CN114239752B - Method, device, equipment and medium for constructing residual life prediction model of relay - Google Patents

Abstract

The embodiment of the present invention discloses a method, device, equipment and medium for constructing a relay remaining life prediction model. The method includes: obtaining the operating parameters of the relay within a period of time, that is, historical operating data; selecting different time windows from the historical operating data The data corresponding to the coefficient is used as the time window data corresponding to the coefficient of different time windows; for each time window data, the preset quality evaluation rules are used to calculate the data quality; the time window data with the best data quality is used as the optimal data set According to the optimal data set, the performance degradation model corresponding to the relay is constructed to obtain the remaining life prediction model of the relay. Therefore, the embodiment of the present invention is based on the extraction of the best-quality time window data, which avoids the need to use all historical operating data in the process of constructing the relay remaining life prediction model, improves the construction speed of the relay remaining life prediction model, and ensures the accuracy of remaining life prediction.

Description

Construction method, device, equipment and medium of relay remaining life prediction model

technical field

The invention relates to the field of life estimation, in particular to a method, device, equipment and medium for constructing a relay remaining life prediction model.

Background technique

As an indispensable component in the automatic control system of the subway, the reliability of the relay directly affects the safety of the subway operation, making the research on the reliability of the relay gradually become the focus of the academic circles, and one of the important research contents of the reliability is the reliability of the relay. life expectancy.

In the existing research on the remaining life prediction of relays, most scholars use all the historical data of relays to construct the life prediction model of relays. However, when the data quality of the historical data of the relay is poor, it will lead to the problems of low prediction accuracy and slow operation speed in the life prediction model of the relay.

Contents of the invention

In view of this, the present invention provides a method, device, equipment and medium for constructing a relay remaining life prediction model, so as to improve the current situation of low prediction accuracy and slow operation speed of the relay life prediction model.

In the first aspect, an embodiment of the present invention provides a method for constructing a relay remaining life prediction model, including:

Obtain the historical operation data of the relay;

Based on a preset number of time window coefficients, the time window data corresponding to each of the time window coefficients is obtained according to the historical operation data, wherein the time window coefficient represents the proportion of the corresponding time window data to the historical operation data ;

Obtaining the data quality of each time window data by using preset quality evaluation rules;

Taking the time window data with the best data quality as the optimal data set;

A performance degradation model corresponding to the relay is constructed according to the optimal data set, and a remaining life prediction model of the relay is obtained.

Optionally, in an implementation manner provided by an embodiment of the present invention, the historical operating data includes a preset number of operating parameters of the relay at each action;

The use of preset quality evaluation rules to obtain the data quality of each of the time window data includes:

For each operation parameter in the time window data, calculate the time series correlation index between each operation parameter and the number of actions based on a preset first preset formula, and calculate each operation based on a second preset formula The monotonicity index of the parameter under the number of actions;

calculating a data quality value corresponding to each of the operating parameters according to the time series correlation index and the monotonicity index of each of the operating parameters;

The time window data with the best data quality as the optimal data set includes:

The time window data containing the operating parameters with the largest data quality value is taken as the optimal data set.

Further, in an implementation manner provided by an embodiment of the present invention, the first preset formula includes:

In the formula, Corr() represents the timing correlation index, X _j represents the parameter value of the jth operating parameter, k represents the number of actions of the relay, T represents the time matrix composed of t _k , and t _k represents the relay’s first At the time of k actions, X _jT (t _k ) represents the trend item of the jth operating parameter at time t _k ;

The second preset formula includes:

In the formula, Mon() represents the monotonicity index, and δ represents the unit step function.

Optionally, in an implementation manner provided by an embodiment of the present invention, the acquiring the historical operation data of the relay includes:

Obtain the original historical operation data of the relay;

Preset wavelet noise reduction processing is performed on the original historical operating data to obtain historical operating data.

Optionally, in an implementation manner provided by an embodiment of the present invention, the data in the historical operation data are sorted in chronological order;

The time window data corresponding to each of the time window coefficients is obtained according to the historical operation data based on the preset number of time window coefficients, including:

Based on the preset number of time window coefficients, the data corresponding to each of the time window coefficients is extracted from the historical operation data in a reverse order;

The data corresponding to each of the time window coefficients is sorted in chronological order to obtain the time window data corresponding to each of the time window coefficients.

Optionally, in an implementation manner provided by an embodiment of the present invention, the historical operating data includes a preset number of operating parameters corresponding to each contact unit;

The step of constructing a performance degradation model corresponding to the relay according to the optimal data set to obtain a remaining life prediction model of the relay includes:

According to the preset number of operating parameters corresponding to each contact unit in the optimal data set, a Wiener degradation model corresponding to the relay is constructed to obtain a remaining life prediction model of the relay.

Further, in an implementation manner provided by an embodiment of the present invention, the contact unit includes at least one normally closed contact and at least one normally open contact, and the preset number of operating parameters includes normally open contact resistance , normally closed contact resistance, pull-in time and release time.

