Flange connection tightness prediction method, equipment, storage medium and product

Abstract

The present invention discloses a method, device, storage medium, and product for predicting the tightness of flange connections. The prediction method comprises obtaining flange joint clearances at different sampling moments; constructing a gap deviation sequence based on the flange joint clearances at different sampling moments; fitting the gap deviation sequence using an ARIMA model to obtain a gap deviation prediction model; and using the gap deviation prediction model to predict the gap deviation at the next sampling moment, calculating the flange joint clearance at the next sampling moment based on the flange joint clearance at the current sampling moment and the gap deviation at the next sampling moment. By performing autoregression and sliding averaging on the gap deviation, the present invention can update the prediction results in real time as factors such as temperature and load change, calculate the flange joint clearance in the future period, and improve prediction accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of flange connection safety technology, and in particular relates to a flange connection tightness prediction method, equipment, storage medium and product. Background Art

[0002] Flange connections are one of the most common connections in industrial equipment, widely used in critical applications such as wind power equipment, aerospace equipment, and petrochemical pipelines. The tightness of flange connections is directly related to the safety, reliability, and service life of the equipment. However, under long-term, high-temperature, high-pressure, or unstable load conditions, flanges can become loose, wear gaskets, or experience gap distortion, leading to potential failures such as leaks, energy waste, and even major safety incidents.

[0003] Existing manual inspections or single-point monitoring methods can only achieve real-time detection of flange connection tightness, but cannot predict the tightness of flange connections in future cycles and cannot prevent the occurrence of safety accidents. Summary of the Invention

[0004] The purpose of the present invention is to provide a flange connection tightness prediction method, equipment, storage medium and product to solve the problem that traditional methods cannot predict the flange connection tightness in future periods, resulting in the inability to provide timely warnings.

[0005] The present invention solves the above technical problems through the following technical solutions: A method for predicting the tightness of flange connection, comprising:

[0006] Step S1: Obtaining the flange joint gap at different sampling times;

[0007] Step S2: constructing a gap deviation sequence based on the flange joint surface gaps at different sampling moments;

[0008] Step S3: fitting the gap deviation sequence using an ARIMA model to obtain a gap deviation prediction model;

[0009] Step S4: using the gap deviation prediction model to predict the gap deviation at the next sampling moment, and calculating the flange joint surface gap at the next sampling moment based on the flange joint surface gap at the current sampling moment and the gap deviation at the next sampling moment;

[0010] Step S5: Repeat step S4 to calculate the flange joint surface clearance at multiple future sampling moments in a rolling manner.

[0011] Furthermore, in step S1, obtaining the flange joint surface gap at different sampling moments specifically includes:

[0012] Step S1.1: collecting temperature, load and flange joint gap at different sampling times;

[0013] Step S1.2: Divide the flange joint surface gap at different sampling moments into sub-segments according to the set time period;

[0014] Step S1.3: Fitting and detecting key inflection points of the flange joint surface gap at different sampling times in each sub-segment, obtaining and saving the gap fitting function and key inflection points of each sub-segment;

[0015] Step S1.4: Reconstruct the flange joint surface gap at different sampling moments within the required time period based on the gap fitting function and key inflection points of each sub-segment.

[0016] Furthermore, in step S1.3, the key inflection point detection is specifically as follows:

[0017] like ,but Recorded as the key inflection point;

[0018] in, represents the flange joint gap at the t-th sampling moment, represents the fitting value of the flange joint gap at the t-th sampling moment, Indicates the inflection point threshold.

[0019] Furthermore, in step S2, a gap deviation sequence is constructed according to the flange joint surface gaps at different sampling moments, specifically including:

[0020] Step S2.1: Perform a primary difference on the flange joint surface gaps at two adjacent sampling moments to obtain the gap deviation after the primary difference, and then obtain a gap deviation sequence; wherein the primary difference formula of the gap deviation is:

[0021] ;

[0022] in, It represents the gap deviation after a difference at the t-th sampling moment, represents the flange joint gap at the t-th sampling moment, represents the flange joint surface clearance at the t-1th sampling time, and N-1 represents the number of known sampling times;

