Pipeline pressure prediction method, device and equipment integrating pressure and vibration information

Abstract

The present disclosure provides a pipeline pressure prediction method, device and equipment that fuses pressure and vibration information, and relates to the field of data processing technology. The method includes: obtaining vibration signals and pressure signals corresponding to multiple detection points on the target pipeline; preprocessing the vibration signals and pressure signals to obtain a vibration matrix and pressure vector aligned in time; performing principal component analysis on the vibration matrix, and taking the principal component components corresponding to the maximum eigenvalue of a preset number as the fusion vector; taking the fusion vector and pressure vector as the input of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point. The technical solution in the present disclosure can improve the accuracy and efficiency of pipeline pressure prediction.

Description

Background Art

[0002] A pressurized pipeline is a system in which the fluid pressure in the pipeline is higher than the external environmental pressure. This pipeline system is widely used to transport liquids (such as water, oil, and chemical products) and gases (such as natural gas and compressed air). In pressurized pipeline systems, monitoring and diagnostic technologies have been widely used to monitor the health of pipelines in real time, especially by extracting pressure signals to reflect the working status of pipelines.

[0003] Although traditional pipeline pressure monitoring technology can achieve intelligent monitoring and diagnosis, its limitation is that it can only provide post-event warning and maintenance guidance, and cannot respond promptly when risks occur in the early stages, resulting in substantial damage once a problem occurs. Therefore, it is necessary to predict the pressure anomalies that the pipeline may face in advance, so that measures can be taken immediately when early risks appear. However, most of the existing pipeline pressure prediction technologies focus on the analysis of the pressure signal itself, and the correlation and regularity between various monitoring data such as vibration and pressure are insufficient. There is little research on pressure prediction, and the accuracy and computational efficiency of existing prediction models need to be further improved.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the embodiments of the present disclosure is to provide a pipeline pressure prediction method that integrates pressure and vibration information, a pipeline pressure prediction device that integrates pressure and vibration information, an electronic device, and a computer-readable storage medium, thereby improving the accuracy and efficiency of pipeline pressure prediction at least to a certain extent.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to a first aspect of an embodiment of the present disclosure, a pipeline pressure prediction method that integrates pressure and vibration information is provided, including: obtaining vibration signals and pressure signals corresponding to multiple detection points on a target pipeline; preprocessing the vibration signals and the pressure signals to obtain a vibration matrix and pressure vector that are aligned in time; performing principal component analysis on the vibration matrix, and taking principal component components corresponding to a preset number of maximum eigenvalues ​​as fusion vectors; taking the fusion vector and the pressure vector as inputs of a pipeline pressure prediction model to obtain a predicted pressure corresponding to a pressure prediction point.

[0008] In some example embodiments of the present disclosure, based on the aforementioned scheme, the process of constructing the pipeline pressure prediction model includes: acquiring multiple historical vibration signals and historical pressure signals of the detection point at different times, and multiple real pressure signals of the prediction point at different times; generating sample data according to the historical vibration signals, the historical pressure signals and the real pressure signals; and training a long short-term memory network based on the sample data to obtain the pipeline pressure prediction model.

[0009] In some example embodiments of the present disclosure, based on the aforementioned scheme, the generating of sample data according to the historical vibration signal, the historical pressure signal and the real pressure signal includes: dividing the historical vibration signal, the historical pressure signal and the real pressure signal based on a preset time step to obtain multiple groups of divided data; preprocessing the historical vibration signal, the historical pressure signal and the real pressure signal in the divided data to obtain a historical vibration matrix and a historical pressure vector aligned in time; performing principal component analysis on the historical vibration matrix, and taking the principal component component corresponding to the preset number of maximum eigenvalues ​​as a historical fusion vector; generating sample data based on the historical fusion vector, the historical pressure vector and the real pressure signal.

[0010] In some example embodiments of the present disclosure, based on the aforementioned scheme, the obtaining of vibration signals and pressure signals corresponding to multiple detection points on the target pipeline includes: setting multiple vibration detection points and at least one pressure detection point in the vicinity of the pressure prediction point on the target pipeline; obtaining acceleration data corresponding to the vibration detection points, and using the acceleration data as the vibration signal; obtaining pressure data corresponding to the pressure detection points, and using the pressure data as the vibration signal.

[0011] In some example embodiments of the present disclosure, based on the aforementioned scheme, a plurality of vibration detection points and at least one pressure detection point are set in the vicinity of the pressure prediction point on the target pipeline, including: a plurality of vibration detection points are set on a plurality of pipe fittings of the target pipeline, and the pressure prediction point is located between the pipe fittings; at least one pressure detection point is set on the pipe body of the target pipeline, and the pressure prediction point is located between the pressure detection points.

[0012] In some example embodiments of the present disclosure, based on the aforementioned scheme, the vibration signal and the pressure signal are preprocessed to obtain a vibration matrix and pressure vector that are aligned in time, including: respectively obtaining the mean and variance corresponding to the vibration signal and the pressure signal; and preprocessing the vibration signal and the pressure signal based on the standard score conversion method according to the mean and variance to obtain a vibration matrix and pressure vector that are aligned in time.

[0013] In some example embodiments of the present disclosure, based on the aforementioned solution, the pipeline pressure prediction method integrating pressure and vibration information further includes: when the predicted pressure is greater than a pressure threshold, issuing a pressure warning.

[0014] In some example embodiments of the present disclosure, based on the aforementioned scheme, a plurality of pressure prediction points are set on the target pipeline, and the pipeline pressure prediction method integrating pressure and vibration information further includes: establishing corresponding pipeline pressure prediction models for the plurality of pressure prediction points, respectively, to realize pressure prediction for the plurality of pressure prediction points.

