Bridge monitoring method based on cloud platform

Abstract

The invention discloses a bridge monitoring method based on a cloud platform, and relates to the technical field of bridge monitoring. The method comprises the steps of firstly building a hardware monitoring system, and selecting a multimode sensor, a data processing module, a 4G transmission module and a power supply module; a software monitoring system is constructed, a modern Web application development architecture is adopted, functions of user login management, data display and downloading, bridge inclination prediction and the like are set at a Web end, and data interaction is realized by means of an Ali cloud IOT platform and an AMQP protocol. Bridge deflection prediction adopts a single-layer linear regression method, and an inclination angle is calculated according to wind speed and temperature data and is displayed at a webpage end in real time. According to the method, the network transmission technology, the multi-sensor fusion technology and the artificial intelligence technology are fused, the bridge monitoring method based on the wireless sensor network is formed, and the healthy surrounding information and the bridge health state of a target bridge can be accurately monitored in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge monitoring, and more specifically, it relates to a bridge monitoring method based on a cloud platform. Background Art

[0002] Bridges are vulnerable to various factors during design, construction and service life, and damage accumulation will threaten the structural safety and even cause serious consequences. Therefore, it is crucial to conduct all-weather information-based monitoring on bridge structure damage.

[0003] Traditional bridge monitoring using wired methods has disadvantages such as complex wiring, high cost and difficult maintenance. While wireless sensor networks are widely used in many fields, with obvious advantages and great potential. At the same time, industries such as railways also attach importance to the application of detection and monitoring technologies and incorporate them into the framework of important technical systems to promote development.

[0004] Many domestic and foreign teams have carried out practices on bridge monitoring systems. However, some systems have deficiencies, such as limited measurement range and low efficiency. In view of this, the present invention is committed to integrating multiple technologies to construct a wireless sensor bridge monitoring system, which has advantages such as a long information transmission distance, the ability to combine deep learning to judge the bridge inclination degree and real-time display. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a bridge monitoring method based on a cloud platform.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A bridge monitoring method based on a cloud platform, comprising the following steps:

[0008] Build a hardware monitoring system, including selecting a multi-mode sensor module, a data processing module, a 4G transmission module and a power module; the multi-mode sensor collects real-time bridge information and transmits it to the single-chip microcomputer, and after the single-chip microcomputer processes it, the data is transmitted to the IOT platform of Alibaba Cloud through the 4G module in accordance with the MQTT protocol;

[0009] Build a software monitoring system, adopting a modern Web application development technology architecture covering front-end and back-end technology stacks; realize data interaction with the hardware monitoring system through the Alibaba Cloud IOT platform and the AMQP protocol;

[0010] Use single-layer linear regression to predict the bridge deflection, calculate the predicted bridge inclination angle based on the received wind speed and temperature data and display it in real time on the web page.

[0011] Preferably, the multi-mode sensor includes an ADXL tilt sensor, an AHT10 temperature and humidity sensor, a wind speed sensor, and a strain sensor; the data processing module selects the single-chip microcomputer STM32F411CEU6; the 4G communication module selects the Air780E communication module; the power module consists of a HU9196 charging module and an 18650 lithium battery.

[0012] Preferably, the Alibaba Cloud IOT platform combines the AMQP protocol to achieve message communication between IoT devices and the backend system; the Web front-end part of the software monitoring system is responsible for processing the display and interaction logic of the user interface; the Web back-end part of the software monitoring system is mainly responsible for processing business logic, data storage, and interaction with the front-end; the web page of the software monitoring system is designed with user login management, real-time data and chart display, monitoring data download, historical monitoring data query, and bridge tilt prediction.

[0013] Preferably, the Web front-end part includes HTML, CSS, JavaScript, Axios, and Webpack; the Web back-end part includes Node.js, Express, and Mongo; Mongo includes Mongoose and MongoDB; HTML is used to define the specific content of the system, CSS is used for the layout and style design of the web page, JavaScript is used for the dynamic interaction function of the web page, Axios is used for data interaction with the back-end, and Webpack is used for packaging and optimizing front-end resources; Node.js provides a high-performance running environment for the entire system, Express is used to build Web services, and Mongoose and MongoDB are used for data storage and management.

