Method for judging abnormal abrasion of end face seal and computer program product

Abstract

The invention discloses a method for judging abnormal abrasion of an end face seal and a computer program product, and belongs to the field of abnormal sealing monitoring. A plurality of monitoring values such as the end face sealing compression amount, a temperature monitoring point, a leakage amount monitoring point and a pressure monitoring point in the end face sealing test system are used as characteristic values, the collected temperature parameter rising rate is used as a basis, and the fluctuation conditions of signals of the end face sealing compression amount, the leakage amount monitoring point and the pressure monitoring point are considered; according to the method, the current temperature rising condition is corrected, finally the current temperature rising condition is compared with a preset temperature rising speed threshold value for judgment, multiple analysis and judgment are conducted, the accuracy of temperature anomaly judgment is improved, and meanwhile the method is based on an infrared end face sealing measurement test device, and the accuracy of sealing face temperature rise measurement is improved.

Description

Technical Field

[0001] The present invention belongs to the field of seal fault diagnosis, and particularly relates to a method for judging abnormal wear of end face seals based on multi-parameter linear programming. Background Art

[0002] In the field of liquid rocket engine turbopumps, various forms of end face seals are widely used, such as contact seals and non-contact seals, etc. To monitor the stability of the seal operation state, temperature, flow rate, and power measurement points are often arranged in the seal test device, and the data is collected and stored through an automated system. Once the seal abnormally wears, it will cause serious damage to the seal surface of the product, resulting in seal failure, and in severe cases, it will cause damage to the operation test system.

[0003] The prior art often judges whether the seal is operating normally by manually or using equipment to determine the maximum threshold of temperature parameters. On the one hand, the setting of a single threshold does not exclude the influence of temperature rise caused by ambient temperature and excessive specific pressure of the product itself. On the other hand, this method insufficiently considers the additional influence of the compression amount of the end face seal product itself on the temperature rise. At the same time, the temperature rise caused by the wear of the end face seal mainly occurs on the surface of the seal contact pair. During the operation of the end face seal, if abnormal wear occurs, the contact situation between the stationary ring assembly and the rotating ring of the end face seal should be a friction state of rapid cyclic contact and separation from contact to separation and then to contact, which will cause fluctuations in power and leakage parameters. Therefore, the monitoring scheme using a single air temperature threshold in the cavity for judgment has disadvantages such as slow monitoring and early warning and inaccurate parameter judgment, and is not suitable for judging abnormal wear.

[0004] Therefore, we propose a method for judging abnormal wear of end face seals based on multi-parameter linear programming to solve some problems existing in the prior art. Summary of the Invention

[0005] The technical problem solved by this application is: overcoming the deficiencies of the prior art, providing a method for judging abnormal wear of end face seals based on multi-parameter linear programming. Based on the surface temperature of the end face seal obtained by infrared temperature measurement, by considering the product compression amount parameter, the fluctuation of the operating power parameter, and the fluctuation of the leakage amount parameter, it effectively reduces the influence of product specific pressure factors, product compression amount factors, ambient temperature, and untimely reflection of cavity temperature, which is beneficial to improving the accuracy of judging the temperature rise of the end face seal, discovering abnormal wear during the operation of the end face seal in advance, and protecting the product and the test system to a certain extent.

[0006] The technical solution provided by this application is as follows:

[0007] A method for judging abnormal wear of end face seals, comprising:

[0008] S1: Obtain data of the seal cavity during the normal operation of the end face seal x times;

[0009] From the data of the seal cavity during each normal operation of the end face seal, intercept the numerical values of multiple measurement parameters of the temperature rise signal segments corresponding to different time points. The multiple measurement parameters are the target parameter and n decision parameters. The target parameter is the temperature parameter, and x groups of samples are obtained. Each group of samples contains n + 1 measurement parameters, and each measurement parameter contains m data;

[0010] S2: According to the temperature parameter data in S1, calculate the temperature rise slope data of the temperature parameter x times; according to the x temperature rise slope data, calculate the average value of the temperature rise slope and the root mean square value of the temperature rise. Calculate the temperature rise slope judgment threshold [K] based on the average value of the temperature rise slope and the root mean square value of the temperature rise;

[0011] S3: According to the decision parameter data in S1, obtain the normalized decision parameter data matrix AN and the normalized target parameter matrix TN, and obtain the prediction weight coefficient matrix [n 1 n 2 ...n n and the normalized target parameter matrix TN'. The weight coefficient matrix should satisfy the constraint conditions.

[0012] S4: Process the data to be judged. The data to be judged includes the data of multiple measurement parameters. The multiple measurement parameters include the target parameter and n decision parameters. The target parameter is the temperature parameter, and each measurement parameter contains m data within the time window length Δt;

[0013] According to step S3 and the data to be judged, obtain the normalized temperature rise slope adjustment empirical value K 2 ', and perform anti-normalization calculation on the normalized temperature rise slope adjustment empirical value K 2 ' within the sample data range to obtain the temperature rise slope adjustment empirical value K2;

[0014] According to the m data included in the temperature parameter, calculate the current temperature rise slope K1;

[0015] S5: According to the temperature rise slope adjustment empirical value K2 and the current temperature rise slope K1, obtain the temperature rise slope judgment value K; when K is greater than the temperature rise slope judgment threshold [K], it is determined that the end face seal has abnormal wear, otherwise it is normal.

