Method and system for distributed optical fiber monitoring of settlement of foundations of large equipment in ports

Abstract

The application relates to the technical field of port engineering and foundation settlement monitoring, and discloses a distributed optical fiber monitoring method and system for port large equipment foundation settlement, which comprises the following steps: a multi-physical quantity distributed optical fiber sensing network is arranged in the port large equipment foundation to sense strain, temperature and pore water pressure, the multi-physical quantity signals are synchronously collected, the data is converted into discrete settlement data of each monitoring point of the foundation through a multi-physical quantity fusion settlement inversion model, spatial interpolation is performed on the discrete data, a continuous settlement distribution atlas is generated, uneven settlement areas are analyzed based on the atlas, and corresponding equipment operation control instructions are generated. Through fusion of distributed optical fiber strain, temperature and pore water pressure sensing data, a nonlinear settlement inversion model is constructed, multi-dimensional data can be converted into accurate settlement, and immediate settlement caused by equipment load and long-term settlement caused by foundation soil consolidation can be distinguished.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of port engineering and foundation settlement monitoring, in particular to a distributed optical fiber monitoring method and system for settlement of large equipment foundation in a port. BACKGROUND

[0002] A port is an important transportation hub, and large equipment such as shore cranes, yard cranes, and bucket wheel machines is the core guarantee for efficient operation of the port. These devices are heavy and will exert dynamic and complex loads on the foundation during operation. In particular, ports located on soft soil foundations along the coast are prone to uneven settlement of the equipment foundation due to long-term heavy loads. Such settlement not only affects the smooth operation and service life of the equipment, but also can cause serious damage to the track structure and even overturn the equipment, resulting in huge economic losses and safety hazards. Therefore, continuous and accurate monitoring and intelligent management of the settlement of large equipment foundations is the key to ensuring the safety of port operations.

[0003] Existing technologies have developed a variety of monitoring methods. For example, traditional geodetic surveying methods such as leveling and static leveling can measure the settlement of specific points with very high precision, providing accurate static data for structural safety assessment. In addition, some technical solutions also begin to use various sensors such as strain gauges and pressure cells to monitor changes in physical quantities at specific locations of the equipment foundation, assisting engineers in point analysis. These technologies indeed provide basic support for the maintenance and safety management of port engineering in their respective application scenarios, helping staff to understand the state changes of certain key points.

[0004] However, these prior art solutions still have obvious deficiencies in practical application, and have not completely solved the fundamental problem of monitoring the settlement of the foundation of large-scale equipment in the port. First, the spatial continuity of data acquisition is poor. The traditional method only obtains settlement data at discrete points and cannot fully reflect the settlement distribution of the entire foundation area, resulting in a large blank area between monitoring points. Once a local sudden settlement occurs, a monitoring blind area will be formed, causing a major safety hazard. Second, the single dimension of the data limits the accuracy of the analysis. Most existing solutions only collect a single physical quantity, such as strain, and ignore the coupling relationship between temperature, pore water pressure and other physical quantities closely related to settlement. The single dimension of information cannot effectively separate the contributions of various factors, resulting in low accuracy of the settlement data obtained by inversion and making it difficult to make correct judgments about the root cause of the settlement. Finally, the entire monitoring and control process lacks real-time and automation. Existing technologies usually rely on manual periodic measurement and offline data analysis, and there is a serious time lag from data collection to conclusion, making it impossible for managers to respond to settlement abnormalities in a timely manner, let alone achieve closed-loop management of automatically adjusting the operating state of the equipment based on real-time data. Therefore, the technical personnel in the field propose a distributed optical fiber monitoring method and system for the settlement of the foundation of large-scale equipment in the port to solve the above problems. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a distributed optical fiber monitoring method and system for the settlement of the foundation of large-scale equipment in the port, which solves the problems of discontinuous spatial coverage, single data dimension and delayed response that cannot achieve real-time intelligent control in the prior art.

[0006] To achieve the above purpose, in a first aspect, the embodiments of the present application provide a distributed optical fiber monitoring method for the settlement of the foundation of large-scale equipment in the port, comprising the following steps:

[0007] A multi-physical quantity distributed optical fiber sensing network is laid out inside the foundation of large-scale equipment in the port, and the multi-physical quantity distributed optical fiber sensing network is used to sense strain data, temperature data and pore water pressure data of the foundation of large-scale equipment in the port;

[0008] Synchronously collect multi-physical quantity signals of the strain data, temperature data and pore water pressure data carried in the multi-physical quantity distributed optical fiber sensing network;

[0009] Process the multi-physical quantity signals, and convert the strain data, temperature data and pore water pressure data into discrete settlement data of each monitoring point of the foundation of large-scale equipment in the port by a multi-physical quantity fusion settlement inversion model;

[0010] Perform spatial interpolation processing on the discrete settlement data to generate a continuous settlement distribution map representing the overall settlement state of the foundation.

[0011] analyzing the uneven settlement area of the port large equipment foundation based on the continuous settlement distribution map, and generating equipment operation control instructions sent to the port equipment management system according to the analysis result.

[0012] In a further technical solution, the step of laying the multi-physical quantity distributed optical fiber sensing network in the port large equipment foundation comprises:

[0013] The armored cable containing the strain sensing optical fiber, the temperature sensing optical fiber and the pore water pressure sensing optical fiber is layered and laid along the longitudinal direction, the transverse direction and the depth direction of the port large equipment foundation, forming a three-dimensional monitoring grid.

[0014] In a further technical solution, the step of synchronously collecting the multi-physical quantity signals of the multi-physical quantity distributed optical fiber sensing network comprises:

[0015] The Brillouin frequency shift optical signal or the Rayleigh scattering optical signal in the strain sensing optical fiber and the temperature sensing optical fiber is collected by means of Brillouin optical time domain reflection or Rayleigh scattering;

[0016] And the reflected light wavelength drift signal caused by pressure change in the pore water pressure sensing optical fiber is collected by the fiber grating sensor.

