Structural health monitoring device for civil engineering

Abstract

The invention discloses a structural health monitoring device for civil engineering. According to the technical scheme, the structural health monitoring device is characterized by comprising a multi-parameter sensing module used for collecting strain, acceleration, displacement, temperature, humidity, vibration frequency and structural surface crack image parameters of a civil engineering structure; the multi-parameter sensing module comprises a strain sensor group, an acceleration sensor group, a displacement sensor group, a temperature and humidity sensor, a vibration sensor and an image acquisition unit; the strain sensor group is composed of a plurality of fiber Bragg grating strain sensors which are installed at key stress parts of the structure in a distributed arrangement mode. According to the structural health monitoring device for civil engineering provided by the invention, by virtue of the unique multi-parameter sensing module, the advanced data processing module and the flexible and reliable data transmission module, various excellent effects are shown in the field of civil engineering structural health monitoring.

Description

Technical Field

[0001] The present invention relates to the field of civil engineering technology, and in particular to a structural health monitoring device for civil engineering, which is particularly suitable for real-time health monitoring and status assessment of large civil engineering structures such as bridges, high-rise buildings, and large venues. Background Art

[0002] With the rapid development of civil engineering, an increasing number of large and complex structures are emerging. Over the long term, these structures are subject to a variety of factors, including the natural environment (such as wind, earthquakes, and temperature fluctuations), loads (such as vehicle and personnel loads), and material aging. This can lead to gradual degradation of structural performance and damage such as cracks and deformation. Timely and accurate monitoring of the health of these structures is crucial to ensuring their safety and extending their service life.

[0003] Currently, common methods for monitoring the health of civil engineering structures mainly include manual inspections and traditional sensor monitoring. Manual inspections have disadvantages such as low efficiency, strong subjectivity, and difficulty in detecting internal damage, and cannot meet the needs of real-time, continuous monitoring of large structures. Traditional sensor monitoring, such as strain gauges and accelerometers, although able to obtain some structural response data, has problems such as single monitoring parameters, complex installation and maintenance, and susceptibility to environmental interference. In addition, existing monitoring devices also have deficiencies in data processing, transmission, and analysis, and are unable to quickly and accurately conduct comprehensive assessments and early warnings of structural health status. Therefore, there is an urgent need to develop a civil engineering structure health monitoring device with more comprehensive functions, higher intelligence, and greater reliability. Summary of the Invention

[0004] In response to the issues mentioned in the background art, the present invention aims to provide a structural health monitoring device for civil engineering projects, addressing existing issues such as limited monitoring parameters, complex installation and maintenance, and insufficient data processing and analysis capabilities. This device can comprehensively monitor multiple health parameters of civil engineering structures in real time, efficiently process and intelligently analyze the monitoring data, and accurately assess and provide early warning of structural health.

[0005] The above technical objectives of the present invention are achieved through the following technical solutions:

[0006] A structural health monitoring device for civil engineering, comprising a multi-parameter sensing module for collecting strain, acceleration, displacement, temperature, humidity, vibration frequency, and surface crack image parameters of a civil engineering structure; the multi-parameter sensing module comprises a strain sensor group, an acceleration sensor group, a displacement sensor group, a temperature and humidity sensor, a vibration sensor, and an image acquisition unit;

[0007] The strain sensor group is composed of multiple fiber Bragg grating strain sensors, which are installed in a distributed manner at key stress-bearing parts of the structure; the acceleration sensor group is composed of multiple triaxial acceleration sensors, which are installed at different floors or mid-span positions of the structure; the displacement sensor group adopts a combination of laser displacement sensors and capacitive displacement sensors, the laser displacement sensors are installed at fixed positions outside the structure, and the capacitive displacement sensors are installed at key nodes inside the structure; the temperature and humidity sensors are high-precision digital temperature and humidity sensors, which are distributed and installed in different areas inside the building structure; the vibration sensors are piezoelectric vibration sensors, which are arranged at parts of the building structure that are prone to vibration; the image acquisition unit includes multiple high-definition cameras and image acquisition cards, and the high-definition cameras are installed on the surface of the structure;

[0008] The multi-parameter sensing module is connected to a data processing module, which is used to receive the raw data collected by the multi-parameter sensing module and perform preprocessing, feature extraction and fusion analysis on the data. The data processing module includes a preprocessing unit, a feature extraction unit and a data fusion unit;

[0009] It also includes: a data transmission module for transmitting the data processed by the data processing module to the remote monitoring center, the data transmission module adopts a combination of wireless transmission and wired transmission, and includes a wireless transmission unit and a wired transmission unit;

[0010] Remote monitoring center, which includes a data storage server, a data analysis and evaluation system, and an early warning system;

[0011] A power management module, which provides a stable power supply for the monitoring device and includes a solar panel, a battery, and a power conversion circuit;

[0012] An intelligent perception fusion layer is added between the multi-parameter perception module and the data processing module, and the intelligent perception fusion layer includes:

[0013] Heterogeneous sensor spatiotemporal alignment unit: This unit uses a Kalman filter algorithm based on timestamp synchronization to perform spatiotemporal alignment on sensor data from fiber Bragg grating strain sensors and high-definition cameras with different sampling frequencies, eliminating monitoring errors caused by asynchronous data collection.