In the second aspect, an embodiment of the present invention provides a device for constructing a relay remaining life prediction model, including:

The obtaining module is used to obtain the historical operation data of the relay;

The window data acquisition module is used to obtain the time window data corresponding to each of the time window coefficients according to the historical operation data based on the preset number of time window coefficients, wherein the time window coefficients indicate that the corresponding time window data occupies The proportion of the historical operating data;

An evaluation module, configured to use preset quality evaluation rules to obtain the data quality of each of the time window data;

Selecting a module for using the time window data with the best data quality as the optimal data set;

A modeling module, configured to construct a performance degradation model corresponding to the relay according to the optimal data set, and obtain a prediction model of the remaining life of the relay.

In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, the memory stores a computer program, and when the computer program runs on the processor, it executes the relay remaining life prediction model disclosed in any one of the first aspect The construction method.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program runs on a processor, it executes the remaining life of the relay as disclosed in any one of the first aspect. How to build predictive models.

In the construction method of the relay remaining life prediction model provided by the embodiment of the present invention, the operating parameters of the relay within a certain period of time, that is, the historical operating data, are first obtained; then from the historical operating data, data of different proportions are selected as different time window coefficients The corresponding time window data; then, for each time window data, use the preset quality evaluation rules to calculate the data quality; then use the time window data with the best data quality as the optimal data set, and construct The performance degradation model corresponding to the relay is used to obtain the remaining life prediction model of the relay, which avoids the need to use all historical operating data in the construction process of the remaining life prediction model of the relay.

Therefore, the embodiment of the present invention is based on the extraction of the best quality time window data, which avoids the need to use all historical data in the process of constructing the remaining life prediction model of the relay, and improves the construction speed of the remaining life prediction model of the relay. Not only that, since the data for constructing the relay remaining life prediction model is the time window data with the best data quality among all time window data, the relay remaining life prediction model can be supported by effective data, thereby ensuring the accuracy of the relay remaining life prediction sex.

Description of drawings

In order to illustrate the technical solution of the present invention more clearly, the following drawings will be briefly introduced in the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as It is regarded as limiting the protection scope of the present invention. In the respective drawings, similar components are given similar reference numerals.

FIG. 1 shows a schematic flowchart of a method for constructing a first relay remaining life prediction model provided by an embodiment of the present invention;

FIG. 2 shows a schematic flowchart of a second method for constructing a relay remaining life prediction model provided by an embodiment of the present invention;

FIG. 3 shows a schematic flowchart of a method for constructing a third relay remaining life prediction model provided by an embodiment of the present invention;

FIG. 4 shows a schematic structural diagram of a device for constructing a relay remaining life prediction model provided by an embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention.

The components of the embodiments of the invention generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

Hereinafter, the terms "comprising", "having" and their cognates that may be used in various embodiments of the present invention are only intended to represent specific features, numbers, steps, operations, elements, components or combinations of the foregoing, And it should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing or adding one or more features, numbers, steps, operations, elements, components or a combination of the foregoing possibilities.

In addition, the terms "first", "second", "third", etc. are only used for distinguishing descriptions, and should not be construed as indicating or implying relative importance.

Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which various embodiments of the present invention belong. The terms (such as those defined in commonly used dictionaries) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or an overly formal meaning, Unless clearly defined in various embodiments of the present invention.

Referring to FIG. 1, FIG. 1 shows a schematic flowchart of the construction of the first relay remaining life prediction model provided by the embodiment of the present invention, that is, the construction of the relay remaining life prediction model provided by the embodiment of the present invention includes:

S110, acquiring historical operation data of the relay.

It can be understood that the historical operation data represents the working status of the relay over a period of time, and can indicate the state change of the relay, so the remaining life prediction model of the relay can be constructed based on the historical operation data.

It can also be understood that the data components of the historical operation data can be set according to actual conditions. For example, in a feasible manner, the historical operation data includes the coil voltage and coil resistance corresponding to the contacts.

In another feasible manner, the historical operation data includes normally open contact resistance, normally closed contact resistance, pull-in time and release time of the relay at each action.

It should be noted that the embodiment of the present invention does not limit the specific acquisition method of the historical operation data of the relay, and the specific acquisition method of the relay can be selected according to the actual situation. For example, in a feasible way, the historical operation data can be obtained by installing detection equipment in the relay to monitor and collect the data of the relay in real time.

In addition, it is not difficult to understand that the acquired historical operation data may include noise. Therefore, in order to improve the quality of historical operating data, in an implementation mode provided by an embodiment of the present invention, specific reference may be made to FIG. 2 , which shows the construction of a second relay remaining life prediction model provided by an embodiment of the present invention. The schematic flow chart of the method, that is, S110 in this implementation mode includes:

S111, obtaining the original historical operation data of the relay;

S112. Perform preset wavelet noise reduction processing on the original historical operating data to obtain historical operating data.

