A port water conservancy construction facility structure strain intelligent sensing device

Abstract

The application discloses a kind of port hydraulic construction facility structure strain intelligent sensing device, including data acquisition module, environmental temperature measurement module, data processing module, data analysis module, early warning module and visualization module.The technical scheme presented in the application, by collecting the stress data value inside parapet, temperature data value and the environmental temperature data of the outer surface of parapet under different media, the processing result of the collected data is used as the input and output vector of model to construct BP neural network analysis model, the stress value of each point inside parapet under different environmental temperature is predicted, according to the prediction result, the change trend and influence mechanism caused by environmental temperature to structure stress are explored, and the temperature stress of parapet of gravity type wharf can be intelligently perceived.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of port water engineering construction facility detection, and particularly relates to a port water engineering construction facility structure strain intelligent sensing device. BACKGROUND

[0002] The super heavy wharf is one of the main structural forms of the wharf, also known as the gravity wharf, which resists the sliding and overturning of the building by the weight of the structure itself and the filler thereon, at the same time, the weight of the structure and the filler thereon and various loads generate pressure on the foundation, which requires the foundation to have a certain strength. The gravity wharf is composed of an upper breast wall, a wall body, a foundation and backfilling material behind the wall. The main function of the breast wall of the gravity wharf is to support various parts of the wharf and bear the horizontal load of the wharf. The stability of the breast wall is a key factor affecting the safe operation of the entire gravity wharf.

[0003] At present, in order to prevent the breast wall of the gravity wharf from being damaged due to the increase of structural stress caused by impact load, strain sensors are pre-embedded during the pouring of the breast wall to collect stress values of different sections of the breast wall, so as to realize the measurement of the running stability of the breast wall.

[0004] Generally, there are mainly two types of temperature stresses affecting the building: the first type is that due to the hydration heat release of the cement in the initial period of mixing and setting of the newly poured concrete, the volume of the concrete expands, cracks appear on the surface of the component, and the concrete shrinks in the process of temperature reduction and hardening strength improvement after the final setting of the concrete; the second type is that due to the large change of the ambient temperature of the structure in the normal use process, the structure appears cold shrinkage and thermal expansion, and when the ambient temperature changes greatly, the corresponding temperature stress is also large.

[0005] However, in the prior art, the method for measuring the breast wall structure stress by pre-embedding strain sensors can only collect the stress values of fixed positions, and cannot analyze the change trend and influence mechanism of the structural stress caused by the environmental temperature, so it is difficult to know the overall situation of the temperature stress of the breast wall through the strain measurement data of a small number of monitoring points when the environmental temperature changes greatly. Therefore, we propose a port water engineering construction facility structure strain intelligent sensing device. SUMMARY

[0006] The main purpose of the present application is to provide a port water engineering construction facility structure strain intelligent sensing device, which collects the stress data values, temperature data values and environmental temperature data of the outer surface of the breast wall under different media, uses the processing results of the collected data as the input and output vectors of the model to construct a BP neural network analysis model, and predicts the stress values of each point inside the breast wall under different environmental temperatures, which can effectively solve the problems in the background art.

[0007] To achieve the above object, the technical scheme adopted by the present application is:

[0008] A port water construction facility structure strain intelligent sensing device, the sensing device comprises:

[0009] A data acquisition module, which is embedded and installed inside the parapet of the gravity wharf, is used to collect stress data values and temperature data values inside the parapet.

[0010] An ambient temperature measurement module, which is randomly distributed on the outer facade of the parapet, is used to collect ambient temperature data of the outer facade of the parapet under different media, including water and air.

[0011] A data processing module, which is in communication connection with the data acquisition module and the ambient temperature measurement module, is used to acquire and preprocess stress data values and temperature data values inside the parapet and ambient temperature data of the outer facade of the parapet.

[0012] A data analysis module, which acquires the prediction results of stress values at each point inside the parapet under different ambient temperatures by constructing a BP neural network analysis model according to the data processing results of the data processing module.

[0013] An early warning module, which generates a warning level according to the prediction results of the data analysis module.