[0014] Preferably, single-layer linear regression is used to predict the bridge deflection, and the basic formula of the regression model is: ŷ = β0 + β1x1 + β2x2 + … β n x n , where ŷ is the predicted value (in this case, the tilt angle of the bridge), β0 is the intercept, β1, β2, …, β n are the regression coefficients, which respectively correspond to the features x1, x2, …, x n ; x1, x2, …, x n are the input features (in this system, they are wind speed and temperature); in the bridge tilt prediction, the training process is carried out through model.fit(X train , y train ), where X train is the feature matrix of the training data containing wind speed and temperature, and y trainis the target variable of the corresponding bridge tilt angle; when the Web end of the software monitoring system receives new wind speed and temperature data, it makes a prediction through model.predict(input_data), calculates the predicted bridge tilt angle and displays it in real time on the web page.

[0015] Preferably, the real-time measurement data of the hardware monitoring system is transmitted to the IOT platform of Alibaba Cloud through a 4G module via the MQTT protocol; the software monitoring system transmits the data of the IOT platform to the Web end through the AMQP protocol to realize real-time interaction and processing of the data; to achieve data interaction and user experience, functions such as user login management, real-time display of data, download of bridge monitoring data, and query of historical data information are designed in the Web end of the software monitoring system.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. In the present invention, multi-mode sensors such as ADXL tilt sensors, AHT10 temperature and humidity sensors, wind speed sensors, and strain sensors are integrated, which can comprehensively obtain various key parameters of the bridge, such as tilt, temperature and humidity, wind speed, and strain conditions, etc., and characterize the bridge health status from multiple dimensions, improving the accuracy of the bridge structure state assessment.

[0018] 2. In the present invention, with the support of the MQTT protocol and the Alibaba Cloud IOT platform supported by the 4G communication module (Air780E communication module), the hardware monitoring system can transmit the measurement data to the cloud platform in real time through the 4G public network, and the software monitoring system then obtains the data from the cloud platform through the AMQP protocol and displays it on the Web end, breaking through geographical restrictions and realizing remote real-time monitoring of the bridge condition, facilitating the management personnel to master the bridge information anytime and anywhere.

[0019] 3. In the present invention, rich functions such as user login management, real-time data and chart display, monitoring data download, historical monitoring data query, and bridge tilt prediction are designed in the Web end of the software monitoring system. User login management ensures system security and usage permission control; real-time data display is convenient for intuitively viewing the latest state of the bridge; data download and query functions are convenient for subsequent in-depth analysis and archiving; bridge tilt prediction helps to discover potential safety hazards in advance.

[0020] 4. In the present invention, in terms of hardware, the single-chip microcomputer supports multiple programming methods such as Micro Python and Arduino programming, and Flash pads are reserved for large data storage and function expansion; in terms of software, a modern Web application development technology architecture is adopted, which is easy to integrate new function modules or interface with other relevant systems according to actual needs in the future, and can adapt to the changing requirements of bridge monitoring. The sensors have the ability to work stably in harsh environments. For example, the AHT10 temperature and humidity sensor has strong anti-interference ability, and the wind speed sensor is suitable for long-term outdoor use, etc., enabling the entire monitoring system to operate reliably in the complex bridge service environment and having a wide range of applications.

[0021] 5. In the present invention, through the real-time monitoring, analysis of various aspects of bridge data and prediction of the tilt angle, bridge structure damage and potential safety risks can be discovered in a timely manner, providing scientific and accurate data support for decision-making such as bridge maintenance, repair, and management, helping to reasonably arrange resources, extend the service life of the bridge, ensure the safe operation of the bridge, and reduce economic losses and casualties caused by sudden bridge accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic structural diagram of a bridge monitoring method based on a cloud platform proposed by the present invention;

[0023] Figure 2 It is a framework diagram of the hardware monitoring system of a bridge monitoring method based on a cloud platform proposed by the present invention;

[0024] Figure 3 It is a framework diagram of the software monitoring system of a bridge monitoring method based on a cloud platform proposed by the present invention;

[0025] Figure 4 It is a data processing code diagram proposed by the present invention;

[0026] Figure 5 It is a schematic diagram of the working principle of information interaction proposed by the present invention;

[0027] Figure 6 It is a schematic diagram of the peripheral circuit of the ADXL tilt sensor proposed by the present invention;

[0028] Figure 7 It is an effect diagram of the login page proposed by the present invention;

[0029] Figure 8 It is an effect diagram of the real-time data chart display proposed by the present invention;

[0030] Figure 9 It is an effect diagram of the real-time data large screen display proposed by the present invention;

[0031] Figure 10The effect diagram of real-time data query and download proposed by the present invention;

[0032] Figure 11 The effect diagram of real-time data acquisition proposed by the present invention. Specific implementation manners

[0033] The embodiments further illustrate a bridge monitoring method based on a cloud platform proposed by the present invention.