[0017] In a further technical solution, the step of processing the multi-physical quantity signals comprises:

[0018] The signal collected by the strain sensing optical fiber is temperature compensated based on the temperature data collected by the temperature sensing optical fiber, to obtain real strain data generated only by foundation deformation;

[0019] And the real strain data, the pore water pressure data and the temperature data are input into the multi-physical quantity fusion settlement inversion model to calculate the discrete settlement data.

[0020] In a further technical solution, the step of spatially interpolating the discrete settlement data comprises:

[0021] The Kriging interpolation method is adopted, the discrete settlement data of each monitoring point and its spatial position relationship are used to calculate the weight coefficient through the semi-variance function, the settlement of the unmonitored area is optimally linearly and unbiasedly predicted, to form the continuous settlement distribution map.

[0022] In a further technical solution, the step of generating the equipment operation control instructions sent to the port equipment management system comprises:

[0023] The settlement overrun area or the settlement rate abnormal area in the continuous settlement distribution map is identified;

[0024] According to the location, range and severity of the settlement amount exceeding limit area or the settlement rate abnormal area, at least one regulation instruction of adjusting the equipment operation path, limiting the equipment operation load or reducing the equipment operation speed is generated through intelligent decision logic constructed based on preset expert rules or reinforcement learning algorithm.

[0025] In a further technical solution, the step of temperature compensation of the signal collected by the strain sensing optical fiber is implemented through the following relationship:

[0026]

[0027] Wherein, is the position coordinate of the optical fiber, is the real strain at the position is the real strain at the position is the measured Brillouin frequency shift at the position is the measured Brillouin frequency shift at the position is the reference Brillouin frequency shift, is the measured temperature at the position is the measured temperature at the position is the strain coefficient of the optical fiber, is the temperature coefficient of the optical fiber.

[0028] In a further technical solution, the multi-physical quantity fusion settlement inversion model is a nonlinear model constructed based on the foundation consolidation theory and fused with machine learning algorithm, which takes the real strain data, pore water pressure data and temperature data as independent input variables, and dynamically distinguishes the immediate settlement caused by equipment load and the long-term settlement caused by foundation soil consolidation by establishing a nonlinear mapping relationship between multi-physical quantity input and settlement response.

[0029] In a further technical solution, the settlement amount of the unmonitored area is predicted by using the Kriging interpolation method, which is implemented through the following relationship:

[0030]

[0031] Wherein, is the settlement amount prediction value of the unmonitored area, is the number of adjacent monitoring points, is the discrete settlement amount of the th monitoring point, is the weight coefficient determined based on the semi-variance function.

[0032] In a second aspect, the embodiments of the present application provide a distributed optical fiber monitoring system for port large equipment foundation settlement, which comprises:

[0033] A multi-physical quantity distributed optical fiber sensing network is arranged in a port large equipment foundation to sense strain data, temperature data and pore water pressure data of the port large equipment foundation;

[0034] An optical signal acquisition unit is configured to synchronously acquire multi-physical quantity signals of the strain data, temperature data and pore water pressure data carried in the multi-physical quantity distributed optical fiber sensing network;

[0035] A data processing terminal is configured to process the multi-physical quantity signals, convert the signals into discrete settlement data of each monitoring point of the port large equipment foundation by a multi-physical quantity fusion settlement inversion model;

[0036] A spatial analysis module is configured to perform spatial interpolation processing on the discrete settlement data output by the data processing terminal to generate a continuous settlement distribution map representing the overall settlement state of the port large equipment foundation;

[0037] An intelligent regulation and control unit is configured to generate equipment operation regulation and control instructions based on the continuous settlement distribution map and send the instructions to a port equipment management system.

[0038] The present application provides a distributed optical fiber monitoring method and system for settlement of a port large equipment foundation, which has the following advantages:

[0039] 1. The present application solves the problem that the single physical quantity monitoring method in the prior art cannot comprehensively reflect the complex foundation settlement mechanism. By fusing distributed optical fiber strain, temperature and pore water pressure sensing data, the present application constructs a nonlinear settlement inversion model, which can convert multi-dimensional data into accurate settlement amount and distinguish between immediate settlement caused by equipment load and long-term settlement caused by consolidation of foundation soil. Compared with the prior art which only relies on single strain data for judgment, the present application avoids misjudgment due to incomplete data and provides a comprehensive technical basis for foundation stability evaluation.

[0040] 2. The present application establishes predictive maintenance capability through dynamic analysis of multi-physical quantity data. The method can predict the future consolidation settlement amount that may occur in advance according to the change trend of pore water pressure and settlement rate, so that the system can change from passive overrun alarm to active risk warning, solving the deficiency of passive response after the problem occurs in the prior art, thereby providing an intervention opportunity before the settlement problem worsens.

[0041] 3、The present application realizes the deep linkage of the monitoring system and the equipment operation management system, directly converts the analysis result of the settlement distribution map into specific equipment operation control instructions, such as adjusting the equipment walking path, limiting the load or speed, and automatically sends to the port equipment management system, the automatic conversion and execution of the data to the instruction solves the problem that the monitoring system and the control system are independent of each other in the prior art, and the response delay needs manual intervention, forms a closed-loop management mode of "monitoring-analysis-feedback-control", and ensures the safety of port operation. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is a system structure schematic diagram of the embodiment of the present application.

[0043] Figure 2 It is a method flow schematic diagram of the embodiment of the present application.

