Differential pressure static level instrument with integrated attitude monitoring function and method thereof

By integrating acceleration sensors and data processors in the static level and using artificial intelligence technology for data analysis, the problem that the differential pressure static level cannot perceive posture changes is solved, and tilt detection and early warning in the absence of settlement are achieved , improve measurement accuracy and monitoring capabilities.

CN118776523BActive Publication Date: 2024-11-19LANZUN TECH (SHANDONG) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411280707.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-13
Publication Date
2024-11-19
Estimated Expiration
2044-09-13

AI Technical Summary

Technical Problem

The existing differential pressure static level cannot perceive the posture changes of the static level itself without settlement, resulting in the lack of function of the settlement monitoring system, especially in slope monitoring, it is difficult to early warning of the collapse situation in advance.

Method used

By integrating acceleration sensors and data processors in static level, a time series of acceleration and pressure data is obtained, fine-grained timing encoding and semantic extraction is used to use artificial intelligence technology to adaptively query matching features to optimize pressure measurements and correct pressure in real time Data, detects tilt without settlement and provides early warning.

Benefits of technology

Improves measurement accuracy, detects tilt in the absence of settlement, provides early warning for possible mutation settlement, and enhances monitoring capabilities for structures that are prone to collapse.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118776523B_ABST
    Figure CN118776523B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of differential pressure static level, and specifically discloses a differential pressure static level with integrated posture monitoring function and a method thereof. The method comprises: obtaining the time series of acceleration data and the time series of pressure data, and using data processing and analysis technology based on artificial intelligence to perform fine-grained time series encoding and time series semantic extraction on the acceleration data and the pressure data, so as to intelligently obtain the optimized pressure measurement value according to the adaptive query matching features of the acceleration data and the pressure data in the time series. In this way, the posture change of the static level can be monitored by the acceleration sensor, so as to correct the pressure measurement data in real time, thereby eliminating the error caused by the posture change and improving the accuracy of the measurement. At the same time, the tilt in the absence of settlement can be detected, providing early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of differential pressure static level instruments, and in particular, to a differential pressure static level instrument integrating a posture monitoring function and a method thereof. Background Art

[0002] Static levels are mainly used to form settlement monitoring systems and monitor the settlement of railways, bridges, slopes, etc. However, as a differential pressure static level using the principle of liquid communicating vessels, only when the distance between the liquid level of the liquid storage tank and the lowest liquid level of the static level storage container at the measuring point changes, can the pressure acting on the pressure core inside the static level change, and then be collected, amplified and output. When the static level tilts based on the horizontal plane without settlement, such as left and right tilting, front and back tilting, etc., since the distance between the liquid storage tank and the lowest liquid level in the static level storage container at the measuring point does not change, and the pressure acting on the pressure core inside the static level does not change, the settlement monitoring system cannot perceive the change in the posture of the static level itself. However, this tilt is often a precursor to the occurrence of sudden settlement on the surface of the monitored object. For example, in slope monitoring, before landslides occur, the static level first vibrates and tilts. Therefore, under current technical conditions, the static level used in the settlement monitoring system has certain functional deficiencies.

[0003] Therefore, an optimized differential pressure hydrostatic level is desired. Summary of the invention

[0004] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] In a first aspect, the present application provides a pressure differential static level with an integrated posture monitoring function, the pressure differential static level comprising: a static level body, the static level body comprising a data processor; an acceleration sensor and a pressure sensor installed on the static level body; a data transmission unit installed on the static level body, the data transmission unit being communicatively connected to the data processor; wherein the data processor comprises: an acceleration-pressure data acquisition module for acquiring a time series of acceleration data and a time series of pressure data; a pressure data time series measurement fluctuation semantic extraction module for inputting the time series of pressure data into a sequence encoder to obtain a pressure data time series measurement fluctuation semantic feature vector; a static level; an acceleration sensor and a pressure sensor installed on the static level body; a data transmission unit being communicatively connected to the data processor; wherein the data processor comprises: an acceleration-pressure data acquisition module for acquiring a time series of acceleration data and a time series of pressure data; a pressure data time series measurement fluctuation semantic extraction module for inputting the time series of pressure data into a sequence encoder to obtain a pressure data time series measurement fluctuation semantic feature vector; A leveler attitude time series feature extraction module is used to input the time series of the acceleration data into a posture feature extractor to obtain a sequence of fine-grained time series feature vectors of the static level attitude; a pressure measurement-attitude correction query matching module is used to use the pressure data time series measurement fluctuation semantic feature vector as a query feature vector, and input the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level attitude fine-grained time series feature vector into a feature query response adaptive encoder based on fine-grained decoupling to obtain a pressure measurement-attitude correction query matching feature matrix as a pressure measurement-attitude correction query matching feature; an optimized pressure measurement value generation module is used to obtain an optimized pressure measurement value based on the pressure measurement-attitude correction query matching feature.

[0006] Optionally, the pressure data time series measurement fluctuation semantic extraction module includes: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain the pressure data time series measurement fluctuation semantic feature vector.

[0007] Optionally, the static level instrument attitude time series feature extraction module is used to: input the time series of the acceleration data into a posture feature extractor based on a Bi-LSTM model to obtain a sequence of fine-grained time series feature vectors of the static level instrument attitude.