[0034] Refer to Figure 1 , a bridge monitoring method based on a cloud platform includes the following steps:

[0035] Build a hardware monitoring system, including selecting a multi-mode sensor module, a data processing module, a 4G transmission module, and a power supply module; the multi-mode sensor collects real-time information of the bridge and transmits it to the single-chip microcomputer, and after being processed by the single-chip microcomputer, the data is transmitted to the IOT platform of Alibaba Cloud through the 4G module using the MQTT protocol;

[0036] Build a software monitoring system, adopting a modern Web application development technology architecture covering front-end and back-end technology stacks; design user login management, real-time data and chart display, monitoring data download, historical monitoring data query, and bridge tilt prediction functions on the Web side, and realize data interaction with the hardware monitoring system through the Alibaba Cloud IOT platform and the AMQP protocol;

[0037] Adopt single-layer linear regression to predict the bridge deflection, calculate the predicted bridge tilt angle according to the received wind speed and temperature data, and display it in real time on the web page.

[0038] Refer to Figure 2 , powered by the power supply module, long press the POW key of the 4G communication module until the green light flashes, at this time the communication module is successfully turned on, then press the BOOTo key of the single-chip microcomputer STM32 once, the red light flashes, at this time the entire hardware monitoring system is successfully powered on, each module is successfully powered, the multi-sensor module starts to work, collects the real-time information of the bridge and transmits it to the single-chip microcomputer STM32 in the form of digital signals, the microcontroller sequentially reads the corresponding data of the multi-mode sensor through the serial bus, converts the digital signal into digital data, stores the data in the external SRAM through the DMA method, and then sends it to the 4G communication module through the UART serial acquisition circuit, and sends it to the IOT platform of Alibaba Cloud using the 4G public network.

[0039] The multi-mode sensor module includes an ADXL tilt sensor, an AHT10 temperature and humidity sensor, a wind speed sensor, and a strain sensor;

[0040] The health condition of the bridge is mostly characterized by parameters such as inclination. After selection, the ADXL inclination sensor is finally chosen. This sensor is a small-sized, thin, low-power, and complete three-axis accelerometer that provides a signal-conditioned voltage output and can measure acceleration with a full-scale range of at least ±3g. It can measure the static gravitational acceleration in tilt detection applications, as well as the dynamic acceleration caused by motion, shock, or vibration. The bandwidth of this accelerometer is selected using the capacitors X_OUT, Y_OUT, and Z_OUT on the CX, CY, and CZ pins. The bandwidth range of the X and Y axes is 0.5Hz to 1600Hz, and the bandwidth range of the Z axis is 0.5Hz to 550Hz. The inclination parameter is set as the moving distance of the Z axis in the X axis. SCL is connected to the B10 serial port of the single-chip microcomputer STM32, SDA is connected to the B3 serial port of the single-chip microcomputer, and SDO is connected to the power supply module.

[0041] The strain sensor selected is a foil strain gauge of model BF350-3AA, with a nominal resistance of 120Ω, a limit strain of 1%, and a sensitivity coefficient of 2% ± 1%. The measurement circuit is a Wheatstone bridge, which can convert the change in the resistance value of the resistance strain gauge (caused by force) into a change in the output voltage of the bridge, thereby effectively realizing the conversion of strain into voltage for easy measurement and acquisition. The output voltage of the single-arm bridge of the strain gauge is Connect the OUT of this sensor to the A6 serial port of the single-chip microcomputer STM32, and connect VCC to the power supply module.