[0016] Preferably, the decision parameter is the operating power parameter, the seal product leakage parameter or the seal product assembly compression amount; the m data of the temperature parameter and the decision parameter are sequentially collected at different time points within the sampling time.

[0017] Preferably, in S1, the number of samples x ≥ 10 × (the number of items of measurement parameters ÷ the qualified rate of end face seal products).

[0018] Preferably, in S2, the time window length Δt is set as ±2.5 s around the time point with the maximum slope of the temperature rise data, and the data of the measurement parameters are taken within the time window length Δt.

[0019] Preferably, in S2, according to the temperature parameter data in S1, the temperature rise slope data of the temperature parameters are calculated x times; according to the x temperature rise slope data, the slope mean value and the root mean square value are calculated, and the temperature rise slope judgment threshold [K] is calculated according to the slope mean value and the root mean square value, including:

[0020] According to the temperature parameter data in S1, the temperature slope values of the temperature parameter data of each group of samples are calculated within the time window Δt, and the temperature slope value matrix K = [K 1 K 2 ...K x ' is obtained;

[0021] Then, according to the temperature slope value matrix K = [K 1 K 2 ...K x ', the average temperature rise slope and the root mean square value σ of the temperature parameter data of x groups of samples are obtained, and the temperature rise slope judgment threshold

[0022] Preferably, in S3, according to the decision parameter data in S1, the normalized decision parameter data matrix AN and the normalized target parameter matrix TN are obtained, and the prediction weight coefficient matrix [n 1 n 2 ...n n and the normalized target parameter matrix TN' are obtained according to the decision parameter data matrix AN and the normalized target parameter matrix TN. The weight coefficient matrix should satisfy the constraint conditions, including:

[0023] Among the n decision parameters, there are l predetermined parameters and k operation fluctuation parameters. The predetermined parameters are the parameters that do not change after the assembly of the sealing component, and the operation fluctuation parameters are the parameters that fluctuate with the operation; for the predetermined parameters and the operation fluctuation parameters expressed in matrix format, the predetermined parameter matrix of the yth group of samples is obtained and the coefficient of variation matrix of the operation fluctuation parameters

[0024] The coefficient of variation matrix of the operation fluctuation parameters of the yth group of samples and the predetermined parameter matrix are combined to obtain the decision parameter data matrix of the yth group of samples as Perform normalization on AN y within the range of 0 to 1 to obtain the normalized decision parameter data matrix AN;

[0025] According to the temperature parameter data, calculate to obtain a temperature slope value matrix K = [K 1 K 2 ...K y ', where K y represents the temperature slope of the y-th group of samples within the time window length Δt. After normalizing the temperature slope value matrix K in the range of 0 to 1, a normalized target parameter matrix TN = [TN 1 TN 2 ...TN y ' is obtained;

[0026] The weight coefficient matrix is [n 1 n 2 ...n n . Predict the normalized target parameter matrix TN' = AN * [n 1 n 2 ...n n '. The weight coefficient matrix should satisfy the constraint condition to minimize the difference between TN' and TN.

[0027] Preferably, the predetermined parameters and the operation fluctuation parameters are expressed in matrix format, and a predetermined parameter matrix and a coefficient of variation matrix of the operation fluctuation parameters are obtained, including:

[0028] For l predetermined parameters, for each of the x groups of data, among the m data within the time window length Δt, the predetermined matrix predetermined parameter matrix

[0029] For k operation fluctuation parameters, for each of the x groups of data, among the m time series within the time window length Δt, find the standard deviation matrix 1 a 2 ...a k of the operation fluctuation parameters a2 = [a and the mean matrix of the operation fluctuation parameters, where the m-sequence matrix of the g-th operation fluctuation parameter of the y-th group is The standard deviation matrix of the operation fluctuation parameters is The mean matrix of the operation fluctuation parameters is According to the quotient of the standard deviation matrix of the operation fluctuation parameters and the mean matrix of the operation fluctuation parameters, obtain the coefficient of variation matrix of the operation fluctuation parameters. The coefficient of variation matrix of the operation fluctuation parameters is:

[0030] Preferably, the weight coefficient matrix satisfies the constraint condition:

[0031] The error value is the smallest, i.e., o = min(TN' - TN);

[0032] n 1 n 2 ...n n They are all evenly distributed between 0 and 1;

[0033] The weights corresponding to l predetermined parameters are the same, i.e., n 1 、n 2 ……n l They are all equal;

[0034] The weights corresponding to k operating fluctuation parameters are the same, i.e., n l+1 ……n l+k They are all equal, n = l + k;

[0035] The sum of the weights corresponding to the predetermined parameters is equal to the sum of the weights corresponding to the operating fluctuation parameters.

[0036] Preferably, in S4, according to the steps of S3 and the data to be judged, the normalized temperature rise slope adjustment empirical value K 2 ' is obtained, including:

[0037] According to the decision parameter data of the steps of S3 and the data to be judged, a decision parameter data matrix is obtained, and the decision parameter data matrix is normalized within the sample data range to obtain a predicted normalized decision parameter matrix TN';

[0038] According to the predicted normalized target parameter matrix TN' in S3, combined with the temperature parameter data of the data to be judged, the normalized temperature rise slope adjustment empirical value K 2 ' is obtained, K 2 ' = AN i *[n 1 n 2 ...n n '.

