LNG storage tank structure health monitoring method, device, equipment, medium and product

Abstract

The invention discloses an LNG storage tank structure health monitoring method, device, equipment, medium and product, and relates to the field of LNG equipment monitoring, the method comprises the following steps: collecting precooling period data and operation period data of an LNG storage tank by adopting a layered sensor network; integrating the precooling period data and the operation period data by adopting a unified timestamp to obtain monitoring data; acquiring a strain displacement monitoring base line, and calibrating the strain displacement monitoring base line in real time based on the temperature field data in combination with the shrinkage characteristics of the inner wall of the LNG storage tank to obtain a pre-cooling working condition dynamic base line; on the basis of the pre-cooling working condition dynamic base line, distinguishing abnormal structure displacement of the structure deformation data through a three-dimensional motion coupling model to obtain three-dimensional coupling data; the strain temperature displacement parameters are associated based on the three-dimensional coupling data, the temperature field data and the structural deformation data, and a monitoring value is obtained; and generating a health monitoring result based on the monitoring value and the pre-cooling working condition dynamic baseline. According to the invention, the real-time performance and the accuracy of the health monitoring of the LNG storage tank structure can be improved.

Description

Technical Field

[0001] This application relates to the field of LNG storage tank monitoring, and in particular to a method, apparatus, equipment, medium and product for monitoring the structural health of LNG storage tanks. Background Technology

[0002] Liquefied Natural Gas (LNG) storage tank health monitoring is a systematic project that comprehensively utilizes sensing technology, data acquisition, and intelligent analysis to monitor and assess the structural status and safety indicators of LNG storage tanks in real time under critical operating conditions. Existing monitoring systems particularly focus on two core hazardous conditions: first, the pre-cooling process during the initial commissioning of the tank, where temperature sensors deployed on the tank wall closely monitor the temperature gradient to prevent material damage caused by excessive thermal stress due to sudden cooling; and second, foundation settlement throughout the entire operating cycle, where a high-precision settlement monitoring system tracks uneven settlement trends in real time and provides early warnings of potential structural instability risks. Through these targeted monitoring efforts, combined with data analysis and digital twin technology, crucial decision-making support can be provided to ensure the structural integrity and long-term operational safety of the storage tanks.

[0003] However, existing technologies for LNG storage tank structural health monitoring often rely on independent monitoring of a single physical parameter, lacking the coordinated integration of multi-dimensional data such as vertical settlement, horizontal displacement, and temperature field. Furthermore, the benchmark thresholds are mostly fixed and cannot be dynamically adjusted according to different operating conditions such as pre-cooling and operation, resulting in a lag or risk of misjudgment in the identification of subtle structural anomalies. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this application provides a method, apparatus, equipment, medium, and product for monitoring the structural health of LNG storage tanks.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for monitoring the structural health of an LNG storage tank, including: A hierarchical sensor network is used to collect data on the pre-cooling period and the operation period of the LNG storage tank; both the pre-cooling period data and the operation period data include temperature field data, structural deformation data and tilt angle. The pre-cooling period data and the operation period data are integrated using a unified timestamp to obtain monitoring data; Obtain the strain-displacement monitoring baseline, and based on the temperature field data, calibrate the strain-displacement monitoring baseline in real time in conjunction with the shrinkage characteristics of the inner wall of the LNG storage tank to obtain the dynamic baseline of the pre-cooling condition. Based on the dynamic baseline of the pre-cooling condition, the abnormal structural displacements of the structural deformation data are distinguished by a three-dimensional motion coupling model to obtain three-dimensional coupling data; abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank. Based on the three-dimensional coupling data, the temperature field data, and the structural deformation data, strain-temperature-displacement parameters are correlated to obtain monitored values; the strain-temperature-displacement parameters are determined based on the monitored data. Health monitoring results are generated based on the monitored values ​​and the dynamic baseline of the pre-cooling condition.