That is, after the original historical operation data is obtained, noise reduction processing is performed on the historical operation data to remove the noise data contained in the data, so as to obtain a data set with a relatively obvious trend of quantity change.

The process of wavelet denoising processing mentioned in the embodiment of the present invention may include: according to the characteristics of different intensity distributions corresponding to wavelet decomposition coefficients of noise and signals in different frequency bands, removing the wavelet decomposition coefficients corresponding to noise in each frequency band, to The wavelet decomposition coefficient of the original signal is retained, and then the wavelet reconstruction operation is performed on the wavelet decomposition coefficient of the retained original signal to obtain a pure signal, thereby obtaining historical operating data.

S120, based on a preset number of time window coefficients, according to the historical operating data, obtain the time window data corresponding to each of the time window coefficients, wherein the time window coefficient indicates that the corresponding time window data accounts for the historical operating data proportion.

That is, according to the size of each time window coefficient in the embodiment of the present invention, a corresponding proportion of data is extracted from historical operating data.

It can be understood that the time window coefficient can be set according to the actual situation. For example, in a feasible manner, the preset number of time window coefficients includes the first time window coefficient with a size of 0.1, the second time window coefficient with a size of 0.5, and The third time window coefficient with a size of 1, and then the process of obtaining the corresponding time window data according to the first time window coefficient, the second time window coefficient and the third time window coefficient is: 0.1 in the historical operating data, that is, 10% The data of the first time window coefficient is used as the time window data corresponding to the first time window coefficient, 50% of the data in the historical operation data is used as the time window data corresponding to the second time window coefficient, and 100% of the historical operation data, that is, all data is used as the third time window data. The time window data corresponding to the time window coefficient.

In another feasible manner, the preset number of time window coefficients is a set whose set elements are [0.1, 0.2, 0.3, . . . , 1], and the difference between the sizes of two adjacent time window coefficients is 0.1.

Optionally, in order to better reflect the data change trend in the time window data and improve the prediction accuracy of the remaining life prediction model of the relay, in a feasible implementation manner, the data in the historical operation data are chronologically order sorting;

The S120 includes:

Based on the preset number of time window coefficients, the data corresponding to each of the time window coefficients is extracted from the historical operation data in a reverse order;

The data corresponding to each of the time window coefficients is sorted in chronological order to obtain the time window data corresponding to each of the time window coefficients.

That is, in this embodiment, the data corresponding to each time window coefficient will be extracted from the historical operation data in a reverse order, and the data corresponding to each time window coefficient will be sorted in forward order to obtain the time window data.

Exemplarily, when the size of a time window coefficient is 0.5 and the amount of historical operating data is 1000, when extracting the time window data, it will start from the last data of the historical operating data, that is, the 1000th data, and start from the Extract 500 data forward, and then sort the extracted 500 data in chronological order, so as to obtain the time window data corresponding to the time window coefficient of 0.5, that is, the 501-1000th data in the historical operation data as the time window data.

It should be understood that obtaining the time window data through the extraction method provided in this implementation manner can relatively effectively reflect future changes in the data.

For example, if the 1-500th data in the historical operation data is used as the time window data, and the remaining life prediction model of the relay is obtained based on the time window data, even if the relay remaining life prediction model can accurately predict the 501-1000th follow-up There is no guarantee that the remaining life prediction model of the relay can accurately predict the future changes of the data. And if there is an error between the predicted 501-1000th follow-up data and the actual 501-1000th data, the error of the future change of the data predicted by the remaining life prediction model of the relay will be even greater.

Moreover, if the remaining life prediction model of the relay cannot accurately predict the 501-1000th follow-up data, it indicates that the change trend of the first 500 data and the last 500 data of the relay is inconsistent, and the future change trend of the data cannot be predicted.

Therefore, the embodiment of the present invention completes the extraction of time window data based on this implementation, so that the relay remaining life prediction model can predict the future changes of the data, and reduces the amount of data required to build the relay remaining life prediction model, improving the Model training efficiency.

S130. Obtain the data quality of each time window data by using a preset quality evaluation rule.

That is, the embodiment of the present invention will calculate the data quality of each time window data to determine the pros and cons of historical operating data under different time window coefficients.

It can be understood that the specific calculation process of the data quality of each time window data can be set according to the actual situation, for example, in a feasible manner, the calculation process of the data quality of each time window data includes: for each time window data Carry out the preset data cleaning operation, and calculate the data retention rate according to the data volume of the time window data after cleaning and the data volume of the time window data before cleaning, that is, to obtain the data quality of each time window data.

S140, taking the time window data with the best data quality as an optimal data set.

That is, after obtaining the quality of each time window data, use the time window data with the best quality, that is, the optimal data set to obtain the remaining life prediction model of the relay.

If the size of the time window coefficient corresponding to the time window data is not greater than 1, the amount of data required to obtain the remaining life prediction model of the relay will be reduced, thereby improving the generation efficiency of the model.

