Multi-parameter optical fiber sensing monitoring system based on cloud platform

Abstract

The invention provides a multi-parameter optical fiber sensing monitoring system based on a cloud platform. The system comprises a sensing layer, a network transmission layer, a platform service layer and an application service layer, the sensing layer is used for sensing and collecting environment information and performing information interaction with the cloud platform; the network transmission layer is used for realizing data interaction between the sensing layer and a cloud platform; the platform service layer is used for analyzing and storing the data uploaded by the sensing layer and providing an open API (Application Program Interface); and the application service layer is used for developing a cloud platform by adopting a front-end and back-end separation mode based on a B / S (Browser / Server) architecture so as to realize data visual display. The intelligent level and the operation and maintenance efficiency of the monitoring system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of structural health monitoring, and in particular relates to a multi-parameter optical fiber sensing monitoring system based on a cloud platform. Background Art

[0002] Over the long term, large infrastructure projects such as bridges, tunnels, and dams are susceptible to environmental erosion, loads, and other factors, leading to structural degradation and potentially posing safety risks. Therefore, real-time health monitoring of these infrastructure projects is crucial. Traditional electrical sensors have limitations in structural health monitoring, such as high power supply requirements, the need for decentralized power supply, corrosion susceptibility, and susceptibility to electromagnetic interference. These issues increase failure rates and make maintenance more difficult, severely restricting the effectiveness of long-term monitoring.

[0003] Fiber Bragg Grating (FBG) sensors offer advantages such as simultaneous multi-parameter sensing, immunity to electromagnetic interference, and long-distance transmission, providing a new approach for infrastructure health monitoring. With the development of IoT technology, monitoring systems are evolving towards distributed sensing and centralized management. The wide-range sensing characteristics of fiber optic sensing technology are well-suited to the needs of building large-scale IoT sensing systems. However, existing third-party cloud platforms only enable data upload for health monitoring and lack functionality such as demodulator parameter configuration, data analysis, and sensor degradation monitoring, making them difficult to meet the requirements of intelligent operations and maintenance. Therefore, there is an urgent need to develop a high-precision, highly stable fiber Bragg grating demodulation system capable of communicating with the cloud to enable remote monitoring. Summary of the Invention

[0004] The purpose of the present invention is to provide a multi-parameter fiber optic sensing monitoring system based on a cloud platform to solve the problems existing in the prior art, enhance the system's remote online monitoring capabilities, and realize multi-parameter real-time monitoring and intelligent analysis of large-scale infrastructure.

[0005] The technical solutions of the present invention are as follows:

[0006] A multi-parameter optical fiber sensing monitoring system based on a cloud platform, comprising: a perception layer, a network transmission layer, a platform service layer, and an application service layer

[0007] The perception layer is used to perceive and collect environmental information and exchange information with the cloud platform;

[0008] The network transmission layer is used to realize data interaction between the perception layer and the cloud platform;

[0009] The platform service layer is used to parse and store the data uploaded by the perception layer and provide an open API interface;

[0010] The application service layer is used to develop a cloud platform based on the B / S architecture and adopt a front-end and back-end separation mode to achieve data visualization.

[0011] Furthermore, the sensing layer includes an FBG sensor and a fiber Bragg grating demodulator;

[0012] The FBG sensor is used to sense changes in external physical quantities and convert the physical quantities into wavelength signals;

[0013] The fiber Bragg grating demodulator is used to collect FBG reflection spectrum data, calculate wavelength drift, receive cloud platform configuration commands to control the start and stop of data collection, and interact with the cloud platform for data.

[0014] Furthermore, the network transmission layer adopts the MQTT protocol to realize data interaction between the perception layer and the cloud platform.

[0015] Furthermore, the fiber Bragg grating demodulator includes:

[0016] The light source module consists of an ASE broadband light source, an optical isolator, an FP filter, and an optical splitter; the driver module generates a driving signal through a sawtooth wave generator to control the working state of the FP filter; the wavelength calibration module is used to calibrate the FBG center wavelength; the sensor module consists of multiple FBG sensors to measure changes in external physical quantities; the data acquisition and processing module transmits A / D sampling data to the host computer through the microcontroller using serial port communication, and the host computer calculates the FBG center wavelength.

[0017] Furthermore, the wavelength calibration module includes an FP etalon to perform dynamic wavelength calibration on the tunable FP filter.

[0018] Furthermore, the cloud platform is developed based on the B / S architecture and adopts a front-end and back-end separation mode. The front-end uses the Vue framework to build the UI interface, and the back-end uses the SpringBoot framework to implement business logic, supporting remote configuration and real-time monitoring functions.

[0019] Furthermore, the platform service layer adopts a hybrid storage architecture that combines MySQL and InfluxDB, where InfluxDB is used to store wavelength and spectral data with strong time series, and MySQL is used to store device configuration information and alarm rule structured data, achieving high-reliability storage and low-latency access to data.