[0006] Secondly, this application provides an LNG storage tank structural health monitoring device, comprising: A hierarchical sensor network is used to collect data on the pre-cooling period and the operation period of LNG storage tanks; both the pre-cooling period data and the operation period data include temperature field data, structural deformation data, and tilt angle. The data synchronization and acquisition module is used to integrate the pre-cooling period data and the operation period data using a unified timestamp to obtain monitoring data. The baseline calibration module is used to acquire the strain-displacement monitoring baseline and, based on the temperature field data and the shrinkage characteristics of the inner wall of the LNG storage tank, calibrate the strain-displacement monitoring baseline in real time to obtain the dynamic baseline for the pre-cooling condition. The three-dimensional coupling module is used to distinguish abnormal structural displacements in the structural deformation data based on the dynamic baseline of the pre-cooling condition through a three-dimensional motion coupling model, thereby obtaining three-dimensional coupling data; abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank. The monitoring value determination module is used to obtain the monitoring value by associating strain-temperature-displacement parameters with the three-dimensional coupling data, the temperature field data, and the structural deformation data; the strain-temperature-displacement parameters are determined based on the monitoring data. The monitoring result generation module is used to generate health monitoring results based on the monitoring values ​​and the dynamic baseline of the pre-cooling condition.

[0007] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the LNG storage tank structural health monitoring method provided above.

[0008] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the LNG storage tank structural health monitoring method provided above.

[0009] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the LNG storage tank structural health monitoring method described above.

[0010] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, apparatus, equipment, medium, and product for monitoring the structural health of LNG storage tanks. By employing a hierarchical sensor network to collect pre-cooling and operational data of LNG storage tanks, including temperature field data, structural deformation data, and tilt angle data, and integrating this data using a unified timestamp, it addresses the lack of collaborative integration of multi-dimensional data such as vertical settlement, horizontal displacement, and temperature field in existing technologies. Furthermore, by acquiring a strain-displacement monitoring baseline and calibrating it in real-time based on temperature field data and the shrinkage characteristics of the LNG storage tank's inner wall, a dynamic baseline for pre-cooling conditions is obtained. This allows for dynamic adjustment of the monitoring benchmark according to different operating conditions such as pre-cooling and operation, thus solving the problem that benchmark thresholds are often fixed and cannot be dynamically adjusted according to different operating conditions. Based on the dynamic baseline for pre-cooling conditions, three-dimensional coupling data obtained through a three-dimensional motion coupling model is correlated with strain, temperature, and displacement parameters to obtain monitoring values. Based on these monitoring values ​​and the dynamic baseline for pre-cooling conditions, a health monitoring result is generated, improving the real-time performance and accuracy of LNG storage tank structural health monitoring. This addresses the problems of lag or misjudgment in the identification of subtle structural anomalies in existing technologies. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a method for monitoring the structural health of an LNG storage tank, provided as an embodiment of this application; Figure 2 This is a schematic diagram of the functional modules of an LNG storage tank structural health monitoring device provided in one embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0015] In one exemplary embodiment, this application provides a method for monitoring the structural health of an LNG storage tank. This method is executed by a computer device, specifically a terminal or server, or both. In this embodiment, the method is described using a server as an example. Figure 1 As shown, the method includes: Step 100: Collect pre-cooling and operational data of the LNG storage tank using a hierarchical sensor network. Both pre-cooling and operational data include temperature field data, structural deformation data, and tilt angle. By employing a hierarchical sensor network, structural changes can be captured from multiple physical levels, such as vertical settlement, horizontal displacement, and temperature field distribution, avoiding the limitations of single-parameter monitoring and providing a more comprehensive reflection of the tank's true condition under different operating conditions.

[0016] Step 101: Integrate pre-cooling period data and operational period data using a unified timestamp to obtain monitoring data. Based on this step, this application can achieve multi-dimensional, all-time perception of the tank structure status by integrating monitoring data from multiple sensors in a hierarchical sensor network.

[0017] Step 102: Obtain the strain displacement monitoring baseline, and based on the temperature field data, calibrate the strain displacement monitoring baseline in real time in combination with the shrinkage characteristics of the inner wall of the LNG storage tank to obtain the dynamic baseline of the pre-cooling condition.

[0018] Step 103: Based on the dynamic baseline of the pre-cooling condition, the abnormal structural displacements in the structural deformation data are distinguished using a three-dimensional motion coupling model to obtain the three-dimensional coupled data. Abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank.