[0008] Optionally, the pressure measurement-posture correction query matching module includes: a fine-grained decoupling unit, which is used to perform fine-grained decoupling on the sequence of the static level instrument posture fine-grained time series feature vectors to obtain a set of subsequences of the static level instrument posture fine-grained time series feature vectors; a static level instrument posture fine-grained time series fine-grained local association encoding unit, which is used to input each subsequence of the static level instrument posture fine-grained time series feature vector in the set of subsequences of the static level instrument posture fine-grained time series feature vectors into a subsequence semantic reinforcement encoder to obtain a set of static level instrument posture fine-grained time series fine-grained decoupled local association feature vectors; a covariance matrix calculation unit, which is used to calculate the covariance matrix between the query feature vector and each static level instrument posture fine-grained time series fine-grained decoupled local association feature vector in the set of the static level instrument posture fine-grained time series fine-grained decoupled local association feature vectors to obtain a query static level instrument posture fine-grained time series response covariance matrix. A set; a static level instrument attitude fine-grained temporal response trace measurement factor calculation unit, which is used to calculate the matrix traces of each query static level instrument attitude fine-grained temporal response covariance matrix in the set of query static level instrument attitude fine-grained temporal response covariance matrices to obtain a set of query static level instrument attitude fine-grained temporal response trace measurement factors; a static level instrument attitude fine-grained temporal response trace measurement factor activation unit, which is used to activate the set of query static level instrument attitude fine-grained temporal response trace measurement factors to obtain a set of query static level instrument attitude fine-grained temporal response trace attention weights; a weighted sum calculation unit, which is used to use each query static level instrument attitude fine-grained temporal response trace attention weight in the set of query static level instrument attitude fine-grained temporal response trace attention weights as weights, and calculate the weighted sum of the set of query static level instrument attitude fine-grained temporal response covariance matrices to obtain the pressure measurement-posture correction query matching feature matrix.

[0009] Optionally, the static level attitude fine-grained temporal fine-grained local association encoding unit includes: inputting each subsequence of the static level attitude fine-grained temporal feature vector in the set of subsequences of the static level attitude fine-grained temporal feature vectors into a subsequence semantic reinforcement encoder based on the Bi-LSTM model to obtain a set of the static level attitude fine-grained temporal fine-grained decoupled local association feature vectors.

[0010] Optionally, the static level instrument attitude fine-grained timing response trace measurement factor calculation unit is used to: respectively calculate the sum of the elements on the diagonal of each queried static level instrument attitude fine-grained timing response covariance matrix in the set of queried static level instrument attitude fine-grained timing response covariance matrices to obtain the set of queried static level instrument attitude fine-grained timing response trace measurement factors.

[0011] Optionally, the static level instrument posture fine-grained temporal response trace measurement factor activation unit is used to: input the set of measurement factors of the query static level instrument posture fine-grained temporal response trace into a Sigmoid activation function to obtain a set of attention weights of the query static level instrument posture fine-grained temporal response trace.

[0012] Optionally, the optimized pressure measurement value generating module is used to: input the pressure measurement-posture correction query matching feature matrix into a decoder-based pressure measurement optimizer to obtain an optimized pressure measurement decoding value.

[0013] In the second aspect, the present application provides an optimization method for a pressure differential static level that integrates a posture monitoring function, the method comprising: acquiring a time series of acceleration data and a time series of pressure data; inputting the time series of pressure data into a sequence encoder to obtain a semantic feature vector of pressure data time series measurement fluctuations; inputting the time series of acceleration data into a posture feature extractor to obtain a sequence of fine-grained time series feature vectors of static level posture; using the semantic feature vector of pressure data time series measurement fluctuations as a query feature vector, and inputting the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level posture fine-grained time series feature vector into a feature query response adaptive encoder based on fine-grained decoupling to obtain a pressure measurement-posture correction query matching feature matrix as a pressure measurement-posture correction query matching feature; based on the pressure measurement-posture correction query matching feature, obtaining an optimized pressure measurement value.

[0014] Optionally, inputting the time series of the pressure data into a sequence encoder to obtain a semantic feature vector of the pressure data time series measurement fluctuation, including: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain a semantic feature vector of the pressure data time series measurement fluctuation.

[0015] By adopting the above technical solution, the time series of acceleration data and the time series of pressure data are obtained, and the data processing and analysis technology based on artificial intelligence is used to perform fine-grained time series encoding and time series semantic extraction on the acceleration data and the pressure data, so as to intelligently obtain the optimized pressure measurement value according to the adaptive query matching characteristics of the acceleration data and the pressure data in the time series. In this way, the posture change of the static level can be monitored by the acceleration sensor, so as to make real-time corrections to the pressure measurement data, thereby eliminating the errors caused by the posture change and improving the accuracy of the measurement. At the same time, the tilt in the absence of settlement can be detected, providing early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.

[0016] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0018] In the attached picture: Figure 1 The present invention is a schematic diagram of the settlement monitoring principle of a static level instrument according to the prior art.

[0019] Figure 2 It is a block diagram of a differential pressure static level instrument integrated with a posture monitoring function according to an exemplary embodiment.

[0020] Figure 3 It is a schematic structural diagram of a pressure differential static level according to an exemplary embodiment.

[0021] Figure 4 is a block diagram of a differential pressure static level according to an exemplary embodiment.

[0022] Figure 5 It is a schematic diagram of the structure of the mainframe part of a pressure differential static level according to an exemplary embodiment.

[0023] Figure 6 The present invention is a flow chart showing a method for optimizing a differential pressure static level integrating a posture monitoring function according to an exemplary embodiment.