[0020] Compared with the prior art, the present invention has the following advantages:

[0021] The cloud-based multi-parameter fiber optic sensing monitoring system provided by the present invention realizes real-time monitoring of multiple parameters of large-scale infrastructure, including physical quantities such as temperature, strain, and pressure, providing comprehensive data support for structural health status assessment.

[0022] The fiber Bragg grating interrogator adopts tunable FP filter demodulation method and FP standard dynamic calibration technology to improve the demodulation accuracy and stability. The demodulation accuracy is better than 3pm, meeting the needs of high-precision monitoring.

[0023] The cloud platform adopts a hybrid storage architecture and MQTT communication protocol to achieve efficient storage and reliable transmission of massive data, support remote configuration, real-time monitoring and intelligent analysis functions, and improve the intelligence level and operation and maintenance efficiency of the monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings illustrate various embodiments generally by way of example and not limitation, and together with the description and claims, serve to explain embodiments of the invention. Where appropriate, the same reference numerals are used throughout the drawings to refer to the same or similar parts. Such embodiments are illustrative and are not intended to be exhaustive or exclusive of the embodiments of the present apparatus or method.

[0025] Figure 1 shows a schematic diagram of the system architecture of the present invention;

[0026] Figure 2 Shows the overall schematic diagram of the fiber Bragg grating demodulation system of the present invention;

[0027] Figure 3 Shows a schematic diagram of the optical path design of the demodulator of the present invention;

[0028] Figure 4 Shown is a schematic diagram of the system hardware design of the present invention. DETAILED DESCRIPTION

[0029] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0030] An embodiment of the present invention provides a multi-parameter optical fiber sensing monitoring system based on a cloud platform, comprising:

[0031] Perception layer, network transmission layer, platform service layer and application service layer, such as Figure 1 As shown, combined with the needs of online monitoring of optical fiber sensing technology, the contents of each layer are introduced in detail below.

[0032] (1) Perception layer: The perception layer mainly includes FBG sensors and fiber Bragg grating demodulators, which work together to realize the perception and collection of environmental information. The wavelength drift of the FBG sensor is calculated in the demodulator, and the relevant configuration commands from the cloud platform are received to start or stop data collection, and the FBG center wavelength and spectrum data are uploaded.

[0033] (2) Network transmission layer: The system uses 4G / 5G network as the communication basis to ensure the stable operation of the fiber optic sensing system equipment. In view of the large amount of data in the fiber optic sensing system, the system achieves secure networking through WIFI or 5G network, and uses its fast and stable connection capabilities to ensure the efficiency of data transmission. At the same time, the system uses the MQTT protocol to realize data interaction between the demodulator and the cloud platform. By subscribing to relevant topics, the system ensures the accurate transmission of messages and meets the system's data upload and command issuance requirements.

[0034] (3) Platform service layer: It mainly realizes the analysis and storage of demodulator data by the cloud platform, uses the internal EMQX server to realize server-side communication through message subscription, provides an open API interface, realizes the server and demodulator to set and subscribe to the same topic, and completes the demodulator data transmission and platform control instructions.

[0035] (4) Application service layer: The system's operating environment is built and configured through a cloud server. The cloud platform adopts a system architecture with front-end and back-end separation. The front-end page is built using VUE, and the back-end service is implemented using Springboot. On the back-end of the server, topics subscribed to by the demodulator and the cloud platform are established to implement message transmission, and the data sent by the MQTT server is parsed and stored in a time series database. The communication interface between the front-end and back-end is established. The front-end designs the UI interface of the cloud platform and displays it in the browser. The back-end transmits the received data and demodulator status information to the front-end, which displays the corresponding data in the browser.

[0036] Demodulator overall architecture

[0037] The present invention uses a tunable FP filter and a demodulation method based on dynamic calibration of FP etalon to build a complete FBG demodulation system. The system is mainly divided into five parts. The light source module is the core component of the system, which is composed of devices such as ASE light source, optical isolator and FP filter; the driving module generates a driving signal through a sawtooth wave generator to control the working state of the FP filter; the wavelength calibration module is mainly a FP etalon, which is used to calibrate the FBG center wavelength; the sensing module is composed of multiple FBG sensors, which are used to measure changes in external physical quantities; the data acquisition and processing module uses serial communication through the single-chip microcomputer to transmit the A / D sampling data to the host computer, and the FBG center wavelength is calculated by software demodulation in the host computer. The overall design of the demodulation system is as follows: Figure 2 shown.