[0014] Self-calibration sensor network: A micro-electromechanical system calibration unit is embedded in the strain sensor group, and a self-calibration program is periodically triggered by the built-in standard strain gauge to achieve self-correction of sensor accuracy;

[0015] The data processing module includes an edge computing server, which is integrated with:

[0016] Lightweight convolutional neural network: A lightweight model is deployed in the feature extraction unit to perform real-time semantic segmentation of crack images and complete crack width and length measurement for a single image within 50ms.

[0017] Federated learning module: The federated learning module supports multiple monitoring devices to form a distributed computing network, collaboratively training the structural damage identification model without sharing the original data, protecting user data privacy while improving the model's generalization ability.

[0018] Preferably, the data analysis and evaluation system of the remote monitoring center adopts a deep learning algorithm to optimize the structural health status evaluation model by learning and training historical monitoring data.

[0019] Preferably, the power management module also includes a power monitoring unit, which monitors the battery power in real time and transmits the power information to the data processing module. When the power is lower than a set threshold, a charging reminder signal is sent to the remote monitoring center through the data transmission module.

[0020] Preferably, the multi-parameter sensing module further includes a stress sensor, which is used to monitor the stress distribution inside the structure. The stress sensor is a piezoresistive stress sensor and is installed at a key position of the building structure.

[0021] Preferably, the data processing module is connected to a calibration module, which is used to regularly calibrate the sensors in the multi-parameter sensing module. The calibration module includes a standard signal generator and a calibration control unit. The standard signal generator generates known standard strain, acceleration, and displacement signals. The calibration control unit controls the output of the standard signal and compares the measured value of the sensor with the standard value, and calibrates and corrects the sensor according to the comparison result.

[0022] Preferably, the data transmission module further includes a data encryption unit, and the data encryption unit encrypts the transmitted data by combining a symmetric encryption algorithm and an asymmetric encryption algorithm.

[0023] Preferably, the wireless transmission unit adopts 4G / 5G communication technology and Bluetooth technology. When the structure is located in a location with good network coverage, the data is uploaded to the remote monitoring center through the 4G / 5G communication module. When the structure is in an area with weak network signals or no network coverage, Bluetooth technology is used to transmit the data to a nearby smart terminal device, and then the data is uploaded to the remote monitoring center through the smart terminal device; the wired transmission unit adopts fiber optic communication technology, laying fiber optic lines inside the structure and directly connecting the data processing module with the remote monitoring center.

[0024] Preferably, the fiber Bragg grating strain sensors in the strain sensor group adopt an array packaging structure, adjacent fiber Bragg grating strain sensors are connected by an optical wavelength division multiplexer, and each fiber Bragg grating strain sensor is provided with an independent temperature-compensated fiber Bragg grating, the reflection center wavelength of the temperature-compensated fiber Bragg grating differs from the reflection center wavelength of the corresponding strain sensor by 5-10nm. By jointly calculating the wavelength drift of the temperature-compensated fiber Bragg grating and the strain sensor, the influence of temperature change on the structural strain measurement result is eliminated.

[0025] Preferably, the three-axis acceleration sensor is also integrated with a data preprocessor, which can perform fast Fourier transform on the filtered acceleration signal, calculate the main frequency, secondary frequency and amplitude of each frequency component of the structural vibration in real time, and encode the calculation results according to a preset data format to reduce the amount of data transmission and improve data transmission efficiency.

[0026] Preferably, the feature extraction unit in the data processing module adopts a hybrid neural network architecture that combines a convolutional neural network and a long short-term memory network. The convolutional neural network is used to extract local features in the structural health parameter data, and the long short-term memory network is used to capture the time series characteristics of the data; the hybrid neural network architecture automatically learns and extracts structural health feature parameters by performing supervised training on a large amount of historical structural health monitoring data; the hybrid neural network architecture is also provided with an attention mechanism module, which can dynamically adjust the weight of each parameter feature according to the importance of different structural health parameters to the structural health status assessment.