[0019] Step 104: Based on the three-dimensional coupled data, temperature field data, and structural deformation data, correlate the strain-temperature-displacement parameters to obtain the monitored values. The strain-temperature-displacement parameters are determined based on the monitored data.

[0020] Step 105: Generate health monitoring results based on monitoring values ​​and the dynamic baseline of pre-cooling conditions.

[0021] By implementing steps 100-105 above, this application can effectively improve the ability to identify subtle structural anomalies, detect potential risks at an early stage, and provide a more accurate basis for judgment on structural safety protection.

[0022] In an exemplary embodiment of this application, to ensure that the monitoring threshold always matches the actual operating conditions, the monitoring benchmark can be dynamically adjusted, taking into account the low-temperature characteristics of the pre-cooling stage and the structural stability requirements during normal operation. Based on this, step 102 above, which involves calibrating the strain-displacement monitoring baseline in real time based on temperature field data and the shrinkage characteristics of the LNG storage tank inner wall to obtain the dynamic baseline for the pre-cooling condition, includes: Step 1. Data Acquisition and Analysis: During the pre-cooling phase of commissioning, based on the set temperature field (e.g., -169℃) in the pre-cooling data, the strain displacement caused by low-temperature contraction of the LNG tank inner wall (such as the contraction amount derived from the material's thermal expansion coefficient and temperature gradient) is determined. Structural deformation data corresponding to the strain displacement are then filtered out to obtain compensated structural deformation data. This step primarily aims to achieve deviation compensation and eliminate initial low-temperature deviations. Specifically, the strain displacement caused by inner wall contraction can be used as an initial deviation and subtracted from the monitoring data.

[0023] Step 2: Based on the compensated structural deformation data, calibrate the strain-displacement monitoring baseline in real time to obtain the dynamic baseline for the pre-cooling condition. This step is mainly to generate a corrected baseline. Using the compensated structural deformation data as the new baseline (i.e., the dynamic baseline for the pre-cooling condition) can eliminate the interference of low temperature on structural monitoring and ensure that the monitoring thresholds in the subsequent operation phase are consistent with the actual structural state.

[0024] Furthermore, in subsequent processing, this pre-cooling condition dynamic baseline can determine the benchmark threshold for monitoring during the subsequent operation period. Based on this, and combining the descriptions of steps 1 and 2 above, the baseline threshold can include a temperature threshold and a settlement threshold. Therefore, the process for determining the baseline threshold of the pre-cooling condition dynamic baseline provided in this application can be described as follows: In the pre-cooling stage of the pre-cooling condition dynamic baseline, the temperature threshold is set to -165℃ to -160℃, and the settlement threshold is dynamically adjusted downwards according to a linear function and a set rule as the temperature decreases. In the operation stage of the pre-cooling condition dynamic baseline, both the temperature threshold and the settlement threshold are set multiples of the average temperature and settlement values ​​over the 24 hours after the pre-cooling stabilization period.

[0025] In practical applications, the process of determining the baseline threshold based on the dynamic baseline of the pre-cooling condition may include: (1) Pre-cooling stage: lasts for several hours (e.g., 96 hours), with one set of baseline samples collected every hour. The temperature threshold is set to -165℃ to -160℃. The sedimentation threshold is linearly reduced as the temperature decreases (e.g., the sedimentation threshold decreases by 0.8 mm for every 10℃ decrease in temperature).

[0026] (2) During normal operation: The threshold is dynamically adjusted 3 times a day and locked at 1.2 times the average value of 24 hours after the pre-cooling stabilization period.

[0027] (3) Constraints: The cooling rate during the pre-cooling stage is ≤5℃ / h, the data validity rate is ≥98%, and the calibration must be performed when there is no liquid inflow or outflow in the storage tank.

[0028] In one exemplary embodiment of this application, the process of constructing the three-dimensional motion coupling model used in step 103 above may include: Step 1: Obtain the circumferential rotation angle of the LNG storage tank.

[0029] Step 2: Determine the vertical differential settlement and horizontal shrinkage displacement based on structural deformation data.

