Long-piled wharf structure full-life lightweight monitoring method

Abstract

The invention discloses a full-life lightweight monitoring method for a long-pile wharf structure. The method comprises the steps that S1, monitoring nodes are selected and optimized; s2, deploying a lightweight array, and collecting monitoring data on the monitoring nodes based on LoRaWAN low-power networking; s3, deploying a lightweight edge server, and performing cleaning, feature extraction and storage on the monitoring data in a manner of removing abnormal values, filling missing data and performing clustering recognition; and S4, analyzing the monitoring data based on a DTW algorithm, judging the real-time health state of the structure, and displaying a judgment result in real time through a three-dimensional visual platform. According to the invention, the whole-life health monitoring cost is reduced, and meanwhile, the data automatic identification and structure safety evaluation functions are perfected.

Description

TECHNICAL FIELD

[0001] The application belongs to a kind of high-pile wharf structure full life lightweight monitoring method. BACKGROUND

[0002] With the surge in global port transportation demand, the bearing capacity, safety and durability of wharf structure are facing severe challenges. Traditional manual inspection and static detection methods cannot meet the full life cycle management needs, especially in the context of extreme weather, high-frequency operation and structural aging, potential safety hazards are increasingly prominent. By building an intelligent and real-time health monitoring system, dynamic perception and scientific evaluation of wharf structure state are realized, which is a key measure to ensure port safety operation, prolong the service life of facilities and optimize operation and maintenance costs.

[0003] At present, many high-pile wharf structures in China have applied health monitoring systems to monitor the wharf in real time, but the existing monitoring point layout scheme has a large number of sensors, high cost, and in the traditional scheme, according to finite element analysis, most of the measuring points are located in the low stress area, which cannot reasonably cover the entire wharf area; secondly, the early warning model has limitations, with a high false alarm rate.

[0004] Therefore, a high-pile wharf structure full life lightweight monitoring method is provided. SUMMARY

[0005] To solve the above problems existing in the prior art, the application provides a high-pile wharf structure full life lightweight monitoring method, which reduces the cost of full life health monitoring, and improves the functions of automatic data identification and structure safety evaluation.

[0006] The technical solution to achieve the above purpose is: A high-pile wharf structure full life lightweight monitoring method, comprising: Step S1, selecting and optimizing monitoring nodes; Step S2, deploying a lightweight array, and collecting monitoring data on the monitoring nodes based on LoRaWAN (LoRa technology-based Internet of Things communication protocol) low-power networking; Step S3, deploying a lightweight edge server, cleaning, feature extraction and storage of monitoring data by removing outliers, filling missing data and clustering identification; Step S4, analyzing the monitoring data based on the DTW (Dynamic Time Warping) algorithm to determine the real-time health status of the structure, and displaying the determination result in real time through a three-dimensional visualization platform.

[0007] Preferably, in step S1, selecting and optimizing monitoring nodes, comprising: By inputting the geometric parameters, material parameters and environmental loads of the wharf structure, a dynamic numerical simulation model is constructed, and by setting the stiffness, friction coefficient and soil resistance parameters of the pile-soil interface, the bending moment, shear force and axial force transmission law of the pile below the mud surface are simulated. The maximum bending moment point, shear force mutation point and displacement sensitive point are extracted from the stress cloud map to determine the monitoring node position.

[0008] Preferably, in step S1, the geometric parameters include but are not limited to pile length, cross-sectional size, bent spacing; the material parameters include but are not limited to concrete strength, reinforcement ratio; the environmental loads include but are not limited to ship impact force, wave force, surcharge.

[0009] Preferably, in step S2, the lightweight array is composed of fiber Bragg grating strain gauges, GNSS displacement meters, static leveling instruments and temperature sensors, and the data collected by the above sensors together constitute the monitoring data.