[0027] In summary, the present invention mainly has the following beneficial effects:

[0028] The structural health monitoring device for civil engineering proposed in the present invention has demonstrated excellent effects in many aspects in the field of structural health monitoring of civil engineering by virtue of its unique multi-parameter sensing module, advanced data processing module, flexible and reliable data transmission module, powerful remote monitoring center and stable power management module. The present invention comprehensively and accurately collects multiple parameters, providing deep insights into the structural status; efficiently processes and integrates data to improve the accuracy of monitoring and analysis; adopts dual modes to achieve reliable data transmission to ensure real-time and stable data transmission; adopts intelligent analysis and early warning to achieve scientific management of the structural health status; and can achieve stable and sustainable power supply to ensure long-term reliable operation of the device. The perception layer improves data quality through self-calibration and spatiotemporal alignment, the edge computing node realizes localized intelligent analysis to reduce data transmission pressure, and three-mode communication ensures reliable data transmission, forming a complete innovative link of "high-precision acquisition-intelligent processing-reliable transmission-scientific decision-making". The present invention adopts fiber optic sensing + deep learning: combining the distributed measurement characteristics of fiber Bragg gratings with the image feature extraction capabilities of convolutional neural networks to solve the problem that traditional strain monitoring cannot intuitively associate structural surface damage. This invention adopts edge computing + Internet of Things module: deploying a lightweight AI model on the monitoring device side to realize the hierarchical processing mode of "edge computing + cloud collaboration", breaking through the computing power bottleneck of existing solutions relying on cloud servers, and reducing data transmission traffic by more than 50%. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is one of the system block diagrams of the present invention;

[0030] Figure 2 This is the second system block diagram of the present invention;

[0031] Figure 3 This is the third system block diagram of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments 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 are within the scope of protection of the present invention.

[0033] Example 1

[0034] refer to Figure 1-Figure 3A structural health monitoring device for civil engineering includes a multi-parameter sensing module for collecting strain, acceleration, displacement, temperature, humidity, vibration frequency, and surface crack image parameters of the civil engineering structure; the multi-parameter sensing module includes a strain sensor group, an acceleration sensor group, a displacement sensor group, a temperature and humidity sensor, a vibration sensor, and an image acquisition unit;

[0035] The strain sensor group consists of multiple fiber Bragg grating strain sensors, which are installed in a distributed manner at key stress-bearing locations of the structure. The acceleration sensor group consists of multiple triaxial acceleration sensors, which are installed at different floors or mid-span locations of the structure. The displacement sensor group uses a combination of laser displacement sensors and capacitive displacement sensors, with the laser displacement sensors installed at fixed locations outside the structure and the capacitive displacement sensors installed at key nodes inside the structure. The temperature and humidity sensors use high-precision digital temperature and humidity sensors, which are distributed and installed in different areas inside the building structure. The vibration sensor uses a piezoelectric vibration sensor, which is arranged at locations where vibration is likely to occur in the building structure. The image acquisition unit includes multiple high-definition cameras and image acquisition cards, and the high-definition cameras are installed on the surface of the structure.

[0036] The multi-parameter perception module is connected to a data processing module, which is used to receive the raw data collected by the multi-parameter perception module and perform preprocessing, feature extraction and fusion analysis on the data. The data processing module includes a preprocessing unit, a feature extraction unit and a data fusion unit;

[0037] It also includes: a data transmission module for transmitting the data processed by the data processing module to the remote monitoring center, the data transmission module adopts a combination of wireless transmission and wired transmission, and includes a wireless transmission unit and a wired transmission unit;

[0038] Remote monitoring center, which includes data storage server, data analysis and evaluation system, and early warning system;

[0039] Power management module, which provides a stable power supply for the monitoring device. The power management module includes a solar panel, a battery, and a power conversion circuit;

[0040] An intelligent perception fusion layer is added between the multi-parameter perception module and the data processing module. The intelligent perception fusion layer includes:

[0041] Heterogeneous sensor spatiotemporal alignment unit: This unit uses a Kalman filter algorithm based on timestamp synchronization to perform spatiotemporal alignment on sensor data from fiber Bragg grating strain sensors and high-definition cameras with different sampling frequencies, eliminating monitoring errors caused by asynchronous data collection.

[0042] Self-calibrating sensor network: A micro-electromechanical system calibration unit is embedded in the strain sensor group. The built-in standard strain gauge regularly triggers the self-calibration process to achieve self-correction of sensor accuracy.

[0043] The data processing module includes an edge computing server, which integrates:

[0044] Lightweight convolutional neural network: A lightweight model is deployed in the feature extraction unit to perform real-time semantic segmentation of crack images, completing crack width and length measurement for a single image within 50ms.