[0038] Demodulator performance indicators

[0039] To meet the monitoring requirements of temperature, strain, magnetic field, and pressure, and to meet the high-precision demodulation of sensor reflectance spectra across multiple channels and maximize the use of light intensity by sensors in a quasi-distributed network, the STM32F429 chip was selected to control the data acquisition module and package the spectral data. The communication module then transmits the data to the host computer for processing, display, analysis, and image rendering, enabling the detection of external environmental parameters such as temperature, strain, magnetic field, and pressure. The expected technical specifications of the demodulation system are shown in Table 1.

[0040] Table 1 Demodulator performance indicators

[0041]

[0042]

[0043] Key factors affecting system demodulation performance

[0044] (1) Optical path

[0045] The optical architecture of the demodulation system primarily consists of the following core components: an ASE broadband light source, a FP etalon, an FP filter, an FBG sensor, and a photodetector. Key performance characteristics of the ASE source, such as spectral flatness and output optical power, directly impact the overall characteristics of the optical system and the quality of sensor data acquisition. The maximum drive frequency of the FP filter affects the demodulation rate of the demodulation system. The denser the comb structure of the etalon, the better the calibration and the higher the demodulation accuracy.

[0046] (2) Circuit part

[0047] The demodulation system's circuitry primarily consists of an FP filter driver circuit, a photodetector, and data acquisition and transmission circuits. The voltage provided by the FP filter driver circuit determines the number of peaks detected by the FP etalon in the spectrum. A higher drive voltage results in a greater number of peaks and improved demodulation accuracy. However, the drive voltage must not exceed the operating range of the FP filter. The drive circuit's frequency directly affects the system's demodulation rate. Assuming the maximum drive frequency of the FP filter is exceeded, the data acquisition rate also affects the system's demodulation speed. To ensure signal integrity, a matching digital filter is required to suppress noise interference.

[0048] (3) Data processing algorithm

[0049] The demodulation system's data processing algorithm consists of three components: a denoising algorithm, a multi-peak detection algorithm, and a wavelength calibration algorithm. The effectiveness of the denoising algorithm directly impacts subsequent multi-peak detection and etalon addressing, ultimately affecting demodulation accuracy. Dynamic calibration and addressing of the FP etalon's mark peaks are crucial steps in the demodulation algorithm implementation. The selection and quality of the data points at both ends of the multi-peak detection algorithm also impact computational efficiency.

[0050] Optical Path Design of Grating Fiber Demodulator

[0051] The optical path design of the demodulation system based on the tunable FP filter is as follows: Figure 3 The system uses an ASE broadband light source as its spectral input, with an optical isolator preventing light reflections from entering the source. A single-chip microcontroller controls the scanning of the FP filter, converting the broadband spectrum into a narrowband spectrum at a specific wavelength. The filtered narrowband optical signal is split into two paths by a fiber optic beam splitter: one path enters the sensing optical path to obtain the FBG reflection spectrum, and the other path enters the reference optical path to obtain the transmission spectrum of the FP etalon. The reflected light from both paths is converted into electrical signals by a photodetector and then combined with the FP etalon to accurately calibrate the FBG center wavelength.

[0052] According to the system design and Figure 3 The system's optical path design diagram constructs the demodulator's optical path. The performance parameters of the optical components and their coordinated matching directly affect the performance indicators of the demodulation system. The following details the performance requirements and selection criteria for each component of the system's optical path. This provides theoretical support for subsequent system integration and performance optimization, while also ensuring that the wavelength coverage, response speed, and accuracy of the optical components are well matched.

[0053] (1) FBG sensor

[0054] FBG sensors, as the sensing unit of the demodulation system, are responsible for sensing external physical quantities. The sensitivity of the FBG sensor directly determines the detection accuracy of that physical quantity. This study used an FBG temperature sensor independently developed by the research team. Its temperature measurement range is -60 to 50°C, and its central wavelength is 1550nm.

[0055] (2) Broadband light source

[0056] A broadband light source is the signal source for the demodulation system. Its optical stability, including power and wavelength stability, as well as its spectral characteristics, directly determines the system's demodulation wavelength range and measurement accuracy. During optical transmission, the use of wavelength-selective devices such as tunable FP filters and FP etalons, as well as sensors, significantly increases optical power loss. Therefore, the light source must possess sufficiently high output optical power to ensure stable system operation. The demodulation system designed in this article utilizes a C BandASE Source with a wavelength range of 1520-1575 nm, a center wavelength of 1547.5 nm, and a bandwidth of 55 nm, meeting the demodulation system's performance requirements.