[0023] Step S2.2: Perform a stationarity test on the gap deviation sequence. If the gap deviation sequence is stationary, obtain the gap deviation sequence; if the gap deviation sequence is non-stationary, proceed to step S2.3;

[0024] Step S2.3: Perform a quadratic difference on each gap deviation in the gap deviation sequence to obtain the gap deviation after the quadratic difference, and then obtain the gap deviation sequence, and proceed to step S2.2; wherein, the quadratic difference formula of the gap deviation is:

[0025] ;

[0026] in, It represents the gap deviation after the second difference at the t-th sampling moment, Indicates the gap deviation after a difference at the t-1th sampling time.

[0027] Furthermore, in step S3, the gap deviation sequence is fitted using an ARIMA model, specifically including:

[0028] Based on the ARIMA model, a gap deviation fitting model is constructed. The specific expression is:

[0029] ;

[0030] in, 、 、 Respectively represent the gap deviation at the t, t-1, and t-2 sampling moments, 、 are autoregressive coefficients, represents the sliding mean coefficient, 、 Represent the white noise error terms at the tth and t-1th sampling moments respectively;

[0031] Initializing the autoregressive coefficient and the sliding average coefficient, and estimating the autoregressive coefficient and the sliding average coefficient according to the gap deviation sequence to obtain estimated values of the autoregressive coefficient and the sliding average coefficient;

[0032] Substituting the estimated values of the autoregressive coefficient and the sliding average coefficient into the gap deviation fitting model, a gap deviation prediction model is obtained.

[0033] Furthermore, the prediction method also includes performing a multi-level warning based on the predicted flange joint surface gap and gap threshold range at the next sampling moment, specifically including:

[0034] like , a level one warning is issued;

[0035] like And the duration reaches , a second-level warning is issued;

[0036] like , a level three warning will be issued;

[0037] in, represents the flange joint gap at the t-th sampling moment, Indicates the predicted flange joint surface clearance at the next sampling moment, Indicates the allowable gap variation value, Indicates the lower limit of the gap threshold range, Indicates the upper limit of the gap threshold range, Indicates the time threshold.

[0038] Furthermore, before the multi-level warning, the gap threshold range is dynamically updated according to the temperature and load corresponding to the flange joint gap at the current sampling moment. The specific update formula is:

[0039] ;

[0040] ;

[0041] ;

[0042] in, 、 Respectively represent the lower limit of the gap threshold range at the tth and t-1th sampling moments, 、 Respectively represent the upper limit of the gap threshold range at the tth and t-1th sampling moments, Indicates the flange joint surface clearance correction value, 、 Both represent smoothing coefficients, Indicates the reference value of the flange joint gap. 、 Both represent coupling coefficients, represents the temperature at the t-th sampling moment, Indicates the temperature reference value, represents the load at the tth sampling moment, Indicates the load reference value.

[0043] Based on the same concept, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program / instruction stored in the memory, wherein the processor executes the computer program / instruction to implement the flange connection tightness prediction method as described above.

[0044] Based on the same concept, the present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon, which implements the flange connection tightness prediction method as described above when the computer program / instruction is executed by a processor.

[0045] Based on the same concept, the present invention also provides a computer program product, including a computer program / instruction, which implements the flange connection tightness prediction method as described above when executed by a processor.

[0046] Compared with the prior art, the advantages of the present invention are:

[0047] The present invention uses the ARIMA model to predict the tightness of flange connections. The ARIMA model is specially designed to process data with time series characteristics, can capture trends and periodicity in historical data, and can be applied to predicting changes in flange joint gaps over time. By performing autoregression and sliding average on the gap deviation, the ARIMA model can update its prediction results in real time as factors such as temperature and load change, and then calculate the flange joint gap in future periods, thereby improving prediction accuracy. Maintenance and inspection plans can be optimized based on the flange joint gap in future periods, thus avoiding unnecessary maintenance and reducing operation and maintenance costs. At the same time, early warning can be achieved based on the flange joint gap in future periods, thus timely avoiding the occurrence of safety accidents and improving operational safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only one embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 It is a flow chart of the flange connection tightness prediction method in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0051] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0052] Example 1

[0053] Traditional manual inspections or single-point monitoring methods can only detect the tightness of the current flange connection, but cannot predict the tightness of the flange connection in future cycles, nor can they prevent the occurrence of safety accidents. Based on this, an embodiment of the present invention provides a flange connection tightness prediction method. The flange includes an upper flange, a lower flange, and bolts connecting the upper and lower flanges. The flange connection tightness refers to the gap between the flange mating surfaces, that is, the gap between the upper and lower flanges.