[0039] A computer program product includes a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method described in any one of the above are implemented.

[0040] In summary, the present application at least includes the following beneficial technical effects:

[0041] By considering multiple parameters such as predetermined parameters such as product assembly parameters other than temperature parameters, operating fluctuation parameters such as product power and leakage, and fusing them, the current temperature rise situation of the end face seal is comprehensively judged, the temperature rise caused by the operation of the end face seal under actual different working conditions is obtained, whether the current state is worn is judged, the misjudgment of the product wear state caused by a single temperature parameter is reduced, and the accuracy of judging abnormal wear of the end face seal product during operation is improved. Description of the Drawings

[0042] Figure 1 It is an overview of the process of the method for judging abnormal wear of end face seals based on multi - parameter linear programming, mainly including fitting historical data and processing and judging the data to be judged;

[0043] Figure 2 It is the processing flow of the data to be judged in the method for judging abnormal wear of end face seals based on multi - parameter linear programming.

[0044] Figure 3 It is a measuring device for obtaining the temperature parameter of the end face seal based on infrared temperature measurement. The temperature of the end face seal product is obtained by infrared temperature measurement, and at the same time, the leakage amount and power during the operation of the end face seal product are obtained.

[0045] Explanation of the reference numerals in the drawings: 1. Infrared temperature - measuring sensor; 2. Temperature - measuring base; 3. Visualization lens; 4. Temperature - measuring adapter; 5. Axle - head locking; 6. Rotating ring; 7. Stationary ring; 8. Main shaft; 9. Leakage - amount nozzle; 10. End cover; 11. Shell. Detailed Embodiments

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will further describe in detail the disclosed embodiments of the present invention with reference to the drawings.

[0047] The embodiments of the present application disclose a method for judging abnormal wear of end face seals based on multi - parameter linear programming. As shown in Figure 1 and Figure 2 , the judging method includes the following steps:

[0048] S1: Obtain the data of the seal cavity during the normal operation of the end face seal x times;

[0049] In the data of the seal cavity during each normal operation of the end face seal, intercept the numerical values of multiple measurement parameters of the temperature - rising signal segments corresponding to different time points. The multiple measurement parameters are the target parameter and n decision parameters. The decision parameters are other data that can be collected or are required, such as the operating power parameter, the leakage - amount parameter of the seal product, or the assembly compression amount of the seal product. The target parameter is the temperature parameter. Each group of temperature parameters and decision parameters contains m data, and the m data are sequentially collected at different time points within the sampling time;

[0050] Intercept the data of the seal cavity during each normal operation of the end face seal to obtain x groups of samples, and each group of samples contains (n + 1) measurement parameters;

[0051] S2 Calculate the temperature rise slope data of the temperature parameter for the x times described in S1 based on the temperature parameter samples in S1; calculate the mean value of the temperature rise slope and the root mean square value of the temperature rise based on the x temperature rise slope data, and calculate the temperature rise slope judgment threshold [K] based on the mean value and the root mean square value of the temperature rise slope;

[0052] S3 Obtain the normalized decision parameter data matrix AN and the normalized target parameter matrix TN based on the decision parameter data in S1, and obtain the prediction weight coefficient matrix [n 1 n 2 ...n n and the normalized target parameter matrix TN', and the weight coefficient matrix should satisfy the constraint conditions.

[0053] S4 Process the data to be judged. The data to be judged includes data of multiple measurement parameters. The multiple measurement parameters include the compression amount δ, leakage amount Q q , operating power P, and other n parameter data that can be collected or are required; each measurement parameter includes m data with a time window length of Δt; according to step S3 and the data to be judged, obtain the normalized temperature rise slope adjustment empirical value K 2 ', and perform inverse normalization calculation on the normalized temperature rise slope adjustment empirical value K 2 ' within the sample data range to obtain the temperature rise slope adjustment empirical value K2;

[0054] For the newly collected temperature rise parameters of the sealing cavity, obtain the corresponding current temperature rise slope K1;

[0055] S5 Calculate the temperature rise slope judgment value K of the temperature rise slope adjustment empirical value K2 and the current temperature rise slope K1; when K is greater than the temperature rise slope judgment threshold [K], it is determined that the end face seal has abnormal wear, otherwise it is normal.

[0056] Preferably, S1 specifically includes: The collected data may include temperature parameters, operating power parameters, leakage amount parameters of the sealing product, and assembly compression amount of the sealing product, as well as other data that can be collected or are required.

[0057] Preferably, S1 specifically includes: The parameter data involved need to be ensured to be collected simultaneously, that is, the temperature parameter and the decision parameter data have the same time point.

[0058] Preferably, the S1 specifically includes: the number of selected measurement parameter samples should conform to the EVP (events per variable) principle, that is, the number x of normal operation data samples of the end face seal test selected is not less than ten times the quotient of the number of selected measurement parameters and the qualified rate of the end face seal product; the selected measurement parameters include but are not limited to temperature parameters, operating power parameters, seal product leakage parameters, and seal product assembly compression.