[0024] Figure 7 is a system software flow chart according to an exemplary embodiment.

[0025] Figure 8 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0026] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0027] It should be understood that the various steps described in the method implementation of the present application can be performed in different orders and / or performed in parallel. In addition, the method implementation may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0028] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0029] It should be noted that the concepts such as "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0030] It should be noted that the modifications of "one" and "plurality" mentioned in the present application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0031] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0032] The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

[0033] Static levels are mainly used for automatic monitoring of settlement in large structures, subways, bridges, tunnels, railways, water conservancy projects, dams and goaf areas. The static level settlement system consists of multiple static level sensors of the same model. The liquid storage tanks are connected by air pipes and liquid pipes. The reference point is placed at a stable horizontal base point. When the measuring point rises or falls relative to the reference point, it will cause changes in pressure at each point. By measuring the changes in sensor pressure, the rise and fall changes of each measuring point relative to the horizontal base point can be calculated. Figure 1 As shown, levels 1-3 are measuring points, and the benchmark has the level as the base point. The vertical displacements of all measuring points (level 1, level 2, level 3) are relative to the benchmark level (base point). The vertical displacement of this point is relatively constant or can be accurately determined by other means, so that the settlement changes of each measuring point of the static level system can be accurately calculated.

[0034] Measuring point settlement calculation formula: Base point: initial measurement value - current measurement value = base point change value (1).

[0035] Measuring point: initial measurement value - current measurement value = settlement change value of measuring point⑵.

[0036] The settlement observation value of the measuring point hi = ⑴-⑵=reference point change value-measuring point settlement change value; when hi<0, it is settlement; when hi>0, it is uplift.

[0037] According to the working principle of the static level, under the existing technology, the static level is mainly used to form a settlement monitoring system to monitor the settlement of railways, bridges, slopes, etc. However, as a differential pressure static level using the principle of liquid communicating vessels, only when the distance between the liquid level of the liquid storage tank and the lowest liquid level of the static level storage container at the measuring point changes, can the pressure acting on the pressure core inside the static level change, and then be collected, amplified and output. When the static level has a lateral tilt based on the horizontal plane without settlement, such as left and right tilt, front and back tilt, etc. without settlement, since the distance between the liquid storage tank and the lowest liquid level in the static level storage container at the measuring point has not changed, and the pressure acting on the pressure core inside the static level has not changed, the settlement monitoring system cannot perceive the change in the posture of the static level itself. However, this tilt is often a precursor to the occurrence of a sudden settlement on the surface of the monitored object. For example, in slope monitoring, before landslides occur, the static level first vibrates and tilts. Therefore, under current technical conditions, the static level used in the settlement monitoring system has certain functional deficiencies.

[0038] Figure 2 FIG. 1 is a block diagram of a differential pressure static level instrument with an integrated attitude monitoring function according to an exemplary embodiment. Figure 2 As shown, the pressure differential static level 100 includes: a static level body 11 , and the static level body 11 includes a data processor 12 .

[0039] An acceleration sensor 13 and a pressure sensor 14 are installed on the static level body 11 .

[0040] A data transmission unit 15 is installed on the static level body 11 , and the data transmission unit 15 is communicatively connected to the data processor 12 .

[0041] The data processor 12 includes: an acceleration and pressure data acquisition module 101, which is used to acquire the time series of acceleration data and the time series of pressure data.

[0042] The pressure data time series measurement fluctuation semantic extraction module 102 is used to input the time series of the pressure data into a sequence encoder to obtain a pressure data time series measurement fluctuation semantic feature vector.

[0043] The static level instrument attitude time series feature extraction module 103 is used to input the time series of the acceleration data into the attitude feature extractor to obtain a sequence of fine-grained time series feature vectors of the static level instrument attitude.

[0044] The pressure measurement-posture correction query matching module 104 is used to use the pressure data time series measurement fluctuation semantic feature vector as the query feature vector, and input the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector into a feature query response adaptive encoder based on fine-grained decoupling to obtain a pressure measurement-posture correction query matching feature matrix as the pressure measurement-posture correction query matching feature.

[0045] The optimized pressure measurement value generating module 105 is used to obtain the optimized pressure measurement value based on the pressure measurement-posture correction query matching feature.

[0046] It should be understood that the static level measures the pressure difference of the liquid, and the attitude (such as tilt angle) will affect the reading of the pressure sensor. Therefore, by monitoring the attitude, the pressure measurement data can be corrected to eliminate the error caused by the attitude change.

[0047] Correspondingly, in the above-mentioned data processor, it obtains the time series of acceleration data and the time series of pressure data, and adopts data processing and analysis technology based on artificial intelligence to perform fine-grained time series encoding and time series semantic extraction on the acceleration data and the pressure data, so as to intelligently obtain the optimized pressure measurement value according to the adaptive query matching characteristics of the acceleration data and the pressure data in the time series. In this way, the posture change of the static level can be monitored by the acceleration sensor, so as to correct the pressure measurement data in real time, thereby eliminating the error caused by the posture change and improving the accuracy of the measurement. At the same time, the tilt in the absence of settlement can be detected, providing early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.

[0048] Specifically, in the technical solution of the present application, first, the time series of acceleration data and the time series of pressure data are obtained. Then, the temporal volatility and patterns of the pressure data in different time ranges are considered to be different. Therefore, in order to better capture and mine the temporal variation rules and fluctuations of the pressure data at different time points, in the technical solution of the present application, the time series of the pressure data is input into a sequence encoder based on the RNN model to obtain the semantic feature vector of the temporal measurement fluctuation of the pressure data.