[0044] Figure 3 It is a multi-physical quantity distributed optical fiber sensing network layout schematic diagram of the embodiment of the present application.

[0045] Figure 4 It is a functional module and data flow schematic diagram in the data processing terminal of the embodiment of the present application.

[0046] Figure 5 It is a flow schematic diagram of the spatial analysis module for data space interpolation processing of the embodiment of the present application.

[0047] Figure 6 It is a curve graph of the settlement of the monitoring point changing with time of the embodiment of the present application.

[0048] Figure 7 It is a two-dimensional thermal diagram of continuous settlement distribution of the embodiment of the present application.

[0049] Figure 8 It is a three-dimensional surface graph of continuous settlement distribution of the embodiment of the present application.

[0050] EXPLANATION OF DRAWINGS:

[0051] 10, multi-physical quantity distributed optical fiber sensing network; 20, optical signal acquisition unit; 30, data processing terminal; 40, spatial analysis module; 50, intelligent control unit. DETAILED DESCRIPTION

[0052] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] Figure 1 is a schematic diagram of a distributed optical fiber monitoring system for port large equipment foundation settlement according to an embodiment of the present application. As shown, the embodiment of the present application provides a distributed optical fiber monitoring system for port large equipment foundation settlement, which can include a multi-physical quantity distributed optical fiber sensing network 10, an optical signal acquisition unit 20, a data processing terminal 30, a spatial analysis module 40, and an intelligent control unit 50. Figure 1

[0054] The multi-physical quantity distributed optical fiber sensing network 10 is laid in the interior of the port large equipment foundation, and is used to sense strain data, temperature data, and pore water pressure data of the foundation. The multi-physical quantity distributed optical fiber sensing network 10 is composed of armored optical cables, which internally integrate strain sensing optical fibers, temperature sensing optical fibers, and pore water pressure sensing optical fibers. The strain sensing optical fibers are used to sense the axial deformation of the foundation structure generated in the settlement process; the temperature sensing optical fibers are used to measure the temperature of the optical fibers along the line and the surrounding environment; and the pore water pressure sensing optical fibers are used to measure the pore water pressure of the foundation soil (i.e. the pressure of water in the pores between soil particles, which reflects the consolidation and compression state of the soil). The armored optical cables are laid in layers along the longitudinal, transverse, and depth directions of the equipment foundation, forming a three-dimensional monitoring grid.

[0055] In another embodiment of the present application, the armored optical cables can also internally integrate acceleration or vibration sensing optical fibers. Such optical fibers are used to assist in analyzing the running stability of the equipment and the dynamic response of the foundation by monitoring the vibration frequency and amplitude of the foundation under equipment operation or external disturbance, to provide more dynamic information for settlement inversion.

[0056] The optical signal acquisition unit 20 is connected with the multi-physical quantity distributed optical fiber sensing network 10, and is used to synchronously acquire the multi-physical quantity signals carried in the multi-physical quantity distributed optical fiber sensing network 10. The optical signal acquisition unit 20 includes a laser emitter, an optical coupler, a photodetector, and a signal demodulator. The laser emitter is used to emit a probe light signal to the multi-physical quantity distributed optical fiber sensing network 10. The optical coupler is used to split the probe light signal emitted by the laser emitter, so that it can simultaneously enter each optical fiber branch of the multi-physical quantity distributed optical fiber sensing network 10, and is used to converge the scattered light or reflected light returned from the optical fiber network to the photodetector. The photodetector is used to receive the light signal returned from the multi-physical quantity distributed optical fiber sensing network 10. The signal demodulator is used to demodulate the received light signal to obtain raw data representing strain, temperature, and pore water pressure.

[0057] ​The data processing terminal 30 is connected with the optical signal acquisition unit 20, and is used for processing the multi-physical quantity signals. The data processing terminal 30 is internally provided with a plurality of software modules, including a temperature compensation module and a multi-physical quantity fusion settlement inversion model. The temperature compensation module is used for compensating the signals collected by the strain sensing optical fiber based on the temperature data collected by the temperature sensing optical fiber, so as to obtain real strain data generated only by the basic deformation. The multi-physical quantity fusion settlement inversion model is used for receiving the real strain data, the pore water pressure data and the temperature data, and converting the real strain data, the pore water pressure data and the temperature data into discrete settlement data of the basic monitoring points.

[0058] The spatial analysis module 40 is connected with the data processing terminal 30, and is used for performing spatial interpolation processing on the discrete settlement data. The spatial analysis module 40 generates a continuous settlement distribution map representing the basic overall settlement state based on the discrete settlement data of the monitoring points and the spatial position relationship thereof. The continuous settlement distribution map can be presented in the form of a two-dimensional heat map or a three-dimensional surface graph (as shown in Figure 7 and Figure 8 ).

[0059] The intelligent control unit 50 is connected with the spatial analysis module 40, and is used for generating a device operation control instruction sent to the port device management system based on the analysis result of the continuous settlement distribution map. The intelligent control unit 50 is internally provided with an intelligent control decision module, which generates a corresponding control instruction according to the over-limit settlement area or the abnormal settlement rate area identified in the continuous settlement distribution map. The control instruction is sent to the port device management system through a communication interface, and the control instruction is used for remotely adjusting the operation path, load or speed of the device. The communication interface can adopt a general industrial Ethernet protocol.

[0060] The functional modules of the data processing terminal 30, the spatial analysis module 40 and the intelligent control unit 50 can be logically divided, but can be physically integrated on the same industrial computer or server hardware. The hardware platform is configured with a processor for performing high-speed data operation, a memory for storing historical monitoring data and a nonlinear fusion model, and a communication interface for data exchange with other systems.