[0054] Figure 1 The flow chart of the flange connection tightness prediction method provided by the present invention is shown as follows: Figure 1 As shown in FIG, the flange connection tightness prediction method includes the following steps:

[0055] Step S1: Obtain the flange joint surface gap at different sampling times.

[0056] In a specific embodiment of the present invention, obtaining the flange joint surface gap at different sampling moments specifically includes:

[0057] Step S1.1: Collect the temperature, load and flange joint gap at different sampling times.

[0058] Taking the steam pipe flange of a chemical company as an example, the flange has a design pressure of 2.0 MPa and a year-round operating temperature of 150±30°C. It must cope with seasonal changes, load fluctuations, and thermal expansion and contraction. Temperature sensors, load sensors, and gap sensors are used to collect temperature, load, and flange joint gap at different sampling times. In this embodiment, the temperature sensor uses a K-type or E-type thermocouple and is attached to the flange body or bolt surface; the load sensor uses a strain gauge or load cell and is placed on the bolt to dynamically sense the bolt preload; and the gap sensor uses a laser ranging sensor or eddy current displacement sensor and is placed near the flange gasket contact area to monitor the tiny gap between the upper and lower flanges.

[0059] In this embodiment, the temperature, load, and flange joint clearance are collected every 5 to 15 minutes, and the temperature, load, and flange joint clearance are collected for 8 weeks.

[0060] Step S1.2: Divide the flange joint surface gap at different sampling moments according to the set time period to obtain sub-segments.

[0061] To reduce data storage, the present invention compresses and stores the flange joint clearance at different sampling times in segments. In this embodiment, in step S1.1, eight weeks of flange joint clearance are collected. The eight weeks of flange joint clearance are divided into one-week time periods, and each sub-segment contains one week of flange joint clearance.

[0062] Step S1.3: Fitting and detecting key inflection points of the flange joint surface gap at different sampling moments in each sub-segment are performed to obtain and save the gap fitting function and key inflection points of each sub-segment.

[0063] In this embodiment, the least squares method is used to perform linear fitting on the flange joint surface gap at different sampling times in each sub-segment. The specific formula is:

[0064] (1)

[0065] This gives the gap fitting function for each sub-segment:

[0066] (2)

[0067] in, represents the flange joint gap at the t-th sampling moment, represents the fitting value of the flange joint gap at the t-th sampling moment, 、 represents the fitting coefficient.

[0068] In order to improve the accuracy of the subsequent reconstructed flange joint gap, key inflection point detection is also performed based on the flange joint gap at different sampling moments in each sub-segment and the linear fitting function. Specifically, if ,but is recorded as the key inflection point, where In this embodiment, the inflection point threshold Set to 0.015mm.

[0069] The gap fitting function and key inflection points of each sub-segment are stored in the InfluxDB time series database, and the gap fitting function and key inflection points of the sub-segment obtained from the newly collected data are stored in the InfluxDB time series database every week. The key inflection points are used to identify the key gap points in the gap time series that deviate abnormally from the linear fitting trend. The key inflection points represent the important twists and turns and changes in each sub-segment, ensuring the accuracy when reconstructing the data. By only storing the gap fitting function and key inflection points of each sub-segment, the present invention greatly reduces the data storage space and transmission burden, and accurately reflects the real dynamic changes of the flange joint surface gap, providing a data basis for subsequent flange joint surface gap prediction and early warning.

[0070] Step S1.4: Reconstruct the flange joint surface gap at different sampling moments within the required time period based on the gap fitting function and key inflection points of each sub-segment.