[0059] Preferably, the S1 specifically includes: the time range of the data of the selected measurement parameters should be ±2.5 s at the time point with the maximum slope of the temperature rise data as the time window length Δt of the corresponding sample.

[0060] Preferably, the S2 specifically includes: the temperature rise slope judgment threshold [K] is determined by the 3σ principle. First, the temperature slope values of the temperature parameter data of each group of samples in the time window Δt are obtained, and the temperature slope value matrix K = [K 1 K 2 ...K x ' is obtained. Then, the average temperature rise slope and the root mean square value σ of the temperature parameter data of x groups of samples are obtained, and the temperature rise slope judgment threshold

[0061] Preferably, the S3 specifically includes decision parameters, target parameters, and weight coefficients:

[0062] Among them, the n decision parameters are divided into l predetermined parameters and k operating fluctuation parameters. The predetermined parameters are the parameters that do not change after the assembly of the seal component, such as the product compression parameter; the operating fluctuation parameters are the parameters that fluctuate with the operation, such as the operating power parameter and the seal product leakage parameter.

[0063] For the l predetermined parameters, for each of the x groups of data, among the m data in the time window length Δt, the predetermined matrix Each predetermined parameter value remains the same among the m data. Therefore, the first group of data is used as the predetermined matrix value and no preprocessing is performed. The obtained predetermined parameter matrix For the k operating fluctuation parameters, for each of the x groups of data, among the m time series in the time window length Δt, preprocessing is required; the preprocessing method is to find the k operating fluctuation parameters of the operating fluctuation parameter standard deviation matrix and the operating fluctuation parameter mean matrix Among them, the m time series matrix of the gth operating fluctuation parameter of the yth group is The standard deviation and mean are calculated for each of them to obtain the operating fluctuation parameter standard deviation matrix as The operating fluctuation parameter mean matrix is According to the standard deviation matrix of operation fluctuation parameters and the mean matrix of operation fluctuation parameters to obtain the coefficient of variation matrix of operation fluctuation parameters; as shown in the formula in the following table; the coefficient of variation matrix of operation fluctuation parameters

[0064] The coefficient of variation matrix of operation fluctuation parameters is:

[0065] The coefficient of variation matrix of operation fluctuation parameters of the y-th group of samples is combined with the predetermined parameter matrix to obtain the decision parameter data matrix as

[0066] For n decision parameters (AN y ), perform normalization on the 0-1 range to obtain the decision parameter data matrix as AN.

[0067] Among them, the target parameter is the evaluation parameter for prediction, and the temperature slope value matrix K = [K 1 K 2 ...K y ' is obtained, where K y represents the temperature slope of the y-th group of samples in the time window length Δt. After performing normalization processing on the temperature slope value matrix K in the 0-1 range, the normalized target parameter matrix TN = [TN 1 TN 2 ...TN y ' is obtained.

[0068] Among them, the weight coefficient is the optimal weight of the linear programming target parameter obtained under the weight constraint conditions of all samples. [n 1 n 2 ...n n is the weight coefficient matrix, and weight distribution is performed for each normalized decision parameter. The product of the weight coefficient and the normalized processing parameter matrix gives the predicted normalized target parameter matrix TN' = AN * [n 1 n 2 ...n n '.

[0069] Preferably, the weight coefficient matrix should meet the following constraint conditions. The constraint conditions include: the linear programming function o = TN' - TN, satisfying the minimum error value, that is, o = min(TN' - TN); the weight coefficient matrix n 1 n 2 ...n n is distributed between 0 and 1; the weights corresponding to l predetermined parameters are the same (n 1 、n 2 ……n l are all equal), and the weights corresponding to k operation fluctuation parameters are the same (n l+1……n l+k are all equal, n = l + k), and the sum of the weights corresponding to the predetermined parameters is equal to the sum of the weights corresponding to the operation fluctuation parameters. Under the constraint conditions, a weight coefficient matrix with a dimension of 1×n is obtained.

[0070] Preferably, the S4 specifically includes: processing the data to be judged, the acquisition parameter interval of the data to be judged is Δt, and the number of data included in the data to be judged is m;

[0071] Preferably, the S4 specifically includes: within the time window Δt of the data to be judged, obtaining the predetermined parameter matrix according to S3; at the same time, solving the coefficient of variation matrix of the operation fluctuation parameters for k operation fluctuation parameters among the n decision parameters in the manner of S3, and obtaining the coefficient of variation matrix of the operation fluctuation parameters of the data to be judged.

[0073] Preferably, the S4 specifically includes: within the time window Δt of the data to be judged, obtaining the decision parameter data matrix respectively according to S3, and normalizing the n groups of decision parameter matrices within the corresponding sample data range of S3 (the corresponding normalization range of x groups of data), and obtaining the predicted normalized decision parameter matrix AN i = Normalize

[0074] Preferably, the S4 specifically includes substituting and obtaining the normalized temperature rise slope adjustment empirical value K 2 ': K 2 ' = AN i *[n 1 n 2 ...n n ', and performing inverse normalization calculation on the normalized temperature rise slope adjustment empirical value K 2 ' within the corresponding sample data range of S3 (the corresponding normalization range of x groups of data) to obtain the temperature rise slope adjustment empirical value K2.

