A system for measuring the inclination of agricultural building wall structures

Abstract

The present invention relates to the technical field of agricultural engineering structure monitoring and discloses a system for measuring the inclination of agricultural building wall structures. The system includes a tilt data acquisition module, a structural benchmark construction module, a tilt calculation module, an anomaly identification and classification module, and a visualization and early warning module. The system acquires spatial coordinates and initial posture through laser ranging and posture fusion, constructs a reference reference plane using point cloud fitting, calculates the inclination angle between the current posture vector and the reference in real time, and establishes a structural drift model to extract plane offset data. The system utilizes a tilt state clustering algorithm to cluster tilt fluctuation characteristics and combines dynamic thresholds to determine the current state category, achieving classified identification and multi-level early warning of wall states. Finally, it graphically displays tilt hotspot areas in the form of a three-dimensional heat map and remotely transmits early warning signals. The system of the present invention has high accuracy, fast response, and strong adaptability, making it suitable for intelligent monitoring and safety early warning of agricultural building structures.

Description

Technical Field

[0001] The invention relates to the technical field of agricultural engineering structure monitoring, and in particular to an agricultural building wall structure inclination measurement system. Background Art

[0002] In modern agricultural building structures, particularly those in greenhouses, livestock and poultry farms, and grain storage facilities, wall tilt or structural deformation can directly impact building safety and production efficiency. Because these buildings are often located in environmentally volatile areas and subject to the combined effects of wind loads, foundation settlement, and water vapor erosion, their walls are susceptible to irreversible deformation. Traditional inclination measurement methods rely primarily on manual visual inspection or fixed-point monitoring with handheld goniometers. These methods suffer from low monitoring frequency, high data dispersion, and a reliance on experience, making them unable to meet the requirements for continuous, precise, and intelligent monitoring of large-scale structures.

[0003] Currently, some research attempts to deploy electronic devices such as inclinometers and accelerometers on walls for point-based posture monitoring. While this improves automated data collection capabilities, it is limited by the data dimensions and spatial distribution of a single or small number of points, making it difficult to accurately reconstruct the spatial form and posture evolution of the entire wall structure. Furthermore, existing solutions often use fixed thresholds for anomaly detection, lacking dynamic adaptability to changes in wall conditions, resulting in high false alarm rates and poor generalization capabilities, limiting their deployment and application in real-world agricultural scenarios.

[0004] On the other hand, while some high-precision industrial structure monitoring solutions have incorporated technologies such as point cloud fitting and 3D modeling, these solutions are difficult to implement in agricultural buildings due to high equipment costs, high power consumption, and complex algorithms. Furthermore, these solutions often overlook the deep correlation between data fluctuation characteristics and tilt behavior classification, resulting in a lack of a complete closed-loop monitoring architecture encompassing data acquisition, structural modeling, posture analysis, anomaly classification, and early warning linkage. Therefore, there is an urgent need for an agricultural building wall tilt measurement system that integrates multi-point fusion perception, 3D modeling and analysis, intelligent clustering and classification, and remote visual early warning. Summary of the Invention

[0005] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a system for measuring the inclination of an agricultural building wall structure, comprising the following modules:

[0006] Tilt data acquisition module, used to obtain the spatial coordinate data of each measurement point on the target wall within a set time interval and initial value of tilt angle ;

[0007] Structural benchmark building module for Generate wall structure reference plane and the vertical reference vector ;

[0008] Tilt calculation module for vertical reference vector Calculate the current inclination angle of each measuring point on the wall with the current posture vector , and calculate the current structural plane offset data ;

[0009] Anomaly recognition and classification module, used to compare the tilt angle and the initial value of the tilt angle , combined with time series and fluctuation characteristics , determine and classify wall tilt behavior;

[0010] Visualization and warning module, used to display the current tilt angle , fluctuation characteristics and classification results Display the graph and Generate early warning signals .

[0011] Preferably, the tilt data acquisition module includes:

[0012] Laser distance measuring unit for setting frequency Get the wall surface The spatial coordinate data of a fixed measuring point in a three-dimensional coordinate system ;

[0013] Attitude estimation unit, used to combine gyroscope and accelerometer signals through quaternion transformation Calculate the initial value of the wall inclination angle , the quaternion transformation The posture fusion calculation is performed using the following formula:

[0014] ;

[0015] in, is the rotation axis vector, obtained jointly by the accelerometer and gyroscope, is the attitude angle increment of the current sampling period, Improved the accuracy and stability of wall posture recognition.