[0045] Federated learning module: The federated learning module supports multiple monitoring devices to form a distributed computing network, collaboratively training the structural damage identification model without sharing the original data, protecting user data privacy while improving the model generalization ability.

[0046] The perception layer improves data quality through self-calibration and spatiotemporal alignment, the edge computing nodes implement localized intelligent analysis, reduce data transmission pressure, and three-mode communication ensures reliable data transmission, forming a complete innovative chain of "high-precision acquisition-intelligent processing-reliable transmission-scientific decision-making". The present invention adopts fiber optic sensing + deep learning: combining the distributed measurement characteristics of fiber Bragg gratings with the image feature extraction capabilities of convolutional neural networks to solve the problem that traditional strain monitoring cannot intuitively associate structural surface damage. The present invention adopts edge computing + Internet of Things modules: deploying lightweight AI models at the monitoring device end to realize the hierarchical processing mode of "edge computing + cloud collaboration", breaking through the computing power bottleneck of existing solutions that rely on cloud servers, and reducing data transmission traffic by more than 50%.

[0047] Among them, the data analysis and evaluation system of the remote monitoring center adopts a deep learning algorithm to optimize the structural health status assessment model through learning and training of historical monitoring data.

[0048] Among them, the power management module also includes a power monitoring unit, which monitors the battery power in real time and transmits the power information to the data processing module. When the power is lower than the set threshold, a charging reminder signal is sent to the remote monitoring center through the data transmission module.

[0049] Among them, the multi-parameter perception module also includes a stress sensor, which is used to monitor the stress distribution inside the structure. The stress sensor adopts a piezoresistive stress sensor and is installed in key parts of the building structure.

[0050] Among them, the data processing module is connected to a calibration module, which is used to regularly calibrate the sensors in the multi-parameter sensing module. The calibration module includes a standard signal generator and a calibration control unit. The standard signal generator generates known standard strain, acceleration, and displacement signals. The calibration control unit controls the output of the standard signal and compares the measured value of the sensor with the standard value. The sensor is calibrated and corrected according to the comparison result.

[0051] The data transmission module further includes a data encryption unit, which encrypts the transmitted data by combining a symmetric encryption algorithm with an asymmetric encryption algorithm.

[0052] Among them, the wireless transmission unit adopts 4G / 5G communication technology and Bluetooth technology. When the structure is located in a location with good network coverage, the data is uploaded to the remote monitoring center through the 4G / 5G communication module. When the structure is in an area with weak network signals or no network coverage, Bluetooth technology is used to transmit data to nearby smart terminal devices, and then the data is uploaded to the remote monitoring center through the smart terminal devices; the wired transmission unit adopts fiber optic communication technology, laying fiber optic lines inside the structure and directly connecting the data processing module with the remote monitoring center.

[0053] Among them, the fiber Bragg grating strain sensors in the strain sensor group adopt an array packaging structure. Adjacent fiber Bragg grating strain sensors are connected by an optical wavelength division multiplexer, and each fiber Bragg grating strain sensor is equipped with an independent temperature-compensated fiber Bragg grating. The reflection center wavelength of the temperature-compensated fiber Bragg grating differs from the reflection center wavelength of the corresponding strain sensor by 5-10nm. By jointly calculating the wavelength drift of the temperature-compensated fiber Bragg grating and the strain sensor, the influence of temperature changes on the structural strain measurement results is eliminated.

[0054] Among them, the three-axis acceleration sensor is also integrated with a data preprocessor, which can perform fast Fourier transform on the filtered acceleration signal, calculate the main frequency, secondary frequency and amplitude of each frequency component of the structural vibration in real time, and encode the calculation results according to the preset data format to reduce the data transmission volume and improve data transmission efficiency.

[0055] Among them, the feature extraction unit in the data processing module adopts a hybrid neural network architecture that combines a convolutional neural network and a long short-term memory network. The convolutional neural network is used to extract local features in the structural health parameter data, and the long short-term memory network is used to capture the time series characteristics of the data; the hybrid neural network architecture automatically learns and extracts structural health feature parameters through supervised training on a large amount of historical structural health monitoring data; the hybrid neural network architecture is also equipped with an attention mechanism module, which can dynamically adjust the weight of each parameter feature according to the importance of different structural health parameters to the structural health status assessment.