[0075] Preferably, the S4 specifically includes: within the time window Δt of the data to be judged, obtaining the current temperature rise slope K1 according to the m data, and K1 is the temperature rise slope value of the seal cavity temperature T in the unit time window length during the acquisition.

[0076] Preferably, the S5 specifically includes that the temperature rise slope judgment value K is obtained by averaging the temperature rise slope adjustment value K2 and the current temperature rise slope K1.

[0077] Preferably, the S5 specifically includes comparing the temperature rise slope judgment value K with the temperature rise slope judgment threshold [K]; when K is greater than the temperature rise slope judgment threshold [K], it is determined that the end face seal has abnormal wear, otherwise it is normal.

[0078]

[0079]

[0080] K1 Current temperature rise slope (15)

[0081] K = (K1 + K2) / 2 Temperature rise slope judgment value (16)

[0082] As Figure 3 shown, an infrared end-face seal temperature measurement test device is further provided in an embodiment of the present invention. The system is used to monitor the end-face seal structure. The end-face seal structure includes a housing 11, a main shaft 8, a dynamic seal ring 6, a static seal ring 7, and an end cover 10. The main shaft 8 is rotatably connected inside the housing 11. The static seal ring 7 is connected to the housing 11. The dynamic seal ring 6 is connected to the main shaft 8. The dynamic seal ring 6 and the static seal ring 7 are in contact with each other. The end cover 10 is fixedly connected to the end of the housing 11 to achieve end-face sealing. When the main shaft 8 and the housing 11 are installed, a fixed assembly compression amount of the sealed product is generated.

[0083] The test device includes: a sealed operation test device, an end cover for visual temperature measurement, and a measurement system.

[0084] Preferably, the sealed operation test device is a test tooling that provides a rated speed and pressure for the end-face seal. Specifically, it includes a main shaft, a dynamic seal ring, a static seal ring assembly, and a housing. The main shaft and the dynamic seal ring are fixedly connected by a locking method and rotate together with the shafting. The static seal ring is fastened to the housing. The sealing surface of the static seal ring cooperates with the sealing surface of the dynamic seal ring and moves relative to each other. A sealed space is formed among the static seal ring, the housing, and the shafting. A medium with a corresponding pressure is introduced through a nozzle. The shafting generates rotation through power input.

[0085] Preferably, the visual temperature measurement end cover provides a sealed environment for the end-face seal while ensuring the visual function of infrared temperature measurement. Specifically, it includes an end cover, a leakage measurement nozzle, a temperature measurement adapter seat, a visual lens, and a temperature measurement base.

[0086] The end cover 10 is installed with a temperature measurement adapter seat 4. The temperature measurement adapter seat 4 is connected to a temperature measurement base 2. A visual lens 3 is provided between the temperature measurement base 2 and the temperature measurement adapter seat 4. The temperature measurement base 2 is used to connect an infrared temperature sensor 1. Temperature parameters are obtained through the infrared temperature sensor 1.

[0087] Preferably, the measurement system specifically includes infrared temperature measurement, leakage measurement, compression amount measurement, and power measurement equipment. Among them, the infrared temperature sensor can measure the temperature of the end-face seal surface in the pressure vessel cavity through the visual lens. The measurement system can collect and record data.

[0088] The end cover 10 is installed with a leakage rate nozzle 9, and the leakage rate nozzle 9 is used to connect a leakage rate measuring device, and the leakage rate parameter of the sealed product is obtained through the leakage rate measuring device.

[0089] The main shaft 8 is connected with a power measuring device, and the operating power parameter is obtained through the power measuring device.

[0090] The end of the main shaft 8 is connected with a shaft head locking 5 structure.

[0091] Embodiment 1

[0092] The working principle and characteristics of the present invention are further described below by taking a certain end face seal product as an example. In this embodiment, a method for judging abnormal wear of an end face seal based on multi-parameter linear programming includes a seal test device, and an infrared temperature sensor, a leakage rate measuring flow sensor, and a power sensor are installed on the test device. The judgment method includes the following steps:

[0093] S1: Obtain the data of the seal cavity during the normal operation of the end face seal x times;

[0094] In the data of the seal cavity during each normal operation of the end face seal, intercept the numerical values of multiple measurement parameters of the temperature rise signal segments corresponding to multiple different time points. The multiple measurement parameters are the target parameter and n decision parameters. The decision parameters are other data that can be collected or required, such as the operating power parameter, the leakage rate parameter of the sealed product, or the assembly compression amount of the sealed product. The target parameter is the temperature parameter. Each set of temperature parameters and decision parameters contains m data, and the m data are sequentially collected at different time points within the sampling time;

[0095] Intercept the data of the seal cavity during each normal operation of the end face seal to obtain x sets of samples, and each set of samples contains (n + 1) measurement parameters;

[0096] S2: According to the temperature parameter samples in S1, calculate the temperature rise slope data of the temperature parameters in the x times in S1; according to the x temperature rise slope data, calculate the average value of the temperature rise slope and the root mean square value of the temperature rise, and calculate the temperature rise slope judgment threshold [K] according to the average value of the temperature rise slope and the root mean square value;