[0015] According to a second aspect of an embodiment of the present disclosure, a pipeline pressure prediction device that integrates pressure and vibration information is provided, including: a signal acquisition module, which can be used to acquire vibration signals and pressure signals corresponding to multiple detection points on a target pipeline; a signal processing module, which can be used to preprocess the vibration signal and the pressure signal to obtain a vibration matrix and pressure vector that are aligned in time; a matrix analysis module, which can be used to perform principal component analysis on the vibration matrix, and use the principal component components corresponding to a preset number of maximum eigenvalues ​​as fusion vectors; a pressure prediction module, which can be used to use the fusion vector and the pressure vector as inputs of a pipeline pressure prediction model to obtain a predicted pressure corresponding to a pressure prediction point.

[0016] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the pipeline pressure prediction method integrating pressure and vibration information described above is implemented.

[0017] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the pipeline pressure prediction method based on the above-mentioned fusion pressure and vibration information is implemented.

[0018] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0019] The pipeline pressure prediction method that integrates pressure and vibration information in the example embodiment of the present disclosure, on the one hand, obtains the vibration signal and pressure signal on the pipeline. The vibration signal can reflect the dynamic changes of the fluid state inside the pipeline, while the pressure signal is directly related to the pressure level of the pipeline. The combination of two different types of data enhances the richness of the input information of the pipeline prediction model, enables the model to more comprehensively understand the physical state of the pipeline, and improves the accuracy of the prediction; on the other hand, the principal component analysis is performed on the vibration signal, and the principal component is extracted as the fusion vector, which effectively reduces the data dimension and eliminates redundant information, while also retaining the key features in the signal, improving data processing efficiency and prediction efficiency; on the other hand, the vibration signal and the pressure signal are preprocessed to ensure the temporal alignment of the two, laying a foundation for subsequent data fusion and analysis research on pressurized pipelines and providing an effective idea.

[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0022] Figure 1 A schematic diagram showing an exemplary system architecture of a pipeline pressure prediction method that integrates pressure and vibration information, to which an embodiment of the present disclosure can be applied, is shown.

[0023] Figure 2 A flow chart of a pipeline pressure prediction method integrating pressure and vibration information according to some embodiments of the present disclosure is schematically shown.

[0024] Figure 3 A schematic diagram of setting positions of vibration detection points and pressure detection points according to some embodiments of the present disclosure is schematically shown.

[0025] Figure 4 Another schematic diagram of setting positions of vibration detection points and pressure detection points according to some embodiments of the present disclosure is schematically shown.

[0026] Figure 5 A schematic diagram of the setting positions of another vibration detection point and a pressure detection point according to some embodiments of the present disclosure is schematically shown.

[0027] Figure 6A schematic diagram of a PCA fusion process according to some embodiments of the present disclosure is schematically shown.

[0028] Figure 7 A schematic diagram of the structure of a pipeline pressure prediction model according to some embodiments of the present disclosure is schematically shown.

[0029] Figure 8 A schematic diagram schematically shows the composition of sample data of a pipeline pressure prediction model according to some embodiments of the present disclosure.

[0030] Fig. 9 A block diagram of a pipeline pressure prediction device integrating pressure and vibration information according to some embodiments of the present disclosure is schematically shown.

[0031] Fig.10 A schematic diagram of the structure of a computer system of an electronic device according to some embodiments of the present disclosure is schematically shown.

[0032] Fig.11 A schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure is schematically shown.

[0033] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0034] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this specification. Instead, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0035] The terms used in this specification are for the purpose of describing specific embodiments only and are not intended to limit this specification. The singular forms "a", "the" and "the" used in this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0036] It should be understood that although the terms first, second, third, etc. may be used in this specification to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this specification, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0037] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0038] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the present disclosure.

[0039] Furthermore, the drawings are only schematic illustrations and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0040] At present, monitoring and diagnosis methods based on pressure signals have been widely used in various types of pressurized water pipelines and related pumping stations. Such methods usually extract signals from one or more pressure measurement points, establish a mapping relationship between signal characteristic values ​​and pipeline operation status, and realize intelligent monitoring and diagnosis functions. However, monitoring and diagnosis technology also has obvious defects. It can only carry out "post-warning" and "post-maintenance" of pipelines based on monitoring information, and cannot timely judge the early risks of pipelines. When problems have occurred, substantial losses have already occurred. Therefore, it is necessary to predict the pressure anomalies that the pipeline may face in advance, so that immediate measures can be taken when early risks appear.

[0041] However, in the construction of prediction models, traditional methods mostly use statistical models, such as autoregressive moving average models, which can predict single-series pressure signals, but are not very effective in processing non-stationary signals or complex signal patterns. With the development of machine learning technology, shallow neural networks such as support vector machines and back propagation neural networks have been introduced, which can better handle regression problems, but the model's generalization ability is limited when faced with large-scale, high-noise, and variable pipeline data.

[0042] In addition, the relevant pipeline pressure prediction methods mostly focus on the analysis of the pressure signal itself, ignoring the close connection between the pipeline vibration characteristics and the internal pressure state, such as the resonance effect of external vibration on the pipeline, the impact of vibration on pressure measurement equipment, and the vibration mechanism caused by special phenomena such as water hammer. These factors have seriously affected the accuracy of the prediction method. When using deep learning models to predict pipeline pressure, a large amount of data needs to be processed, the calculation efficiency is low, and the utilization rate of multi-source data such as pressure vibration is insufficient, which leads to insufficient accuracy and efficiency of pipeline pressure prediction.

[0043] In order to solve all or part of the technical problems in the above-mentioned related technologies, in an exemplary embodiment of the present disclosure, a pipeline pressure prediction method integrating pressure and vibration information is firstly provided. Figure 1 A schematic diagram showing an exemplary system architecture of a pipeline pressure prediction method that integrates pressure and vibration information, to which an embodiment of the present disclosure can be applied, is shown.