[0071] During prediction, the gap fitting functions and key inflection points of the sub-segments in the latest InfluxDB time series database are used to reconstruct the flange joint clearance at different sampling times. For example, the latest InfluxDB time series database stores the gap fitting functions and key inflection points for 10 sub-segments (i.e., 10 weeks of historical flange joint clearance). If the flange joint clearance at different sampling times within an 8-week period is needed for prediction, the gap fitting functions and key inflection points of the 3rd to 10th sub-segments are used to reconstruct the flange joint clearance at different sampling times within the 8-week period. Each segment corresponds to a one-week period.

[0072] The reconstructed flange joint clearances at different sampling times were then de-identified, with outliers removed and possible interpolation or smoothing performed. The flange joint clearances at different sampling times over an eight-week period were divided into a training set and a test set. The training set was used to determine the estimated autoregressive and moving average coefficients in the ARIMA model, while the test set was used to test the performance of the gap deviation prediction model.

[0073] Step S2: constructing a gap deviation sequence based on the flange joint surface gaps at different sampling moments.

[0074] In a specific embodiment of the present invention, a gap deviation sequence is constructed based on the flange joint surface gaps at different sampling moments, specifically including:

[0075] Step S2.1: Perform a difference on the flange joint surface gap at two adjacent sampling moments to obtain the gap deviation after the difference, and then obtain the gap deviation sequence ; Among them, the first difference formula of gap deviation is:

[0076] (3)

[0077] in, It represents the gap deviation after a difference at the t-th sampling moment, represents the flange joint gap at the t-th sampling moment, represents the flange joint surface clearance at the t-1th sampling time, and N-1 represents the number of known sampling times.

[0078] Step S2.2: Gap deviation sequence Perform a stationarity test. If the gap deviation sequence is stationary, there is no need to perform a second difference to obtain the gap deviation sequence. If the gap deviation sequence is non-stationary, a second difference is required and proceed to step S2.3.

[0079] In this embodiment, the ADF (Augmented Dickey-Fuller) or KPSS method is used to test the stationarity of the gap deviation sequence. If the gap deviation sequence is non-stationary, it is necessary to perform further differentiation until the final gap deviation sequence is stationary. in represents the gap deviation at the t-th sampling moment, Can be or .

[0080] Step S2.3: Perform a secondary difference on each gap deviation in the gap deviation sequence to obtain the gap deviation after secondary difference, and then obtain the gap deviation sequence , and go to step S2.2. The quadratic difference formula of the gap deviation is:

[0081] (4)

[0082] in, It represents the gap deviation after the second difference at the t-th sampling moment, Indicates the gap deviation after a difference at the t-1th sampling time.

[0083] The gap deviation sequence constructed by the present invention is a stationary sequence. The statistical characteristics (mean, variance, and autocovariance) of the stationary sequence do not change over time, so that modeling can be performed more effectively. For example, moving average or autoregressive models usually assume that the residuals are independently distributed and have a constant variance, which may not be true in non-stationary data. Regression analysis in non-stationary data may lead to pseudo-regression, that is, the model appears to be effective, but in fact there is no actual correlation. The stabilization of the gap deviation sequence ensures the validity and reliability of subsequent model results. After the gap deviation sequence is stabilized, the trend and seasonal factors of the gap deviation sequence can be reduced, the analysis focus becomes more concentrated, and the analysis complexity is simplified. The present invention uses a stabilized gap deviation sequence for modeling, which can reduce errors and improve the prediction accuracy of the model.

[0084] Step S3: Use the ARIMA model to fit the gap deviation series to obtain a gap deviation prediction model.

[0085] In a specific embodiment of the present invention, an ARIMA model (Autoregressive Integrated Moving Average Model) is used to fit the gap deviation series, specifically including:

[0086] Step S3.1: Based on the ARIMA model, a gap deviation fitting model is constructed. The specific expression is:

[0087] (5)

[0088] in, 、 、 Respectively represent the gap deviation at the t, t-1, and t-2 sampling moments, 、 are autoregressive coefficients, represents the sliding mean coefficient, 、 Respectively represent the white noise error term at sampling time t and t-1. In this embodiment, the ARIMA model uses ARIMA (2,1,1).