[0061] Figure 2 is a flowchart of a distributed optical fiber monitoring method for foundation settlement of a large-scale port device according to an embodiment of the present application. As shown in Figure 2 , the method realizes accurate monitoring and active control of the foundation settlement of the large-scale device through fusion and analysis of multi-physical quantity data. The method includes the following steps:

[0062] S1: A multi-physical quantity distributed optical fiber sensing network 10 is laid in a large-scale equipment foundation of a port, and the multi-physical quantity distributed optical fiber sensing network 10 is used to sense strain data, temperature data and pore water pressure data of the foundation.

[0063] S2: Multi-physical quantity signals including strain data, temperature data and pore water pressure data carried in the multi-physical quantity distributed optical fiber sensing network 10 are synchronously collected.

[0064] S3: The multi-physical quantity signals are processed, and the strain data, temperature data and pore water pressure data are converted into discrete settlement data of each monitoring point of the foundation by a multi-physical quantity fusion settlement inversion model.

[0065] S4: The discrete settlement data are subjected to spatial interpolation processing to generate a continuous settlement distribution map representing the overall settlement state of the foundation.

[0066] S5: Based on the continuous settlement distribution map, an uneven settlement area of the foundation is analyzed, and equipment operation control instructions are generated and sent to a port equipment management system. Specifically, the continuous settlement distribution map is analyzed to determine whether there is an over-limit settlement area or an abnormal settlement rate area. If the determination result is no, the step returns to step S2, and if the determination result is yes, the equipment operation control instructions are generated and sent to the port equipment management system.

[0067] The steps of the embodiment of the application will be described in detail below.

[0068] Referring to Figure 3 , Figure 3 It is a multi-physical quantity distributed optical fiber sensing network layout schematic diagram in a large-scale equipment foundation of a port according to an embodiment of the application.

[0069] The multi-physical quantity distributed optical fiber sensing network 10 of step S1 is laid in the large-scale equipment foundation of the port, and the multi-physical quantity distributed optical fiber sensing network 10 is used to sense strain data, temperature data and pore water pressure data of the foundation, specifically including:

[0070] According to the structure type of the large equipment foundation of the port and the geological exploration report, a layout scheme of a three-dimensional monitoring grid is designed. The three-dimensional monitoring grid is used to comprehensively capture the uneven deformation of the foundation structure in the horizontal direction and the vertical direction. The layout scheme considers different structural parts of the foundation, such as pile foundation, pile cap, pile cap beam, etc. The armored optical cable containing strain sensing optical fiber, temperature sensing optical fiber and pore water pressure sensing optical fiber is layered buried. The strain sensing optical fiber and the temperature sensing optical fiber are laid on the reinforcement net frame of the pile cap or the pile cap beam to ensure reliable strain transmission between them and the structure body. The pore water pressure sensing optical fiber is laid in the foundation soil layer below the foundation, and the laying depth is determined according to the thickness of the soft soil layer in the geological report to obtain the pore water pressure data at different depths.

[0071] In the specific construction process, the armored optical cable is fixed at the specified position by pre-buried or post-positioned method. Before pouring concrete, the armored optical cable is fixed on the steel reinforcement framework along the predetermined path, and the bending radius of the armored optical cable is ensured to meet the technical specification to avoid damage to the performance of the optical fiber. In the soil layer where the pore water pressure needs to be measured, the pore water pressure sensing optical fiber in the armored optical cable is fixed after being deeply inserted to the specified depth through the probe with a water filtering device. The probe can filter soil particles to ensure that the pore water can enter and effectively act on the flexible diaphragm of the pore water pressure sensing optical fiber.

[0072] The strain sensing optical fiber, the temperature sensing optical fiber and the pore water pressure sensing optical fiber in the multi-physical quantity distributed optical fiber sensing network 10 are respectively used to sense different physical quantities. The strain sensing optical fiber and the temperature sensing optical fiber realize the sensing of physical quantities through their material properties, that is, the changes of strain and temperature will cause the changes of Brillouin frequency shift or Rayleigh scattering signal in the optical fiber. The pore water pressure sensing optical fiber is packaged in the probe with a flexible diaphragm, and the change of pore water pressure causes the deformation of the diaphragm, which in turn causes the change of the periodic structure of the fiber grating sensor packaged therein, and finally causes the wavelength of the reflected light to drift. The wavelength drift is linearly related to the pressure.

[0073] The multi-physical quantity signals of strain, temperature and pore water pressure carried in the multi-physical quantity distributed optical fiber sensing network 10 in step S2 are synchronously collected, which is specifically realized by the optical signal collection unit 20, and the process includes:

[0074] The optical signal collection unit 20 emits a probe laser pulse to the multi-physical quantity distributed optical fiber sensing network 10. When the laser pulse propagates in the optical fiber, Brillouin scattering or Rayleigh scattering occurs with the optical fiber material. The optical signal collection unit 20 receives the scattered light signal returned from the multi-physical quantity distributed optical fiber sensing network 10 through a photodetector.

[0075] For the strain sensing fiber and the temperature sensing fiber, the signal demodulator in the optical signal acquisition unit 20 determines the strain and temperature by analyzing the frequency shift of the scattered light signal. Specifically, the frequency shift of the Brillouin scattered light in the optical fiber has a linear relationship with the strain and temperature suffered by the optical fiber, which can be expressed as:

[0076]

[0077] is the coordinate of the optical fiber position, is the measured Brillouin frequency shift at the position, is the strain coefficient of the optical fiber, is the temperature coefficient of the optical fiber, is the strain at the position, is the temperature at the position, is the reference Brillouin frequency shift. The signal demodulator synchronously obtains the strain and temperature data of the strain sensing fiber and the temperature sensing fiber along the line by analyzing the frequency shift.