[0057] (3) Tunable FP filter

[0058] According to the tunable FP filter demodulation principle, the performance parameters of the FP filter have a decisive influence on the system, which is mainly reflected in the following two aspects. First, the maximum demodulation speed of the system is limited by the maximum driving frequency of the FP filter, and second, the wavelength resolution of the system is affected by the filtering capacity of the FP filter. Under ideal circumstances, the driving voltage of the FP filter has a good linear relationship with the wavelength, but due to the physical properties of the piezoelectric ceramic material and the mechanical structure limitations of the device, there is an inevitable hysteresis phenomenon. Therefore, it is necessary to offset this effect by adding a reference optical path. In the demodulation system, when the 3dB bandwidth of the FBG is equal to the half-maximum full width of the FP filter, the wavelength detection sensitivity of the system can reach its maximum value. Moreover, the closer these two values ​​are to each other, the more ideal the demodulation effect of the system. The following is an analysis of the main characteristics of the FP filter:

[0059] The free spectral range (FSR) is a core parameter of FP filters. It is the frequency or wavelength interval between two adjacent interference maxima or minima, and therefore represents the maximum spectral range that the filter can independently resolve. A larger FSR indicates a stronger spectral resolution. Furthermore, the filter's drive voltage range is determined by both its FSR and the bandwidth of the ASE light source. The FP filter used in this article has an effective drive voltage range of approximately 5 to 18 V.

[0060] The demodulator in this article uses a tunable FP filter. Fiber optic cables extend from each end of the filter for connecting to the fiber isolator and fiber coupler. Below this, two power cables connect to the FP filter driver module. Parameters such as wavelength range, FSR, finesse, and tunable voltage are shown in Table 2.

[0061] Table 3.3 FP filter performance parameters

[0062]

[0063]

[0064] (4) FP etalon

[0065] Due to the hysteresis characteristics of piezoelectric ceramics, creep effects, and environmental factors, the driving voltage of a tunable FP filter exhibits a nonlinear relationship with the actual cavity length. Therefore, the FBG center wavelength obtained directly using the FP filter will have significant errors. To improve wavelength demodulation accuracy, an FP etalon is introduced as a wavelength calibration module. Key performance parameters of an etalon include 3dB bandwidth, spectral coverage, and FSR. During wavelength calibration, the etalon's mark peak position must be located, followed by multi-peak fitting and addressing of all the etalon's comb peaks. Therefore, as a peak width indicator, the 3dB bandwidth directly affects the accuracy of fitting and peak addressing. When the etalon's 3dB bandwidth is equal to the FWHM of the tunable FP filter, the system demodulation error is minimized. A higher bandwidth match results in higher demodulation accuracy. Furthermore, the etalon must fully cover the output spectral range of a broadband light source to ensure the effectiveness of full-band wavelength calibration. The etalon's FSR is the wavelength spacing between adjacent interference peaks. A smaller etalon's FSR indicates a greater number of interference peaks within the same spectral range, more precise addressing, and more accurate FBG wavelengths obtained after linear interpolation.

[0066] The FP etalon used in this article is from Prius Optoelectronics. It employs a fixed-length optical FP cavity structure. Two cables equipped with fiber adapters extend from the right side of the etalon, one end connected to a circulator and the other to a photodetector. A characteristic peak on the far right of the spectrum, significantly lower in intensity than the other interference peaks, is known as the mark peak. Using the position of the mark peak and the etalon's FSR, the wavelength corresponding to each peak can be calculated, allowing for wavelength calibration of the system's FBG.

[0067] Circuit Design of Fiber Bragg Grating Demodulator

[0068] The hardware of the fiber Bragg grating demodulator mainly consists of a power supply module, an FP filter driving circuit, an optoelectronic conversion module, and a data acquisition and transmission module. First, the entire demodulation system was built using modular components to verify the feasibility of the proposed solution. Then a highly integrated circuit board was designed to reduce costs and size. The core functional circuits include: a power supply module that provides a stable operating voltage for each circuit unit; an FP filter driving circuit that can realize accurate voltage scanning to generate triangular waves and sawtooth waves to control the FP filter; the optoelectronic conversion module mainly includes a high-speed PIN photodiode to realize optical signal reception, a multi-stage amplification circuit, and an integrated fourth-order Butterworth low-pass filter. The data acquisition and transmission module uses a 16-bit high-precision ADC chip. The data transmission is based on the USART2 and DMA of the STM32F429. The data is first placed in the buffer, and then transmitted to the host computer through the serial port in turn, and the host computer implements software demodulation. The system hardware design diagram is shown below. Figure 4 shown.

[0069] Data acquisition control module

[0070] The demodulator's control module uses the STM32F429 as its control device, primarily responsible for controlling AD data acquisition, the FP filter driver circuit, and data transmission. The minimum system, centered around the STM32F429IGT6 chip, comprises a download circuit, an external clock, and a reset circuit. The download circuit, as the interface between the development board and the computer, is crucial in the system. It typically consists of a serial-to-USB converter chip, enabling functions such as program burning and testing. During this process, the BOOT pin selects the boot mode, while the NRST pin performs a hardware reset. Both are essential for the proper functioning of the download circuit. To ensure stable operation of the STM32 chip's internal clock, the minimum system must include an external clock source, providing the necessary time reference for stable chip operation. The reset circuit is a crucial component of the STM32 minimum system, primarily implementing the hardware reset function. In practical applications, a simple RC delay circuit is often used to implement both software and hardware reset functions, and this circuit must be connected to the NRST pin. When the system encounters an abnormality or needs to be restarted, the reset circuit can take effect in time to restore the system to its initial state, ensuring the stability and reliability of the entire system.