[0044] like Figure 1 As shown, the system architecture 100 may include a pipeline 101, one or more of a vibration detection device 102 and a pressure detection device 103, a network 104 and a server 105. The vibration detection device 102 may be deployed on the plane area of ​​the connecting components such as flanges and joints of the pipeline, and the pressure detection device 103 may be deployed on the pipeline 101. The network 104 is used to provide a medium for a communication link between the vibration detection device 102, the pressure detection device 103 and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or optical fiber cables, etc. The vibration detection device may be various devices for collecting vibration signals, including but not limited to acceleration sensors, laser Doppler vibrometers, etc. The pressure detection device is various devices for collecting pressure signals, including but not limited to pressure sensors, piezoelectric materials, etc. Among them, the vibration detection device 102 and the pressure detection device 103 may be set on pipe fittings on the pipeline 101 or other suitable locations such as the outer wall of the pipe body. It should be understood that Figure 1 The number of vibration detection devices, pressure detection devices, networks, servers and pipelines in the embodiment is only for illustration. Any number of vibration detection devices, pressure detection devices, networks, servers and pipelines may be provided according to the implementation requirements.

[0045] The pipeline pressure prediction method for integrating pressure and vibration information provided in the embodiment of the present disclosure can be executed based on the above-mentioned detection device and server. For example, the vibration signal and pressure signal on the pipeline 101 can be obtained by the preset vibration detection device 102 and the pressure detection device 103, respectively, and the vibration signal and pressure signal are sent to the server 105 through the network 104. After receiving the vibration signal and the pressure signal, the server 105 pre-processes the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time, and then performs principal component analysis on the vibration matrix, and uses the principal component components corresponding to the maximum eigenvalue of the preset number as the fusion vector, and uses the fusion vector and the pressure vector as the input of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point. Finally, when the predicted pressure is greater than the pressure threshold, a pressure warning is issued. However, it is easy for those skilled in the art to understand that the above-mentioned application scenario is only used for example, and is not limited to this in this exemplary embodiment.

[0046] Figure 2 The following schematically shows a flow chart of a pipeline pressure prediction method integrating pressure and vibration information according to some embodiments of the present disclosure. Figure 2 As shown, the pipeline pressure prediction method integrating pressure and vibration information may include the following steps:

[0047] Step S210, obtaining vibration signals and pressure signals corresponding to multiple detection points on the target pipeline;

[0048] Step S220, preprocessing the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time;

[0049] Step S230, performing principal component analysis on the vibration matrix, and taking the principal component components corresponding to the preset number of maximum eigenvalues ​​as fusion vectors;

[0050] Step S240: Using the fusion vector and the pressure vector as inputs of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point.

[0051] According to the pipeline pressure prediction method that integrates pressure and vibration information in the above-mentioned example embodiment, on the one hand, by acquiring the vibration signal and pressure signal on the pipeline, the vibration signal can reflect the dynamic changes of the fluid state inside the pipeline, and the pressure signal is directly related to the pressure level of the pipeline. The combination of two different types of data enhances the richness of the input information of the pipeline prediction model, enables the model to understand the physical state of the pipeline more comprehensively, and improves the accuracy of the prediction; on the other hand, the principal component analysis is performed on the vibration signal, and the principal component is extracted as the fusion vector, which effectively reduces the data dimension and eliminates redundant information, while also retaining the key features in the signal, improving data processing efficiency and prediction efficiency; on the other hand, the vibration signal and the pressure signal are preprocessed to ensure the temporal alignment of the two, which lays the foundation for the subsequent data fusion and analysis research of pressurized flow pipelines and provides an effective idea.

[0052] The pipeline pressure prediction method integrating pressure and vibration information in the above exemplary embodiment will be further described below.

[0053] In step S210, vibration signals and pressure signals corresponding to a plurality of detection points on the target pipeline are obtained.

[0054] Among them, by obtaining vibration signals and pressure signals corresponding to multiple detection points, vibration and pressure data covering multiple pipeline parts can be collected, rather than just limited to a local area, thereby improving the comprehensiveness of the data. Moreover, through vibration signals and pressure signals, the pipeline can be comprehensively analyzed from the perspectives of mechanics and fluid mechanics, improving the accuracy of pressure prediction.

[0055] Specifically, the target pipeline may represent a pressurized pipeline selected for pressure status monitoring, which may be a variety of pipelines, such as water supply pipelines, oil pipelines, natural gas pipelines, process pipelines in chemical plants, thermal pipelines, or hydraulic or fuel pipelines on ships and airplanes, etc. The target pipeline may be composed of straight pipe bodies, elbows, valves, flanges and other components.

[0056] The vibration signal may represent an electrical signal converted from the mechanical vibration generated by the pipeline and its components during operation, and such vibration may be caused by a variety of factors, such as fluid flow inside the pipeline, wear of mechanical components, structural defects, thermal expansion and contraction, etc. For example, the vibration signal may be acquired by an acceleration sensor installed on the pipeline.

[0057] The pressure signal may represent an electrical signal or digital signal that can reflect the pressure change inside the pipeline. The pressure signal can reflect the state of the fluid inside the pipeline, such as flow rate, flow rate, pressure fluctuation, etc. It is an important indicator for monitoring whether the pipeline is working properly. The abnormality of the pressure signal may indicate that the pipeline is blocked or leaking or other faults. For example, the pressure signal can be obtained by a pressure sensor installed on the pipeline.

[0058] In some embodiments, obtaining vibration signals and pressure signals corresponding to multiple detection points on the target pipeline specifically includes the following steps: setting multiple vibration detection points and at least one pressure detection point in the vicinity of the pressure prediction point on the target pipeline; obtaining acceleration data corresponding to the vibration detection point, and using the acceleration data as the vibration signal; obtaining pressure data corresponding to the pressure detection point, and using the pressure data as the vibration signal. The selected pressure prediction point needs to be selected from the pressure detection points where the pressure detection device is set, and the pressure detection point here can include the pressure prediction point itself.