[0049] In one embodiment of the present application, the pressure data time series measurement fluctuation semantic extraction module includes: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain the pressure data time series measurement fluctuation semantic feature vector.

[0050] Then, considering that the timing correlation patterns of the acceleration data between different time spans are different, in the technical solution of the present application, a posture feature extractor based on a Bi-LSTM model that can capture bidirectional time dependencies in the time series is used to perform fine-grained timing correlation capture and extraction on the time series of the acceleration data to obtain a sequence of fine-grained timing feature vectors of the static level attitude.

[0051] In one embodiment of the present application, the static level instrument attitude time series feature extraction module is used to: input the time series of the acceleration data into the attitude feature extractor based on the Bi-LSTM model to obtain a sequence of fine-grained time series feature vectors of the static level instrument attitude.

[0052] Furthermore, in order to further and more meticulously improve the matching interaction between the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector, so as to more accurately correct the pressure measurement data, in the technical solution of the present application, the pressure data time series measurement fluctuation semantic feature vector is used as the query feature vector, and the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector is input into the feature query response adaptive encoder based on fine-grained decoupling to obtain the pressure measurement-posture correction query matching feature matrix. It is worth mentioning that the feature query response adaptive encoder based on fine-grained decoupling improves the response accuracy between the pressure data and the static level instrument posture through fine-grained decoupling and adaptive coding. Specifically, firstly, each static level instrument posture fine-grained time series feature vector is fine-grainedly decoupled to decompose the sequence of vectors, so as to capture more refined local information and analyze and understand the data at a more detailed level. Next, the decomposed subsequences are input into the subsequence semantic enhancement encoder based on the Bi-LSTM model to capture the contextual temporal correlation between each subsequence and enhance the semantic expression and mutual correlation of each static level instrument posture fine-grained subsequence. Subsequently, the covariance matrix between the query feature vector and the local correlation feature obtained after Bi-LSTM encoding is calculated to quantify the local temporal correlation between the pressure data and the static level instrument posture. Furthermore, the trace of the covariance matrix, that is, the sum of the elements on the diagonal, is calculated to evaluate the overall temporal correlation between the two. Next, the trace of each covariance matrix is ​​input as a metric factor into the sigmoid function to adaptively adjust the weight of each factor to obtain the attention weight of the query static level instrument posture fine-grained temporal response trace. Finally, the weighted sum of the set of each covariance matrix is ​​calculated using each attention weight as the weight to comprehensively adjust the importance of each feature to obtain the pressure measurement-posture correction query matching feature matrix.

[0053] In one embodiment of the present application, the pressure measurement-posture correction query matching module includes: a fine-grained decoupling unit, which is used to perform fine-grained decoupling on the sequence of the static level instrument posture fine-grained time series feature vectors to obtain a set of subsequences of the static level instrument posture fine-grained time series feature vectors; a static level instrument posture fine-grained time series fine-grained local association encoding unit, which is used to input each subsequence of the static level instrument posture fine-grained time series feature vector in the set of subsequences of the static level instrument posture fine-grained time series feature vectors into a subsequence semantic reinforcement encoder to obtain a set of static level instrument posture fine-grained time series fine-grained decoupled local association feature vectors; a covariance matrix calculation unit, which is used to calculate the covariance matrix between the query feature vector and each static level instrument posture fine-grained time series fine-grained decoupled local association feature vector in the set of the static level instrument posture fine-grained time series fine-grained decoupled local association feature vectors to obtain the query static level instrument posture fine-grained time series response covariance matrix. A set of difference matrices; a static level instrument attitude fine-grained temporal response trace metric factor calculation unit, used to calculate the matrix traces of each query static level instrument attitude fine-grained temporal response covariance matrix in the set of query static level instrument attitude fine-grained temporal response covariance matrices to obtain a set of query static level instrument attitude fine-grained temporal response trace metric factors; a static level instrument attitude fine-grained temporal response trace metric factor activation unit, used to activate the set of query static level instrument attitude fine-grained temporal response trace metric factors to obtain a set of query static level instrument attitude fine-grained temporal response trace attention weights; a weighted sum calculation unit, used to use each query static level instrument attitude fine-grained temporal response trace attention weight in the set of query static level instrument attitude fine-grained temporal response trace attention weights as weights, and calculate the weighted sum of the set of query static level instrument attitude fine-grained temporal response covariance matrices to obtain the pressure measurement-posture correction query matching feature matrix.

[0054] Furthermore, in one embodiment of the present application, the static level instrument attitude fine-grained temporal fine-grained local association encoding unit includes: inputting each subsequence of the static level instrument attitude fine-grained temporal feature vector in the set of subsequences of the static level instrument attitude fine-grained temporal feature vectors into a subsequence semantic reinforcement encoder based on the Bi-LSTM model to obtain a set of the static level instrument attitude fine-grained temporal fine-grained decoupled local association feature vectors.

[0055] Furthermore, in one embodiment of the present application, the static level instrument attitude fine-grained timing response trace measurement factor calculation unit is used to: respectively calculate the sum of the elements on the diagonal of each query static level instrument attitude fine-grained timing response covariance matrix in the set of query static level instrument attitude fine-grained timing response covariance matrices to obtain the set of query static level instrument attitude fine-grained timing response trace measurement factors.