Moreover, because the data quality in the optimal data set is the best, it can provide effective data support for the construction process of the relay remaining life prediction model and ensure the effective training of the relay remaining life prediction model.

S150. Construct a performance degradation model corresponding to the relay according to the optimal data set, and obtain a remaining life prediction model of the relay.

That is, the performance degradation model of the relay is constructed according to the data in the optimal data set, that is, the model expression that the performance of the relay gradually degrades with time or the number of actions, so as to realize the prediction of the remaining life of the relay and obtain the prediction model of the remaining life of the relay.

It can be understood that the construction process of the performance degradation model corresponding to the relay is relatively complete, that is, the construction process of the performance degradation model corresponding to the relay can be set as required. For example, in a feasible way, the construction process of the performance degradation model includes: inputting the optimal data set into the preset neural network model, and training the neural network model to obtain the trained performance degradation model, that is, the relay residual Life Prediction Model.

In a feasible implementation manner, the historical operating data includes a preset number of operating parameters corresponding to each contact unit;

Further, the S150 includes:

According to the preset number of operating parameters corresponding to each contact unit in the optimal data set, a Wiener degradation model corresponding to the relay is constructed to obtain a remaining life prediction model of the relay.

That is to say, the historical operating state of the relay in this embodiment is described by a preset number of operating parameters corresponding to each contact unit. Further, in the embodiment of the present invention, a corresponding Wiener degradation model is constructed based on a preset number of operating parameters corresponding to each contact unit, so as to obtain a relay remaining life prediction model.

It should be understood that the Wiener degradation model is often used in the remaining life prediction research, which can effectively describe the state changes of equipment or devices.

It also needs to be understood that the working status of the relay is actually related to the working status of any contact. Failure of any contact will cause the relay to fail to work normally, and then the remaining life of the relay can be understood as all contacts of the relay. Among the contact unit life corresponding to the unit, the minimum contact unit life.

It can be understood that the data components of the data in the historical operation data can be set according to the actual situation. For example, in the aforementioned feasible manner, the historical operation data includes the coil voltage and coil resistance corresponding to the contact.

In another feasible manner, the contact unit includes at least one normally closed contact and at least one normally open contact, and the preset number of operating parameters includes normally open contact resistance, normally closed contact resistance, Pick-up time and release time.

Among them, the pull-in time of the relay refers to the time required from the time when the coil of the relay is energized until all the contacts of the relay reach the working state. The release time of a relay is defined as the time required for all contacts to transition to the released state from the moment the coil is de-energized. The pull-in time and release time of the relay can effectively express the switching speed of the working state of the relay, and then can effectively characterize the aging condition of the relay.

Both the normally open contact resistance and the normally closed contact resistance change due to the degree of oxidation and roughness of the contact surface, and the degree of oxidation and roughness are related to the service time and number of actions of the relay, and then the resistance of the normally open contact and the normal The closed contact resistance can express the working condition of the relay, so the remaining life of the relay can be calculated according to the normally open contact resistance and the normally closed contact resistance.

Therefore, the embodiment of the present invention completes the construction of the remaining life prediction model of the relay based on the pull-in time, the release time, the normally open contact resistance and the normally closed contact resistance, so that the remaining life prediction model of the relay can accurately predict the remaining life of the relay.

In the construction method of the relay remaining life prediction model provided by the embodiment of the present invention, the operating parameters of the relay within a certain period of time, that is, the historical operating data, are first obtained; then from the historical operating data, data of different proportions are selected as different time window coefficients The corresponding time window data; then, for each time window data, use the preset quality evaluation rules to calculate the data quality; then use the time window data with the best data quality as the optimal data set, and construct The performance degradation model corresponding to the relay is used to obtain the remaining life prediction model of the relay, which avoids the need to use all historical operating data in the construction process of the remaining life prediction model of the relay.

Therefore, the embodiment of the present invention is based on the extraction of the best quality time window data, which avoids the need to use all historical data in the process of constructing the remaining life prediction model of the relay, and improves the construction speed of the remaining life prediction model of the relay. Not only that, since the data for constructing the relay remaining life prediction model is the time window data with the best data quality among all time window data, the relay remaining life prediction model can be supported by effective data, thereby ensuring the accuracy of the relay remaining life prediction sex.

Optionally, in an implementation manner provided by an embodiment of the present invention, please refer to FIG. 3 for details. FIG. 3 shows a schematic flowchart of a method for constructing a third relay remaining life prediction model provided by an embodiment of the present invention, that is The historical operating data in this embodiment includes a preset number of operating parameters of the relay at each action;

Further, the S130 includes:

S131, for each operating parameter in the time window data, calculate the time series correlation index between each operating parameter and the number of actions based on a preset first preset formula, and calculate each of the operating parameters based on a second preset formula The monotonicity index of above-mentioned operation parameter under described action number of times;

S132. Calculate a data quality value corresponding to each of the operating parameters according to the time series correlation index and the monotonicity index of each of the operating parameters;

Further, the S140 includes:

S141, taking the time window data including the operating parameter with the largest data quality value as an optimal data set.