[0089] Step S3.2: Autoregressive coefficients 、 and the sliding average coefficient Initialize and adjust the autoregressive coefficients according to the gap deviation sequence constructed in step S2 、 and the sliding average coefficient Estimation is performed to obtain the estimated values of the autoregressive coefficient and the moving average coefficient.

[0090] In this embodiment, the maximum likelihood or least squares method is used to estimate the autoregressive coefficients. 、 and the sliding average coefficient Perform estimation, specifically, calculate the likelihood function or the residual sum of squares, and then iteratively update using gradient descent or Newton's method until convergence to obtain the optimal estimated values of the autoregressive coefficients and sliding average coefficients.

[0091] Step S3.3: Substitute the estimated values of the autoregressive coefficient and the sliding average coefficient into the gap deviation fitting model (i.e., formula (5)) to obtain the gap deviation prediction model.

[0092] For example, 、 、 , then the gap deviation prediction model is:

[0093] (6)

[0094] The performance of the gap deviation prediction model was evaluated, specifically by calculating the AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) metrics. If the AIC and BIC were less than the set thresholds and the residual sequence approximated white noise, the gap deviation prediction model was effective. The Ljung-Box test confirmed the absence of significant autocorrelation in the residual sequence. The gap deviation prediction model was tested on a test set, with a mean square error (MSE) of 0.0021, a mean absolute error (MAE) of 0.038 mm, and a mean absolute percentage error (MAPE) of 4.7%, demonstrating that the gap deviation prediction model exhibits good performance.

[0095] Step S4: using the gap deviation prediction model to predict the gap deviation at the next sampling moment, and calculating the flange joint surface gap at the next sampling moment based on the flange joint surface gap at the current sampling moment and the gap deviation at the next sampling moment.

[0096] The formula for predicting the gap deviation at the next sampling moment using the gap deviation prediction model is:

[0097] (7)

[0098] The calculation formula for the flange joint gap at the next sampling moment is:

[0099] (8)

[0100] in, represents the gap deviation of the predicted t+1th sampling moment, represents the predicted flange joint surface clearance at the t+1th sampling moment, They represent the white noise error terms at the t+1th sampling moment respectively.

[0101] Step S5: Repeat step S4 to calculate the flange joint clearance at multiple sampling moments in the future. The specific formula is:

[0102] (9)

[0103] (10)

[0104] A prediction curve is drawn based on the predicted flange joint surface clearance at multiple future sampling moments, and the prediction curve is displayed.

[0105] Step S6: Based on the predicted flange joint clearance at the next sampling moment and gap threshold range for multi-level warning.

[0106] In a specific embodiment of the present invention, the flange joint clearance at the next sampling moment is predicted. Multi-level warnings are provided based on the gap threshold range, including:

[0107] like , it indicates that the actual gap value exceeds the predicted gap value. Temperature, load and other factors are relatively severe during this period. The tightness of the flange connection deteriorates and an early warning is needed. Therefore, a first-level early warning is issued and prompted on the operation interface, suggesting that inspections be strengthened.

[0108] like And the duration reaches , a second-level warning is issued, and the operation and maintenance personnel are notified via SMS / email, suggesting that the flange be tightened or the gasket be evaluated within 24 to 48 hours;

[0109] like , it indicates that the actual gap value exceeds the gap threshold range, and a third-level warning is issued, automatically linking the control room to execute the pressure reduction or shutdown procedure to prevent the failure from spreading;

[0110] in, Indicates the allowable gap variation value, Indicates the lower limit of the gap threshold range, Indicates the upper limit of the gap threshold range, Indicates the time threshold. In this embodiment, , .

[0111] Once an early warning is triggered, it will be recorded in the log table for subsequent tracing and maintenance optimization.

[0112] In order to improve the accuracy of the early warning, the flange joint gap at the current sampling moment is The corresponding temperature and load For the gap threshold range ( , ) is dynamically updated, and the specific update formula is:

[0113] (11)

[0114] (12)

[0115] (13)

[0116] in, 、 Respectively represent the lower limit of the gap threshold range at the tth and t-1th sampling moments, 、 Respectively represent the upper limit of the gap threshold range at the tth and t-1th sampling moments, Indicates the flange joint surface clearance correction value, 、 Both represent smoothing coefficients, Indicates the reference value of the flange joint gap. 、 Both represent coupling coefficients, represents the temperature at the t-th sampling moment, Indicates the temperature reference value, represents the load at the tth sampling moment, Indicates the load reference value.