[0010] Preferably, in step S2, the fiber Bragg grating strain gauge is used to monitor the strain change on the surface or inside the structure, reflecting the stress state. The GNSS displacement meter is used to monitor the three-dimensional displacement of the structure, including horizontal displacement and vertical displacement. The static leveling instrument is used to monitor uneven settlement or inclination. The temperature sensor is used to monitor the environment or structure temperature for temperature compensation and thermal stress analysis.

[0011] Preferably, in step S4, based on the analysis of the monitoring data by the DTW algorithm, the real-time health status of the structure is determined, including: Step S41, collect time series data under historical health status to form multiple reference curves and construct a reference sequence library; Step S42, denoising, alignment and standardization processing are performed on the real-time collected monitoring data; Step S43, DTW matching is performed between the real-time sequence and each reference sequence to calculate the "distance" value between them; Step S44, set a threshold value, if the DTW distance between the real-time sequence and the reference sequence exceeds the threshold value, it is judged as abnormal, and the health status of the structure is comprehensively judged combined with the multi-sensor data; Step S45, the determination result is displayed in real time through a three-dimensional visualization platform to assist operation and decision-making.

[0012] Preferably, in step S43, if there are time series in the data set, denoted as wherein, For two time series and The alignment path is: ; Two time series and The DTW distance value is expressed by the formula: ; In the formula, This is the set of all alignment paths between two sequences that meet the constraints.

[0013] Compared with existing technologies, the beneficial effects of this invention are as follows: By deploying a lightweight array, this invention reduces the number of measurement points by 50% compared with traditional solutions, and uses DTW-based algorithm to analyze monitoring data to determine the real-time health status of the structure, thereby improving the accuracy of damage identification, reducing the false alarm rate, and reducing the error in remaining lifetime prediction; LoRaWAN networking reduces equipment energy consumption; the lightweight sensor overall layout scheme proposed in this invention reduces the cost of full life cycle health monitoring, while improving the functions of automated data identification and structural safety assessment. Attached Figure Description

[0014] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of a lightweight monitoring method for the entire life cycle of a high-pile wharf structure according to the present invention; Figure 2 This is a flowchart illustrating the process of analyzing monitoring data and determining the real-time health status of a structure based on the DTW algorithm in this invention. Figure 3 This is a schematic diagram of the stress monitoring curve I time series of the PHC pile in this invention on October 1st of a certain year; Figure 4 This is a schematic diagram of the stress monitoring curve II time series of the PHC pile in this invention on October 14th of a certain year; Figure 5 This is a schematic diagram of the stress monitoring curve III time series of the PHC pile in this invention on November 29th of a certain year; Figure 6 This is a schematic diagram of the stress monitoring curve IV time series of the PHC pile in this invention on December 19th of a certain year. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] like Figure 1 As shown, a method for lightweight monitoring of the entire life cycle of a high-pile wharf structure includes: Step S1: Select and optimize monitoring nodes.

[0017] In this embodiment, the selection and optimization of monitoring nodes includes: By inputting the geometric parameters of the wharf structure (such as pile length, cross-sectional dimensions, and spacing between supports), material parameters (such as concrete strength and steel reinforcement ratio), and environmental loads (such as ship impact force, wave force, and surcharge), a dynamically adjustable numerical simulation model is constructed. This model simulates the mechanical response of the wharf structure under different load conditions, provides a scientific basis for identifying key monitoring nodes, avoids reliance on experience-based monitoring point placement, and improves the targeting and economy of the monitoring system. By setting the stiffness, friction coefficient, and soil resistance parameters of the pile-soil interface, the system simulates the transmission patterns of bending moment, shear force, and axial force in the pile shaft below the mud surface; accurately reflecting the stress state of the pile foundation in actual soil. This provides a basis for identifying potentially high-risk areas below the mud surface (such as the point of maximum bending moment), reduces reliance on deep sensors, and enables "lightweight" monitoring by inferring the stress state below from the upper monitoring data.