[0030] Step 3: Establish the coupling relationship between vertical differential settlement, horizontal contraction displacement and circumferential rotation angle and the weight load of the LNG storage tank, and use the coupling relationship as a three-dimensional motion coupling model.

[0031] The core purpose of constructing the three-dimensional motion coupling model in this application is to establish the vertical differential settlement (Δh) and horizontal contraction displacement (Δh). l The coupling relationship between the circumferential rotation angle (θ) and the LNG weight load (e.g., Δh=f(Δ)). l Based on this, the distinction between abnormal structural displacement and normal deformation can be described as follows: Calculate the deformation range under normal operating conditions (e.g., Δh≤5mm, θ≤0.2°) according to the tank design parameters and LNG weight. Compare the actual monitored coupling data with the normal range; any deviation exceeding this range is considered abnormal structural displacement, while any deviation below this range is considered normal deformation caused by the LNG weight.

[0032] In an exemplary embodiment of this application, in order to automatically generate a graded maintenance plan based on real-time monitoring results, this application combines the coupling effect of structural parameters to determine the degree of anomaly, thereby avoiding the lag and subjectivity of traditional manual inspections. Based on this, the implementation process of step 105 above may include: Step 1: Determine the baseline threshold based on the dynamic baseline of the pre-cooling condition.

[0033] Step 2: Determine whether the monitored value is greater than the baseline threshold.

[0034] Step 3: When the monitored value exceeds the baseline threshold, the health monitoring result is abnormal, which can trigger an early warning and generate abnormal monitoring information based on the monitored value. Maintenance decisions are then generated based on the abnormal monitoring information.

[0035] Step 4: When the monitored value is less than or equal to the baseline threshold, the health monitoring result is normal.

[0036] Based on the above description, the LNG storage tank structural health monitoring method provided in this application significantly improves the initiative and accuracy of storage tank health management through multi-parameter collaborative analysis and intelligent response mechanisms. This dynamically adaptable monitoring and decision-making mode not only reduces unnecessary downtime for maintenance and lowers operating costs, but also enables rapid response when critical risks occur, ensuring the safety and stability of LNG storage and providing reliable support for the safety management of the storage tank throughout its entire life cycle. Furthermore, by using temperature field data for dynamic correction as provided in step 102 above, this application also has at least the following advantages compared to existing technologies: 1) Eliminate initial low temperature deviation: During the pre-cooling stage, the low temperature of LNG (such as -169℃) will cause the inner wall to shrink. Temperature field data can accurately capture this characteristic. The corrected baseline (i.e., the dynamic baseline of the pre-cooling condition) can reflect the actual structural deformation rather than the initial effect of low temperature shrinkage.

[0037] 2) Matching actual working conditions: Avoiding the limitations of traditional fixed thresholds that cannot adapt to different working conditions of pre-cooling / operation, improving the ability to identify subtle structural anomalies, and reducing the risk of lag or misjudgment.

[0038] 3) Multi-dimensional collaboration: Combining temperature field and strain displacement data, it more comprehensively reflects the structural state of the storage tank at low temperature, providing an accurate benchmark for subsequent monitoring.

[0039] Based on the same inventive concept, this application also provides an apparatus for implementing the LNG tank structural health monitoring method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the LNG tank structural health monitoring method apparatus provided below can be found in the limitations of the LNG tank structural health monitoring method described above, and will not be repeated here.

[0040] In one exemplary embodiment, such as Figure 2 As shown, an LNG storage tank structural health monitoring method and device is provided, comprising: a hierarchical sensor network, a data synchronization acquisition module, a baseline calibration module, a three-dimensional coupling module, a monitoring value determination module, and a monitoring result generation module.

[0041] A hierarchical sensor network is used to collect data from LNG storage tanks during the pre-cooling and operational periods. Both pre-cooling and operational data include temperature field data, structural deformation data, and tilt angle.

[0042] The data synchronization and acquisition module is used to integrate pre-cooling period data and operation period data using a unified timestamp to obtain monitoring data.