[0059] Among them, the pressure prediction point can represent the position selected on the target pipeline where pressure prediction is required, but the pressure detection point needs to be equipped with a pressure detection device. The pressure prediction point can be selected from the pressure detection points where pressure sensors are deployed. The acquired data of the pressure prediction point is the historical pressure signal. The pressure signal of the prediction point in the future can be predicted based on the multi-channel historical vibration signal value of the area near a certain pressure prediction point in the previous period and the historical pressure signal of the point. The pressure prediction point can be determined based on factors such as pipeline layout, fluid properties, and historical fault records, but the location of the pressure prediction point needs to be equipped with a pressure sensor, otherwise the result cannot be verified and it is difficult to calculate accurately. The pressure prediction point is mainly set in key areas on the pipeline wall, such as elbows, tees, pipelines near valves, or pipelines near key interfaces. The adjacent area can represent the area on the target pipeline within a certain range around the pressure prediction point that may have a direct impact on the pressure change at the prediction point. The acceleration data can represent the vibration intensity information measured by the acceleration detection device installed on the pipeline, which can reflect the amplitude and frequency of pipeline vibration and is an important indicator for evaluating the structural integrity and operating status of the pipeline. The pressure data may represent the pressure information measured by a pressure detection device installed on the pipeline, which may reflect the pressure state of the fluid inside the pipeline.

[0060] Furthermore, by setting up multiple detection points in the vicinity of the pressure prediction point, accurate monitoring of key areas can be achieved. This helps capture the vibration signals and pressure signals that are most directly related to the pressure changes at the pressure prediction point, improves the pertinence and effectiveness of data collection, and thus more accurately predicts and analyzes pressure change trends. Generally speaking, the more data sources that can be relied upon during prediction, the higher the prediction precision and accuracy.

[0061] In some embodiments, multiple vibration detection points and at least one pressure detection point are set in the vicinity of the pressure prediction point on the target pipeline, specifically including the following steps: multiple vibration detection points are set on multiple pipe fittings of the target pipeline, and the pressure prediction points are located between the pipe fittings; at least one pressure detection point is set on the pipe body of the target pipeline, and the pressure prediction point is located between the pressure detection points. In addition, in some embodiments, the pressure prediction point can also be located on one side of the pipe fitting, and the pressure prediction point can be located on one side of the pressure detection point.

[0062] Among them, setting vibration detection points on pipe fittings can specifically monitor areas in the pipeline that are prone to vibration, such as flanges, elbows, tees, valves and other pipe fittings. These parts are often more prone to abnormal vibration due to fluid dynamic effects or structural characteristics. By setting vibration detection points in these special parts, potential fault signals can be captured earlier and accurate monitoring of the health status of the pipeline can be achieved. Setting multiple pressure detection points on the pipe body and ensuring that the pressure prediction point is located between these detection points can make the acquired pressure data more representative. At the same time, these pipe fittings usually have obvious planes, which is conducive to the deployment of vibration detection devices or related sensors.

[0063] Specifically, in some embodiments, reference Figure 3 As shown, the pressure prediction point is located between the pipe fittings, and the pressure prediction point is located between the pressure detection points. On the pressurized pipeline 10, a predicted pressure point 20 is set, and the front flange 11 of the measuring point and the rear flange 12 of the measuring point are respectively located on both sides of the predicted pressure point 20, and the water flow direction is set to flow from the front flange 11 of the measuring point to the rear flange 12 of the measuring point. Four vibration detection points are respectively set on the front flange 11 of the measuring point and the rear flange 12 of the measuring point, which are respectively located above and below the pipeline, specifically including: a horizontal measuring point 31 above the front flange and a vertical measuring point 32 above the front flange, a horizontal measuring point 33 below the front flange and a vertical measuring point 34 below the front flange, a horizontal measuring point 41 above the rear flange and a vertical measuring point 42 above the rear flange, a horizontal measuring point 43 below the rear flange and a vertical measuring point 44 below the rear flange. The surface near the flange is relatively smooth and has a small curvature, which is easy to implement sensor deployment. In addition, a first pressure detection point 21 and a second pressure detection point 22 are respectively set on both sides of the pressure prediction point. The vibration detection point is provided with an acceleration sensor, and the pressure detection point is provided with a pressure sensor, so that 2 groups of pressure signals and 8 groups of vibration signals near the pressure prediction point can be obtained.

[0064] Optionally, when the first pressure detection point 21 or the second pressure detection point 22 cannot be met, the pressure signal of the pressure prediction point 20 can also be used as the pressure detection point. At this time, 1 to 2 groups of pressure signals and 8 groups of vibration signals near the pressure prediction point can be obtained.

[0065] In some embodiments, reference Figure 4 As shown, a predicted pressure point 20 is provided on the pressurized pipeline 10, and the front end flange 11 of the measuring point is located on the left side of the predicted pressure point 20. The water flow direction is set to flow from the left side to the right side of the front end flange 11 of the measuring point. Four vibration detection points are provided on the front end flange 11 of the measuring point, which are respectively located above and below the pipeline, specifically including: a horizontal measuring point 31 above the front end flange and a vertical measuring point 32 above the front end flange, a horizontal measuring point 33 below the front end flange and a vertical measuring point 34 below the front end flange. The surface near the flange is relatively smooth and has a small curvature, which is easy to implement sensor deployment. The first pressure detection point 21 is located on the left side of the front end flange 11 of the measuring point. The vibration detection point is provided with an acceleration sensor, and the first pressure detection point 21 and the pressure prediction point 20 are provided with pressure sensors, so that two groups of pressure signals and four groups of vibration signals near the pressure prediction point can be obtained.

[0066] Optionally, when the first pressure detection point 21 cannot be met, the pressure signal of the pressure prediction point 20 can also be used as the pressure detection point. In this case, one group of pressure signals and four groups of vibration signals near the pressure prediction point can be obtained.