[0056] The structural health monitoring device for civil engineering proposed in this invention, with its unique multi-parameter sensing module, advanced data processing module, flexible and reliable data transmission module, powerful remote monitoring center, and stable power management module, has demonstrated excellent results in many aspects of civil engineering structural health monitoring. The specific results can be analyzed from the following aspects:

[0057] 1. Comprehensive and accurate collection of multiple parameters, deep insight into structural status:

[0058] The multi-parameter sensing module integrates various sensor types to comprehensively collect a wide range of health parameters of civil engineering structures, including strain, acceleration, displacement, temperature, humidity, vibration frequency, and images of surface cracks. The strain sensor group utilizes fiber Bragg grating (FBG) strain sensors. Based on the strain-wavelength modulation principle of FBGs, the grating length is 5-10 mm, the reflection center wavelength is between 1530 and 1565 nm, and the strain sensitivity coefficient reaches 1.2 pm / με. It offers strong resistance to electromagnetic interference and high accuracy, enabling distributed measurement. It can accurately capture subtle strain changes in different parts of the structure under various loads and environmental factors, providing critical data for structural stress analysis. The acceleration sensor group, comprised of triaxial accelerometers, simultaneously measures acceleration responses in three mutually perpendicular directions. Analysis of this data accurately determines the structure's vibration characteristics, such as natural frequency and mode shape, enabling timely detection of abnormal vibrations and providing strong support for structural dynamics analysis.

[0059] The displacement sensor assembly combines laser displacement sensors and capacitive displacement sensors. The former, installed at fixed locations on the exterior of the structure, precisely measures overall displacement. The capacitive displacement sensors, installed at key internal nodes, accurately measure relative displacement between these nodes. Together, these sensors enable comprehensive, high-precision monitoring of the structure's displacement status. Temperature and humidity sensors monitor real-time changes in the structure's ambient temperature and humidity. Since temperature and humidity significantly impact the properties of structural materials, this data can be used to refine structural health assessments, making them more accurate. The vibration sensor, a piezoelectric sensor, is located in vibration-prone areas of the structure. It quickly and accurately converts structural vibration signals into electrical signals, capturing parameters such as amplitude and frequency, providing a basis for assessing the structure's vibration status. The image acquisition unit's high-definition camera features autofocus and optical image stabilization, with a resolution of at least 4K and a frame rate of at least 30 fps. This allows for clear, all-around coverage of the structure's surface, capturing high-quality images of cracks and other information, facilitating intuitive observation and analysis of surface damage. The coordinated operation of these sensors enables multi-dimensional, comprehensive, and accurate monitoring of the health of civil engineering structures, providing in-depth insight into structural changes under various operating conditions.

[0060] 2. Efficient data processing and integration to improve the accuracy of monitoring and analysis:

[0061] The data processing module efficiently processes the raw data collected by the multi-parameter sensing module. The preprocessing unit uses technologies such as wavelet filtering algorithms to effectively remove noise and interference signals from the data. For example, filtering the data collected by the fiber Bragg grating strain sensor can eliminate high-frequency noise introduced by external electromagnetic interference. It also performs grayscale and noise reduction operations on image data to improve data quality and reliability. The feature extraction unit extracts key characteristic parameters from the preprocessed data. For strain data, it extracts strain peak value, strain change rate, etc. For vibration data, it extracts natural frequency, damping ratio, etc. For image data, it uses edge detection, crack width calculation and other algorithms to extract crack-related characteristic parameters. These characteristic parameters can effectively reflect the health status of the structure.

[0062] The data fusion unit utilizes an improved DS evidence theory and introduces a weighted allocation strategy for evidence conflict to fuse data collected by different sensor types, such as strain, acceleration, and displacement. This strategy fully considers the reliability and relevance of each sensor's data, avoiding conflicts and misjudgments during the data fusion process. This allows the fused data to more comprehensively and accurately reflect structural health information, significantly improving the accuracy of structural health assessments. Compared to traditional single-sensor monitoring or simple data processing methods, it can more accurately determine the presence, extent, and location of structural damage.

[0063] 3. Dual-mode reliable data transmission to ensure real-time and stable data transmission:

[0064] The data transmission module utilizes a combination of wireless and wired transmission to ensure reliable and real-time data transmission. The wireless transmission unit integrates 4G / 5G communication technology and Bluetooth. When the structure is located in an area with good network coverage, the 4G / 5G communication module uploads data to the remote monitoring center at a high and stable transmission rate, enabling real-time remote data transmission and meeting the need for real-time monitoring of the structural health status. When the structure is in an area with weak network signals or no network coverage, Bluetooth technology can transmit data to a nearby smart terminal device, which then uploads the data to ensure continuous data transmission.