[0078] For the pore water pressure sensing fiber, the optical signal acquisition unit 20 collects the signal by the fiber grating sensing technology. The pore water pressure sensing fiber is packaged in a probe with a flexible diaphragm. The flexible diaphragm transmits the pore water pressure in the soil layer to the fiber grating. The periodic structure of the fiber grating changes due to the pore water pressure, resulting in a shift in the center wavelength of its reflected light. The signal demodulator in the optical signal acquisition unit 20 measures the wavelength shift. The wavelength shift has a linear relationship with the pore water pressure:

[0079]

[0080]

[0081] is the measured reflected light wavelength shift, is the applied pore water pressure, is the pressure sensitivity coefficient of the fiber grating, is the reference wavelength shift.

[0082] By integrating hardware and software with different signal demodulation functions (such as Brillouin demodulation and fiber grating demodulation) in the optical signal acquisition unit 20, the embodiment of the present application realizes the synchronous collection of different physical quantity signals. The synchronization in the collection process is realized through a unified trigger clock and data timestamp mechanism, ensuring the correspondence of the strain, temperature and pore water pressure data in time.

[0083] As Figure 4 ​​​​​​​​As shown, Figure 4 This is a schematic diagram of the functional modules and data flow inside a data processing terminal 30 according to an embodiment of the present invention.

[0084] Step S3 involves processing the multi-physical quantity signals. The strain, temperature, and pore water pressure data are converted into discrete settlement data for each monitoring point on the foundation using a multi-physical quantity fusion settlement inversion model. This process is specifically completed in the data processing terminal 30 and includes:

[0085] First, the data processing terminal 30 receives the demodulated raw multi-physical quantity signal from the optical signal acquisition unit 20. The raw multi-physical quantity signal includes Brillouin frequency shift, Rayleigh scattering signal, and reflected light wavelength drift data from the pore water pressure sensing fiber at various locations along the multi-physical quantity distributed optical fiber sensing network 10. The data processing terminal 30 performs data preprocessing on the raw signal, including signal denoising and timestamp-based data alignment, to form a time-synchronized dataset of strain data, temperature data, and pore water pressure data.

[0086] Subsequently, the temperature compensation module in the data processing terminal 30 calls the time-synchronized dataset. Based on the temperature data acquired by the temperature sensing fiber optic cable, the temperature compensation module separates the temperature effects from the mixed signals acquired by the strain sensing fiber optic cable, thereby calculating the true strain data generated solely by the deformation of the foundation structure. The calculation process follows the following formula:

[0087]

[0088] in, These are the coordinates of the fiber optic location. For position Real response in the situation For position The measured Brillouin frequency shift, For reference to Brillouin frequency shift, For position The actual measured temperature The strain coefficient of the optical fiber is denoted as . This represents the temperature coefficient of the optical fiber.

[0089] Then, the multi-physical quantity fusion settlement inversion model in the data processing terminal 30 receives real strain data output from the temperature compensation module, pore water pressure data and temperature data from the optical signal acquisition unit. This multi-physical quantity fusion settlement inversion model is a nonlinear model that deeply integrates foundation consolidation theory and machine learning algorithms. It can overcome the limitations of traditional single physical quantity or simple linear models, and achieve a comprehensive, high-precision, and multi-dimensional inversion of the settlement mechanism of large port equipment foundations.

[0090] The multi-physical fusion settlement inversion model uses real strain data pore water pressure data and temperature data An external input variable representing equipment load is introduced as a major independent input variable (supplied by the port equipment management system, reflecting the real-time position and working state of the equipment). The settlement of each monitoring point is calculated by establishing the following complex nonlinear function relationship :

[0091]

[0092] wherein, is the calculated settlement, is the real strain data, is the pore water pressure data, is the temperature data, is the equipment load data, represents a multi-physical quantity fusion settlement inversion model, which is a comprehensive function that takes multiple coupled physical quantities (real strain pore water pressure data temperature data and equipment load as independent variables, and outputs the settlement through a highly nonlinear mapping relationship.

[0093] This nonlinear model deeply integrates the foundation consolidation theory in structure and logic to accurately distinguish the settlement components from different sources.

[0094] Instantaneous settlement (elastic deformation), which is mainly caused by the elastic deformation of the structure and the instantaneous deformation of the soil body caused by the equipment load . In the model, real strain data is used as a key indicator reflecting the instantaneous response of the structure and the near-field soil, combined with equipment load for analysis to analyze this part of the settlement. The model learns the response relationship between strain and load in the short term to obtain the characteristics of instantaneous settlement.

[0095] Long-term settlement (consolidation settlement), which is mainly caused by consolidation caused by the dissipation of pore water pressure of the foundation soil. Pore water pressure data is the core variable for analyzing long-term consolidation settlement, and the model quantifies this part of the settlement by monitoring the change trend (such as the dissipation process) and its lagging effect on settlement. The model contains a sub-module to simulate the consolidation process of the soil body, or learns the internal law between pore pressure change and long-term settlement through machine learning algorithms.

[0096] temperature data The present application is used for accurate correction of thermal deformation of soil and structure. Due to thermal expansion and contraction of materials, measurement errors or actual deformation may occur. The nonlinear model compensates for temperature data readings of and pore water pressure data or corrects the settlement amount at the output end of the model, ensuring the accuracy of settlement inversion is not disturbed by temperature effects.