[0067] In some embodiments, reference Figure 5 As shown, a predicted pressure point 20 is provided on the pressurized pipeline 10, and the rear end flange 12 of the measuring point is located on the right side of the predicted pressure point 20. The water flow direction is set to flow from the left side to the right side of the rear end flange 12 of the measuring point. Four vibration detection points are provided on the rear end flange 12 of the measuring point, which are respectively located above and below the pipeline, specifically including: a horizontal measuring point 41 above the rear end flange and a vertical measuring point 42 above the rear end flange, a horizontal measuring point 43 below the rear end flange and a vertical measuring point 44 below the rear end flange. The surface near the flange is relatively smooth and has a small curvature, which is easy to implement sensor deployment. The first pressure detection point 21 is located on the right side of the rear end flange 12 of the measuring point. The vibration detection point is provided with an acceleration sensor, and the first pressure detection point 21 and the pressure prediction point 20 are provided with pressure sensors, so that two groups of pressure signals and four groups of vibration signals near the pressure prediction point can be obtained.

[0068] Optionally, when the first pressure detection point 21 cannot be met, the pressure signal of the pressure prediction point 20 can also be used as the pressure detection point. In this case, one group of pressure signals and four groups of vibration signals near the pressure prediction point can be obtained.

[0069] In summary, in other embodiments of the present disclosure, the number of vibration detection points may be greater than or equal to 4, and the number of pressure detection points may be greater than or equal to 1, wherein one pressure detection point may also be the pressure prediction point itself.

[0070] In step S220, the vibration signal and the pressure signal are preprocessed to obtain a vibration matrix and a pressure vector that are aligned in time.

[0071] The vibration matrix can represent a two-dimensional array consisting of vibration signals recorded by multiple vibration sensors in the same time interval. Each row can represent the vibration data of a sensor in a period of time, and each column can represent the vibration data of all sensors at a certain moment. The pressure vector can represent a one-dimensional array consisting of pressure values ​​recorded by the pressure sensor in a time series, and each element corresponds to the pressure data at a point in time.

[0072] In some embodiments, the vibration signal and the pressure signal are preprocessed to obtain a vibration matrix and a pressure vector that are aligned in time, specifically including the following steps: respectively obtaining the mean and variance corresponding to the vibration signal and the pressure signal; based on the mean and the variance, the vibration signal and the pressure signal are preprocessed based on the standard fractional conversion method to obtain a vibration matrix and a pressure vector that are aligned in time.

[0073] Among them, the standard score conversion can also be called Z-score standardization or Z-score conversion, which can be used to standardize a set of data to a standard normal distribution with zero mean and unit variance. Specifically, the vibration signal and pressure signal are centered and standardized by the standard score conversion method, and the calculation process is shown in formula (1):

[0074] (1)

[0075] in, Indicates vibration signal or pressure signal, subscript t Indicates a point in time, represents the mean value of the signal. represents the variance of the signal. The data is after preprocessing.

[0076] Furthermore, by preprocessing the vibration signal and pressure signal through the standard score conversion method, the signal offset can be eliminated, ensuring the comparability between different signals or features, thereby improving the convergence of the subsequent pressure prediction model and avoiding the gradient descent problem caused by the difference in feature scales.

[0077] In step S230, principal component analysis is performed on the vibration matrix, and principal component components corresponding to a preset number of maximum eigenvalues ​​are used as fusion vectors.

[0078] Among them, principal component analysis can represent a method for identifying patterns in data and reducing the dimension of data while retaining data features. The eigenvalue can represent the numerical value corresponding to the eigenvector in the principal component analysis, which can represent the variance along the direction of the eigenvector. A larger eigenvalue means that the principal component explains more changes in the data. The principal component component can represent a new orthogonal variable calculated by PCA, which is a linear combination of the original variables. The fusion vector can be composed of eigenvectors corresponding to a preset number of eigenvalues. For example, the first three principal component components with the highest contribution rate, that is, the largest eigenvalue, can be selected as the fusion signal of the vibration signal.

[0079] Figure 6 The figure shows a schematic diagram of the PCA fusion process of data, which contains n historical vibration signals and m historical pressure signals. Specifically, the historical vibration signals of n vibration detection points from time t1 to time t2 are obtained, and the vibration matrix is ​​obtained based on the historical vibration signals. The vibration matrix is ​​subjected to principal component analysis, and the principal component components corresponding to the top 3 largest eigenvalues ​​are used as fusion vectors. The three principal component components are respectively , and The historical pressure signals of the m vibration detection points from time t1 to time t2 are obtained, and the pressure vector is obtained according to the historical pressure signals.

[0080] Furthermore, principal component analysis can effectively eliminate redundant information and reduce the amount of computational data, and data fusion can reduce the difficulty of signal prediction.

[0081] Specifically, the vibration signals that have been decentralized and standardized are combined into a matrix, and the principal component analysis method is used to analyze them to obtain the principal component components. The matrix , after standardization preprocessing according to formula (1), the matrix . After principal component analysis is performed using the following formula (2), It can be broken down into:

[0082] (2)

[0083] Where T represents the principal component score matrix, represents the principal component loading matrix, represents the pivot score vector, In step S240, the fusion vector and the pressure vector are used as inputs of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point.

[0084] Specifically, the principal component components corresponding to the top three largest eigenvalues ​​when performing principal component analysis on the vibration matrix can be selected as the fusion vector. The three principal component components are , and At the same time, m pressure detection points are set on the target pipeline, and the pressure signals obtained are recorded as , … .

[0085] The fusion vector and the pressure vector are input into the pipeline pressure prediction model. For example, the pipeline pressure prediction model predicts the pressure data corresponding to 10 pressure prediction points based on the multi-source data of 100 consecutive times, that is, the fusion vector and the pressure vector. The prediction time is recorded as T, and the pressure data corresponding to the pressure prediction points at the 10 time points to be predicted can be recorded as ,predict The data used can be: fusion vector of 100 consecutive time points , and , and the pressure vector , , .in, Represents the pressure vector corresponding to the pressure prediction point.