[0071] Power Module

[0072] The hardware system is powered using a combination of switching and linear power supplies, with a 12V DC power supply serving the input circuit. Based on the system's design specifications and hardware architecture, a partitioned power supply strategy was adopted, dividing the system power supply into two parts: the STM32 minimum system power supply and the functional module power supply. The minimum system power supply provides stable power for the STM32F429 main control chip and peripheral circuits. It supports dual 3.3V / 5V inputs and uses LDO regulators for power filtering, enhancing the stability and anti-interference capabilities of the minimum system. The functional module power supply includes the AD acquisition module, FP filter driver circuit, broadband light source, and photoelectric conversion circuit. The broadband light source and photoelectric conversion circuit utilize independent 5V DC power supplies, using linear regulators for high-precision voltage regulation to ensure light source stability and reduce power supply ripple interference on optical signal acquisition. The AD acquisition section requires a 5V analog power supply and a 5V digital power supply. The analog power supply powers the AD7606 signal conditioning circuit, while the digital power supply drives the chip's logic units. The FP filter driving circuit power supply needs to generate ±15V voltage and sawtooth wave or triangle wave signal for driving, and the boost output is achieved through the boost module. At the same time, the stability of the triangle wave generating circuit needs to be ensured.

[0073] Tunable FP filter driver circuit

[0074] Based on the principle of narrowband optical wavelength scanning, the FP filter's output wavelength exhibits a linear relationship with the applied voltage. The system uses a DDS chip to generate a high-precision triangle or sawtooth wave signal, which is then driven by a voltage amplifier circuit to achieve filter scanning. When the drive voltage ranges from 0 to 18V, the filter's transmission spectrum covers the entire FSR. Because the effective spectral width of the ASE broadband light source must cover as many peaks as possible of the FP etalon and the corresponding triangle wave positions, the actual drive voltage range is selected to be 15V.

[0075] AD sampling circuit

[0076] Based on the characteristics of the full-spectrum demodulation mode, the design of the demodulation system's AD acquisition module must meet the following technical requirements: To ensure time alignment of signals from different channels, the acquisition module must support a synchronous sampling mechanism capable of simultaneously acquiring the periodic voltage signal from the FP filter driver circuit, the spectrum of the FP etalon, and the reflectance spectrum of the FBG. The AD7606 chip, with its built-in 8-channel synchronous sample-and-hold circuit, enables parallel acquisition of multi-channel signals, meeting the requirement for simultaneous data acquisition from all three channels. To quickly acquire the FBG and etalon light intensity signals while preserving the complete spectral characteristics, the data acquisition module must have at least 16-bit resolution and a conversion rate exceeding 1,000 samples per second. Finally, it is important to consider whether the voltage amplitude of the sensor signal matches the voltage range of the data acquisition module. If the selected AD chip has a too large a voltage range, the sampling accuracy will be affected; if the voltage amplitude is too small, it will not meet the sensor signal sampling requirements. Based on the above analysis, the AD7606 chip was selected, with parameters shown in Table 3.

[0077] Table 3 AD7606 chip parameters

[0078] parameter Numerical unit Number of channels 8 Bit width 16 Bit Input voltage range ±5 or ±10 V Internal reference voltage 2.49～2.505 V Throughput 200 KSPS Input clamp protection circuit ±16.5 V Analog input impedance 1 MΩ Oversampling ratio 1、2、6、8、16、32、64 times

[0079] The STM32F429 chip adjusts the AD7606's oversampling ratio via a control pin. Connecting the chip's ground pin to ground via a filter capacitor and pull-down resistor reduces noise interference and ensures glitches in data acquisition. When a rising edge occurs on the STM32F429 chip's signal, multiple channels initiate simultaneous sampling. From a hardware design perspective, the AD chip operates within a voltage range of –10V to +10V. Test results show that within this voltage range, the AD7606 achieves sampling accuracy better than 0.5mV, meeting the system's data acquisition accuracy requirements.

