as to obtain the remaining sub-life of the target component corresponding to any target test item.

[0120] In a possible implementation, the method for determining the remaining life of a target component according to the remaining sub-life corresponding to at least one target test item may be: determining the minimum value of the remaining sub-life corresponding to at least one target test item as the remaining life of the target component.

[0121] In this embodiment, for different target test items, the corresponding test data sets are different, and the corresponding feature indicator sets are also different, so the remaining life obtained is also different. The smallest remaining sub-life is taken as the remaining life of the target component. The aging time can be determined according to the remaining life, and according to the production line inspection plan, the inspection closest to the failure time is selected to perform maintenance on the target component, which can avoid failure of the target component during use.

[0122] By using the above method for determining the remaining life of the target component based on the test data set, the test data set is continuously updated in the process of continuously acquiring the test data, thereby continuously outputting the remaining life of the target component, and realizing the dynamic update of the remaining life of the target component, so that after the alarm, the remaining life of the target component can be always paid attention to. And there is no need to collect a large amount of failure data for the target component in advance to specify the service life, which can reduce costs.

[0123] Figure 6 FIG. 1 is a flow chart showing another method for determining the remaining life of a component according to an exemplary embodiment. Figure 6 As shown, in step S601, a target test item is determined.

[0124] In step S602, a test data subset corresponding to each target test item is obtained.

[0125] In step S603, feature index extraction is performed for the test data subset of each target test item.

[0126] In step S604, the status monitoring data is initialized, the interval alarm counter l is recorded as 0, and the continuous alarm counter A is recorded as l =0.

[0127] In step S605, second characteristic indicators corresponding to a plurality of consecutive historical monitoring moments are determined.

[0128] In step S606, an alarm threshold is determined.

[0129] In step S607, the current first characteristic index is determined.

[0130] In step S608, the alarm threshold is adaptively updated and it is determined whether to alarm.

[0131] In step S609, the interval alarm counter 1 is updated, and the continuous alarm counter A is updated. l .

[0132] In step S610, it is determined whether there is A l -A l-1 ≥2, if yes, execute step S611, if no, execute step S612.

[0133] In step S611, there is no need for remaining life prediction, and the process returns to step S608.

[0134] In step S612, a degradation model is constructed, and then a life prediction model is constructed.

[0135] In step S613, maximum likelihood estimation of parameters is performed on the target data set.

[0136] In step S614, a degradation result is obtained through Kalman filter state estimation.

[0137] In step S615, the remaining life is calculated.

[0138] In step S616, predictive maintenance of the target component is performed based on the remaining life.

[0139] Figure 7 is a schematic diagram showing a prediction effect according to an exemplary embodiment. Figure 8 FIG. 1 is a schematic diagram showing another prediction effect according to an exemplary embodiment. Taking the target component as the probe as an example, the result obtained is as follows: Figure 7 and Figure 8 As shown, among which, Figure 7 It can be seen that the method for determining the remaining life of components proposed in the present disclosure can issue an alarm for probe abnormalities more than 400 test cycles before the probe fails, and has the ability to keenly identify abnormal probe states. Figure 8 It can be seen that when the calculation of the remaining life is triggered, the remaining life determination method of the component proposed in the present invention can iteratively perform parameter estimation, state estimation and remaining life calculation based on the test data acquired in real time, and the output remaining life prediction result is constantly approaching the actual remaining life of the probe, thereby providing the decision-making information necessary for the predictive maintenance operation of the probe.

[0140] Fig. 9 is a block diagram of a device for determining the remaining life of a component according to an exemplary embodiment. Fig. 9 The device 900 for determining the remaining life of a component includes an acquisition module 901 , a feature extraction module 902 and a first determination module 903 .

[0141] The acquisition module 901 is configured to acquire a test data set corresponding to a target component;

[0142] The feature extraction module 902 is configured to extract features from the test data set to obtain a feature index set, wherein the feature index set includes feature indexes corresponding to multiple monitoring moments;

[0143] The first determination module 903 is configured to determine the remaining life of the target component according to the test data set when the characteristic indicator in the characteristic indicator set meets the alarm condition.

[0144] Optionally, the test data set includes test data subsets corresponding to multiple target test items;

[0145] The feature extraction module 902 includes:

[0146] The feature extraction submodule is configured to perform feature extraction on the test data subset corresponding to each target test item to obtain a feature index set corresponding to each target test item;

[0147] The first determining module 903 includes:

[0148] A first determination submodule is configured to determine, for a characteristic indicator set corresponding to any target test item, a remaining sub-life of the target component corresponding to the any target test item according to a test data subset corresponding to the any target test item when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy an alarm condition;

[0149] The second determination submodule is configured to determine the remaining life of the target component according to the remaining sub-life corresponding to at least one target test item.