[0086] Furthermore, using the fusion vector and pressure vector as the input of the pipeline pressure prediction model can not only combine the fusion vector and the pressure vector to capture the potential correlation between vibration and pressure and enhance the prediction performance of the model, but also integrate multiple types of feature data to more comprehensively reflect the status of the pipeline and improve the accuracy and reliability of the prediction.

[0087] In some embodiments, after the predicted pressure corresponding to the pressure prediction point is obtained, when the predicted pressure is greater than the pressure threshold, a pressure warning is issued.

[0088] Specifically, the predicted pressure is compared with the pre-set pressure threshold to determine whether to trigger the early warning mechanism. When the predicted pressure exceeds the pressure threshold, the early warning process is immediately initiated. For example, warnings can be issued to relevant personnel through sound, light, text message, email, etc. At the same time, the occurrence time of the early warning, the predicted pressure and the comparison with the pressure threshold are automatically recorded to provide data support for subsequent analysis and improvement. In addition, the automatic control device can also perform early warning response operations, such as adjusting the valve opening, starting backup equipment, etc., to reduce pressure or restore normal operation.

[0089] In some embodiments, the process of constructing a pipeline pressure prediction model specifically includes the following steps: obtaining multiple historical vibration signals and historical pressure signals of the detection point at different times, and multiple real pressure signals of the prediction point at different times; generating sample data based on the historical vibration signals, historical pressure signals and real pressure signals; and training a long short-term memory network based on the sample data to obtain a pipeline pressure prediction model.

[0090] Among them, the long short-term memory network is a special type of recurrent neural network that can effectively capture long-term dependencies when processing long sequence data. The pipeline pressure prediction model based on the long short-term memory network can accurately predict the pressure data of the pressure prediction point.

[0091] The network structure of the pipeline pressure prediction model can be as follows: Figure 7 As shown in the figure, the model inputs 600 parameters at each moment and outputs 10 parameters. The pipeline pressure prediction model consists of an input layer, four LSTM networks, two Dropout layers, a fully connected layer, and an output layer. All parameters are initialized before model training. The model training is based on the training set and the validation set, and the model performance is evaluated based on the test set. During the model training process, the grid search cross-validation method can be used to optimize parameters such as the training set size, batch, and number of iterations to improve the prediction effect of the model. In the input layer, the input data dimension is (100, 6), and each set of input data consists of 100 time steps and 6 signal vectors, including 3 vibration principal components and 3 pressure vectors; the output data dimension is also (100, 6). The input data dimension of the first LSTM layer is (100, 6), and the output data dimension is (100, 300), which contains 50 hidden units; the input data dimension of the second LSTM layer is (100, 300), and the output data dimension is (100, 3000), which contains 10 hidden units; the input data dimension of the third and fourth LSTM layers are both (100, 3000), and the output data dimension is (100, 3000); the retention rate of the first Dropout layer is set to 0.4, the retention rate of the second Dropout layer is set to 0.2, and the data dimension of the output layer is (10, 1).

[0092] In some embodiments, based on the sample data of historical vibration signals, historical pressure signals and real pressure signals, the following steps are specifically included: dividing the historical vibration signals, historical pressure signals and real pressure signals based on a preset time step to obtain multiple groups of divided data; preprocessing the historical vibration signals, historical pressure signals and real pressure signals in the divided data to obtain a historical vibration matrix and a historical pressure vector aligned in time; performing principal component analysis on the historical vibration matrix, and taking the principal component components corresponding to a preset number of maximum eigenvalues ​​as historical fusion vectors; generating sample data based on the historical fusion vectors, historical pressure vectors and real pressure signals.

[0093] Specifically, refer to Figure 8 As shown, n historical vibration signals corresponding to the vibration detection points from time t1 to time t2, historical pressure signals corresponding to the pressure prediction points, and m historical pressure signals corresponding to the pressure detection points are obtained, and used as the basis for pressure prediction and the input data source for training the pressure prediction model. Then, the real pressure signal corresponding to the pressure prediction point from time t2 to time t3 is used as the prediction target. In addition, the above historical vibration signals, historical pressure signals, and real pressure signals all come from historical records, but the real pressure signal refers to the signal that appears after the historical vibration signal and the historical pressure signal appear, and its data length is shorter than the historical vibration signal and the historical pressure signal; the real pressure signal is connected to the historical pressure signal in time, and is the real data of the historical pressure signal for a period of time; from the logic of pressure prediction, the established prediction model can predict the real pressure signal based on the historical vibration signal and the historical pressure signal (which are converted into fusion vectors and pressure vectors when input into the model).

[0094] Among them, the time step can be used to intercept signal segments according to a certain fixed time interval, thereby forming a series of temporally continuous data segments. For example, the time step can be set to 100 seconds, then from the historical vibration signal and the historical pressure signal, a segment of data is intercepted every 100 seconds as a group of partitioned data, and 10 seconds after 100 seconds of the historical vibration signal is selected as the real pressure signal. The above 100 seconds of historical vibration signal, historical pressure signal and 10 seconds of real pressure signal together constitute a set of input and output data of the prediction model, that is, a set of training data. Based on this, the dynamic characteristics of the signal changing over time can be captured, and the data can be aligned in time, which is convenient for subsequent analysis and modeling. In addition, in the preprocessing stage, the historical vibration signal and the historical pressure signal are converted into matrix and vector forms to ensure that they are aligned in time, so that the starting point and end point of each signal segment on the time axis are the same, which is conducive to maintaining data consistency and subsequent feature extraction. The historical fusion vector and the historical pressure vector are used as input, and the real pressure signal is used as output to train the long short-term memory network to obtain the final pipeline pressure prediction model.