[0065] The wired transmission unit utilizes fiber-optic communication technology, laying fiber optic lines within the structure to directly connect the data processing module with the remote monitoring center. Fiber-optic communication offers the advantages of fast transmission speeds and strong anti-interference capabilities, ensuring that data transmission is unaffected by external electromagnetic interference and signal attenuation. It is particularly suitable for applications requiring high data transmission stability and reliability, such as long-term monitoring of large, critical civil engineering structures, providing a solid foundation for stable, high-speed data transmission. The dual-mode data transmission method enables the monitoring device to adapt to a variety of complex environmental conditions, ensuring timely and accurate data transmission to the remote monitoring center, supporting subsequent data analysis and decision-making.

[0066] 4. Intelligent analysis and early warning to achieve scientific management of structural health status:

[0067] The remote monitoring center has powerful data analysis and evaluation capabilities. The data storage server uses distributed storage technology to securely and reliably store the raw data collected by the multi-parameter sensing module and the result data processed by the data processing module, facilitating rapid data retrieval and query, laying the foundation for the accumulation and analysis of long-term monitoring data. The data analysis and evaluation system is based on big data analysis and deep learning algorithms. Through learning and training historical monitoring data, it continuously optimizes structural health assessment models, such as the neural network-based structural damage identification model. The system is capable of in-depth analysis and processing of stored data, not only to assess the health status of the structure in real time, but also to predict the health status change trend of the structure in the next 1-3 months through established prediction models, such as a hybrid prediction model based on the gray prediction model and the Markov chain model. The prediction error does not exceed 10%, allowing for early detection of potential structural safety hazards.

[0068] Based on the results of the data analysis and assessment system, the early warning system promptly issues warning signals in various forms, such as audible alarms, text message alarms, and email alarms, to notify relevant personnel when abnormalities in the structural health status occur. For example, if the monitored structural strain exceeds a set threshold, the vibration frequency changes abnormally, or the crack width continues to increase, the early warning system will quickly issue an alarm, allowing relevant personnel to take timely measures such as structural repair and reinforcement. This achieves scientific management of the health status of civil engineering structures, effectively ensures structural safety, and avoids major safety accidents. It also provides a scientific basis for structural maintenance and management, improving the efficiency and accuracy of maintenance and management.

[0069] 5. Stable and sustainable power supply to ensure long-term reliable operation of the device:

[0070] The power management module provides a stable power supply for the entire monitoring device. Solar panels, mounted in a well-lit location outside the structure, convert solar energy into electricity. A dual-axis automatic tracking system, including horizontal and vertical rotation mechanisms, uses a sun position sensor to monitor the sun's azimuth and altitude in real time, ensuring the solar panels remain perpendicular to the sun's rays. Tracking accuracy is no less than 0.5°, increasing power generation efficiency by at least 25% and providing ample renewable energy for the device.

[0071] The battery stores the electricity generated by the solar panels and powers the monitoring device during periods of low sunlight or at night. It utilizes high-performance lithium-ion batteries with large capacity, long life, and low self-discharge, ensuring stable operation at all times. The power conversion circuit converts the electricity output by the solar panels or stored in the batteries into voltages and currents suitable for each module, ensuring proper functioning. The power monitoring unit monitors the battery charge in real time and transmits this information to the data processing module. When the charge falls below a set threshold, a charging reminder is sent to the remote monitoring center via the data transmission module, enabling timely charging measures. This stable and sustainable power supply eliminates the need for external conventional power supply, reducing operating costs and reliance on external power, ensuring the long-term, reliable operation of the monitoring device and meeting the requirements for long-term health monitoring of civil engineering structures.

[0072] In summary, the civil engineering structure health monitoring device of the present invention demonstrates significant advantages in data acquisition, processing, transmission, analysis, and power supply through the collaborative work of multiple modules and innovative design. It can realize real-time, accurate, and comprehensive monitoring and scientific management of the health status of civil engineering structures, providing strong technical support for ensuring the safety and reliability of civil engineering structures. It has broad application prospects and important social and economic value.

[0073] Example 2

[0074] refer to Figures 1 to 3 The difference from Example 1 is that this example provides a case study, using a large bridge as an example to illustrate the specific implementation of the present invention. Fiber Bragg grating strain sensors are distributedly installed at key locations of the bridge, such as the main beam, piers, and abutments, to form a strain sensor group that monitors the strain changes of the bridge under the influence of vehicle loads, wind forces, and other factors in real time. Triaxial acceleration sensors are installed at different mid-span locations and on the tops of the piers to form an acceleration sensor group that monitors the vibration characteristics of the bridge. Laser displacement sensors are installed at fixed locations on both sides of the bridge, and capacitive displacement sensors are installed at key nodes within the bridge to form a displacement sensor group that monitors the overall displacement of the bridge and the relative displacement of the nodes. High-precision digital temperature and humidity sensors are installed in different areas within the bridge to monitor the temperature and humidity changes in the environment in which the bridge is located. Piezoelectric vibration sensors are installed at locations on the bridge that are prone to vibration to monitor the vibration status of the bridge. Multiple high-definition cameras are installed on the surface of the bridge to form an image acquisition unit that captures images of cracks on the bridge surface in real time.