[0080] Data transmission circuit

[0081] In the design of a fiber Bragg grating (FBG) demodulation system, spectral data sampled by the AD is first stored in a data buffer. Once a certain amount of data accumulates in the buffer, the data is transmitted to the host computer via a serial port. The baud rate for data transmission can be set according to actual needs. The default baud rate of the demodulator in this article is 500,000. During serial transmission, to ensure accurate data reception and parsing, each 8-bit binary byte must be appended with a start bit and a stop bit to form a complete frame structure. When transmitting spectral data, it is important to ensure that the AD sampling rate matches the serial port communication rate. If the sampling rate is significantly higher than the serial port transmission speed, the data buffer will overflow, resulting in data loss or parsing errors, directly affecting the accuracy of subsequent demodulation and even causing abnormal results.

[0082] To address serial port data throughput bottlenecks and transmission integrity requirements, the system optimizes serial port DMA and frame-based transmission mechanisms. Specifically, spectral data is first temporarily stored in a buffer queue and then segmented into blocks of specific sizes according to preset rules. Idle time is then inserted between each block to ensure the receiver has ample time to process the data and reset the receive buffer. To match frame-based transmission with the communication frequency, two counters are designed: a pixel counter and a frame counter. The pixel counter tracks the number of pixels transmitted within the current frame. Each frame contains 552 pixels. When the pixel counter reaches 552, the frame counter increments. When the frame counter reaches a threshold (e.g., 10 frames), the DMA engine initiates a complete frame transfer and then resets the counter, forming a periodic transmission cycle. Each data set in spectral data transmission follows the USART2 format of a start bit, eight data bits, and a stop bit. This approach balances data throughput with receiver processing load, ensuring data integrity while avoiding the risk of overload.

[0083] This section primarily completes the development of a fiber optic sensing cloud platform. Using a fiber grating interrogator, the wavelength and spectral data from the FBG sensor are uploaded to the cloud platform for storage and display. Monitoring data from different channels can be viewed in real time on the webpage. The system provides model training and call functions, enabling the selection of a data segment from a time series database as a dataset for online model training or call-based model analysis of sensor data to enable alarm prediction. The system supports historical data queries and can display current data based on different time periods and query methods. Ultimately, the constructed fiber optic sensing cloud platform integrates multiple functions, including data acquisition, parsing, storage, analysis, and display, to meet the application requirements of online fiber optic sensing monitoring and analysis.

[0084] Cloud platform development architecture

[0085] Front-end and back-end development framework

[0086] When designing the fiber-optic sensing cloud platform system, the performance and future functional expansion requirements of the cloud platform were considered, and a front-end and back-end separation development model based on a B / S architecture was adopted. This B / S architecture offers the advantages of rapid deployment, with subsequent maintenance and updates completed through a configuration center. Access is possible through common browsers without the need for client installation. Furthermore, access to the demodulation device is a core function of the cloud platform. The demodulation device establishes two-way communication with the platform via the MQTT protocol, storing collected data on the cloud platform. This enables real-time upload of sensor data to the cloud and issuance of control commands. This provides data support for other related functional modules, enabling device management and control.

[0087] The front end of the system is developed based on the Vue2 framework. The front end mainly implements the design of functional pages and the interface design of the back end. In order to improve the maintainability and scalability of the system, the front end adopts a component-based development model to decouple various functional modules. The view layer uses a responsive design and adopts CSS layout to achieve multi-terminal adaptation. Vue integrates module packaging tools such as Webpack to compress code and merge files to reduce loading time. Vuex is used for state management, maintain cross-component data flow, and ensure data consistency. LocalStorage and sessionStorage are used to cache commonly used data to improve the efficiency of secondary access. Echarts is used to design data visualization to display real-time data of each channel of the demodulator. For asynchronous communication requests, the front-end framework designs a standardized data exchange interface based on the HTTP protocol to ensure efficient and accurate interaction between the front and back ends. Unified identity authentication and data encryption transmission are achieved through request interceptors.

[0088] The system backend was developed using the popular SpringBoot framework, primarily responsible for handling business logic and implementing functionality. SpringBoot significantly simplifies the complex XML configuration typically associated with traditional Java project development. Using Maven management, system configuration files can be quickly completed. SpringBoot employs the classic MVC architecture, tightly linking each layer through annotations. The container manages object creation, and the Lombok plugin simplifies the code, improving its structure and readability for easier maintenance. To standardize request and exception handling, interceptors are implemented in the system, reducing code coupling. The control layer primarily handles interface design. Leveraging the Jackson tool embedded in the SpringBoot framework, processing results are serialized into JSON format for easy browser display. For data persistence, the system is based on the Mybatis framework and incorporates the Mybatis Plus plugin to further enhance database development efficiency. To ensure system stability and scalability, a hybrid data storage system uses MySQL, Redis, and InfluxDB.

[0089] Communication between cloud platform and demodulator

[0090] Data communication between the cloud platform and the demodulator primarily relies on the MQTT protocol for bidirectional communication. MQTT is a lightweight communication protocol that supports data communication between clients and servers and is widely used in the Internet of Things. This protocol transmits the FBG sensor's central wavelength obtained by the demodulator to the EMQX server, where the system parses the data and stores it in a database.