[0014] A visualization module, which is in communication connection with the data analysis module and the early warning module, is used to visually display the prediction results of the stress values of the parapet and the ambient temperature and the warning level.

[0015] Further, the data acquisition module comprises strain sensors and parapet internal temperature sensors, which are arranged in a grid-like equidistant uniform manner along the longitudinal section and the transverse section of the parapet along the center position of the parapet section of the gravity wharf, wherein the longitudinal section of the parapet is arranged in one direction, and the transverse section is arranged in two directions.

[0016] Further, the ambient temperature measurement module comprises outer facade temperature sensors, the number of which in different media is not less than two.

[0017] Further, the preprocessing step of the data processing module is:

[0018] S1, acquiring all collected data of the data acquisition module and the ambient temperature measurement module.

[0019] S2, calculate the average of all the measurements of the outer facade temperature sensors in different media to obtain the ambient temperature data Ta and Tw of the parapet outer facade in different media, wherein, Ta is the ambient temperature data of the parapet outer facade in air medium, Tw is the ambient temperature data of the parapet outer facade in water medium, n is the number of outer facade temperature sensors distributed in air medium, and m is the number of outer facade temperature sensors distributed in water medium;

[0020] S3, the strain sensors and the parapet inner temperature sensors are numbered as A ij and B ij , respectively, and the numbering rules are:

[0021] a) define the direction from the parapet outer facade to the center position of the longitudinal section as the first positive direction, and j represents the layer number of the sensor arranged along the first positive direction, wherein the larger j is, the closer to the center position of the longitudinal section;

[0022] b) define the direction from the top of the parapet to the bottom as the second positive direction, and i represents the layer number of the sensor arranged along the second positive direction, wherein the larger i is, the closer to the bottom of the parapet;

[0023] S4, construct data matrices Y1, Y2, X1, X2 of collected data according to the numbering results of the strain sensors and the parapet inner temperature sensors, wherein, , , wherein a ij is the data value collected by the strain sensor A ij at time t, b ij is the data value collected by the parapet inner temperature sensor B ij at time t, and k is the layer number of the sensor closest to the water surface in the second positive direction;

[0024] S5, normalize each element value in the data matrices Y1, Y2, X1, X2 to obtain normalized data matrices Y1', Y2', X1', X2'.

[0025] Further, the BP neural network analysis model is constructed with data matrices X1' and X2' as input vectors and Y1' and Y2' as output vectors.

[0026] Further, the method for dividing the warning level is:

[0027] Step 1, obtain the prediction result data P j of the parapet stress value of each point;

[0028] Step 2, use the prediction result data P jThe numerical values create a sample set, denoted as {P1, P2,..., P n};

[0029] Step 3, the mean and standard deviation in the sample set are obtained, and the data is normalized by using the mean and standard deviation, and the normalization formula is In the formula, z is a standard parameter, sigma is the variance of the sample data, and mu is the mean of the sample data;

[0030] Step 4, after completing the standardization, the standard parameter is used The numerical interval is adjusted to [0, 1], and the function value of f(k) is used to classify the early warning level, and the classification mechanism is:

[0031] When f(k) < 0.3, the early warning level is classified as level one; When 0.3 < f(k) < 0.7, the early warning level is classified as level two.

[0032]

[0033] The present application has the following beneficial effects:

[0034] 1) The technical scheme provided by the present application, by collecting the stress data value, temperature data value and the environment temperature data of the outer surface of the parapet under different media, using the processing results of the collected data as the input and output vectors of the model to construct the BP neural network analysis model, the stress value of each point in the parapet under different environment temperatures is predicted, and according to the prediction result, the change trend and influence mechanism of the structure stress caused by the environment temperature are explored, the intelligent perception of the temperature stress of the parapet of the gravity wharf can be realized. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a structural block diagram of the port water construction facility structure strain intelligent perception device of the present application;

[0036] Figure 2 It is a distribution schematic diagram of the data acquisition module on the longitudinal section of the parapet;

[0037] Figure 3 It is a distribution schematic diagram of the data acquisition module on the transverse section of the parapet;

[0038] Figure 4 It is a distribution schematic diagram of the environment temperature measurement module.