[0095] In addition, the long short-term memory network requires a rich historical data set during the training process. The training set data should include historical data from the long-term operation of the pipeline section. The richness of the data determines the prediction ability of the model. Exemplarily, the sample data should include at least 6,000 groups of data, that is, at least 6,000 sample data at different times are selected. According to the above example, at least 5,900 groups of training data can be formed. Combined with the comprehensive consideration of the model parameters, the model training requirements can basically be met. The effective range of the model prediction is the interval between the maximum and minimum values ​​of the real pressure signal selected in the training set, that is, the effective range is In the interval, p ( t ) represents the real pressure signal corresponding to the pressure prediction point. The data set is divided into training set, validation set and test set, and the division ratio can be 7:1:2. The training set can be used to train or fit the model, the validation set can be used to verify the model parameters, and the test set can be used to evaluate the final performance of the model. In addition, when the flow direction of the liquid in the pipe is reversed, the pipeline pressure prediction model needs to be re-established for prediction, and the original model cannot be applied.

[0096] In some embodiments, a plurality of pressure prediction points are set on the target pipeline, and corresponding pipeline pressure prediction models can be established for the plurality of pressure prediction points respectively to achieve pressure prediction for the plurality of pressure prediction points. In addition, by setting a plurality of pressure prediction points on the target pipeline and establishing an independent pipeline pressure prediction model for each prediction point, the prediction accuracy and reliability can be effectively improved.

[0097] The pipeline pressure prediction method that integrates pressure and vibration information in the above example embodiment, on the one hand, by obtaining the vibration signal and pressure signal on the pipeline, the vibration signal can reflect the dynamic changes of the fluid state inside the pipeline, and the pressure signal is directly related to the pressure level of the pipeline. The combination of two different types of data enhances the richness of the input information of the pipeline prediction model, enables the model to more comprehensively understand the physical state of the pipeline, and improves the accuracy of the prediction; on the other hand, the principal component analysis is performed on the vibration signal, and the principal component components are extracted as the fusion vector, which effectively reduces the data dimension and eliminates redundant information. At the same time, the key features in the signal are retained, which improves the data processing efficiency and the prediction efficiency; on the other hand, the vibration signal and the pressure signal are preprocessed to ensure the temporal alignment of the two, which lays the foundation for the data fusion research of pressurized pipelines and provides an effective idea.

[0098] Furthermore, in the exemplary embodiment of the present disclosure, the vibration signal is used as the supplementary information of the pressure signal, which provides more comprehensive data for pipeline monitoring, improves the diversity and richness of data collection, helps to more accurately evaluate the health status of the pipeline, establishes the association between the vibration signal and the pressure signal, and makes up for the shortcomings of the related technology in relying on a single information source and ignoring pipeline vibration. In addition, the multi-channel vibration signal and the pressure signal are combined to obtain more other factors related to the pressure, which is equivalent to considering the interference of many factors such as the external environment, the stress condition, and the material properties in the model, which lays the foundation for improving the accuracy and robustness of the prediction model. Moreover, by using a deep learning model based on a long short-term memory network designed specifically for pipeline pressure prediction, the long short-term memory network can effectively solve the gradient disappearance and gradient explosion problems when processing long sequence data, and can capture long-distance dependencies in time series data, so it is suitable for application objects such as pipelines that require long-term monitoring and a large number of applications. In addition, preprocessing the data by principal component analysis before prediction can reduce the dimension and complexity of multi-channel sensor data, thereby greatly improving the computational efficiency and reducing the interference of noise and redundant information. In addition, this method is applicable to various types of pressurized water pipelines and can be configured in combination with thresholds and alarm standards in specific engineering scenarios, so that it can implement effective early warning functions for different application environments.

[0099] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0100] Next, in the embodiment of the present disclosure, a pipeline pressure prediction device integrating pressure and vibration information is also provided. Fig. 9 As shown in , a pipeline pressure prediction device 900 integrating pressure and vibration information can be composed of a signal acquisition module 901, a signal processing module 902, a matrix analysis module 903 and a pressure prediction module 904, wherein: the signal acquisition module can be used to obtain vibration signals and pressure signals corresponding to multiple detection points on the target pipeline; the signal processing module can be used to preprocess the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time; the matrix analysis module can be used to perform principal component analysis on the vibration matrix, and use the principal component components corresponding to a preset number of maximum eigenvalues ​​as fusion vectors; the pressure prediction module can be used to use the fusion vector and the pressure vector as inputs of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point.

[0101] It should be noted that the specific details of each part of the above-mentioned pipeline pressure prediction device that integrates pressure and vibration information have been described in detail in the implementation method of the pipeline pressure prediction method that integrates pressure and vibration information. The undisclosed details can be found in the implementation method content of the method part, and will not be repeated here.

[0102] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above-mentioned pipeline pressure prediction method integrating pressure and vibration information is also provided.

[0103] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware embodiments, complete software embodiments (including firmware, microcode, etc.), or embodiments combining hardware and software aspects, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0104] Refer to the following Fig.10 hereinafter describes the electronic device 1000 according to such an embodiment of the present disclosure. Fig.10 The electronic device 1000 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0105] like Fig.10 As shown, the electronic device 1000 is in the form of a general computing device. The components of the electronic device 1000 may include but are not limited to: the at least one processing unit 1010, the at least one storage unit 1020, a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010), and a display unit 1040.

[0106] The storage unit stores program codes, which can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 1010 can perform the following steps: Figure 2 In step S210 shown in the figure, vibration signals and pressure signals corresponding to multiple detection points on the target pipeline are obtained; in step S220, the vibration signals and pressure signals are preprocessed to obtain a vibration matrix and pressure vector aligned in time; in step S230, principal component analysis is performed on the vibration matrix, and principal component components corresponding to a preset number of maximum eigenvalues ​​are used as fusion vectors; in step S240, the fusion vector and pressure vector are used as inputs of the pipeline pressure prediction model to obtain the predicted pressure corresponding to the pressure prediction point.

[0107] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 1021 and / or a cache memory unit 1022 , and may further include a read-only memory unit (ROM) 1023 .