[0018] The location of monitoring nodes is determined by extracting the maximum bending moment point (lower part of the pile top), the shear force abrupt change point (end of longitudinal and transverse beams), and the displacement sensitive point (mid-span of the panel) from the stress cloud diagram.

[0019] The stress cloud map is a visualization result generated by the finite element software in the post-processing module, which shows the stress distribution of the structure under each load step. The maximum bending moment point usually corresponds to the darkest area (maximum stress) in the bending moment cloud map. The principles for determining the location of monitoring nodes include: significant mechanical response, great impact on the overall safety of the structure, and feasible and economical sensor deployment. By prioritizing the deployment of sensors in these locations, the number of sensors can be significantly reduced while ensuring monitoring effectiveness.

[0020] Step S2: Deploy a lightweight array and collect monitoring data from the monitoring nodes based on LoRaWAN low-power networking.

[0021] In this embodiment, the lightweight array consists of fiber optic strain gauges, GNSS displacement gauges, hydrostatic levels, and temperature sensors. The data collected by these sensors together constitute the monitoring data.

[0022] In the embodiments, fiber optic strain gauges are used to monitor strain changes on or inside a structure, reflecting the stress state. GNSS displacement gauges are used to monitor three-dimensional displacement of structures, including horizontal and vertical displacement. A hydrostatic level is used to monitor uneven settlement or tilting. Temperature sensors are used to monitor ambient or structural temperatures for temperature compensation and thermal stress analysis.

[0023] The collected monitoring data is wirelessly transmitted to the gateway via the LoRaWAN network and then uploaded to the server, achieving low-power, long-distance, minute-level data collection.

[0024] Step S3: Deploy a lightweight edge server to clean, extract features, and store the monitoring data by removing outliers (Z-score>3σ), filling in missing data (interpolation method), and using clustering identification.

[0025] Step S4: Analyze the monitoring data based on the DTW algorithm to determine the real-time health status of the structure, and display the determination results in real time through a three-dimensional visualization platform; thereby realizing the ability to predict the working status of the structure to a certain extent based solely on real-time monitoring data, achieving the purpose of health monitoring.

[0026] Dynamic Time Warping (DTW) is an algorithm used to measure the similarity between two time series. DTW is based on the idea of ​​dynamic programming. Under specific constraints, it finds the optimal alignment path between the two time series curves and their corresponding distance value within a constructed cost matrix. First, a distance matrix between the two time series is constructed. Then, according to the idea of ​​dynamic programming, the alignment path with the minimum cumulative cost value is found in the distance matrix to match the two series. Finally, the distance of this shortest alignment path is used as the similarity measure between the two time series.

[0027] like Figure 2 As shown, the real-time health status of a structure is determined based on the analysis of monitoring data using the DTW algorithm, including: Step S41: Collect time series data of historical health status, form multiple benchmark curves, and construct a benchmark sequence library.

[0028] Step S42: Denoise, align, and standardize the real-time collected monitoring data.

[0029] Step S43: Perform DTW matching between the real-time sequence and each benchmark sequence to calculate the "distance" value (i.e., similarity index) between them.

[0030] In the example, if the dataset contains A time series, denoted as ,in, For two time series and The alignment path is: ; Two time series and The DTW distance value is expressed by the formula: ; In the formula, This is the set of all alignment paths between two sequences that meet the constraints.

[0031] The following constraints were set during the calculation of the DTW distance: Boundary conditions: The calculation process must start from the leftmost position on the time axis of the sequence and continue to accumulate until the rightmost position on the time axis.

[0032] Continuity condition: During the matching calculation of point pairs, the span of the time distance scale can be at most 1, that is, if the current two point pairs are... Then the next pair of points matched must be , or .

[0033] Monotonicity condition: The matching calculation process of point pairs can only find matching point pairs to the right of the previous matching point pair in the two sequences, that is, at the next time scale. In other words, the alignment path can never be returned.

[0034] Step S44: Set a threshold. If the DTW distance between the real-time sequence and the reference sequence exceeds the threshold, it is judged as abnormal. The structural health status is then comprehensively judged by combining multi-sensor data.