[0016] Data synchronization processing unit, used for coordinate data Initial value of the inclination angle with the wall Perform time stamp alignment and perform time mean filtering on the sampling point data using a sliding window mechanism to output a stable data set. and For use by subsequent modules.

[0017] Preferably, the structural benchmark building module includes:

[0018] Point cloud fitting unit, used to stabilize the coordinates of multiple measurement points Perform plane fitting to construct the current reference plane , and its fitting function is expressed as:

[0019] ;

[0020] in, represents the components of the plane normal vector, is the fitting offset constant;

[0021] The vertical vector generation unit is used to extract the plane normal vector and normalize it to construct the vertical reference vector. , used for subsequent tilt angle judgment, where .

[0022] Preferably, the inclination calculation module includes:

[0023] Attitude deviation analysis unit, used to compare vertical reference vectors at each time With the current posture vector Angle between , the calculation formula is:

[0024] ;

[0025] in, for The current wall posture vector at this moment, representing The wall surface normal vector at the moment is calculated as:

[0026] Get the coordinates of several spatial points on the current wall surface, , which is then obtained through the point cloud fitting unit ,but ;

[0027] Structural drift modeling unit, used to calculate plane offset data based on the deformation trend of consecutive points on the wall surface , using the structural transformation matrix , construct the following formula:

[0028] ;

[0029] in, is the local transformation affine matrix, and Respectively indicate time and time The spatial coordinate data point set, the structural transformation matrix Based on the difference in collected point values, the following affine modeling formula is used:

[0030] ;

[0031] in, 、 is the scaling factor in each direction, 、 is the rotation component, 、 is a translation term, which is used to accurately describe the deformation and tilt change process of the wall surface.

[0032] Preferably, the anomaly identification and classification module includes:

[0033] Tilt behavior clustering unit, used to classify the tilt behavior based on the current tilt angle and the initial value of the tilt angle Constructing tilt fluctuation features , and classified by the tilt state clustering algorithm. The core of the algorithm is expressed as follows:

[0034] ;

[0035] in, is the classification result, represents the principal component transformation matrix, is the K-means-hierarchical hybrid clustering operator, Represents the fluctuation characteristics, , represents the mean value of the tilt angle change, represents the standard deviation of the tilt angle variation, It represents the rate of change of the tilt angle, and the formula is as follows:

[0036] ;

[0037] ;

[0038] ;

[0039] is the length of time the viewport is observed;

[0040] The tilt state clustering algorithm comprises the following steps:

[0041] First, perform feature normalization to fluctuate features , through principal component transformation Will Mapping to the dimensionality reduction feature subspace, and then combining with the K-means clustering algorithm for parallel execution, obtains more stable tilt behavior clustering results and improves the robustness of anomaly recognition.

[0042] Dynamic threshold calculation unit, used to build adaptive alarm thresholds based on historical trends and environmental parameters , and its calculation formula is:

[0043] ;

[0044] in, is the tilt angle The historical average of the data, is the tilt angle The standard deviation of the data, 、 is the adaptive coefficient, satisfying , To accumulate the tilt duration, the system records the ,when If it is determined to have started to tilt when , the difference between the time when the tilt started and the time when the measurement started is recorded as ;

[0045] Abnormal state classification unit, used for value, and The current state of the wall is divided into the following predefined types:

[0046] .

[0047] Preferably, the visualization and early warning module includes:

[0048] Graphic visualization unit, used to display the current tilt angle The value is mapped to the surface of the structural 3D model, and the tilt hotspot area is highlighted in the form of a color heat map, while the fluctuation characteristics are displayed in real time. Curve trend chart;

[0049] Warning signal generating unit, used for and Trigger an alarm when an abnormality occurs, generating an early warning signal ;

[0050] Information return interface unit, used to transmit warning signals The data is packaged in JSON format, sent to the remote monitoring platform, and recorded in the local anomaly database for subsequent query.

[0051] The heat map is based on The distribution adopts pseudo color mapping function and dynamically adjusts the color range. After being triggered, the information will be transmitted to the cloud platform and mobile terminal in real time, realizing remote monitoring and alarm closed loop.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] Realize accurate perception of the wall's global posture and dynamic construction of structural benchmarks: This invention uses laser ranging and posture fusion algorithms to achieve high-frequency acquisition of the spatial coordinates and posture angles of multiple measuring points on the wall, and combines point cloud fitting technology to construct dynamically updated structural reference datum planes and vertical reference vectors, breaking through the limitations of traditional single-point inclination monitoring and improving the spatial modeling accuracy and posture calculation robustness of large-scale wall structures.