[0117] When the equipment is in a healthy state after installation or maintenance, the temperature, load and flange joint gap are collected by sensors for a long time, and the average temperature, average load and average flange joint gap during stable operation are taken as the temperature reference value, load reference value and flange joint gap reference value. Flange joint gap reference value Affected by temperature and load, it is necessary to correct the flange joint surface gap reference value according to formula (11), and then use the flange joint surface gap correction value to update the gap threshold range. Coupling coefficient 、 It can be obtained through finite element simulation and experimental calibration. In this embodiment, 、 .

[0118] In this embodiment, when t=2, 、 Smoothing coefficient 、 Dynamically adjust according to the historical false alarm rate. The weight of the influence of temperature on the flange joint gap is usually 40%~60%, and the weight of the influence of load on the flange joint gap is 30%~50%. The middle value is taken to get the initial ratio. : =4:6, therefore, in this embodiment, 、 .

[0119] In each multi-level warning, the gap threshold range ( , ) for dynamic updates, or you can The fluctuation exceeds the previous sampling moment ±10% of the gap threshold range ( , ) for dynamic updates.

[0120] Example 2

[0121] An embodiment of the present invention further provides an electronic device comprising: a memory, a processor, and a computer program / instructions stored in the memory, wherein the processor executes the computer program / instructions to implement the flange connection tightness prediction method in the embodiment of the present invention.

[0122] Although not shown, the electronic device includes a processor that can perform various appropriate operations and processes based on programs and / or data stored in a read-only memory (ROM) or programs and / or data loaded from a storage portion into a random access memory (RAM). The processor can be a multi-core processor or can include multiple processors. In some embodiments, the processor can include a general-purpose main processor and one or more special coprocessors, such as a central processing unit, a graphics processing unit (GPU), a neural network processor (NPU), a digital signal processor (DSP), etc. Various programs and data required for device operation are also stored in the RAM. The processor, ROM, and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0123] The processor and memory are used together to execute the program / instructions stored in the memory. When the program / instructions are executed by the computer, the methods, steps or functions described in the above embodiments can be implemented.

[0124] Although not shown, an embodiment of the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the flange connection tightness prediction method in the embodiment of the present invention is implemented.

[0125] Computer-readable storage media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0126] Although not shown, an embodiment of the present invention further provides a computer program product, including: a computer program / instruction, which, when executed by a processor, implements the flange connection tightness prediction method in an embodiment of the present invention.

[0127] The above disclosure is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or modifications within the technical scope disclosed in the present invention, and they should all be covered by the scope of protection of the present invention.

Claims

1. A method for predicting flange connection tightness, characterized in that: The prediction method comprises: Step S1: Obtaining the flange joint gap at different sampling times; Step S2: constructing a gap deviation sequence based on the flange joint surface gaps at different sampling moments; Step S3: fitting the gap deviation sequence using an ARIMA model to obtain a gap deviation prediction model; Step S4: using the gap deviation prediction model to predict the gap deviation at the next sampling moment, and calculating the flange joint surface gap at the next sampling moment based on the flange joint surface gap at the current sampling moment and the gap deviation at the next sampling moment; Step S5: Repeat step S4 to calculate the flange joint surface clearance at multiple future sampling moments in a rolling manner; The gap threshold range is dynamically updated based on the temperature and load corresponding to the flange joint gap at the current sampling moment. The specific update formula is: ； ； ； in, 、 Respectively represent the lower limit of the gap threshold range at the tth and t-1th sampling moments, 、 Respectively represent the upper limit of the gap threshold range at the tth and t-1th sampling moments, Indicates the flange joint surface clearance correction value, 、 Both represent smoothing coefficients, Indicates the reference value of the flange joint gap. 、 Both represent coupling coefficients, represents the temperature at the t-th sampling moment, Indicates the temperature reference value, represents the load at the tth sampling moment, Indicates the load reference value; Multi-level warnings are issued based on the predicted flange joint surface gap and gap threshold range at the next sampling moment.