[0042] The temperature and humidity sensor uses the AHT10 temperature and humidity sensor. The power supply range of the AHT10 is 1.8 - 3.6V, and the recommended voltage is 3.3V. The temperature and humidity sensor outputs a calibrated digital signal and outputs it through the IIC communication method. It is equipped with a newly designed ASIC dedicated chip, an improved MEMS semiconductor capacitive humidity sensing element, and a standard on-chip temperature sensing element, which greatly improves the reliability of the sensor, enables it to work stably in harsh environments, has a relative humidity resolution of 0.1%RH, a temperature resolution of about 0.1°C, and strong anti-interference ability in actual applications. Connect VDD to the power supply module, and connect SDA and SCL to the A8 and A9 serial ports of the single-chip microcomputer to complete data transmission.

[0043] The wind speed sensor is small and light, easy to carry and assemble, with a range of 0-60m / s (default 0-30m / s), a resolution of 0.1m / s, and anti-electromagnetic interference processing; the bottom outlet method is adopted to completely eliminate the aging problem of the rubber pad of the aviation plug, and it can still be effectively waterproof after long-term use. The sensor uses high-performance imported bearings, and the rotation resistance is very small in actual application, and the measurement is very accurate; the all-aluminum shell has very high mechanical strength, high hardness in actual application, excellent corrosion resistance, and it will not rust even if it is used in harsh environments, so it can be used for a long time in outdoor environments, and the equipment structure and weight ratio are close to carefully designed and system distribution, and the moment of inertia is small, and the response is very sensitive; it can be applied to both four-wire and three-wire connection methods, the green line is connected to the positive pole of the system, the brown line is connected to the acquisition end, and the black line is connected to the negative pole of the system.

[0044] The data processing module uses the single-chip microcomputer STM32F411CEU6; the single-chip microcomputer uses a 25MHZ high-speed crystal oscillator and a 32.768Khz low-speed crystal oscillator. Both use high-quality metal shell crystal oscillators, which have better oscillation effects than the traditional single-chip microcomputer STM32F103C8T6. It has a reserved Flash pad to meet the needs of large data storage and micro python, provides USB Disk&&FATFFS routines, supports Micro Python programming, and provides available Micro Python firmware, and supports Arduino programming, the MCU version is V3.1, with 3 buttons, reset button, BOOT0 button and user button; STM32F411CEU6 has up to 256 to 512KB of Flash memory and up to 128KB of SRAM; provides a variety of packages from 49 to 100 pins; 3 USARTs, with a speed of up to 12.5Mbit / s, are used to connect to the 4G communication module to complete the function of information transmission, and 2 full-duplex IICs, up to 32 bits / 192KHz, are used to connect to the temperature and humidity sensor and the inclination sensor, and quickly process the temperature and humidity information collected by the sensor and the bridge inclination information into digital form.

[0045] The 4G communication module uses the Air780E communication module of Shanghai Hezhou Company. The total Flash of this module reaches 5MB, the Flash available for user code is greater than 1.6MB, it supports SMS text messages, the power supply voltage is 3.8V, and there are three power consumption modes, namely normal mode, low power mode, and PSM+ mode; this module contains multiple interfaces, and is connected to the A2 and A3 serial ports of the microcontroller STM32 through UART to complete the data transmission function; this module supports the MQTT protocol and the HTTP protocol, and is connected to the IOT platform of Alibaba Cloud through the MQTT protocol, and the information processed by the microcontroller is displayed in real time on the IOT platform. The working status of the hardware monitoring system and the real-time data of each module can be viewed on the platform end.

[0046] In practical applications, since the 4G communication module consumes a large amount of power during operation, in order to achieve a longer battery life, a power supply solution combining a large-capacity lithium battery pack and the HU9196 charging module is adopted. This module is jointly composed of the HU9196 charging module (including a type-c interface) and 18650 lithium batteries (3.7V, 2.6Ah). The charging module can effectively charge the lithium battery while powering the hardware monitoring system in cooperation with the lithium battery.

[0047] Refer to Figure 3 , the software monitoring system adopts a typical modern Web application development technology architecture, covering the front-end and back-end technology stacks. The front-end part of the software monitoring system is responsible for handling the display and interaction logic of the user interface. HTML is used to define the specific content of the system, CSS is used for the layout and style design of the web page, JavaScript is used for the dynamic interaction function of the web page, Axios is used for data interaction with the back-end, and Webpack is used for packaging and optimizing front-end resources. The back-end part of the software monitoring system is mainly responsible for handling business logic, data storage, and interaction with the front-end. Node.js provides a high-performance operating environment for the entire system, Express is used to build web services, Mongoose and MongoDB are used for data storage and management, and the Alibaba Cloud IOT platform and AMQP are used to achieve message communication between IoT devices and the back-end system.