[0097] For the nonlinear function f in the above multi-physical quantity fusion settlement inversion model, a model architecture with nonlinear fitting capability and time series data processing capability can be used, such as deep neural network (DNN), long short-term memory (LSTM), or gated recurrent unit (GRU). The mechanism of the above machine learning algorithm is as follows: through multi-layer nonlinear transformation, the complex nonlinear coupling relationship between multiple physical quantities is effectively captured; through feature extraction and fusion operation, deep features related to settlement are automatically learned and extracted from the original input data, and effective fusion of such deep features is realized; especially for LSTM and GRU models, time series dependency modeling can effectively handle the time series dependency in the process of pore water pressure dissipation over time and settlement development over time. In the parameter determination process of the multi-physical quantity fusion settlement inversion model, the specific parameters of the model need to be fine-tuned and iteratively calibrated in combination with large-scale finite element simulation analysis data and measured data from field calibration load tests; among them, finite element simulation can provide theoretical settlement response data under different working conditions, and field calibration load test can provide high-precision verification data in real application scenarios. The above double training mechanism can ensure that the model has good robustness in terms of generalization ability and calculation accuracy. Finally, the multi-physical quantity fusion settlement inversion model can output discrete settlement data of each monitoring point along the multi-physical quantity distributed optical fiber sensing network 10. The discrete settlement data can accurately reflect the settlement state of the foundation at different time nodes and different spatial positions, and has realized effective decoupling of immediate settlement and long-term consolidation settlement components.

[0098] Referring to Figure 5 , Figure 5 is a flowchart of data space interpolation processing by the spatial analysis module 40 according to an embodiment of the present application.

[0099] Step S4 is to perform spatial interpolation processing on the discrete settlement data to generate a continuous settlement distribution map representing the overall settlement state of the foundation, which is realized in the spatial analysis module 40. The process includes:

[0100] The spatial analysis module 40 receives discrete settlement data from each monitoring point along the multi-physical quantity distributed optical fiber sensor network 10, output from the data processing terminal 30. The discrete settlement data is correlated with the precise coordinate position of each monitoring point in three-dimensional space.

[0101] The spatial analysis module 40 uses Kriging interpolation to process discrete settlement data. Kriging interpolation is an optimal linear unbiased prediction method based on spatial statistics principles. The method first constructs a semivariance function by performing spatial autocorrelation analysis on the discrete settlement data. The semivariance function quantifies the relationship between the difference in settlement values ​​between any two monitoring points and their spatial distance. Through the semivariance function, spatial correlation information between the settlement values ​​of unmonitored areas and the settlement values ​​of surrounding known monitoring points can be obtained.

[0102] Specifically, the core of Kriging interpolation lies in constructing the semivariance function. This process first calculates the empirical value of the semivariance between all pairs of monitoring points, i.e.:

[0103]

[0104] in, Distance The empirical value of the semivariance at that location. The distance is The number of monitoring points and Positions and Settlement at the location.

[0105] Subsequently, the empirical values ​​are fitted using a mathematical model to obtain the theoretical semivariogram function. Fitting models include, but are not limited to, the spherical model, the exponential model, and the Gaussian model. In this embodiment of the invention, the optimal fitting model can be selected based on the characteristics of the actual data. For example, the spherical model is used to describe the spatial correlation of geological variables; it has a clear nugget effect, sill value, and range, and can well reflect the relationship between subsidence and increasing distance.

[0106] Based on the semivariance function, the spatial analysis module 40 calculates the weight coefficients of neighboring monitoring points for each unmonitored area's prediction point. The calculation of these weight coefficients comprehensively considers the distance between the monitoring point and the prediction point, as well as the spatial relationships between the monitoring points. The calculation is obtained by solving the following system of Kriging equations:

[0107]

[0108] wherein, is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above; is the semi-variance value between the first monitoring point and the second monitoring point, which is calculated by the theoretical semi-variance function constructed as mentioned above;

[0109] By solving this equation set, the optimal linear unbiased weight coefficient of each monitoring point for the settlement prediction of the unmonitored area can be obtained .

[0110] The settlement of the unmonitored point is calculated by the following relationship:

[0111]

[0112] wherein, is the settlement prediction value of the unmonitored area, is the number of adjacent monitoring points participating in the calculation, is the discrete settlement of the first monitoring point, is the discrete settlement of the first monitoring point, is the weight coefficient determined based on the semi-variance function. The sum of the weight coefficients is equal to 1, ensuring the unbiasedness of the prediction value.

[0113] By performing Kriging interpolation processing on the entire foundation area, the spatial analysis module 40 generates a complete and continuous settlement data grid. Through data visualization technology, the settlement data grid generates a two-dimensional heat map or a three-dimensional surface map (as shown in Figure 7 and Figure 8 ), which represents the overall settlement state of the large equipment foundation in the port. The continuous settlement distribution map clearly shows the settlement distribution and uneven settlement area in the foundation area.

[0114] Step S5 analyzes the uneven settlement area of the foundation based on the continuous settlement distribution map and generates equipment operation control instructions sent to the port equipment management system, which is specifically implemented in the intelligent control unit 50, and the process includes:

[0115] The intelligent control unit 50 receives the continuous settlement distribution map data generated by the spatial analysis module 40. The intelligent control decision module built in the intelligent control unit 50 performs deep data analysis on the continuous settlement distribution map data to accurately identify the settlement amount over-limit area or the settlement rate abnormal area. The identification process not only compares the settlement values of each area in the map with the preset settlement amount thresholds (such as the first-level warning threshold and the second-level danger threshold), but also compares the dynamically monitored settlement rate with the preset settlement rate threshold. The threshold is set according to the type of port equipment, the specific load working condition, the nature of the foundation soil, and the historical settlement data, and can be dynamically adjusted according to the feedback of the monitoring data to ensure the timeliness and accuracy of the warning.

[0116] When the intelligent control unit 50 identifies the settlement amount over-limit area or the settlement rate abnormal area, its intelligent control decision module will automatically generate and recommend corresponding device operation control instructions according to the accurate position, impact range and specific severity of the abnormal area (e.g., exceeding the first-level threshold or the second-level threshold). These control instructions are conducive to achieving the optimal balance between device operation safety and port operation efficiency.