2. The flange connection tightness prediction method according to claim 1, characterized in that: In step S1, obtaining the flange joint surface gap at different sampling times specifically includes: Step S1.1: collecting temperature, load and flange joint gap at different sampling times; Step S1.2: Divide the flange joint surface gap at different sampling moments into sub-segments according to the set time period; Step S1.3: Fitting and detecting key inflection points of the flange joint surface gap at different sampling times in each sub-segment, obtaining and saving the gap fitting function and key inflection points of each sub-segment; Step S1.4: Reconstruct the flange joint surface gap at different sampling moments within the required time period based on the gap fitting function and key inflection points of each sub-segment.

3. The flange connection tightness prediction method according to claim 2, characterized in that: In step S1.3, the key inflection point detection is specifically as follows: like ,but Recorded as the key inflection point; in, represents the flange joint gap at the t-th sampling moment, represents the fitting value of the flange joint gap at the t-th sampling moment, Indicates the inflection point threshold.

4. The flange connection tightness prediction method according to claim 1, characterized in that: In step S2, a gap deviation sequence is constructed based on the flange joint surface gaps at different sampling moments, specifically including: Step S2.1: Perform a primary difference on the flange joint surface gaps at two adjacent sampling moments to obtain the gap deviation after the primary difference, and then obtain a gap deviation sequence; wherein the primary difference formula of the gap deviation is: ； in, It represents the gap deviation after a difference at the t-th sampling moment, represents the flange joint gap at the t-th sampling moment, represents the flange joint surface clearance at the t-1th sampling time, and N-1 represents the number of known sampling times; Step S2.2: Perform a stationarity test on the gap deviation sequence. If the gap deviation sequence is stationary, obtain the gap deviation sequence; if the gap deviation sequence is non-stationary, proceed to step S2.3; Step S2.3: Perform a quadratic difference on each gap deviation in the gap deviation sequence to obtain the gap deviation after the quadratic difference, and then obtain the gap deviation sequence, and proceed to step S2.2; wherein, the quadratic difference formula of the gap deviation is: ； in, It represents the gap deviation after the second difference at the t-th sampling moment, Indicates the gap deviation after a difference at the t-1th sampling time.

5. The flange connection tightness prediction method according to claim 1, characterized in that: In step S3, the gap deviation sequence is fitted using an ARIMA model, specifically including: Based on the ARIMA model, a gap deviation fitting model is constructed. The specific expression is: ； in, 、 、 Respectively represent the gap deviation at the t, t-1, and t-2 sampling moments, 、 are autoregressive coefficients, represents the sliding mean coefficient, 、 Represent the white noise error terms at the tth and t-1th sampling moments respectively; Initializing the autoregressive coefficient and the sliding average coefficient, and estimating the autoregressive coefficient and the sliding average coefficient according to the gap deviation sequence to obtain estimated values of the autoregressive coefficient and the sliding average coefficient; Substituting the estimated values of the autoregressive coefficient and the sliding average coefficient into the gap deviation fitting model, a gap deviation prediction model is obtained.

6. The flange connection tightness prediction method according to any one of claims 1 to 5, characterized in that: The multi-level warning is performed based on the predicted flange joint surface gap and gap threshold range at the next sampling moment, specifically including: like , a level one warning is issued; like And the duration reaches , a second-level warning is issued; like , a level three warning will be issued; in, represents the flange joint gap at the t-th sampling moment, Indicates the predicted flange joint surface clearance at the next sampling moment, Indicates the allowable gap variation value, Indicates the lower limit of the gap threshold range, Indicates the upper limit of the gap threshold range, Indicates the time threshold.

7. An electronic device comprising a memory, a processor, and a computer program / instruction stored in the memory, characterized in that: The processor executes the computer program / instructions to implement the flange connection tightness prediction method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the flange connection tightness prediction method according to any one of claims 1 to 6 is implemented.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the flange connection tightness prediction method according to any one of claims 1 to 6 is implemented.

Images