[0054] Constructing an evolvable tilt anomaly identification and classification model: This invention introduces a tilt state clustering algorithm to automatically complete cluster modeling based on the tilt angle fluctuation characteristics. At the same time, it combines a dynamic threshold function to achieve adaptive classification and early warning judgment, effectively identifying normal, suspicious and abnormal states, and improving the system's ability to understand complex wall deformation trends and classification accuracy.

[0055] Forming a closed-loop intelligent monitoring system of collection-analysis-display-early warning: The present invention realizes a closed-loop process of wall status from data collection, structural modeling, tilt analysis to classification identification and early warning display through the collaborative work of multiple modules. Combined with three-dimensional thermal maps and remote signal push mechanisms, it has strong real-time, visualization and remote response capabilities, and is suitable for the actual deployment needs of structural health monitoring in agricultural building scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of the system module flow provided for this application;

[0057] Figure 2 Schematic diagram of the tilt data acquisition module provided for this application;

[0058] Figure 3 Schematic diagram of the structural benchmark building blocks provided for this application;

[0059] Figure 4 Schematic diagram of the inclination calculation module provided in this application;

[0060] Figure 5 Schematic diagram of the anomaly identification and classification module provided for this application;

[0061] Figure 6 Schematic diagram of the visualization and warning module provided for this application. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0063] refer to Figures 1-6 The embodiment of the present invention provides an agricultural building wall structure inclination measurement system, including the following modules:

[0064] Tilt data acquisition module, used to obtain the spatial coordinate data of each measurement point on the target wall within a set time interval and initial value of tilt angle .

[0065] In this embodiment, the tilt data acquisition module is deployed on the facade structure of a large-span greenhouse wall agricultural building. To ensure spatial coverage and posture perception accuracy, eight equally spaced fixed measurement points are selected as data collection points. The module consists of three parts: a laser ranging unit, a posture estimation unit, and a data synchronization processing unit. First, the laser ranging unit uses a laser sensor with an accuracy of ±1mm and collects the three-dimensional coordinate values ​​of each measurement point at a frequency of 10Hz. The format is:

[0066] ;

[0067] The acquired coordinates are based on a unified coordinate system, and the sensor position and the original structural surface of the wall have been spatially matched and calibrated during the initial calibration.

[0068] At the same time, the attitude estimation unit integrates a MEMS nine-axis module, which updates the attitude angle once per second through data fusion of the gyroscope and accelerometer. The quaternion fusion algorithm is used for attitude estimation, and its rotation expression is:

[0069] ;

[0070] in, is the instantaneous rotation axis vector composed of the acceleration direction and the angular velocity direction, Increment angle for the attitude.

[0071] Subsequently, the data synchronization processing unit uses the built-in high-precision clock to add time stamps to the sensor data and synchronizes the multiple point data with the initial attitude angle value. Then, the spatial coordinate data and initial attitude values ​​are processed by time mean through a sliding window filter (the window width is set to 5 seconds) to output a stable value. and , providing a highly reliable data foundation for structural benchmark construction.

[0072] Structural benchmark building module for Generate wall structure reference plane and the vertical reference vector .

[0073] In this embodiment, the structural reference construction module is used to generate the spatial reference plane of the wall in real time, which is used as the reference for tilt angle calculation and drift modeling. First, the stable multi-point spatial coordinate data collected is processed by the point cloud fitting unit. .

[0074] The coordinate data of the aforementioned 8 measuring points are selected and input into the least square fitting model. The fitting plane function is as follows:

[0075] ;

[0076] in, are the components of the plane normal vector, is the structural offset constant, obtained by solving the matrix to minimize the sum of squared errors. The system outputs this plane expression in real time as the structural reference plane.

[0077] Subsequently, the vertical vector generation unit normalizes the plane normal vector to construct the reference attitude vector:

[0078] ,in, ;

[0079] vector Used for comparison with the real-time wall posture, it serves as the core benchmark for subsequent tilt angle calculation and drift identification.

[0080] The entire module periodically refreshes the reference plane and normal vector to effectively respond to slight material changes and structural deformations caused by temperature and humidity changes in agricultural buildings, thereby improving the long-term stability and computational reliability of the system.

[0081] Tilt calculation module for vertical reference vector Calculate the current inclination angle of each measuring point on the wall with the current posture vector , and calculate the current structural plane offset data .