It can be understood that both the historical operating data and the time window data obtained according to the historical operating data in this embodiment are composed of a preset number of operating parameters.

It can also be understood that the preset number of operating parameters can be set according to actual conditions. For example, in a feasible manner, the preset number of operating parameters includes the coil current corresponding to each contact. In another feasible manner, the preset number of operating parameters include normally open contact resistance, normally closed contact resistance, pull-in time and release time.

It is not difficult to understand that the assessment of data quality includes not only the assessment of data integrity, standardization and consistency, but also the assessment of the relevance of each data. Therefore, the embodiment of the present invention evaluates the relevance of data based on the time-series correlation index between each operating parameter and the number of actions, and the monotonicity index of each operating parameter under the number of actions, and then evaluates the quality of the data.

Optionally, the calculation methods of the time series correlation index and the monotonicity index, that is, the first preset formula and the second preset formula can be set/selected according to the actual situation, as in an implementation mode provided by an embodiment of the present invention , the first preset formula includes:

In the formula, Corr() represents the timing correlation index, X _j represents the parameter value of the jth operating parameter, k represents the number of actions of the relay, T represents the time matrix composed of t _k , and t _k represents the relay’s first At the time of k actions, X _jT (t _k ) represents the trend item of the jth operating parameter at time t _k ;

The second preset formula includes:

In the formula, Mon() represents the monotonicity index, and δ represents the unit step function.

Optionally, the process of calculating data quality according to the time series correlation index and monotonicity index can be set according to actual needs, for example, in a feasible manner, after obtaining the time series correlation index and monotonicity index of each operating parameter, each The data quality of each operating parameter is the sum of the corresponding time series correlation index and the corresponding monotonicity index.

In another feasible manner, the process of calculating the data quality according to the time series correlation index and the monotonicity index includes: multiplying the time series correlation index of each operating parameter by a preset first weight to obtain the first data; The monotonicity index of each operating parameter is multiplied by a preset second weight to obtain the second data; the first data and the second data are added together to obtain the data quality of the operating parameter.

Therefore, the embodiment of the present invention determines the correlation between each operating parameter and the number of actions in each time window data based on the time series correlation index and monotonicity index of each operating parameter. It is not difficult to understand that the remaining life of the relay can be understood as the minimum number of times that the relay can operate normally under preset conditions, and then it can be known that if the data can effectively reflect the relationship between the relay and the number of operations, the remaining life prediction model of the relay The higher the prediction accuracy is.

Therefore, in the embodiment of the present invention, based on the timing correlation index and monotonicity index of each operating parameter, the time window data with the strongest correlation with the number of actions of the relay, that is, the best data quality is selected to construct the remaining life prediction model of the relay , so as to ensure the prediction accuracy of the relay remaining life prediction model.

In addition, in order to better illustrate the process of obtaining the relay remaining life prediction model, the embodiment of the present invention also provides the derivation process of the relay remaining life prediction model, as follows:

Assuming that the historical operating data includes a preset number of operating parameters corresponding to each contact unit, the preset number of operating parameters include normally open contact resistance, normally closed contact resistance, pull-in time and release time, and assume that Z _j ( t) is the degradation amount of the relay at time t, then the degradation process of the relay can be expressed as:

Z _j (t)＝Z _j (0)+μ _j t+σ _j B _j (t) (1)

Among them, Z _j (0) represents the initial degradation of the j-th operating parameter; μ _j represents the diffusion rate of the j-th operating parameter, that is, the drift parameter; σ _j represents the diffusion parameter of the j-th operating parameter, and is a constant; B _j (t) means that the jth operating parameter follows the standard Brownian motion, which characterizes the dynamic characteristics of the degradation process of the jth operating parameter, and B _j (t) obeys a normal distribution with a mean value of 0 and a variance of t, that is, B _j (t)～N(0,t).

For the degradation process described by formula (1), let the preset failure threshold corresponding to the jth operating parameter be the failure threshold of the relay as D _j , and the life value corresponding to the jth operating parameter is T _j , that is, the degradation amount Z The time when _j (t) reaches the failure threshold D _j for the first time, and the definition formula of the life T _j corresponding to the jth operating parameter is:

T _j ＝inf{t|Z _j (t)＝D _j ,t≥0} (2)

According to the formula (2), it can be seen that the life of the relay obeys the inverse Gaussian distribution, and then the cumulative failure probability function F _j (t) corresponding to the jth operating parameter is:

Therefore, the formula of the reliability R _j (t) corresponding to the jth operating parameter is:

R _j (t) = 1-F _j (t) (4)

It is understandable that the failure of the relay is a competitive failure, that is, the failure of any contact unit in the relay is equivalent to the failure of the relay, then the reliability R′ _i (t) of the i-th unit of the relay is the product of the reliability corresponding to all operating parameters ,Right now:

In the formula, R _ij (t) represents the reliability of the j-th operating parameter corresponding to the i-th unit at time t.