[0056] Furthermore, in one embodiment of the present application, the static level instrument attitude fine-grained temporal response trace measurement factor activation unit is used to: input the set of measurement factors of the query static level instrument attitude fine-grained temporal response trace into a Sigmoid activation function to obtain a set of attention weights of the query static level instrument attitude fine-grained temporal response trace.

[0057] Specifically, the pressure data time series measurement fluctuation semantic feature vector is used as the query feature vector, and the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector is input into the feature query response adaptive encoder based on fine-grained decoupling, and processed with the following response adaptive coding formula to obtain the pressure measurement-posture correction query matching feature matrix; wherein the response adaptive coding formula is: ;in, Each static level instrument attitude fine-grained time series feature vector in the sequence of the static level instrument attitude fine-grained time series feature vectors, To decouple operations at a fine-grained level, A set of subsequences representing the fine-grained time series feature vectors of the static level attitude, A subsequence of each static level instrument attitude fine-grained time series feature vector in a set of subsequences of the static level instrument attitude fine-grained time series feature vectors is represented, represents the Bi-LSTM encoder, is each static level instrument attitude fine-grained time series fine-grained decoupling local correlation feature vector in the set of the static level instrument attitude fine-grained time series fine-grained decoupling local correlation feature vectors, is the query feature vector, represents the operation of calculating the covariance matrix, is the covariance matrix of each query static level instrument attitude fine-grained time series response in the set of the query static level instrument attitude fine-grained time series response covariance matrix, represents the calculation matrix trace, is each query static level instrument attitude fine-grained time series response trace measurement factor in the set of query static level instrument attitude fine-grained time series response trace measurement factors, yes function, is the attention weight of each query static level instrument posture fine-grained temporal response trace in the set of attention weights of the query static level instrument posture fine-grained temporal response trace, is the pressure measurement-posture correction query matching feature matrix.

[0058] Then, the pressure measurement-attitude correction query matching feature matrix is ​​input into the pressure measurement optimizer based on the decoder to obtain the optimized pressure measurement decoding value. That is, the pressure measurement-attitude correction query matching feature obtained by adaptive response encoding using the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level attitude fine-grained time series feature vector is decoded to intelligently obtain the optimized pressure measurement. In this way, the attitude change of the static level can be monitored by the acceleration sensor, and the pressure measurement data can be corrected in real time to eliminate the error caused by the attitude change, thereby improving the accuracy of the measurement. At the same time, the tilt in the absence of settlement can be detected, providing early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.

[0059] In one embodiment of the present application, the optimized pressure measurement value generation module is used to: input the pressure measurement-posture correction query matching feature matrix into a decoder-based pressure measurement optimizer to obtain an optimized pressure measurement decoding value.

[0060] In a preferred embodiment, the present application takes into account that the sequence of the pressure data timing measurement fluctuation semantic feature vector and the static level attitude fine-grained timing feature vector respectively represent the cyclic timing coding characteristics of the pressure data and the short-range and long-range bidirectional timing correlation characteristics of the acceleration data. Therefore, after performing feature query response adaptive encoding based on fine-grained decoupling, the pressure measurement-attitude correction query matching feature matrix will also cause the timing fine-grained structure adaptive response offset caused by the difference in timing feature association dimensions, thereby causing decoding semantic regression offset, making it have insufficient offset semantic coverage, causing outlier decoding regression inference mapping deviation, and affecting the accuracy of the optimized pressure measurement decoding value obtained by the pressure measurement optimizer based on the decoder input of the pressure measurement-attitude correction query matching feature matrix.

[0061] Based on this, when the pressure measurement-posture correction query matching feature matrix is ​​input into a decoder-based pressure measurement optimizer to obtain an optimized pressure measurement decoding value, the pressure measurement-posture correction query matching feature matrix is ​​optimized, and the optimization process specifically includes: calculating the characteristic mean of the pressure measurement-posture correction query matching feature matrix, and dividing the characteristic mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the pressure measurement-posture correction query matching feature matrix to obtain a pressure measurement-posture correction query matching distribution characterization value; subtracting the pressure measurement-posture correction query matching distribution characterization value and dividing it by the pressure measurement-posture correction query matching distribution characterization value to obtain a pressure measurement-posture correction query matching distribution modulation value; activating the pressure measurement-posture correction query matching feature matrix through a probabilistic function to obtain a probabilistic pressure measurement-posture correction query matching feature matrix; and converting the probabilistic pressure measurement-posture correction query matching feature matrix into a probabilistic pressure measurement-posture correction query matching feature matrix. After performing dot-wise subtraction between the matching feature matrix and the pressure measurement-posture correction query matching distribution modulation value, take the absolute value and calculate the negative of the logarithmic value with base 2 to obtain a probabilistic pressure measurement-posture correction query matching distribution modulation information matrix; after dividing the pressure measurement-posture correction query matching distribution characterization value by one minus the difference of each eigenvalue of the probabilistic pressure measurement-posture correction query matching feature matrix, sum all eigenvalues ​​of the probabilistic pressure measurement-posture correction query matching feature matrix and divide by the scale of the pressure measurement-posture correction query matching feature matrix to obtain a probabilistic pressure measurement-posture correction query matching distribution modulation bias value; and perform dot-wise addition on the probabilistic pressure measurement-posture correction query matching distribution modulation information vector and the probabilistic pressure measurement-posture correction query matching distribution modulation bias value multiplied by a weight as a hyperparameter to obtain an optimized pressure measurement-posture correction query matching feature matrix.