[0117] The specific control instruction generation logic can adopt the following intelligent decision based on the expert rule base and / or the intelligent decision based on the deep learning reinforcement learning algorithm.

[0118] The expert rule base-based intelligent decision mode pre-establishes a set of expert rule base, the content of which is based on rich engineering experience, geomechanics theory and international safety standards. For example, when the settlement amount or settlement rate of a certain track area first exceeds the first-level warning threshold, the system immediately generates an instruction to reduce the running speed of the device in that area to X meters / minute, and starts intensive monitoring of that area. If the settlement amount or settlement rate continues to deteriorate and exceeds the second-level danger threshold, higher-level instructions are generated to limit the operation of the device in that area (e.g., prohibit heavy load passing) or completely stop the operation of the device in that area and perform emergency repair. For areas that present local, non-uniform over-limit, the system will generate instructions to adjust the device running path to avoid the affected area.

[0119] In a preferred embodiment of the present application, the intelligent control decision module built-in the intelligent regulation unit 50 adopts intelligent decision based on reinforcement learning algorithm of deep learning to realize autonomous learning and optimized generation of regulation instructions. The reinforcement learning algorithm takes real-time settlement, settlement rate, uneven settlement area range, current load of the equipment, running speed and historical regulation effect in the continuous settlement distribution map as its state input. The executable action set includes specific regulation instructions such as adjusting the equipment speed, limiting the load, changing the running path, pausing the work, etc. The reward mechanism is designed to minimize the settlement risk, maximize the safe operation time of the equipment and minimize the operation interruption. Through large-scale training and iterative optimization of historical data and simulated environment offline or online, the reinforcement learning algorithm can learn to automatically derive and execute the optimal and highly adaptive regulation strategy under different settlement abnormal states. The intelligent decision mode based on the deep learning reinforcement learning algorithm can effectively replace the traditional pre-defined expert rule base, realize more intelligent, more creative and adaptive dynamic decision-making under complex and variable working conditions.

[0120] The intelligent regulation unit 50 sends the generated equipment operation regulation instructions to the port equipment management system in real time through a standardized communication interface (for example, a general industrial Ethernet protocol). After receiving and verifying the regulation instructions, the port equipment management system automatically executes the corresponding equipment operation regulation, such as adjusting the walking path of the equipment, limiting its running load or reducing its running speed. The whole process of generation, sending and execution of regulation instructions does not require manual intervention, realizing real-time, proactive and intelligent management of potential settlement risks.

[0121] In order to further illustrate the cooperative working process of the technical scheme of the embodiment of the present application, a specific working scene example will be described below.

[0122] Figure 6 is a curve diagram of the settlement of the monitoring point changing with time according to an embodiment of the present application. As shown in Figure 6 , the present embodiment takes a port bucket wheel machine as the application object of the large equipment. The bucket wheel machine moves along the track to carry out stacking and unloading work in the port yard. The running track foundation of the bucket wheel machine is a pile-supported pile cap beam structure, which is subject to uneven settlement risk under the long-term action of heavy load, reciprocating movement and vibration.

[0123] In the present embodiment, the armored optical cable in the multi-physical quantity distributed optical fiber sensing network 10 is arranged along the longitudinal direction of the track beam of the bucket wheel machine at the top and the bottom and further extends into the pile foundation of the pile cap. The strain sensing optical fiber is bundled and fixed with the steel mesh frame inside the track beam to ensure that the strain data can be accurately transmitted. The pore water pressure sensing optical fiber is arranged in the soft soil layer below the track foundation through the probe to monitor the change of pore water pressure in the soil body during consolidation.

[0124] When the bucket wheel machine is performing the stacking and reclaiming operation along the track, the optical signal acquisition unit 20 synchronously acquires the strain, temperature and pore water pressure data from the multi-physical quantity distributed optical fiber sensing network 10. The optical signal acquisition unit 20 transmits the data to the data processing terminal 30. The data processing terminal 30 calls the internal temperature compensation module and the multi-physical quantity fusion settlement inversion model to process the data in real time, and generates discrete settlement data of each monitoring point along the track beam.

[0125] The spatial analysis module 40 receives the discrete settlement data and performs Kriging interpolation processing. According to the processing result, the spatial analysis module 40 generates and updates the settlement distribution heat map of the track foundation of the bucket wheel machine in real time. The settlement distribution heat map directly shows the settlement amount distribution along the track beam, and the color change reflects the settlement amount difference in different regions.

[0126] With reference to Figure 6 , the intelligent control unit 50 continuously analyzes the continuous settlement distribution map. When the analysis result shows that the settlement rate of the bucket wheel machine abnormally increases (as shown in the curve slope significantly increases after the time point t1) when the bucket wheel machine runs in a certain track region (for example, position A to position B), and exceeds the preset settlement rate warning threshold, the intelligent control decision module of the intelligent control unit 50 automatically triggers a warning. Figure 6

[0127] The intelligent control unit 50 then generates a control instruction according to the predefined logic. The instruction can include limiting the running speed of the bucket wheel machine to X meters / minute in the XX to XX track section or prohibiting the bucket wheel machine from performing heavy load stacking and reclaiming operation in the XX to XX track section. The intelligent control unit 50 automatically sends the control instruction to the port equipment management system through the communication interface. After receiving the control instruction, the port equipment management system automatically executes the corresponding control program to adjust the running state of the bucket wheel machine. The control process realizes the active intervention of the potential settlement risk, and avoids the equipment running failure or track structure damage caused by the aggravation of uneven settlement.