[0039] In the figure: 1, parapet; 2, strain sensor; 3, parapet inner temperature sensor; 4, outer surface temperature sensor in air medium; 5, outer surface temperature sensor in water medium. DETAILED DESCRIPTION

[0040] ​​The present application is further described below in conjunction with the specific embodiments, wherein the accompanying drawings are only used for illustrative description, represent only schematic diagrams, not physical diagrams, and cannot be understood as limitation to the present application, and in order to better illustrate the specific embodiments of the present application, some components of the drawings are omitted, enlarged or reduced, and do not represent the actual size of the product.

[0041] Embodiment 1

[0042] As Figures 1-4 shown, a port water engineering construction facility structure strain intelligent sensing device, the sensing device comprises:

[0043] A data acquisition module is embedded and installed in the parapet of the gravity type wharf, and is used for acquiring stress data values and temperature data values inside the parapet;

[0044] An environmental temperature measurement module is randomly distributed on the outer facade of the parapet, and is used for acquiring environmental temperature data of the outer facade of the parapet under different media, the media including water and air;

[0045] A data processing module is in communication connection with the data acquisition module and the environmental temperature measurement module, and is used for acquiring and preprocessing the stress data values and temperature data values inside the parapet and the environmental temperature data of the outer facade of the parapet;

[0046] A data analysis module acquires a prediction result of stress values of each point inside the parapet under different environmental temperatures by constructing a BP neural network analysis model according to a data processing result of the data processing module;

[0047] An early warning module generates a warning level according to a prediction result of the data analysis module;

[0048] A visualization module is in communication connection with the data analysis module and the early warning module, and is used for visually displaying the prediction result of the stress values of the parapet and the environmental temperature and the warning level.

[0049] In the embodiment, the stress values of each point inside the parapet under different environmental temperatures of the parapet in the air medium can be predicted by the technical solution of the present application, and the specific implementation steps are as follows:

[0050] Step 1, collecting data of each sensor in the air medium by the data acquisition module and the data processing module, respectively as Ta, b ij , a ij , wherein i≤k, k is the layer number of the sensor closest to and above the water surface in the second positive direction;

[0051] Step 2, the collected data is preprocessed by the data processing module to obtain data matrix Y1 and X1 of the collected data, wherein, , ,

[0052] Step 3, a BP neural network analysis model is constructed by the data analysis module, the normalized X1' matrix of the data matrix X1 is taken as the input vector of the network, the normalized Y1' matrix of the data matrix Y1 is taken as the output vector of the network, and the neural network is constructed by taking the newff function as an example:

[0053] Function form:

[0054] net = newff (PR, [S1, S2, …, SN], {TF1, TF2, …, TFN}, BTF, BLF, PF)

[0055] Wherein, PR: the matrix composed of sample data, composed of an R×2 matrix of maximum value and minimum value; Si: the number of nodes in the i-th layer, a total of N layers;

[0056] TFi: transfer function of nodes in the i-th layer, including linear transfer function purelin; tangent S-type transfer function tansig; logarithmic S-type transfer function logsig, default is "tansig";

[0057] BTF: training function, used for adjusting network weights and thresholds, default training function trainlm based on Levenberg_Marquardt conjugate gradient method.

[0058] BLF: network learning function, including BP learning rule learngd; BP learning rule learngdm with momentum term. Default is "learngdm";

[0059] PF: network performance analysis function, including mean absolute error performance analysis function mae; mean square performance analysis function mse, default is "mse".

[0060] Generally, the first four parameters are set in the use process, and the last two parameters adopt system default parameters.

[0061] Example:

[0062] net = newff([-1,1],[3,1],{'tansig','purelin'});

[0063] Where [-1,1] represents the minimum and maximum values of the input vector, [3,1] represents that the hidden layer has 3 nodes and the output layer has 1 node; tansig, purelin distribution represents the transfer function of the hidden layer and the output layer;

[0064] The prediction results of the stress values of each point inside the parapet under different environmental temperatures are output by the constructed neural network, and through analysis of the prediction results, the change trend of the stress values of each point inside the parapet under different environmental temperatures in the air medium can be obtained.