[0048] Refer to Figure 4 , the convolutional neural network is one of the most widely used artificial neural networks at present, with good non-linear mapping ability, self-learning ability, and fault tolerance ability. It is the core part of the forward neural network and is mainly used in pattern recognition, function approximation, data compression, prediction estimation, classification, etc. In the present invention, single-layer linear regression is used to predict the bridge deflection. The basic formula of the regression model is: ŷ = β0 + β1x1 + β2x2 + … β n x n , where ŷ is the predicted value (in this case, the tilt angle of the bridge), β0 is the intercept, β1, β2, …, β n are the regression coefficients, which respectively correspond to the features x1, x2, …, x n ; x1, x2, …, x nis the input feature (wind speed and temperature in this system); during the code execution, the training model is carried out through model.fit(X_train, y_train), where X_train is the feature matrix of the training data containing wind speed and temperature, and y_train is the target variable of the corresponding bridge tilt angle; the training process will find the optimal β0, β1, β2 (in this case, there are only two features, so there are only two regression coefficients β1 and β2, corresponding to wind speed and temperature); when the Web side of the software monitoring system sends new wind speed and temperature data through a POST request, these data will be used to create a new DataFrame, and then prediction is carried out through model.predict(input_data); this prediction process is to use the above-mentioned linear regression formula, substitute the new wind speed and temperature values into the formula, calculate the predicted bridge tilt angle, and display it in real time on the web page; index.html obtains the interface data through the interface link to Python and then renders it to the display interface.

[0049] Refer to Figure 5 , the real-time measurement data of the hardware monitoring system is transmitted to the IOT platform of Alibaba Cloud through the 4G module via the MQTT protocol, and the software monitoring system transmits the data of the IOT platform to the Web side through the AMQP protocol to achieve real-time interaction and processing of data; to achieve good data interaction and user experience, functions such as user login management, real-time display of data (data display and bar chart), download of bridge monitoring data, and query of historical data information are designed in the Web side of the software monitoring system.

[0050] The software monitoring system first designs the user login management function. Under the login page, the functions of user login management and new user registration can be completed; the user login management function is implemented using the Flask-Login framework. The essence of the Flask-Login framework is to implement the setting and operation of this function based on session. Because session usually has a long timeliness during the system operation, Flask-Login will accurately record the session data information of the logged-in user within the expiration time set by the system; the effect diagram of the login page is as Figure 7 shown.

[0051] The software monitoring system designs the real-time data display page of each sensor and the global display page of the monitoring system to complete the real-time display function of data; two data display methods are set on the real-time data display page, namely, display in the form of a line chart and overall large-screen display. The effect diagrams can be seen respectively in Figure 8 and Figure 9; The figure shows the data change of each sensor in the most recent ten minutes (the time can be adjusted) in real time. The module displays the latest data collected by each sensor. In actual applications, since there is a certain intermittency when the data is collected and uploaded to the Alibaba Cloud IOT platform, there will be a certain time deviation in the displayed data (which can be reduced by setting the collection and upload interval time of the hardware monitoring system); The line chart of real-time data is made by Echarts, and the data required for making the chart is obtained by Axios requesting the server and Node.js.

[0052] On the design page of the bridge monitoring data download and historical data information query function, there is a time selection control. By selecting the time in the time box, the data query work of the specified query time by the user can be realized. The data that can be queried is the historical data monitored and uploaded by each sensor, and it can be presented in the form of a line chart (the data of multiple sensors in a certain period can be displayed in parallel); Here, the data query is also obtained by Axios requesting the server and Node.js; The time selection control uses the daterangepicker time component of Bootstrap, and at the same time, the query of the default time period and the custom time period is also designed; The line chart under this module is also made by Echarts, and the data table is made by the Bootstrap-table plugin in Bootstrap; Based on this plugin, the functions of query, pagination, setting display columns and data download are designed and implemented in the table; When using this plugin to implement the data download function, the excel data download mode is adopted. The effect diagrams of data query and download are as Figure 10 shown.