[0062] The optimized expression of the pressure measurement-posture correction query matching feature matrix is ​​as follows: .

[0063] And among them: ; is the scale of the pressure measurement-posture correction query matching feature matrix, i.e., width multiplied by height, is the eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the maximum eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the minimum eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the pressure measurement-posture correction query matching distribution representation value, is each eigenvalue of the probabilistic pressure measurement-posture correction query matching feature matrix, is the probabilistic pressure measurement-posture correction query matching feature matrix, is the weight of the hyperparameter, represents the logarithmic function with base 2, is the optimized pressure measurement-posture correction query matching feature matrix, It is added by location point. It is subtracted by position point.

[0064] That is, in the above preferred example, the pressure measurement-posture correction query matching feature matrix is ​​used to perform the probability information distribution planning based on the eigenvalue of the pressure measurement-posture correction query matching feature matrix by means of the Bernoulli probability modulation distribution relative to the eigenvalue distribution, and the probability reverse mapping of the probability feature of the pressure measurement-posture correction query matching feature matrix is ​​used as the extended coverage of the set mapping space of the pressure measurement-posture correction query matching feature matrix to independently understand the interactive path between the intuitive probability information distribution and the abstract probability space mapping of the pressure measurement-posture correction query matching feature matrix, so as to improve the accuracy of the optimized pressure measurement decoding value obtained by the pressure measurement optimizer based on the decoder input of the optimized pressure measurement-posture correction query matching feature matrix by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the pressure measurement-posture correction query matching feature matrix to the decoding regression probability. In this way, the posture change of the static level can be monitored by the acceleration sensor, and the pressure measurement data can be corrected in real time to eliminate the error caused by the posture change, thereby improving the accuracy of the measurement. At the same time, it can detect tilt without settlement and provide early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.

[0065] In summary, the above scheme is adopted to obtain the time series of acceleration data and the time series of pressure data, and use artificial intelligence-based data processing and analysis technology to perform fine-grained time series encoding and time series semantic extraction on the acceleration data and the pressure data, so as to intelligently obtain the optimized pressure measurement value according to the adaptive query matching characteristics of the acceleration data and the pressure data in the time series. In this way, the posture change of the static level can be monitored by the acceleration sensor, so as to correct the pressure measurement data in real time, thereby eliminating the error caused by the posture change and improving the accuracy of the measurement. At the same time, the tilt in the absence of settlement can be detected, providing early warning for possible sudden settlement, which is especially important for structures prone to landslides such as slopes.

[0066] In one embodiment of the present application, a pressure differential static level device with attitude monitoring function is designed to address the functional deficiencies of current static levels. A high-precision acceleration sensor is installed inside the device, and any attitude changes and vibrations of the static level itself can be accurately captured, and a settlement warning can be issued in advance when no settlement has occurred. Moreover, since the acceleration sensor is directional, in addition to issuing a settlement warning, it can also grasp the direction and angle of the static level's inclination, providing more accurate data support for determining when settlement occurs.

[0067] The present application integrates a high-precision acceleration sensor inside the static level, which can collect vibrations from all directions and provide reliable compressive vibration data of the monitoring points after filtering out noise through a certain filtering algorithm.

[0068] The key point of this application is that a high-precision acceleration sensor is integrated inside the static level, which greatly improves the function of the static level used for settlement monitoring and the scope of data collection. In addition, the cost of using the acceleration sensor is not high, the structure of the static level does not need to be specially modified, and the energy consumption does not increase significantly, which greatly improves production efficiency and social benefits.

[0069] The static level has two core sensors: one is a pressure sensor for measuring pressure, and the other is a three-axis accelerometer (or inclinometer) for measuring posture. They respectively collect the pressure of the liquid in the device and the posture changes of the device, and upload the data to the server through amplification circuits.

[0070] Figure 3 is a schematic diagram of the structure of a pressure differential static level according to an exemplary embodiment. Figure 3 As shown, the differential pressure static level includes: a housing 21, a pressure core 22, a liquid storage container 23, a power supply and signal 24, a liquid pipe interface 25 and an air pipe interface 26. Figure 4 is a block diagram of a pressure differential static level according to an exemplary embodiment. Figure 4 As shown, the differential pressure static level includes: a static level MCU, a power supply, an acceleration sensor, a liquid pressure core and a data transmission RTU.

[0071] The above mechanical structure provides structure and protection for the sensor to function. The differential pressure static level can be simply divided into two chambers, one is the liquid chamber and the other is the air chamber. It mainly ensures that the pressure-bearing end of the pressure sensor is in contact with the liquid, can sense the pressure of the liquid, and convert it into an electronic signal. At the same time, it is also necessary to ensure that the circuit at the other end of the pressure sensor is not invaded by liquid to prevent short circuit, and to ensure contact with normal atmospheric pressure to provide a reference for normal air pressure for the sensor.

[0072] Figure 5FIG. 1 is a schematic diagram of the structure of the main unit of a pressure differential static level according to an exemplary embodiment. Figure 5 As shown, it includes: an observation window fixing component 31, a transparent acrylic observation window component 32, a waterproof rubber ring 33, a sealing screw column 34, an exhaust hole 35, a sensor clamping ring 36, a sealing ring 37, a circuit sealing cover 38, a fixing screw 39, a sensor adapter 40, a data transmission interface 41, an air pressure balance vent 42, a multi-purpose bracket 43, a liquid injection hole 44, and a static level body 45.