[0097] S3: According to the decision parameter data in S1, obtain the normalized decision parameter data matrix and the normalized target parameter matrix, and obtain the prediction weight coefficient matrix and the normalized target parameter matrix according to the decision parameter data matrix and the normalized target parameter matrix. The weight coefficient matrix should meet the constraint conditions;

[0098] S4: Process the data to be judged. The data to be judged includes the data of multiple measurement parameters. The multiple measurement parameters include the compression amount δ and the leakage rate Q of the worn product q, other n parameter data such as operating power P that can be collected or are required; each measurement parameter includes m data with a time window length of Δt; according to step S3 and the data to be judged, the normalized temperature rise slope adjustment empirical value K is obtained 2 ', the normalized temperature rise slope adjustment empirical value K 2 ' is inversely normalized within the sample data range to obtain the temperature rise slope adjustment empirical value K2;

[0099] For the newly collected temperature rise parameters of the seal cavity, the corresponding current temperature rise slope K1 is obtained;

[0100] S5 calculates the temperature rise slope judgment value K of the temperature rise slope adjustment empirical value K2 and the current temperature rise slope K1; when K is greater than the temperature rise slope judgment threshold [K], it is determined that the end face seal has abnormal wear, otherwise it is normal.

[0101] Preferably, S1 specifically includes: the collected data may include temperature parameters, operating power parameters, seal product leakage parameters, and seal product assembly compression amounts.

[0102] Preferably, S1 specifically includes: the parameter data involved needs to be collected simultaneously, that is, the temperature parameter and decision parameter data have the same time point.

[0103] Preferably, S1 specifically includes: the number of selected measurement parameter samples should conform to the EVP (events per variable) principle, that is, the number of normal operation data samples x of the selected end face seal test is not less than ten times the quotient of the number of selected measurement parameters and the qualified rate of the end face seal product. According to the selected number of parameters being 3 and the product qualified rate being 90%, the selected number of samples should not be less than 34, and the number of samples x is 34.

[0104] Preferably, S1 specifically includes: the time range of the data of the selected measurement parameters should be ±2.5 s at the time point with the maximum temperature rise data slope as the time window length Δt of the corresponding sample, that is, Δt is 5 s.

[0105] Preferably, S2 specifically includes: the temperature rise slope judgment threshold [K] is determined by the 3σ principle. First, the temperature slope values of the temperature parameter data of each group of samples in the time window Δt are obtained to obtain the temperature slope value matrix K = [K 1 K 2 ...K x ', and then the average temperature rise slope of the temperature parameter data of x groups of samples is obtained and the root mean square value σ to obtain the temperature rise slope judgment threshold According to the selected 34 groups of sample data, the average temperature rise slope is obtained is 0.0511, the root mean square value σ is 0.0502, and finally the temperature rise slope judgment threshold [K] is obtained as 0.2017.

[0106] Preferably, the S3 specifically includes decision parameters, target parameters, and weight coefficients:

[0107] Among them, the n decision parameters are divided into l predetermined parameters and k operating fluctuation parameters. The predetermined parameters are the parameters that do not change after the assembly of the sealing component. The selected parameter is the product compression amount δ, that is, l = 1; the operating fluctuation parameters are the parameters that fluctuate with the operation. The selected parameters are the leakage amount Q q and the operating power P, that is, k = 2.

[0108] For 1 predetermined parameter, for each of the 34 groups of data, among the m selected data (m = 500) in the time window length Δt, the predetermined matrix The predetermined parameter values remain the same among the m data. Therefore, the first group of data is used as the predetermined matrix value, and no preprocessing is performed. The obtained predetermined parameter matrix For 2 operating fluctuation parameters, for each of the 34 groups of data, among the m time series of the time window length Δt, preprocessing is required; the preprocessing method is to find the 2 operating fluctuation parameters The standard deviation matrix of the operating fluctuation parameters and the mean matrix of the operating fluctuation parameters The quotient of the y-th group of data, the coefficient of variation matrix of the preprocessed operating fluctuation parameters

[0109] The coefficient of variation matrix of the operating fluctuation parameters is: Combined with the predetermined parameter matrix to obtain the decision parameter data matrix as For the 3 decision parameters (AN y ), the decision parameter data matrix is normalized in the range of 0 to 1 as AN.

[0110] Among them, the target parameter is the evaluation parameter for prediction. Obtain the temperature slope value matrix K = [K 1 K 2 ...K 34 ' of all sample data. After normalizing the temperature slope value matrix K in the range of 0 to 1, the normalized target parameter matrix TN = [TN 1 TN 2 ...TN 34 ' is obtained.

[0111] Among them, the weight coefficient is the optimal weight of the linear programming target parameter obtained under the weight constraint conditions of all samples, [n 1 n 2 n 3is the weight coefficient matrix, which assigns weights to each normalized decision parameter. The product of the weight coefficient and the normalized processing parameter matrix gives the predicted normalized target parameter matrix TN' = AN * [n 1 n 2 n 3 '.