[0065] Embodiment 2

[0066] As shown in Figures 1-4 A port water engineering construction facility structure strain intelligent sensing device, the sensing device comprises:

[0067] A data acquisition module is embedded and installed inside the parapet of the gravity type wharf, and is used for acquiring stress data values and temperature data values inside the parapet;

[0068] An environmental temperature measurement module is randomly distributed on the outer facade of the parapet, and is used for acquiring environmental temperature data of the outer facade of the parapet under different media, wherein the media include water and air;

[0069] A data processing module is in communication connection with the data acquisition module and the environmental temperature measurement module, and is used for acquiring and preprocessing the stress data values and the temperature data values inside the parapet and the environmental temperature data of the outer facade of the parapet;

[0070] A data analysis module acquires prediction results of stress values of each point inside the parapet under different environmental temperatures by constructing a BP neural network analysis model according to the data processing results of the data processing module;

[0071] An early warning module generates an early warning level according to the prediction results of the data analysis module;

[0072] A visualization module is in communication connection with the data analysis module and the early warning module, and is used for visually displaying the prediction results of the stress values of the parapet and the environmental temperature and the early warning level.

[0073] In this embodiment, the stress values of each point inside the parapet under different environmental temperatures of the parapet in the water medium can be predicted by the technical solution of the present application, and the specific implementation steps are as follows:

[0074] Step 1: Collect the data of each sensor in the air medium by the data acquisition module and the data processing module, respectively as Ta, b ij , a ij , wherein i>k, k is the layer number of the sensor closest to and above the water surface in the second positive direction;

[0075] Step 2, the collected data is preprocessed by the data processing module to obtain data matrix Y2 and X2 of the collected data, wherein, , ,

[0076] Step 3, a BP neural network analysis model is constructed by the data analysis module, the X2 matrix after normalization of the data matrix X2 is taken as the input vector of the network, the Y2 matrix after normalization of the data matrix Y2 is taken as the output vector of the network, and the newff function is taken as an example for neural network construction:

[0077] Function form:

[0078] net = newff (PR, [S1, S2, …, SN], {TF1, TF2, …, TFN}, BTF, BLF, PF)

[0079] Wherein, PR: the matrix composed of sample data, composed of an R x 2 matrix of maximum and minimum values; Si: the number of nodes in the i-th layer, a total of N layers;

[0080] TFi: transfer function of the i-th layer node, including linear transfer function purelin; tangent S-type transfer function tansig; logarithmic S-type transfer function logsig, default is "tansig";

[0081] BTF: training function, used for adjusting network weights and thresholds, default training function trainlm based on Levenberg_Marquardt conjugate gradient method.

[0082] BLF: network learning function, including BP learning rule learngd; BP learning rule learngdm with momentum term. Default is "learngdm";

[0083] PF: network performance analysis function, including mean absolute error performance analysis function mae; mean square performance analysis function mse, default is "mse".

[0084] Generally, the first four parameters are set in the use process, and the last two parameters use the system default parameters.

[0085] Example:

[0086] net = newff([-1,1],[3,1],{'tansig','purelin'});

[0087] Where [-1,1] represents the minimum and maximum values of the input vector, [3,1] represents that the hidden layer has 3 nodes and the output layer has 1 node; tansig, purelin distribution represents the transfer function of the hidden layer and the output layer;

[0088] The constructed neural network outputs the predicted stress values ​​of each point inside the breast wall under different ambient temperatures. By analyzing the predicted results, the changing trend of stress values ​​at each point inside the breast wall under different ambient temperatures in a water medium can be obtained.