[0108] The storage unit 1020 may also include a program / utility 1024 having a set (at least one) of program modules 1025, such program modules 1025 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0109] Bus 1030 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0110] The electronic device 1000 may also communicate with one or more external devices 1070 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1000, and / or communicate with any device that enables the electronic device 1000 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 1050. In addition, the electronic device 1000 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 1060. As shown, the network adapter 1060 communicates with other modules of the electronic device 1000 through a bus 1030. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0111] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0112] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0113] refer to Fig.11 As shown, a program product 1100 for implementing the pipeline pressure prediction method integrating pressure and vibration information according to an embodiment of the present disclosure is described, which can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus, or a device.

[0114] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0115] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0116] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, electromagnetic waves, etc., or any suitable combination of the foregoing.

[0117] Program code for performing the disclosed operations may be written in any combination of one or more programming languages, including object-oriented programming languages ​​and procedural programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0118] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0119] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0120] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0121] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A pipeline pressure prediction method integrating pressure and vibration information, characterized in that: include: Obtain vibration signals and pressure signals corresponding to multiple detection points on the target pipeline; Preprocessing the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time; Performing principal component analysis on the vibration matrix, and taking principal component components corresponding to a preset number of maximum eigenvalues ​​as fusion vectors; Using the fusion vector and the pressure vector as inputs of a pipeline pressure prediction model to obtain a predicted pressure corresponding to a pressure prediction point; The construction process of the pipeline pressure prediction model includes: Acquire a plurality of historical vibration signals and historical pressure signals of the detection point at different times, and a plurality of real pressure signals of the prediction point at different times; Dividing the historical vibration signal, the historical pressure signal and the real pressure signal based on a preset time step to obtain multiple groups of divided data; Preprocessing the historical vibration signal, the historical pressure signal and the real pressure signal in the divided data to obtain a historical vibration matrix and a historical pressure vector aligned in time; The historical vibration matrix is ​​subjected to principal component analysis, and the principal component corresponding to the preset number of maximum eigenvalues ​​is used as the historical fusion vector, wherein, according to Performing principal component analysis on the historical vibration matrix, represents the matrix obtained after the historical vibration matrix is ​​standardized and preprocessed, T represents the principal component score matrix, represents the principal component loading matrix, represents the pivot score vector, represents the load vector; generating sample data based on the historical fusion vector, the historical pressure vector and the real pressure signal; The long short-term memory network is trained based on the sample data to obtain the pipeline pressure prediction model.

2. The pipeline pressure prediction method integrating pressure and vibration information according to claim 1 is characterized in that: The step of obtaining vibration signals and pressure signals corresponding to a plurality of detection points on the target pipeline includes: Setting a plurality of vibration detection points and at least one pressure detection point in the vicinity of the pressure prediction point on the target pipeline; Acquire acceleration data corresponding to the vibration detection point, and use the acceleration data as the vibration signal; The pressure data corresponding to the pressure detection point is acquired, and the pressure data is used as the vibration signal.

3. The pipeline pressure prediction method integrating pressure and vibration information according to claim 2 is characterized in that: The setting of a plurality of vibration detection points and at least one pressure detection point in the vicinity of the pressure prediction point on the target pipeline comprises: A plurality of vibration detection points are arranged on a plurality of pipe fittings of the target pipeline, and the pressure prediction point is located between the pipe fittings; At least one pressure detection point is arranged on the pipe body of the target pipe, and the pressure prediction point is located between the pressure detection points.

4. The pipeline pressure prediction method integrating pressure and vibration information according to claim 1 is characterized in that: Preprocessing the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time includes: Respectively obtaining the mean and variance corresponding to the vibration signal and the pressure signal; According to the mean and the variance, the vibration signal and the pressure signal are preprocessed based on a standard fraction conversion method to obtain a vibration matrix and a pressure vector aligned in time.

5. The pipeline pressure prediction method integrating pressure and vibration information according to claim 1 is characterized in that: Also includes: When the predicted pressure is greater than the pressure threshold, a pressure warning is issued.

6. The pipeline pressure prediction method integrating pressure and vibration information according to claim 1 is characterized in that: A plurality of pressure prediction points are arranged on the target pipeline, and the method further comprises: Corresponding pipeline pressure prediction models are established for multiple pressure prediction points to achieve pressure prediction for multiple pressure prediction points.

7. A pipeline pressure prediction device integrating pressure and vibration information, characterized in that: include: A signal acquisition module, used to acquire vibration signals and pressure signals corresponding to multiple detection points on the target pipeline; A signal processing module, used for preprocessing the vibration signal and the pressure signal to obtain a vibration matrix and a pressure vector aligned in time; A matrix analysis module, used for performing principal component analysis on the vibration matrix, and taking principal component components corresponding to a preset number of maximum eigenvalues ​​as fusion vectors; A pressure prediction module, used to use the fusion vector and the pressure vector as inputs of a pipeline pressure prediction model to obtain a predicted pressure corresponding to a pressure prediction point; The construction process of the pipeline pressure prediction model includes: Acquire a plurality of historical vibration signals and historical pressure signals of the detection point at different times, and a plurality of real pressure signals of the prediction point at different times; Dividing the historical vibration signal, the historical pressure signal and the real pressure signal based on a preset time step to obtain multiple groups of divided data; Preprocessing the historical vibration signal, the historical pressure signal and the real pressure signal in the divided data to obtain a historical vibration matrix and a historical pressure vector aligned in time; The historical vibration matrix is ​​subjected to principal component analysis, and the principal component corresponding to the preset number of maximum eigenvalues ​​is used as the historical fusion vector, wherein, according to Performing principal component analysis on the historical vibration matrix, represents the matrix obtained after the historical vibration matrix is ​​standardized and preprocessed, T represents the principal component score matrix, represents the principal component loading matrix, represents the pivot score vector, represents the load vector; generating sample data based on the historical fusion vector, the historical pressure vector and the real pressure signal; The long short-term memory network is trained based on the sample data to obtain the pipeline pressure prediction model.

8. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the pipeline pressure prediction method integrating pressure and vibration information according to any one of claims 1 to 6 by executing the executable instructions.

Images