That is to say, the reliability R _j (t) corresponding to the four operating parameters, namely normally open contact resistance, normally closed contact resistance, pull-in time and release time are multiplied to obtain the reliability R of the i-th unit ' _i (t).

Furthermore, the relay is composed of multiple units, and the units are mostly connected in series, so the overall reliability R(t) of the relay is the product of the reliability of all units, and its expression is:

Wherein, n represents the number of contact units included in the relay.

Therefore, the expression of the failure cumulative failure probability function of the relay is:

F(t)＝1-R(t) (7)

Let t _k be the current moment, that is, the moment when the number of times the relay performs actions has accumulated to k, and t _j is the life value corresponding to the jth operating parameter, then the remaining life value l _jk of the jth operating parameter at the current time t _k The calculation formula is:

l _jk ＝t _j -t _jk (8)

Among them, t _jk represents the moment corresponding to the jth operating parameter when the number of relay actions reaches k;

According to the nature of the Wiener degradation process, combined with formula (1), it can be known that:

Y _j (l _k )＝Y _j (0)+μ _j l _jk +σ _j B _j (l _jk ) (9)

Among them, Y _j (l _k )=Z _j (l _k +t _k )-Z _j (t _k ); T _j (l _k ) represents the degradation amount corresponding to the remaining life value l _k when the number of actions reaches k ; Y _j (0) represents the degradation increment corresponding to the jth unit when the remaining life value l _k is 0, and Y _j (0)=0.

Therefore, the probability density function f(l _jk ) of the remaining life corresponding to the jth operating parameter at the current moment t _k is:

Among them, z _jk represents the degradation amount corresponding to the jth operating parameter at the current moment t _k .

Furthermore, it can be appreciated that each drift parameter and diffusion parameter needs to be estimated in order to accurately estimate the relevant unknown parameters in the degradation model and the model parameters in the Wiener process. That is to estimate the diffusion parameter σ _j and the drift parameter μ _j in formula (2).

It is understandable that for any given moment t+△t>t>0, the degradation increment between t+△t and t obeys a normal distribution, and then:

Among them, ΔZ _j (t) represents the degradation increment corresponding to the jth operating parameter at time t.

According to the nature of the Wiener process, each operating parameter has an independent amount of degradation, and the probability density function f(△Z _j ) of △Z _j (t) can be expressed as:

In the Wiener process, the likelihood function L(μ _j , σ _j ) of the drift parameter μ _j and the diffusion parameter σ _j is:

L(μ _j ,σ _j )=f(△Z _1j ,△Z _2j ,…△Z _nj )=f(△Z _1j )f(△Z _2j )…f(△Z _nj ) (13)

和扩散参数σ_j对应的极大似然估计量

分别为：Take the logarithm of the likelihood function, and derive the likelihood function after taking the logarithm, and obtain the maximum likelihood estimator corresponding to the drift parameter μ _j

The maximum likelihood estimator corresponding to the diffusion parameter σ _j

They are:

Among them, r represents the data volume of the data set for constructing the remaining life prediction model of the relay, that is, the data volume of the optimal data set; △Z _pj represents the degradation increment of the jth operating parameter; △t _pj represents the value of the jth operating parameter amount of time change.

To sum up, the probability density function f(l _jk ) of the remaining life corresponding to the jth operating parameter at the current moment t _k is expected, and the remaining life value E(l _jk of the jth operating parameter at the current moment t _k is obtained )for:

Among them, E(·) represents the expected value of the parameter.

It can be understood that the relay includes multiple contact units, and each contact unit corresponds to four operating parameters, namely normally open contact resistance, normally closed contact resistance, pull-in time and release time. Thus, one relay corresponds to the remaining life values of the 4 operating parameters. Furthermore, due to the existence of competition failures in the relay, that is, when any parameter of any contact unit of the relay exceeds the threshold value, the relay is considered to be invalid. Therefore, take the minimum remaining life value corresponding to each operating parameter as the remaining life value T ₁ of the relay, and the calculation formula is:

T ₁ =min{E ₁ , E ₂ , E ₃ ,...,E _M } (17)

Among them, min{ } represents the minimum value function, that is, the minimum value of the remaining life value of all operating parameters is taken, E _M represents the remaining life value of the Mth operating parameter, and M represents the number of parameters corresponding to the operating parameter of the relay.

The relay is composed of multiple contact units, each unit corresponds to 4 operating parameters, then the number of all operating parameters corresponding to a relay is:

M=4×m (18)

Among them, m represents the number of contact units contained in the relay.