[0053] The bridge monitoring system is close to the actual engineering application, which is convenient for the layout of the hardware monitoring system; The designed hardware monitoring system is used for function verification on the simulated bridge; During the on-site test, the hardware monitoring system is installed on the bridge deck of the bridge to be measured, and the collection time is 30 minutes;

[0054] The measured data is as Figure 11 shown, which are the collected real-time temperature data, wind speed data, and the actual inclination (Tilt1) and predicted inclination (Tilt2) respectively; The actual data is obtained through the interface link Python in the index.html of the software monitoring system, and then rendered to the display interface to complete the functions of prediction and display on the web page.

[0055] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements should also be regarded as within the protection scope of the present invention.

Claims

1. A bridge monitoring method based on a cloud platform, characterized in that: The following steps are involved: Build a hardware monitoring system, including selecting multi-mode sensor modules, data processing modules, 4G transmission modules and power modules; the multi-mode sensor collects real-time information about the bridge and transmits it to the microcontroller, which processes the data and transmits it to Alibaba Cloud's IOT platform via the 4G module using the MQTT protocol; Build a software monitoring system using a modern web application development technology architecture that covers both front-end and back-end technology stacks; Data interaction with the hardware monitoring system is achieved through the Alibaba Cloud IOT platform and the AMQP protocol; Single-layer linear regression is used to predict the bridge deflection. The predicted bridge inclination angle is calculated based on the received wind speed and temperature data and displayed in real time on the web page.

2. A bridge monitoring method based on a cloud platform according to claim 1, characterized in that: The multi-mode sensor includes an ADXL tilt sensor, an AHT10 temperature and humidity sensor, a wind speed sensor and a strain sensor; the data processing module uses a single-chip microcomputer STM32F411CEU6; the 4G communication module uses an Air780E communication module; and the power module consists of a HU9196 charging module and a 18650 lithium battery.

3. A bridge monitoring method based on a cloud platform according to claim 2, characterized in that: The Alibaba Cloud IOT platform is combined with the AMQP protocol to realize message communication between IoT devices and back-end systems; the Web front-end part of the software monitoring system is responsible for processing the display and interaction logic of the user interface; the Web back-end part of the software monitoring system is mainly responsible for processing business logic, data storage and interaction with the front-end; the web page of the software monitoring system is designed with user login management, real-time data and chart display, monitoring data download, historical monitoring data query, and bridge tilt prediction.

4. A bridge monitoring method based on a cloud platform according to claim 3, characterized in that: The Web front-end part includes HTML, CSS, JavaScript, Axios and Webpack; the Web back-end part includes Node.js, Express and Mongo; the Mongo includes Mongoose and MongoDB; the HTML is used to define the specific content of the system, the CSS is used for the layout and style design of the web page, the JavaScript is used for the dynamic interactive function of the web page, the Axios is used for data interaction with the back-end, and the Webpack is used to package and optimize the front-end resources; the Node.js provides a high-performance operating environment for the entire system, the Express is used to build Web services, and the Mongoose and MongoDB are used for data storage and management.

5. A bridge monitoring method based on a cloud platform according to claim 4, characterized in that: Single-layer linear regression is used to predict bridge deflection. The basic formula of the regression model is: y^=β0+β1x1+β2x2+…β n x n , where y^ is the predicted value (in this case, the inclination angle of the bridge), β0 is the intercept, and β1,β2,…,β n are regression coefficients, which correspond to features x1, x2, …, x n ; x1,x2,…,x n are input features (wind speed and temperature in this system); in bridge tilt prediction, the training process is done through model.fit(X train ,y train ) is performed, where X train is the feature matrix containing the training data of wind speed and temperature, y train is the target variable of the corresponding bridge inclination angle; when the web terminal of the software monitoring system receives new wind speed and temperature data, it predicts through model.predict(input_data), calculates the predicted bridge inclination angle and displays it in real time on the web terminal.

6. A bridge monitoring method based on a cloud platform according to claim 5, characterized in that: The real-time measurement data of the hardware monitoring system is transmitted to Alibaba Cloud's IOT platform through the 4G module via the MQTT protocol; the software monitoring system transmits the data of the IOT platform to the Web end through the AMQP protocol to realize real-time interaction and processing of data; in order to realize data interaction and usage experience, user login management, real-time display of data, download of bridge monitoring data, and historical data information query functions are designed in the Web end of the software monitoring system.

Images