[0112] Preferably, the weight coefficient matrix should meet the following constraint conditions, including: the linear programming function o = TN' - TN, satisfying the minimum error value, i.e., o = min(TN' - TN); the weight coefficient matrix [n 1 n 2 n 3 is distributed between 0 and 1; the weights corresponding to 1 predetermined parameter are the same, and the weights corresponding to 2 operation fluctuation parameters are the same (n 2 = n 3 ), and the sum of the weights corresponding to the predetermined parameter is equal to the sum of the weights corresponding to the operation fluctuation parameters, n 1 = n 2 + n 3 . Finally, the weight parameter matrix obtained is [n 1 n 2 n 3 = [0.2104 0.1052 0.1052].

[0113] Preferably, the specific steps of S4 include: processing the data to be judged, the acquisition parameter interval of the data to be judged is Δt, and the number of data included in the data to be judged is m (m = 500);

[0114] Preferably, the specific steps of S4 include: within the time window Δt of the data to be judged, obtaining the predetermined parameter matrix according to S3; at the same time, solving 1 predetermined parameter matrix in the way of S3 for k operation fluctuation parameters among n decision parameters, that is, the current acquisition compression amount δ; at the same time, performing mutation processing on the current power P and the current leakage amount Q q That is, after obtaining the standard deviation and mean of the current acquisition power P and the current leakage amount Q q respectively, according to the process described in S3, obtaining the quotient of the standard deviation and mean of the two parameters to get the mutation coefficient value E p 、E Qq .

[0115] Preferably, the specific steps of S4 include: within the time window Δt of the data to be judged, obtaining the decision parameter data matrix respectively according to S3, and performing normalization processing on the current acquisition compression amount δ and the mutation power E p 、mutation leakage amount E Qq within the sample data range to obtain the normalized compression amount δ 0 、normalized power E P0 、normalized leakage amount EQq0 , to form the corresponding predicted normalized decision parameter matrix AN i = [0.8400 1.9364 0.8481].

[0116] Preferably, step S4 specifically includes substituting into the empirical formula for adjusting the temperature rise slope in step S3 to obtain the normalized empirical value K 2 ' of the adjusted temperature rise slope: K 2 ' = AN i * [n 1 n 2 n 3 ', and finally obtaining K 2 ' as 0.4697; performing denormalization calculation on the normalized empirical value K 2 ' of the adjusted temperature rise slope within the corresponding sample data range (the corresponding normalization range of x groups of data) in step S3 to obtain the empirical value K2 of the adjusted temperature rise slope as 0.0786.

[0117] Preferably, step S4 specifically includes: within the time window Δt of the data to be judged, based on m pieces of data (m = 500), obtaining the current temperature rise slope K1, which is 0.4025.

[0118] Preferably, step S5 specifically includes obtaining the temperature rise slope judgment value K by averaging the temperature rise slope adjustment value K2 and the current temperature rise slope K1, and obtaining K as 0.2799.

[0119] Preferably, step S5 specifically includes comparing the temperature rise slope judgment value K = 0.2799 with the temperature rise slope judgment threshold [K] = 0.2017; when the temperature rise slope judgment value K is greater than the temperature rise slope judgment threshold [K], it is determined that abnormal wear of the end face seal has occurred.

[0120] The content not described in detail in the specification of this application belongs to the well-known technology in the art.

[0121] The above has described this application in detail in combination with specific embodiments and exemplary examples, but these descriptions should not be construed as limitations on this application. Those skilled in the art understand that without departing from the spirit and scope of this application, various equivalent substitutions, modifications, or improvements can be made to the technical solutions and their implementation manners of this application, and these all fall within the scope of this application. The protection scope of this application is subject to the appended claims.

Claims

1. A method for judging abnormal wear of end face seal, characterized in that: include: S1: Obtain the data of the sealing cavity of the end face seal in normal operation for x times; In the data of the sealing cavity of each normal operation of the end face seal, the values ​​of multiple measurement parameters of the temperature rise signal fragments corresponding to multiple different time points are intercepted, and the multiple measurement parameters are the target parameter and n decision parameters, and the target parameter is the temperature parameter, and x groups of samples are obtained, each group of samples contains n+1 measurement parameters, and each measurement parameter contains m data; S2: Calculate the temperature rise slope data of the temperature parameter x times according to the temperature parameter data in S1; Calculate the slope mean and the root mean square value according to the x temperature rise slope data, and calculate the temperature rise slope judgment threshold [K] according to the slope mean and the root mean square value; S3: According to the decision parameter data and temperature parameter data in S1, the normalized decision parameter data matrix AN and the normalized target parameter matrix TN are obtained, and the prediction weight coefficient matrix [n1 n2...n n ] and the normalized target parameter matrix TN', the weight coefficient matrix should satisfy the constraints; S4: Processing the data to be judged, the data to be judged includes data of multiple measurement parameters, the multiple measurement parameters include a target parameter and n decision parameters, the target parameter is a temperature parameter, and each measurement parameter includes m data within a time window length Δt; According to step S3 and the data to be judged, the normalized temperature rise slope adjustment empirical value K2' is obtained, and the normalized temperature rise slope adjustment empirical value K2' is denormalized within the sample data range to obtain the temperature rise slope adjustment empirical value K2; According to the m data included in the temperature parameters, the current temperature rise slope K1 is calculated; S5: According to the temperature rise slope adjustment experience value K2 and the current temperature rise slope K1, the temperature rise slope judgment value K is obtained; when K is greater than the temperature rise slope judgment threshold value [K], it is determined that the end face seal is abnormally worn, otherwise it is normal.