Corresponding to the method for constructing the relay remaining life prediction model provided by the embodiment of the present invention, the embodiment of the present invention also provides a device for constructing the relay remaining life prediction model. Referring to FIG. 4, FIG. 4 shows that the embodiment of the present invention provides Schematic diagram of the structure of the device for constructing the remaining life prediction model of the relay. The device 200 for constructing the remaining life prediction model of the relay provided in the embodiment of the present invention includes:

An acquisition module 210, configured to acquire historical operating data of the relay;

The window data acquisition module 220 is configured to obtain time window data corresponding to each of the time window coefficients according to the historical operating data based on a preset number of time window coefficients, wherein the time window coefficients represent corresponding time window data The proportion of the historical operating data;

An evaluation module 230, configured to use preset quality evaluation rules to obtain the data quality of each of the time window data;

The selection module 240 is used to use the time window data with the best data quality as the optimal data set;

The modeling module 250 is configured to construct a performance degradation model corresponding to the relay according to the optimal data set, and obtain a prediction model of the remaining life of the relay.

Optionally, in an implementation manner provided by an embodiment of the present invention, the historical operating data includes a preset number of operating parameters of the relay at each action;

Furthermore, the evaluation module includes:

The index calculation sub-module is used to calculate the time series correlation index between each of the operating parameters and the number of actions based on the preset first preset formula for each of the operating parameters in the time window data, and based on the second preset The formula calculates the monotonicity index of each of the operating parameters under the number of actions;

The quality calculation submodule is used to calculate the data quality value corresponding to each of the operating parameters according to the time series correlation index and the monotonicity index of each of the operating parameters;

Furthermore, the selection module is further configured to use the time window data containing the operating parameters with the largest data quality value as the optimal data set.

Further, in an implementation manner provided by an embodiment of the present invention, the first preset formula includes:

In the formula, Corr() represents the timing correlation index, X _j represents the parameter value of the jth operating parameter, k represents the number of actions of the relay, T represents the time matrix composed of t _k , and t _k represents the relay’s first At the time of k actions, X _jT (t _k ) represents the trend item of the jth operating parameter at time t _k ;

The second preset formula includes:

In the formula, Mon() represents the monotonicity index, and δ represents the unit step function.

Optionally, in an implementation manner provided by an embodiment of the present invention, the acquisition module includes:

The raw data acquisition sub-module is used to obtain the original historical operation data of the relay;

The noise reduction sub-module is used to perform preset wavelet noise reduction processing on the original historical operation data to obtain the historical operation data.

Optionally, in an implementation manner provided by an embodiment of the present invention, the data in the historical operation data are sorted in chronological order;

Furthermore, the window data acquisition module includes:

The extraction sub-module is used to extract the data corresponding to each of the time window coefficients from the historical operation data in a reverse order based on the preset number of time window coefficients;

The sorting sub-module is configured to sort the data corresponding to each of the time window coefficients in chronological order to obtain the time window data corresponding to each of the time window coefficients.

Optionally, in an implementation manner provided by an embodiment of the present invention, the historical operating data includes a preset number of operating parameters corresponding to each contact unit;

Furthermore, the modeling module is further configured to construct a Wiener degradation model corresponding to the relay according to a preset number of operating parameters corresponding to each contact unit in the optimal data set, so as to obtain a remaining life prediction model of the relay.

Further, in an implementation manner provided by an embodiment of the present invention, the contact unit includes at least one normally closed contact and at least one normally open contact, and the preset number of operating parameters includes normally open contact resistance , normally closed contact resistance, pull-in time and release time.

The device for constructing the remaining life prediction model of the relay provided in the embodiment of the present application can realize the various processes of the construction method of the remaining life prediction model of the relay in the method embodiment disclosed in FIG. 1, and can achieve the same technical effect. To avoid repetition, it is not described here Let me repeat.

The embodiment of the present invention also provides a computer device, including a memory and a processor, the memory stores a computer program, and when the computer program runs on the processor, it executes the construction of the remaining life prediction model of the relay disclosed in the method embodiment corresponding to Figure 1 method.

The embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program runs on the processor, it executes the remaining life prediction model of the relay disclosed in the method embodiment corresponding to Figure 1 The construction method.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods may also be implemented in other ways. The device embodiments described above are only illustrative. For example, the flowcharts and structural diagrams in the accompanying drawings show the possible implementation architecture and functions of devices, methods and computer program products according to multiple embodiments of the present invention. and operation. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or part of code that includes one or more Executable instructions. It should also be noted that, in alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also to be noted that each block of the block diagrams and/or flow diagrams, and combinations of blocks in the block diagrams and/or flow diagrams, can be implemented by a dedicated hardware-based system that performs the specified function or action may be implemented, or may be implemented by a combination of special purpose hardware and computer instructions.

In addition, each functional module or unit in each embodiment of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a smart phone, a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes. .

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. Should be covered within the protection scope of the present invention.