[0073] Figure 6 is a flow chart of a method for optimizing a differential pressure static level integrating a posture monitoring function according to an exemplary embodiment. Figure 6 As shown, the method includes: step S201, acquiring a time series of acceleration data and a time series of pressure data.

[0074] Step S202: Input the time series of the pressure data into a sequence encoder to obtain a semantic feature vector of the pressure data time series measurement fluctuation.

[0075] Step S203: input the time series of the acceleration data into a posture feature extractor to obtain a sequence of fine-grained time series feature vectors of the static level instrument posture.

[0076] Step S204, using the pressure data time series measurement fluctuation semantic feature vector as the query feature vector, and inputting the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector into the feature query response adaptive encoder based on fine-grained decoupling to obtain the pressure measurement-posture correction query matching feature matrix as the pressure measurement-posture correction query matching feature.

[0077] Step S205: obtaining an optimized pressure measurement value based on the pressure measurement-posture correction query matching feature.

[0078] In one embodiment of the present application, the time series of the pressure data is input into a sequence encoder to obtain a semantic feature vector of the pressure data time series measurement fluctuation, including: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain the semantic feature vector of the pressure data time series measurement fluctuation.

[0079] Figure 7 FIG. 1 is a system software flow chart according to an exemplary embodiment. Figure 7 As shown, after the start, the system clock and I0 port are initialized, and then reach the MCU for processing together with the watchdog, followed by temperature and angle reading and conversion, and then pressure reading and conversion, and then reach RS485, and mutual data acquisition and transmission are carried out, and then configuration parameters and data calibration are performed, and then returned to the MCU.

[0080] Reference below Figure 8 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present application. The terminal device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0081] like Figure 8 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0082] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0083] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are executed.

Claims

1. A differential pressure static level with integrated posture monitoring function, characterized in that: include: A static level body, the static level body comprising a data processor; An acceleration sensor and a pressure sensor installed on the body of the static level; A data transmission unit installed on the static level body, the data transmission unit being communicatively connected to the data processor; Wherein, the data processor comprises: An acceleration and pressure data acquisition module is used to acquire a time series of acceleration data and a time series of pressure data; A pressure data time series measurement fluctuation semantic extraction module, used for inputting the time series of the pressure data into a sequence encoder to obtain a pressure data time series measurement fluctuation semantic feature vector; A static level instrument attitude time series feature extraction module, used for inputting the time series of the acceleration data into the attitude feature extractor to obtain a sequence of fine-grained time series feature vectors of the static level instrument attitude; A pressure measurement-posture correction query matching module is used to use the pressure data time series measurement fluctuation semantic feature vector as a query feature vector, and input the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector into a feature query response adaptive encoder based on fine-grained decoupling to obtain a pressure measurement-posture correction query matching feature matrix as a pressure measurement-posture correction query matching feature; An optimized pressure measurement value generation module, used to obtain an optimized pressure measurement value based on the pressure measurement-posture correction query matching feature; Wherein, the pressure measurement-posture correction query matching module includes: A fine-grained decoupling unit, used for performing fine-grained decoupling on the sequence of the static level instrument attitude fine-grained time series feature vectors to obtain a set of subsequences of the static level instrument attitude fine-grained time series feature vectors; A static level instrument attitude fine-grained temporal fine-grained local association encoding unit, used for inputting each subsequence of the static level instrument attitude fine-grained temporal feature vector in the set of subsequences of the static level instrument attitude fine-grained temporal feature vectors into a subsequence semantic enhancement encoder to obtain a set of static level instrument attitude fine-grained temporal fine-grained decoupled local association feature vectors; A covariance matrix calculation unit, used for calculating the covariance matrix between the query feature vector and each static level instrument attitude fine-grained time series fine-grained decoupling local correlation feature vector in the set of the static level instrument attitude fine-grained time series fine-grained decoupling local correlation feature vectors to obtain a set of query static level instrument attitude fine-grained time series response covariance matrices; A static level instrument attitude fine-grained time series response trace metric factor calculation unit, used for calculating the matrix trace of each query static level instrument attitude fine-grained time series response covariance matrix in the set of query static level instrument attitude fine-grained time series response covariance matrices to obtain a set of query static level instrument attitude fine-grained time series response trace metric factors; A static level instrument attitude fine-grained temporal response trace metric factor activation unit, used for activating the set of metric factors of the query static level instrument attitude fine-grained temporal response trace to obtain a set of attention weights of the query static level instrument attitude fine-grained temporal response trace; A weighted sum calculation unit, used for calculating the weighted sum of the set of covariance matrices of the query static level attitude fine-grained temporal response using each query static level attitude fine-grained temporal response trace attention weight in the set of attention weights of the query static level attitude fine-grained temporal response trace as a weight to obtain the pressure measurement-posture correction query matching feature matrix; Wherein, the optimized pressure measurement value generation module is used to: input the pressure measurement-posture correction query matching feature matrix into a decoder-based pressure measurement optimizer to obtain an optimized pressure measurement decoded value; Wherein, when the pressure measurement-posture correction query matching feature matrix is ​​input into a decoder-based pressure measurement optimizer to obtain an optimized pressure measurement decoding value, the pressure measurement-posture correction query matching feature matrix is ​​optimized, and the optimization process specifically includes: Calculating the characteristic mean of the pressure measurement-posture correction query matching characteristic matrix, and dividing the characteristic mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the pressure measurement-posture correction query matching characteristic matrix to obtain a pressure measurement-posture correction query matching distribution characterization value; Subtracting the pressure measurement-posture correction query matching distribution characterization value from one and dividing the result by the pressure measurement-posture correction query matching distribution characterization value to obtain a pressure measurement-posture correction query matching distribution modulation value; Activating the pressure measurement-posture correction query matching feature matrix through a probabilistic function to obtain a probabilistic pressure measurement-posture correction query matching feature matrix; After performing point subtraction between the probabilistic pressure measurement-posture correction query matching feature matrix and the pressure measurement-posture correction query matching distribution modulation value, taking the absolute value and calculating the negative of the logarithmic value with base 2 to obtain the probabilistic pressure measurement-posture correction query matching distribution modulation information matrix; After dividing the pressure measurement-posture correction query matching distribution characterization value by one minus the difference of each eigenvalue of the probabilistic pressure measurement-posture correction query matching feature matrix, summing all eigenvalues ​​of the probabilistic pressure measurement-posture correction query matching feature matrix and dividing by the scale of the pressure measurement-posture correction query matching feature matrix to obtain a probabilistic pressure measurement-posture correction query matching distribution modulation bias value; The product of the probabilistic pressure measurement-posture correction query matching distribution modulation information vector and the probabilistic pressure measurement-posture correction query matching distribution modulation bias value and the weight as a hyperparameter is multiplied to obtain an optimized pressure measurement-posture correction query matching feature matrix; The optimized expression of the pressure measurement-posture correction query matching feature matrix is ​​as follows: ; And among them: ; is the scale of the pressure measurement-posture correction query matching feature matrix, i.e., width multiplied by height, is the eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the maximum eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the minimum eigenvalue of the pressure measurement-posture correction query matching feature matrix, is the pressure measurement-posture correction query matching distribution representation value, is each eigenvalue of the probabilistic pressure measurement-posture correction query matching feature matrix, is the probabilistic pressure measurement-posture correction query matching feature matrix, is the weight of the hyperparameter, represents the logarithmic function with base 2, is the optimized pressure measurement-posture correction query matching feature matrix, It is added by location point. It is subtracted by position point.