[0150] Optionally, the feature extraction submodule includes:

[0151] A first obtaining unit is configured to extract a test data subset corresponding to any target test item through a sliding window to obtain multiple groups of candidate test data, and extract a group of candidate test data at each monitoring moment;

[0152] The feature extraction unit is configured to extract features from each set of candidate test data to obtain feature indicators corresponding to multiple monitoring moments.

[0153] Optionally, the device 900 for determining the remaining life of a component further includes:

[0154] A second determination module is configured to determine a first characteristic indicator and an alarm threshold corresponding to a current monitoring moment according to a characteristic indicator set corresponding to any target test item;

[0155] The third determination module is configured to determine that the first characteristic indicator corresponding to the current monitoring moment meets the alarm condition when the first characteristic indicator is less than the alarm threshold.

[0156] Optionally, the second determining module includes:

[0157] A first acquisition submodule is configured to acquire a first characteristic indicator corresponding to a current monitoring moment from a characteristic indicator set corresponding to any target test item;

[0158] A second acquisition submodule is configured to acquire second characteristic indicators corresponding to a plurality of consecutive historical monitoring moments before the current monitoring moment;

[0159] The third determination submodule is configured to determine the alarm threshold according to the mean and standard deviation of the second characteristic indicator.

[0160] Optionally, the first determining submodule includes:

[0161] A second obtaining unit is configured to determine a degradation model corresponding to the target component and obtain a life prediction model based on the degradation model;

[0162] A third obtaining unit is configured to obtain a plurality of test data obtained after the at least two consecutive monitoring moments in the test data subset corresponding to any target test item to obtain a target data set;

[0163] The first determination unit is configured to determine the remaining sub-lifetime of the target component corresponding to any target test item according to the target data set and the life prediction model.

[0164] Optionally, the first determining unit includes:

[0165] A first obtaining subunit is configured to perform parameter estimation on the target data set to obtain a degradation rate and a degradation volatility intensity;

[0166] A determination subunit, configured to determine degradation results corresponding to different test moments according to the degradation rate and the degradation volatility intensity;

[0167] The second obtaining subunit is configured to process the degradation rate, the degradation volatility intensity and the degradation result through the life prediction model to obtain the remaining sub-life corresponding to the target component for any target test item.

[0168] Optionally, the acquisition module 901 includes:

[0169] A fourth determination submodule is configured to determine at least one target test item related to the target component;

[0170] The third acquisition submodule is configured to respectively acquire a test data set corresponding to each target test item. The test data set corresponding to any target test item is obtained by periodically testing the target component using the test method corresponding to the target test item.

[0171] Optionally, the second determining submodule includes:

[0172] The second determining unit is configured to determine the minimum value of the remaining sub-lifetimes corresponding to the at least one target test item as the remaining lifetime of the target component.

[0173] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0174] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the method for determining the remaining life of a component provided by the present disclosure.

[0175] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program executable by a programmable device. The computer program has a code portion for executing the above-mentioned method for determining the remaining lifetime of a component when executed by the programmable device.

[0176] Fig.10 1 is a block diagram of a device for determining the remaining life of a component according to an exemplary embodiment. For example, the device 1000 for determining the remaining life of a component can be provided as a server. Fig.10 The apparatus 1000 for determining the remaining life of a component includes a processing component 1022, which further includes one or more processors, and a memory resource represented by a memory 1032 for storing instructions executable by the processing component 1022, such as an application. The application stored in the memory 1032 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1022 is configured to execute instructions to perform the above-mentioned method for determining the remaining life of a component.

[0177] The device 1000 for determining the remaining life of a component may further include a power supply component 1026 configured to perform power management of the device 1000 for determining the remaining life of a component, a wired or wireless network interface 1050 configured to connect the device 1000 for determining the remaining life of a component to a network, and an input / output interface 1058. The device 1000 for determining the remaining life of a component may operate based on an operating system stored in the memory 1032, such as Windows Server 2000. TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or similar.

[0178] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0179] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining the remaining life of a component, characterized in that: include: Obtain the test data set corresponding to the target component; Extracting features from the test data set to obtain a feature index set, wherein the feature index set includes feature indexes corresponding to multiple monitoring moments; When the characteristic indicators in the characteristic indicator set meet the alarm condition, the remaining life of the target component is determined according to the test data set.