Claims (8)

1. A method for constructing a residual life prediction model of a relay is characterized by comprising the following steps:

obtaining historical operation data of a relay, wherein the historical operation data comprises a preset number of operation parameters of the relay during each action;

obtaining time window data corresponding to each time window coefficient according to the historical operating data based on a preset number of time window coefficients, wherein the time window coefficients represent the proportion of the corresponding time window data in the historical operating data;

obtaining the data quality of each time window data by using a preset quality evaluation rule;

taking the time window data with the best data quality as an optimal data set;

constructing a performance degradation model corresponding to the relay according to the optimal data set to obtain a residual life prediction model of the relay;

the obtaining the data quality of each time window data by using a preset quality evaluation rule includes:

for the operation parameters in each time window data, calculating a time sequence correlation index of each operation parameter and the action times based on a preset first preset formula, and calculating a monotonicity index of each operation parameter under the action times based on a second preset formula, wherein the first preset formula comprises the following steps:

wherein Corr () represents a time-series correlation index, X _j A parameter value representing the jth operating parameter, k representing the number of actuations of the relay, and T representing the number of actuations represented by T _k The time matrix of composition, t _k Indicating the time of the kth action of the relay, X _jT (t _k ) Indicating the jth operating parameter at time t _k A trend term of;

the second preset formula includes:

in the formula, mon () represents a monotonicity index, and δ represents a unit step function;

calculating a data quality value corresponding to each operating parameter according to the time sequence correlation index and the monotonicity index of each operating parameter;

the taking the time window data with the best data quality as the optimal data set comprises the following steps:

and taking the time window data of the operating parameters with the maximum data quality values as an optimal data set.

2. The method for constructing the residual life prediction model of the relay according to claim 1, wherein the step of obtaining historical operation data of the relay comprises the following steps:

acquiring original historical operating data of the relay;

and carrying out preset wavelet denoising processing on the original historical operating data to obtain historical operating data.

3. The construction method of the relay residual life prediction model according to claim 1, characterized in that data in the historical operating data are sorted according to time sequence;

the obtaining of the time window data corresponding to each time window coefficient according to the historical operating data based on the preset number of time window coefficients includes:

extracting data corresponding to each time window coefficient from historical operating data in a reverse extraction mode based on a preset number of time window coefficients;

and sequencing the data corresponding to each time window coefficient according to the time sequence to obtain the time window data corresponding to each time window coefficient.

4. The method for constructing the residual life prediction model of the relay according to claim 1, wherein the historical operating data comprises a preset number of operating parameters corresponding to each contact unit;

the method for constructing the performance degradation model corresponding to the relay according to the optimal data set to obtain the residual life prediction model of the relay comprises the following steps:

and according to the preset number of operating parameters corresponding to each contact unit of the optimal data set, constructing a wiener degradation model corresponding to the relay to obtain a residual life prediction model of the relay.

5. The method for constructing a model for predicting the residual life of a relay according to claim 4, wherein the contact unit comprises at least one normally closed contact and at least one normally open contact, and the preset number of operating parameters comprise normally open contact resistance, normally closed contact resistance, pull-in time and release time.

6. A device for constructing a residual life prediction model of a relay is characterized by comprising the following components:

the relay control system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring historical operation data of a relay, and the historical operation data comprises a preset number of operation parameters of the relay during each action;

the window data acquisition module is used for acquiring time window data corresponding to each time window coefficient according to the historical operating data based on a preset number of time window coefficients, wherein the time window coefficients represent the proportion of the corresponding time window data in the historical operating data;

the evaluation module is used for obtaining the data quality of each time window data by using a preset quality evaluation rule;

the selection module is used for taking the time window data with the best data quality as an optimal data set;

the modeling module is used for constructing a performance degradation model corresponding to the relay according to the optimal data set to obtain a residual life prediction model of the relay;

the evaluation module comprises:

the index calculation submodule is used for calculating a time sequence correlation index of each operation parameter and action times based on a preset first preset formula and calculating a monotonicity index of each operation parameter under the action times based on a second preset formula aiming at the operation parameters in each time window data, wherein the first preset formula comprises the following steps:

wherein Corr () represents a time-series correlation index, X _j A parameter value representing the jth operating parameter, k representing the number of actuations of the relay, and T representing the number of actuations represented by T _k The time matrix of composition, t _k Indicates the time of the kth action of the relay, X _jT (t _k ) Indicating the j-th operating parameter at time t _k A trend term of;

the second preset formula includes:

where Mon () represents a monotonicity index, and δ represents a unit step function;

the quality calculation submodule is used for calculating a data quality value corresponding to each operating parameter according to the time sequence correlation index and the monotonicity index of each operating parameter;

the selection module is further used for taking the time window data containing the operation parameters with the maximum data quality values as the optimal data set.

7. A computer device, characterized by comprising a memory and a processor, the memory storing a computer program which, when run on the processor, performs the method of constructing a relay residual life prediction model according to any one of claims 1-5.

8. A computer-readable storage medium, having stored thereon a computer program which, when run on a processor, performs the method of constructing a relay residual life prediction model according to any one of claims 1-5.

Images