[0091] Data and commands from the demodulator are transmitted using the MQTT protocol, interacting within an MQTT server. Both the device and the cloud subscribe to topics with the server, with the MQTT server responsible for data forwarding and push. Currently, communication interfaces are divided into two types: reporting and control protocols. The reporting protocol is used to upload a list of wavelengths and intensities for fiber Bragg gratings (FBGs); the control protocol is used for remote spectrum reading and channel parameter setting. Reporting messages are initiated and sent periodically by the device. To reduce data traffic, data packets utilize a data compression algorithm. Given the importance of this data, an MQTT server arrival response mechanism is implemented to ensure stable data upload to the cloud. If the reporting route is congested or interrupted, the data is retained and retransmitted on the device. Control messages are initiated by the cloud platform, using UDP packets based on the OPM protocol as the data payload and encapsulated in MQTT packets, enabling convenient and reliable IoT data transmission. Because this type of data requires high real-time performance, a guaranteed delivery mechanism is not suitable. The implementation is a two-step process: the first step focuses on implementing reporting messages to enable basic cloud platform functionality and obtain FBG wavelengths. If wavelength data is not received for a long period of time, a device offline alarm is triggered. The second step is to implement control messages, regular sensor spectrum extraction, health monitoring and other functions.

[0092] Another approach utilizes the MQTT framework and the JSON protocol format to ensure efficient and reliable message distribution. The system encompasses three types of objects: server, client, and device, all communicating via the MQTT framework. Logically, the system is divided into a management side (server, client) and a device side (device), with message flows divided into downstream (management side → device side) and upstream (device side → management side). The architecture encompasses devices, management servers, user servers (up to three), and local clients. Except for local clients, all other terminals run the MQTT service. The management and user servers can forward both management and application messages, while local clients can only forward application messages. The system supports a variety of typical application scenarios: First, devices remotely report to servers, connecting to a central MQTT server over the network for message exchange; second, clients locally control devices, with devices initiating local MQTT servers for client interaction; and third, clients remotely control devices, forwarding messages via a deployed MQTT management server. Regardless of the scenario, message exchange relies on MQTT server forwarding, the only difference being the server's location. This architecture not only unifies communication protocols across LANs and WANs but also supports simultaneous reporting from multiple central servers and access from multiple local clients in complex scenarios. Furthermore, if the demodulation device has network access, clients can remotely control the device using the cloud platform's MQTT central server, effectively solving the problem of remote maintenance.

[0093] Design and construction of multi-source heterogeneous databases

[0094] Data storage analysis

[0095] The fiber Bragg grating (FBG) demodulation system generates massive amounts of data during operation, driven by the data characteristics of fiber optic sensing. This massive amount of data generated by fiber optic sensing technology places high demands on database write rates and query response speeds. Therefore, the cloud platform's data storage utilizes a hybrid "relational database + time series database" architecture, with a tiered optimization strategy tailored to different data types and access characteristics. The system employs a three-tier storage logic, using the in-memory database Redis to create a cache pool for temporary storage of frequently accessed data, such as instrument operating parameters and device configuration information. Recent data is prioritized to alleviate pressure on the core database. A MySQL database is used to store data that requires relational queries during system operation, including highly correlated data such as device registration information, channel configuration parameters, and device status. The InfluxDB time series database is used to establish a storage module for wavelength and spectral data. Data collected by each demodulator channel is organized into independent measurement tables based on "instrument number + physical channel number," enabling partitioned storage of time series data.

[0096] Database construction

[0097] Based on the data analysis in the previous section, relational databases are primarily used to store data such as device configuration and operating status, as well as related information such as device alarm information, threshold settings, model training calls, location information, device information, and device status, allowing for query and access. Based on these requirements, we designed data tables for each function in conjunction with database table design principles.

[0098] To use the MySQL database, you must first download and install it on the server and perform a series of configurations. You need to configure the database address, character set, storage engine, port number, and more. For database system security, you also need to set the maximum number of connections allowed and the number of connections allowed. After completing the configuration, start the MySQL service. For more convenient database management, tools such as Navicat are often used to manipulate database tables. Navicat offers a wealth of features, making database operations more intuitive. Furthermore, to meet sensor data storage requirements, the time series database InfluxDB must be installed and configured on the server. InfluxDB is designed specifically for processing time series data and offers efficient read and write performance. After installation, you also need to perform relevant configurations and operations, such as creating a database name for the data store and setting the data storage path to ensure secure and persistent data storage. Based on the frequency and volume of sensor data generated, adjust the data retention policy appropriately to avoid excessive disk space usage. Also, configure the network connection parameters between InfluxDB and the server to ensure normal database access.