[0082] This embodiment calculates the actual tilt angle of the wall at each moment by working in conjunction with the posture deviation analysis unit and the structure drift modeling unit. and plane structure offset First, get the coordinates of the current wall measurement point , perform plane fitting through the least squares fitting model to obtain the current normal vector of the wall:

[0083] ;

[0084] The tilt angle is calculated using the space vector angle formula:

[0085] ;

[0086] This angle reflects the degree of deviation between the current wall surface and the ideal vertical reference;

[0087] In order to further characterize the position drift of the wall surface, the structural drift modeling unit introduces the structural transformation matrix , calculate the in-plane structural drift:

[0088] ;

[0089] in, and Respectively indicate time and time The spatial coordinate data point set;

[0090] , 、 is the scaling factor in each direction, 、 is the rotation component, 、 is a translation term, which is used to accurately describe the deformation and tilt change process of the wall surface.

[0091] Anomaly recognition and classification module, used to compare the tilt angle and the initial value of the tilt angle , combined with time series and fluctuation characteristics , judge and classify the wall tilt behavior.

[0092] In the anomaly recognition and classification module, the tilt angle fluctuation feature vector is first extracted: , represents the mean value of the tilt angle change, represents the standard deviation of the tilt angle variation, represents the rate of change of the tilt angle, where:

[0093] ;

[0094] ;

[0095] ;

[0096] is the length of time the viewport is observed;

[0097] Then, principal component analysis is performed on the fluctuation eigenvector to reduce its dimension, and the dimension reduction result is used as input to construct the tilt state clustering algorithm. Classification model, obtain classification results:

[0098] ;

[0099] in, is the classification result, represents the principal component transformation matrix, is the K-means-hierarchical hybrid clustering operator, Indicates fluctuation characteristics;

[0100] At the same time, in order to improve the adaptability of abnormal judgment, the system builds dynamic alarm thresholds:

[0101] ;

[0102] in, is the tilt angle The historical average of the data, is the tilt angle The standard deviation of the data, 、 is the adaptive coefficient, satisfying , To accumulate the tilt duration, the system records the ,when If it is determined to have started to tilt when , the difference between the time when the tilt started and the time when the measurement started is recorded as ;

[0103] Finally, for value, and The current state of the wall is divided into the following predefined types:

[0104] .

[0105] Visualization and warning module, used to display the current tilt angle , fluctuation characteristics and classification results Display the graph and Generate early warning signals .

[0106] In this embodiment, the system realizes the visualization and intelligent warning mechanism of the wall tilt state based on the three-dimensional model. First, the graphic visualization unit Mapped onto the building facade model surface, a dynamic heat map is generated, where the colors are controlled by a pseudo-color mapping function. Tilt hotspots (red areas) are highlighted in real time, allowing maintenance personnel to intuitively identify structural stress concentration points.

[0107] At the same time, the system is based on Parametric curves generate historical fluctuation trend charts, supporting horizontal time and vertical position comparisons. These charts automatically scale axes and support dynamic refresh, making it easy to observe wall stability over time.

[0108] When satisfied When the following conditions are met, the system will trigger an early warning signal:

[0109] ;

[0110] This signal is triggered by the early warning generation unit and then encapsulated into a standard JSON data packet via the information feedback interface and reported to the remote cloud platform and mobile app. The data structure includes timestamp, location number, tilt angle, fluctuation characteristics, and classification tags, supporting compatibility with existing agricultural construction information platforms.

[0111] This module builds a complete link from three-dimensional visualization to intelligent linkage alarm, adapts to the needs of modern agricultural remote supervision, and significantly improves the interactivity, responsiveness and practicality of the tilt monitoring system.

[0112] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0113] Obviously, the embodiments described above are only some embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are given in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present invention specification and drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A system for measuring the inclination of an agricultural building wall structure, characterized in that: Includes the following modules: Tilt data acquisition module, used to obtain the spatial coordinate data of each measurement point on the target wall within a set time interval and initial value of tilt angle ; Structural benchmark building module for Generate wall structure reference plane and the vertical reference vector ; Tilt calculation module for vertical reference vector Calculate the current inclination angle of each measuring point on the wall with the current posture vector , and calculate the current structural plane offset data ; Anomaly recognition and classification module, including tilt behavior clustering unit, is used to classify the anomaly based on the current tilt angle. and the initial value of the tilt angle Constructing tilt fluctuation features , and classified by the tilt state clustering algorithm. The core of the algorithm is expressed as follows: ； in, is the classification result, represents the principal component transformation matrix, is the K-means-hierarchical hybrid clustering operator, Represents the fluctuation characteristics, ; Dynamic threshold calculation unit, used to build adaptive alarm thresholds based on historical trends and environmental parameters , and its calculation formula is: ； in, is the tilt angle The historical average of the data, is the tilt angle The standard deviation of the data, 、 is the adaptive coefficient, satisfying , To accumulate the tilt duration, the system records the ,when If it is determined to have started to tilt when , the difference between the time when the tilt started and the time when the measurement started is recorded as ; Abnormal state classification unit, used for value, and The current state of the wall is divided into the following predefined types: ； Visualization and warning module, used to display the current tilt angle , fluctuation characteristics and classification results Display the graph and Generate early warning signals .