[0089] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A smart sensing device for structural strain of port hydraulic engineering facilities, characterized in that: The sensing device includes: A data acquisition module is pre-embedded inside the breast wall of the gravity wharf to collect stress and temperature data inside the breast wall. An ambient temperature measurement module is randomly distributed on the outer facade of the breast wall to collect ambient temperature data of the outer facade of the breast wall under different media, including water and air. The data processing module is communicatively connected to the data acquisition module and the ambient temperature measurement module, and is used to acquire and preprocess stress data and temperature data inside the breast wall and ambient temperature data of the outer facade of the breast wall. The data analysis module, based on the data processing results of the data processing module, constructs a BP neural network analysis model to obtain the predicted stress values ​​of various points inside the breast wall under different ambient temperatures. An early warning module generates an early warning level based on the prediction results of the data analysis module; The visualization module is communicatively connected to the data analysis module and the early warning module, and is used to visualize the predicted results and early warning levels of the breast wall stress value and the ambient temperature.

2. The intelligent sensing device for structural strain of port hydraulic engineering facilities according to claim 1, characterized in that: The data acquisition module includes strain sensors and temperature sensors inside the breast wall. Both strain sensors and temperature sensors inside the breast wall are arranged in a matrix distribution along the center of the breast wall section of the gravity wharf, and are uniformly spaced in a grid pattern outwards along the longitudinal and transverse sections of the breast wall. The longitudinal section of the breast wall is arranged in one direction, and the transverse section is arranged in two directions.

3. The intelligent sensing device for structural strain of port hydraulic engineering facilities according to claim 1, characterized in that: The ambient temperature measurement module includes exterior facade temperature sensors, and the number of exterior facade temperature sensors distributed in different media is not less than two.

4. The intelligent sensing device for structural strain of port hydraulic engineering facilities according to claim 1, characterized in that: The preprocessing steps of the data processing module are as follows: S1, acquire all data collected by the data acquisition module and the ambient temperature measurement module; S2, calculate the average of the temperature sensor measurements on all exterior facades under different media, and obtain the ambient temperature data Ta and Tw of the breast wall exterior facade under different media, where, Where, Ta is the ambient temperature data of the exterior facade of the breast wall under air medium, Tw is the ambient temperature data of the exterior facade of the breast wall under water medium, n is the number of exterior facade temperature sensors distributed under air medium, and m is the number of exterior facade temperature sensors distributed under water medium. S3, number the strain sensor and the internal temperature sensor of the breast wall as A respectively. ij and B ij The numbering rules are as follows: a) Define the direction from the outer facade of the breast wall to the center of the longitudinal section as the first positive direction, and j represents the number of the layers where the sensors are arranged along the first positive direction. The closer to the center of the longitudinal section, the larger j is. b) Define the direction from the top to the bottom of the breast wall as the second positive direction, where i represents the layer number of the sensors arranged along the second positive direction, and i is larger the closer to the bottom of the breast wall; S4, construct data matrices Y1, Y2, X1, X2 based on the numbering results of the strain sensors and the temperature sensors inside the breast wall, where, , , , Among them, a ij For strain sensor A ij The data value collected at time t, b ij Temperature sensor B inside the breast wall ij The data value collected at time t, where k is the layer number of the sensor that is closest to and above the water surface in the second positive direction; S5, normalize the values ​​of each element in the data matrices Y1, Y2, X1, X2 to obtain the normalized data matrices Y1', Y2', X1', X2'.

5. The intelligent sensing device for structural strain of port hydraulic engineering facilities according to claim 1, characterized in that: The BP neural network analysis model is constructed using data matrices X1' and X2' as input vectors and Y1' and Y2' as output vectors.

6. The intelligent sensing device for structural strain of port hydraulic engineering facilities according to claim 1, characterized in that: The method for classifying the warning levels is as follows: Step 1: Obtain the predicted stress values ​​P of the breast wall at each point. j ; Step 2, using the prediction result data P j A sample set is created from the numerical values, denoted as {P1, P2, ..., P...}. n }; Step 3: Obtain the mean and standard deviation of the sample set, and standardize the data using the mean and standard deviation. The standardization formula is: In this formula, z is the standard parameter, σ is the variance of the sample data, and μ is the mean of the sample data; Step 4: After standardization is completed, utilize the standard parameters. The numerical range is adjusted to [0,1], and the warning level is classified using the function value of f(k). The classification mechanism is as follows: when At that time, the warning level was classified as Level 1; when At that time, the warning level was classified as Level II.

Images