[0099] Real-time data and historical data query

[0100] The platform's real-time monitoring function plays a crucial role in the structural health monitoring system. Through bidirectional communication with the fiber Bragg grating (FBG) demodulation system, the platform displays real-time data on the center wavelength and intensity of FBGs in different channels. Because FBG sensors are sensitive to changes in physical quantities such as temperature and strain, the changes in their wavelength data over time can reflect the structural condition of the building. By analyzing this data, the system can accurately capture subtle structural deformations, the distribution and changes in temperature fields, and the dynamic evolution of stress concentration areas. To meet the needs of subsequent research and retrospective analysis, this data is stored in InfluxDB, ensuring data security and fast query speed. The database stores data in timestamp order, supporting long-term monitoring data management and analysis. The database retains data for 30 days; data older than 30 days is purged to free up disk space. The historical query function allows users to query data within any time range during monitoring. Multiple data display modes and time aggregation methods are provided, allowing users to easily and intuitively view and analyze current data. Users can directly select the processing method on the page and the system will draw the corresponding curve after confirmation, so as to have a deeper understanding of the operating status and historical trends of the equipment, and provide strong support for the system's intelligent decision-making.

[0101] The device management module is a key component of the system, enabling simple device management. This feature clearly displays the current operating status of all connected devices and allows for simple device commands. Furthermore, you can proactively change the status of a device within this interface. By switching a device to off using the on / off button, the system immediately responds by stopping receiving any incoming messages from the device, changing its operating status in the status bar, and updating its online status to offline. The device management module also features an intuitive geolocation display that accurately displays the location of sensor devices, allowing users to quickly locate each monitoring point and understand the monitoring status of different areas. Furthermore, the system continuously monitors device operating parameters and data. If an anomaly is detected, an alert is generated and highlighted within the device management interface, along with the severity and possible cause. This allows users to quickly respond and implement appropriate maintenance or adjustments to ensure stable and reliable operation of the entire monitoring system.

[0102] Model training and prediction implementation

[0103] On the cloud platform, you can select data from a time-series database, create a dataset, and use it to process data or train a pre-trained model. This module supports model updates, enabling online data selection for training, storing the latest model number in the database, and enabling model updates and replacements through configuration. By leveraging the computing power of cloud computing for online model training and updates, deep learning model training and testing can be completed efficiently and conveniently, improving the system's data processing capabilities.

[0104] The above is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, can make equivalent replacements or changes based on the technical solutions and inventive concepts of the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A multi-parameter optical fiber sensing monitoring system based on a cloud platform, characterized in that: The system includes: a perception layer, a network transmission layer, a platform service layer and an application service layer; wherein, The perception layer is used to perceive and collect environmental information and exchange information with the cloud platform; The network transmission layer is used to realize data interaction between the perception layer and the cloud platform; The platform service layer is used to parse and store the data uploaded by the perception layer and provide an open API interface; The application service layer is used to develop a cloud platform based on the B / S architecture and adopt a front-end and back-end separation mode to achieve data visualization.

2. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 1 is characterized in that: The sensing layer includes an FBG sensor and a fiber Bragg grating demodulator; The FBG sensor is used to sense changes in external physical quantities and convert the physical quantities into wavelength signals; The fiber Bragg grating demodulator is used to collect FBG reflection spectrum data, calculate wavelength drift, receive cloud platform configuration commands to control the start and stop of data collection, and interact with the cloud platform for data.

3. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 1 is characterized in that: The network transmission layer adopts the MQTT protocol to realize data interaction between the perception layer and the cloud platform.

4. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 1, characterized in that: The fiber Bragg grating demodulator comprises: The light source module consists of an ASE broadband light source, an optical isolator, an FP filter, and an optical splitter; the driver module generates a driving signal through a sawtooth wave generator to control the working state of the FP filter; the wavelength calibration module is used to calibrate the FBG center wavelength; the sensor module consists of multiple FBG sensors to measure changes in external physical quantities; the data acquisition and processing module transmits A / D sampling data to the host computer through the microcontroller using serial port communication, and the host computer calculates the FBG center wavelength.

5. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 4 is characterized in that: The wavelength calibration module includes an FP etalon, which performs dynamic wavelength calibration on the tunable FP filter.

6. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 1, characterized in that: The cloud platform is developed based on the B / S architecture and adopts the front-end and back-end separation mode. The front-end uses the Vue framework to build the UI interface, and the back-end uses the SpringBoot framework to implement business logic, supporting remote configuration and real-time monitoring functions.

7. The multi-parameter optical fiber sensing monitoring system based on a cloud platform according to claim 1, characterized in that: The platform service layer adopts a hybrid storage architecture that combines MySQL and InfluxDB, where InfluxDB is used to store wavelength and spectral data with strong time series, and MySQL is used to store device configuration information and alarm rule structured data, achieving high-reliability data storage and low-latency access.

Images