2. The agricultural building wall structure inclination measurement system according to claim 1, characterized in that: The tilt data acquisition module includes: Laser distance measuring unit for setting frequency Get the wall surface The spatial coordinate data of a fixed measuring point in a three-dimensional coordinate system ; Attitude estimation unit, used to combine gyroscope and accelerometer signals through quaternion transformation Calculate the initial value of the wall inclination angle ; Data synchronization processing unit, used for coordinate data Initial value of the inclination angle with the wall Perform time stamp alignment and perform time mean filtering on the sampling point data using a sliding window mechanism to output a stable data set. and For use by subsequent modules.

3. The agricultural building wall structure inclination measurement system according to claim 1, characterized in that: The structural benchmark building module includes: Point cloud fitting unit, used to stabilize the coordinates of multiple measurement points Perform plane fitting to construct the current reference plane , and its fitting function is expressed as: ； in, represents the components of the plane normal vector, is the fitting offset constant; The vertical vector generation unit is used to extract the plane normal vector and normalize it to construct the vertical reference vector , used for subsequent tilt angle judgment, where .

4. The agricultural building wall structure inclination measurement system according to claim 1, characterized in that: The inclination calculation module includes: Attitude deviation analysis unit, used to compare vertical reference vectors at each time With the current posture vector Angle between , the calculation formula is: ； in, for The current wall posture vector at this moment, representing The wall surface normal vector at the moment is calculated as: Get the coordinates of several spatial points on the current wall surface, , which is then obtained through the point cloud fitting unit ,but ; Structural drift modeling unit, used to calculate plane offset data based on the deformation trend of consecutive points on the wall surface , using the structural transformation matrix , construct the following formula: ； in, is the local transformation affine matrix, and Respectively indicate time and time A set of spatial coordinate data points.

5. The agricultural building wall structure inclination measurement system according to claim 1, characterized in that: The tilt fluctuation characteristics in the tilt behavior cluster unit Also includes: , represents the mean value of the tilt angle change, represents the standard deviation of the tilt angle variation, It represents the rate of change of the tilt angle, and the formula is as follows: ； ； ； The length of time to observe the viewport.

6. The agricultural building wall structure inclination measurement system according to claim 1, characterized in that: The visualization and early warning module includes: Graphic visualization unit, used to display the current tilt angle The value is mapped to the surface of the structural 3D model, and the tilt hotspot area is highlighted in the form of a color heat map, while the fluctuation characteristics are displayed in real time. Curve trend chart; Warning signal generating unit, used for and Trigger an alarm when an abnormality occurs, generating an early warning signal ; Information return interface unit, used to transmit warning signals The data is packaged in JSON format, sent to the remote monitoring platform, and recorded in the local anomaly database for subsequent query.

7. The agricultural building wall structure inclination measurement system according to claim 2, characterized in that: The quaternion transformation The posture fusion calculation is performed using the following formula: ； in, is the rotation axis vector, obtained jointly by the accelerometer and gyroscope, is the attitude angle increment of the current sampling period, Improved the accuracy and stability of wall posture recognition.

8. The agricultural building wall structure inclination measurement system according to claim 4, characterized in that: The structural transformation matrix Based on the difference in collected point values, the following affine modeling formula is used: ； in, 、 is the scaling factor in each direction, 、 is the rotation component, 、 is a translation term, which is used to accurately describe the deformation and tilt change process of the wall surface.

9. The agricultural building wall structure inclination measurement system according to claim 5, characterized in that: The tilt state clustering algorithm comprises the following steps: First, perform feature normalization to fluctuate features , through principal component transformation Will Mapping to the dimensionality reduction feature subspace, and then combining with the K-means clustering algorithm for parallel execution, obtains more stable tilt behavior clustering results and improves the robustness of anomaly recognition.

10. The agricultural building wall structure inclination measurement system according to claim 6, characterized in that: The heat map is based on The distribution adopts pseudo color mapping function and dynamically adjusts the color range. After being triggered, the information will be transmitted to the cloud platform and mobile terminal in real time, realizing remote monitoring and alarm closed loop.

Images