[0035] Step S45: The judgment result is displayed in real time through a 3D visualization platform to assist in operation and maintenance decision-making. Specific implementation examples: like Figures 3-6 As shown in Table 1, the time series of monitoring data No. I is used as the baseline data. The similarity distances of several monitoring data No. I-II, No. I-III, No. I-IV, etc. are obtained according to the similarity assessment method to assess their similarity. When there is a large change in similarity, the structure can be considered dangerous. The specific results are shown in Table 1. Table 1 After model training and DTW safety prediction are performed in the cloud, the structural safety status can be fed back through a 3D visualization interface.

[0037] Finally, it should be noted that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-pile wharf structure full-life lightweight monitoring method, characterized in that, The application relates to a method for monitoring the health state of a wharf structure, comprising the following steps: Step S1, selecting and optimizing the monitoring nodes; Step S2, deploying a light array and collecting the monitoring data on the monitoring nodes based on the LoRaWAN low-power networking; Step S3, deploying a light edge server, cleaning, extracting features and storing the monitoring data through the methods of removing abnormal values, filling in missing data and clustering identification; Step S4, analyzing the monitoring data based on the DTW algorithm, determining the real-time health state of the structure, and displaying the determination result in real time through a three-dimensional visualization platform.

2. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 1, characterized in that, In the step S1, the monitoring nodes are selected and optimized, comprising the following steps: By inputting the geometric parameters, material parameters and environmental loads of the wharf structure, a dynamic numerical simulation model is constructed, and the rigidity, friction coefficient and soil resistance parameters of the pile-soil interface are set to simulate the bending moment, shear force and axial force transmission law of the pile below the mud surface; The maximum bending moment point, shear force mutation point and displacement sensitive point are extracted from the stress nephogram, and then the monitoring node position is determined.

3. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 2, characterized in that, In the step S1, the geometric parameters include but are not limited to pile length, cross-sectional size and bent spacing; the material parameters include but are not limited to concrete strength and reinforcement ratio; and the environmental loads include but are not limited to ship impact force, wave force and stacking load.

4. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 1, characterized in that, In the step S2, the light array is composed of fiber bragg grating strain gauges, GNSS displacement meters, static leveling instruments and temperature sensors, and the data collected by the above sensors jointly constitute the monitoring data.

5. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 4, characterized in that, In the step S2, the fiber bragg grating strain gauge is used for monitoring the strain change on the surface or inside of the structure and reflecting the stress state; The GNSS displacement meter is used for monitoring the three-dimensional displacement of the structure, including the horizontal displacement and the vertical displacement; The static leveling instrument is used for monitoring the uneven settlement or inclination; The temperature sensor is used for monitoring the environment or structure temperature and is used for temperature compensation and thermal stress analysis.

6. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 1, characterized in that, In the step S4, the analysis of the monitoring data based on the DTW algorithm determines the real-time health state of the structure, comprising the following steps: Step S41, collecting the time series data under the historical health state, forming a plurality of reference curves, and constructing a reference sequence library; Step S42, denoising, aligning and standardizing the real-time collected monitoring data; Step S43, performing DTW matching between the real-time sequence and each reference sequence, and calculating the "distance" value between them; Step S44, setting a threshold value, if the DTW distance between the real-time sequence and the reference sequence exceeds the threshold value, it is judged as abnormal, and the health state of the structure is comprehensively judged in combination with the multi-sensor data; Step S45, displaying the determination result in real time through a three-dimensional visualization platform to assist operation and decision-making.

7. The full-life lightweight monitoring method of a high-pile wharf structure according to claim 6, characterized in that, In the step S43, if there are two time series, denoted as where for two time series and the alignment path is: ； two time series and The DTW distance value between two time series is expressed by the formula: ； wherein is the set of all alignment paths between the two sequences that satisfy the constraints.

Images