2. A method for determining abnormal wear of an end face seal according to claim 1, characterized in that: The decision parameter is an operating power parameter, a sealing product leakage parameter or a sealing product assembly compression parameter; The m data of temperature parameters and decision parameters are collected in sequence at different time points within the sampling time.

3. The method for determining abnormal wear of an end face seal according to claim 1, characterized in that: In S1, the sample quantity x≥10×(the quotient of the number of measurement parameters and the qualified rate of the end face sealing products).

4. A method for determining abnormal wear of an end face seal according to claim 1, characterized in that: In S2, ±2.5s of the time point with the maximum slope of the temperature rise data is used as the time window length Δt, and the data of the measurement parameter are taken within the time window length Δt.

5. The method for determining abnormal wear of an end face seal according to claim 1, characterized in that: In S2, according to the temperature parameter data in S1, the temperature rise slope data of the temperature parameter is calculated x times; according to the x temperature rise slope data, the slope mean and the root mean square value are calculated, and the temperature rise slope judgment threshold [K] is calculated according to the slope mean and the root mean square value, including: According to the temperature parameter data in S1, the temperature slope value of the temperature parameter data of each group of samples in the time window Δt is calculated to obtain the temperature slope value matrix K = [K1 K2 ... K x ]'; Then according to the temperature slope value matrix K = [K1 K2 ... K x ]', and obtain the average temperature rise slope of the temperature parameter data of x groups of samples And the root mean square value σ, get the temperature rise slope judgment threshold 6. A method for determining abnormal wear of an end face seal according to claim 1, characterized in that: In S3, according to the decision parameter data and temperature parameter data in S1, a decision parameter data matrix AN, a normalized target parameter matrix TN and a weight coefficient matrix [n1 n2 ... n n ], and according to the weight coefficient matrix and the decision parameter data matrix AN, the predicted normalized target parameter matrix TN' is obtained, TN' = AN * weight coefficient matrix, including: The n decision parameters are divided into l predetermined parameters and k operating fluctuation parameters. The predetermined parameters are parameters that do not change after the sealing component is assembled, and the operating fluctuation parameters are parameters that fluctuate with the operation. The predetermined parameters and operating fluctuation parameters are expressed in matrix format to obtain the coefficient of variation matrix of the operating fluctuation parameters. With the predetermined parameter matrix Operational fluctuation parameter variation coefficient matrix With the predetermined parameter matrix The decision parameter data matrix is ​​obtained by combining About AN y Normalize the range from 0 to 1 and get the decision parameter data matrix AN; According to the temperature parameter data, the temperature slope value matrix K = [K1 K2 ... K y ]',K y represents the temperature slope of the yth group of samples in the time window length Δt. After normalizing the temperature slope value matrix K in the range of 0 to 1, the normalized target parameter matrix TN = [TN1 TN2 ... TN y ]'; The weight coefficient matrix is ​​[n1 n2...n n ], TN'=AN*[n1 n2...n n ]'.

7. A method for determining abnormal wear of an end face seal according to claim 6, characterized in that: The predetermined parameters and the operating fluctuation parameters are expressed in a matrix format to obtain the operating fluctuation parameter variation coefficient matrix With the predetermined parameter matrix include: For l predetermined parameters, for each of the x sets of data, in the m data in the time window length Δt, the predetermined matrix y∈[1,x]; Predetermined parameter matrix For k operating fluctuation parameters, for each of y groups of data, in m time series with a time window length Δt, find k operating fluctuation parameters a2 = [a1 a2 ... a k ]'s operating fluctuation parameter standard deviation matrix and the mean matrix of operating fluctuation parameters where a k =[a1 a2...a m ] T , the standard deviation matrix of the operating fluctuation parameters is The mean matrix of operating fluctuation parameters is According to the standard deviation matrix of the operating fluctuation parameters and the mean matrix of operating fluctuation parameters The quotient of , obtains the coefficient of variation matrix of the operating fluctuation parameters, which is:

8. A method for determining abnormal wear of an end face seal according to claim 6, characterized in that: The weight coefficient matrix satisfies the constraints: The error value is the smallest, that is, o = min (TN'-TN); n1 n2...n n All are distributed between 0 and 1; The weights corresponding to the l predetermined parameters are the same, i.e. n1, n2, ... n l are equal; k operating fluctuation parameters have the same corresponding weights, that is, n l+1 ……n l+k are all equal, n = l + k; The sum of the weights corresponding to the predetermined parameters is equal to the sum of the weights corresponding to the operating fluctuation parameters.

9. A method for determining abnormal wear of an end face seal according to claim 1, characterized in that: In said S4, according to step S3 and the data to be determined, the normalized temperature rise slope adjustment empirical value K2' is obtained, including: According to step S3 and the decision parameter data of the data to be judged, a decision parameter data matrix is ​​obtained, and the decision parameter data matrix is ​​normalized within the sample data range to obtain a predicted normalized decision parameter matrix; According to the predicted normalized target parameter matrix TN' in S3, combined with the temperature parameter data of the data to be judged, the normalized temperature rise slope adjustment empirical value K2' is obtained, K2'=AN i *[n1 n2...n n ]'.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images