2. The method for determining the remaining life of a component according to claim 1, characterized in that: The test data set includes test data subsets corresponding to multiple target test items; The feature extraction of the test data set to obtain a feature index set includes: Perform feature extraction on the test data subset corresponding to each target test item to obtain a feature indicator set corresponding to each target test item; When the characteristic indicator in the characteristic indicator set meets the alarm condition, determining the remaining life of the target component according to the test data set includes: For a characteristic indicator set corresponding to any target test item, when the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm condition, determine the remaining sub-life of the target component corresponding to the any target test item according to the test data subset corresponding to the any target test item; The remaining life of the target component is determined according to the remaining sub-life corresponding to at least one target test item.

3. The method for determining the remaining life of a component according to claim 2, characterized in that: The feature extraction is performed on the test data subset corresponding to each target test item to obtain the feature indicator set corresponding to each target test item, including: For a test data subset corresponding to any target test item, the test data subset is extracted through a sliding window to obtain multiple groups of candidate test data, and a group of candidate test data is extracted at each monitoring moment; Feature extraction is performed on each set of candidate test data to obtain feature indicators corresponding to multiple monitoring moments.

4. The method for determining the remaining life of a component according to claim 2, characterized in that: When the characteristic indicators corresponding to at least two consecutive monitoring moments satisfy the alarm condition, before determining the remaining sub-life of the target component corresponding to any target test item according to the test data subset corresponding to the any target test item, the method further includes: Determine the first characteristic indicator and the alarm threshold value corresponding to the current monitoring moment according to the characteristic indicator set corresponding to any target test item; When the first characteristic indicator is less than the alarm threshold, it is determined that the first characteristic indicator corresponding to the current monitoring time satisfies the alarm condition.

5. The method for determining the remaining life of a component according to claim 4, characterized in that: The determining, according to the characteristic indicator set corresponding to any one of the target test items, the first characteristic indicator and the alarm threshold corresponding to the current monitoring moment includes: Acquire a first characteristic indicator corresponding to the current monitoring moment from the characteristic indicator set corresponding to any target test item; Obtaining second characteristic indicators corresponding to multiple consecutive historical monitoring moments before the current monitoring moment; The alarm threshold is determined according to the mean and standard deviation of the second characteristic indicator.

6. The method for determining the remaining life of a component according to claim 2, characterized in that: The determining, based on the test data subset corresponding to the any target test item, the remaining sub-lifetime of the target component corresponding to the any target test item includes: Determining a degradation model corresponding to the target component, and obtaining a life prediction model based on the degradation model; Acquire multiple test data acquired after the at least two consecutive monitoring moments in the test data subset corresponding to any target test item to obtain a target data set; The remaining sub-lifetime of the target component corresponding to any target test item is determined according to the target data set and the life prediction model.

7. The method for determining the remaining life of a component according to claim 6, characterized in that: The determining, according to the target data set and the life prediction model, the remaining sub-life of the target component corresponding to any target test item includes: Performing parameter estimation on the target data set to obtain a degradation rate and a degradation volatility intensity; Determining degradation results corresponding to different test moments according to the degradation rate and the degradation volatility intensity; The degradation rate, the degradation volatility intensity and the degradation result are processed by the life prediction model to obtain the remaining sub-life of the target component corresponding to any target test item.

8. The method for determining the remaining life of a component according to any one of claims 2 to 7, characterized in that: The step of obtaining a test data set corresponding to the target component includes: Determining at least one target test item related to the target component; A test data set corresponding to each target test item is obtained respectively. The test data set corresponding to any target test item is obtained by periodically testing the target component using the test method corresponding to the target test item.

9. The method for determining the remaining life of a component according to any one of claims 2 to 7, characterized in that: The determining the remaining life of the target component according to the remaining sub-life corresponding to at least one target test item includes: The minimum value among the remaining sub-lifetimes corresponding to the at least one target test item is determined as the remaining lifetime of the target component.

10. A device for determining the remaining life of a component, characterized in that: include: An acquisition module is configured to acquire a test data set corresponding to a target component; A feature extraction module is configured to extract features from the test data set to obtain a feature index set, wherein the feature index set includes feature indexes corresponding to multiple monitoring moments; The first determination module is configured to determine the remaining life of the target component according to the test data set when the characteristic indicator in the characteristic indicator set meets the alarm condition.

11. A device for determining the remaining life of a component, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the steps of the method for determining the remaining life of a component as recited in any one of claims 1 to 9 when executed.

12. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method for determining the remaining life of a component described in any one of claims 1 to 9 are implemented.

Images