2. The differential pressure static level with integrated posture monitoring function according to claim 1, characterized in that: The pressure data time series measurement fluctuation semantic extraction module includes: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain the pressure data time series measurement fluctuation semantic feature vector.

3. The differential pressure static level with integrated posture monitoring function according to claim 2, characterized in that: The static level instrument attitude time series feature extraction module is used to: input the time series of the acceleration data into the attitude feature extractor based on the Bi-LSTM model to obtain a sequence of fine-grained time series feature vectors of the static level instrument attitude.

4. The differential pressure static level with integrated posture monitoring function according to claim 3 is characterized in that: The static level instrument attitude fine-grained temporal fine-grained local association encoding unit includes: Each subsequence of the static level instrument attitude fine-grained time series feature vector in the set of subsequences of the static level instrument attitude fine-grained time series feature vector is input into the subsequence semantic reinforcement encoder based on the Bi-LSTM model to obtain a set of the static level instrument attitude fine-grained time series fine-grained decoupled local correlation feature vectors.

5. The differential pressure static level with integrated posture monitoring function according to claim 4, characterized in that: The static level instrument attitude fine-grained timing response trace quantity factor calculation unit is used to: respectively calculate the sum of the elements on the diagonal of each query static level instrument attitude fine-grained timing response covariance matrix in the set of query static level instrument attitude fine-grained timing response covariance matrices to obtain the set of query static level instrument attitude fine-grained timing response trace quantity factors.

6. The differential pressure static level with integrated posture monitoring function according to claim 5, characterized in that: The static level instrument posture fine-grained temporal response trace measurement factor activation unit is used to: input the set of measurement factors of the query static level instrument posture fine-grained temporal response trace into the Sigmoid activation function to obtain the set of attention weights of the query static level instrument posture fine-grained temporal response trace.

7. A method for optimizing a differential pressure static level with an integrated posture monitoring function, using the differential pressure static level with an integrated posture monitoring function as claimed in claim 1, characterized in that: include: Obtain the time series of acceleration data and the time series of pressure data; Inputting the time series of the pressure data into a sequence encoder to obtain a pressure data time series measurement fluctuation semantic feature vector; Inputting the time series of the acceleration data into a posture feature extractor to obtain a sequence of fine-grained time series feature vectors of the static level instrument posture; Taking the pressure data time series measurement fluctuation semantic feature vector as the query feature vector, inputting the sequence of the pressure data time series measurement fluctuation semantic feature vector and the static level instrument posture fine-grained time series feature vector into a feature query response adaptive encoder based on fine-grained decoupling to obtain a pressure measurement-posture correction query matching feature matrix as a pressure measurement-posture correction query matching feature; Based on the pressure measurement-posture correction query matching feature, an optimized pressure measurement value is obtained.

8. The optimization method of differential pressure static level integrating posture monitoring function according to claim 7 is characterized in that: Inputting the time series of the pressure data into a sequence encoder to obtain a semantic feature vector of the pressure data time series measurement fluctuation, including: inputting the time series of the pressure data into a sequence encoder based on an RNN model to obtain the semantic feature vector of the pressure data time series measurement fluctuation.