[0128] The above-described embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it cannot be understood as limiting the scope of the patent. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.​

Claims

1. A distributed optical fiber monitoring method for the settlement of foundations of large port equipment, characterized in that, Includes the following steps: A multi-physical quantity distributed optical fiber sensor network is deployed inside the foundation of large port equipment. The multi-physical quantity distributed optical fiber sensor network is used to sense the strain data, temperature data and pore water pressure data of the large port equipment foundation. Simultaneously acquire multi-physical quantity signals, including strain data, temperature data, and pore water pressure data carried in the multi-physical quantity distributed optical fiber sensor network; The multi-physical quantity signals are processed, and the strain data, temperature data, and pore water pressure data are converted into discrete settlement data of each monitoring point of the port large equipment foundation through a multi-physical quantity fusion settlement inversion model. Spatial interpolation is performed on the discrete settlement data to generate a continuous settlement distribution map characterizing the overall settlement state of the foundation; Based on the continuous settlement distribution map, the uneven settlement area of ​​the port's large equipment foundation is analyzed, and equipment operation control instructions are generated and sent to the port equipment management system based on the analysis results. The step of deploying a multi-physical quantity distributed optical fiber sensor network inside the foundation of large port equipment includes: Armored optical cables containing strain sensing fibers, temperature sensing fibers, and pore water pressure sensing fibers are laid out in layers along the longitudinal, transverse, and depth directions of the large port equipment foundation to form a three-dimensional monitoring grid. The steps for synchronously acquiring the multi-physical quantity signals of the multi-physical quantity distributed optical fiber sensor network include: Brillouin frequency-shifted optical signals or Rayleigh scattered optical signals in the strain sensing fiber and the temperature sensing fiber are acquired by Brillouin optical time-domain reflection or Rayleigh scattering. And the wavelength shift signal of reflected light caused by pressure change in the pore water pressure sensing fiber is collected by a fiber optic grating sensor. The steps for processing the multi-physical quantity signal include: Based on the temperature data collected by the temperature sensing fiber, temperature compensation is performed on the signal collected by the strain sensing fiber to obtain the real strain data generated only by the deformation of the foundation. The real strain data, pore water pressure data, and temperature data are input into the multi-physical quantity fusion settlement inversion model to calculate the discrete settlement data.

2. The distributed optical fiber monitoring method for settlement of large port equipment foundations according to claim 1, characterized in that, The steps for spatial interpolation of the discrete settlement data include: Using the Kriging interpolation method, based on the discrete settlement data of each monitoring point and their spatial relationship, the weighting coefficients are calculated through the semivariance function to perform optimal linear unbiased prediction of the settlement in the unmonitored area, so as to form the continuous settlement distribution map.

3. The distributed optical fiber monitoring method for settlement of large port equipment foundations according to claim 1, characterized in that, The step of generating and sending equipment operation control instructions to the port equipment management system includes: Identify areas of excessive settlement or abnormal settlement rate in the continuous settlement distribution map; Based on the location, range, and severity of the settlement excess area or the settlement rate abnormal area, at least one control command is generated through intelligent decision-making logic constructed based on preset expert rules or reinforcement learning algorithms, which adjusts the equipment operating path, limits the equipment operating load, or reduces the equipment operating speed.

4. The distributed optical fiber monitoring method for settlement of large port equipment foundations according to claim 1, characterized in that, The step of temperature compensation for the signal acquired by the strain sensing fiber is achieved through the following relationship: ； in, These are the coordinates of the fiber optic location. For position Real response in the situation For position The measured Brillouin frequency shift, For reference to Brillouin frequency shift, For position The actual measured temperature The strain coefficient of the optical fiber is denoted as . This represents the temperature coefficient of the optical fiber.

5. The distributed optical fiber monitoring method for settlement of large port equipment foundations according to claim 1, characterized in that, The multi-physical quantity fusion settlement inversion model is a nonlinear model constructed based on the foundation consolidation theory and incorporating machine learning algorithms. The multi-physical quantity fusion settlement inversion model uses the real strain data, pore water pressure data, and temperature data as independent input variables. By establishing a nonlinear mapping relationship between the multi-physical quantity input and the settlement response, it dynamically distinguishes between instantaneous settlement caused by equipment load and long-term settlement caused by foundation soil consolidation.

6. The distributed optical fiber monitoring method for settlement of large port equipment foundations according to claim 2, characterized in that, The prediction of settlement in unmonitored areas using Kriging interpolation is achieved through the following formula: ； in, The predicted settlement value for the unmonitored area. The number of neighboring monitoring points. For the first Discrete settlement at each monitoring point These are the weighting coefficients determined based on the semivariance function.

7. A distributed fiber optic monitoring system for the settlement of foundations of large port equipment, characterized in that, The distributed optical fiber monitoring system is applied to the distributed optical fiber monitoring method for the settlement of large port equipment foundations as described in any one of claims 1-6, and the distributed optical fiber monitoring system comprises: A multi-physical quantity distributed optical fiber sensor network is deployed inside the foundation of large port equipment to sense strain data, temperature data, and pore water pressure data of the large port equipment foundation. The optical signal acquisition unit is used to synchronously acquire multi-physical quantity signals, including strain data, temperature data, and pore water pressure data, carried in the multi-physical quantity distributed optical fiber sensing network. The data processing terminal is used to process the multi-physical quantity signals and convert the signals into discrete settlement data of each monitoring point of the port's large equipment foundation through a multi-physical quantity fusion settlement inversion model. The spatial analysis module is used to perform spatial interpolation processing on the discrete settlement data output by the data processing terminal to generate a continuous settlement distribution map characterizing the overall settlement state of the port's large equipment foundation. The intelligent control unit is used to generate equipment operation control instructions based on the continuous settlement distribution map and send them to the port equipment management system.

Images