[0075] The raw data collected by the multi-parameter sensing module is transmitted to the data processing module. The preprocessing unit first filters and denoises the data to remove noise interference. The feature extraction unit extracts characteristic parameters such as strain peak value, natural frequency, and crack width from the preprocessed data. The data fusion unit uses the improved DS evidence theory to fuse data from different types of sensors to obtain more accurate bridge health information.

[0076] Data processed by the data processing module is transmitted to the remote monitoring center via the data transmission module. When the network signal is good, the data is uploaded to the remote monitoring center via the 4G / 5G communication module. When the network signal is poor, the data is transmitted to a nearby smart terminal device via Bluetooth technology, which then uploads the data. Data can also be stably transmitted to the remote monitoring center via fiber optic lines laid within the bridge using fiber optic communication technology.

[0077] The data storage server of the remote monitoring center stores the collected raw data and processed result data; the data analysis and evaluation system analyzes and processes the data based on deep learning algorithms, and evaluates the health status of the bridge in real time through the established bridge health status assessment model; when the bridge health status is abnormal, the early warning system promptly issues early warning signals in various forms such as sound, text messages, and emails to notify relevant maintenance personnel to handle the situation.

[0078] The power management module's solar panels, mounted in a well-lit location atop the bridge, convert solar energy into electricity and store it in batteries. The power conversion circuit converts the battery energy into a voltage and current suitable for each module, powering the entire monitoring system. The power monitoring unit monitors the battery charge in real time. When the charge falls below a set threshold, a charging reminder is sent to the remote monitoring center via the data transmission module.

[0079] The calibration module regularly calibrates the sensors in the multi-parameter sensing module. A standard signal generator produces known standard strain, acceleration, displacement, and other signals. The calibration control unit controls the output of the standard signal and compares the sensor's measured value with the standard value. Based on the comparison results, the sensor is calibrated and corrected to ensure the accuracy of the sensor's measurement data.

[0080] The data encryption unit in the data transmission module uses a combination of symmetric encryption algorithm and asymmetric encryption algorithm to encrypt the transmitted data to prevent the data from being stolen or tampered with during transmission, thereby ensuring the security of data transmission.

[0081] The structural health monitoring device for civil engineering provided by the present invention realizes the comprehensive collection of multiple health parameters of the structure through a multi-parameter sensing module, analyzes the data using advanced data processing and fusion technologies, and transmits the data to a remote monitoring center in combination with a reliable data transmission method, thereby realizing real-time and accurate monitoring of the health status of civil engineering structures.

[0082] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A structural health monitoring device for civil engineering, characterized in that: It includes a multi-parameter sensing module for collecting strain, acceleration, displacement, temperature, humidity, vibration frequency and surface crack image parameters of civil engineering structures; the multi-parameter sensing module includes a strain sensor group, an acceleration sensor group, a displacement sensor group, a temperature and humidity sensor, a vibration sensor and an image acquisition unit; The strain sensor group is composed of multiple fiber Bragg grating strain sensors, which are installed in a distributed manner at key stress-bearing parts of the structure; the acceleration sensor group is composed of multiple triaxial acceleration sensors, which are installed at different floors or mid-span positions of the structure; the displacement sensor group adopts a combination of laser displacement sensors and capacitive displacement sensors, the laser displacement sensors are installed at fixed positions outside the structure, and the capacitive displacement sensors are installed at key nodes inside the structure; the temperature and humidity sensors are high-precision digital temperature and humidity sensors, which are distributed and installed in different areas inside the building structure; the vibration sensors are piezoelectric vibration sensors, which are arranged at parts of the building structure that are prone to vibration; the image acquisition unit includes multiple high-definition cameras and image acquisition cards, and the high-definition cameras are installed on the surface of the structure; The multi-parameter sensing module is connected to a data processing module, which is used to receive the raw data collected by the multi-parameter sensing module and perform preprocessing, feature extraction and fusion analysis on the data. The data processing module includes a preprocessing unit, a feature extraction unit and a data fusion unit; It also includes: a data transmission module for transmitting the data processed by the data processing module to the remote monitoring center, the data transmission module adopts a combination of wireless transmission and wired transmission, and includes a wireless transmission unit and a wired transmission unit; Remote monitoring center, which includes a data storage server, a data analysis and evaluation system, and an early warning system; A power management module, which provides a stable power supply for the monitoring device and includes a solar panel, a battery, and a power conversion circuit; An intelligent perception fusion layer is added between the multi-parameter perception module and the data processing module, and the intelligent perception fusion layer includes: Heterogeneous sensor spatiotemporal alignment unit: This unit uses a Kalman filter algorithm based on timestamp synchronization to perform spatiotemporal alignment on sensor data from fiber Bragg grating strain sensors and high-definition cameras with different sampling frequencies, eliminating monitoring errors caused by asynchronous data collection. Self-calibration sensor network: A micro-electromechanical system calibration unit is embedded in the strain sensor group, and a self-calibration program is periodically triggered by the built-in standard strain gauge to achieve self-correction of sensor accuracy; The data processing module includes an edge computing server, which is integrated with: Lightweight convolutional neural network: A lightweight model is deployed in the feature extraction unit to perform real-time semantic segmentation of crack images and complete crack width and length measurement for a single image within 50ms. Federated learning module: The federated learning module supports multiple monitoring devices to form a distributed computing network, collaboratively training the structural damage identification model without sharing the original data, protecting user data privacy while improving the model's generalization ability.

2. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The data analysis and evaluation system of the remote monitoring center adopts a deep learning algorithm to optimize the structural health status evaluation model by learning and training historical monitoring data.

3. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The power management module also includes a power monitoring unit, which monitors the battery power in real time and transmits the power information to the data processing module. When the power is lower than the set threshold, a charging reminder signal is sent to the remote monitoring center through the data transmission module.

4. The structural health monitoring device for civil engineering according to claim 3, characterized in that: The multi-parameter sensing module also includes a stress sensor, which is used to monitor the stress distribution inside the structure. The stress sensor is a piezoresistive stress sensor and is installed at a key position of the building structure.

5. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The data processing module is connected to a calibration module, which is used to regularly calibrate the sensors in the multi-parameter sensing module. The calibration module includes a standard signal generator and a calibration control unit. The standard signal generator generates known standard strain, acceleration, and displacement signals. The calibration control unit controls the output of the standard signal and compares the measured value of the sensor with the standard value. The sensor is calibrated and corrected according to the comparison result.

6. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The data transmission module further includes a data encryption unit, which encrypts the transmitted data by combining a symmetric encryption algorithm with an asymmetric encryption algorithm.

7. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The wireless transmission unit adopts 4G / 5G communication technology and Bluetooth technology. When the structure is located in a location with good network coverage, the data is uploaded to the remote monitoring center through the 4G / 5G communication module. When the structure is in an area with weak network signals or no network coverage, Bluetooth technology is used to transmit the data to a nearby smart terminal device, and then the data is uploaded to the remote monitoring center through the smart terminal device; the wired transmission unit adopts optical fiber communication technology, laying optical fiber lines inside the structure and directly connecting the data processing module with the remote monitoring center.

8. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The fiber Bragg grating strain sensors in the strain sensor group adopt an array packaging structure. Adjacent fiber Bragg grating strain sensors are connected by an optical wavelength division multiplexer, and each fiber Bragg grating strain sensor is provided with an independent temperature-compensated fiber Bragg grating. The reflection center wavelength of the temperature-compensated fiber Bragg grating differs from the reflection center wavelength of the corresponding strain sensor by 5-10nm. By jointly calculating the wavelength drift of the temperature-compensated fiber Bragg grating and the strain sensor, the influence of temperature changes on the structural strain measurement results is eliminated.

9. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The three-axis acceleration sensor is also integrated with a data preprocessor, which can perform fast Fourier transform on the filtered acceleration signal, calculate the main frequency, secondary frequency and amplitude of each frequency component of the structural vibration in real time, and encode the calculation results according to a preset data format to reduce the data transmission volume and improve data transmission efficiency.

10. The structural health monitoring device for civil engineering according to claim 1, characterized in that: The feature extraction unit in the data processing module adopts a hybrid neural network architecture that combines a convolutional neural network and a long short-term memory network. The convolutional neural network is used to extract local features in the structural health parameter data, and the long short-term memory network is used to capture the time series characteristics of the data; the hybrid neural network architecture automatically learns and extracts structural health feature parameters through supervised training on a large amount of historical structural health monitoring data; the hybrid neural network architecture is also provided with an attention mechanism module, which can dynamically adjust the weight of each parameter feature according to the importance of different structural health parameters to the structural health status assessment.

Images