[0043] The baseline calibration module is used to acquire the strain-displacement monitoring baseline and, based on temperature field data and combined with the shrinkage characteristics of the inner wall of the LNG storage tank, calibrate the strain-displacement monitoring baseline in real time to obtain the dynamic baseline for pre-cooling conditions.

[0044] The 3D coupling module is used to distinguish abnormal structural displacements in structural deformation data based on the dynamic baseline under pre-cooling conditions, and obtain 3D coupled data through a 3D motion coupling model. Abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank.

[0045] The monitoring value determination module is used to correlate strain, temperature, and displacement parameters based on three-dimensional coupled data, temperature field data, and structural deformation data to obtain monitoring values. These strain, temperature, and displacement parameters are determined based on the monitoring data.

[0046] The monitoring results generation module is used to generate health monitoring results based on monitoring values ​​and the dynamic baseline of pre-cooling conditions.

[0047] As an alternative implementation, the hierarchical sensing network includes: vibrating wire strain gauges, fiber optic sensors, distributed fiber optic temperature sensors, and electric tiltmeters.

[0048] Vibrating wire strain gauges are installed on the bottom plate of the LNG storage tank to collect the settlement deformation of the LNG storage tank in real time.

[0049] Fiber optic sensors are arranged in a ring around the inner wall of the LNG storage tank to collect real-time data on the tank's vertical differential settlement level, contraction displacement, and circumferential rotation angle. Settlement deformation, vertical differential settlement level, and contraction displacement are all structural deformation data.

[0050] Distributed fiber optic temperature sensors are deployed on the inner wall of the LNG storage tank to collect temperature field data in real time.

[0051] An electric tilt meter is installed on the outer wall of the LNG storage tank to collect the tilt angle of the LNG storage tank in real time.

[0052] As an optional implementation, the LNG storage tank structural health monitoring device provided in this application may further include: a maintenance decision system. This system is used to visually display maintenance decisions. The maintenance decisions are generated based on abnormal monitoring information. The abnormal monitoring information is generated using monitoring values ​​that are greater than the baseline threshold determined by the dynamic baseline of the pre-cooling condition.

[0053] In an exemplary embodiment of this application, based on the system architecture provided above, the specific implementation process of the LNG storage tank structural health monitoring method provided in this application is described. This specific implementation process includes: S1: Deploy a layered sensor network. Based on the structural characteristics of the double-walled annular space and concrete bottom plate of the storage tank, install vibrating wire strain gauges on the bottom plate, install fiber optic sensors in the annular space, arrange distributed fiber optic temperature sensors on the outer side of the inner wall, and install electric tilting meters on the outer wall. S2: Perform multimodal data synchronous acquisition. Integrate data collected by various sensors deployed in S1 using a unified timestamp. Real-time acquisition of strain values, temperature field distribution, settlement, and tilt angle during pre-cooling and operation periods. Transmit these data to the processing center to provide a time-aligned dataset for pre-cooling calibration in step S3 and multi-dimensional monitoring in step S4. The processing center includes a baseline calibration module, a three-dimensional coupling module, a monitoring value determination module, and a monitoring result generation module.

[0054] S3: Calibrate the strain displacement monitoring baseline. During the pre-cooling stage of production, based on the -169℃ temperature field data collected in step S2, and combined with the inner wall shrinkage characteristics, the strain displacement monitoring baseline is dynamically corrected to eliminate the initial low temperature deviation. This calibration result will be used as the benchmark threshold for monitoring during the operation period in step S4 below.

[0055] S4: Perform multi-dimensional coupling monitoring in the annular space. Based on the strain displacement monitoring baseline calibrated in step S3 (i.e., the dynamic baseline of the pre-cooling condition), use fiber optic sensors to synchronously monitor vertical differential settlement, horizontal contraction displacement and circumferential rotation angle. Use a three-dimensional motion coupling model to distinguish between normal deformation and abnormal structural displacement caused by LNG weight, and transmit the analysis results (i.e., three-dimensional coupling data) to step S5 below in real time.

[0056] The raw data collected by the fiber optic sensors installed in the annular space also needs to undergo the following processing: (1) Spatiotemporal alignment: Kalman filtering (the noise covariance of the state equation is diag[0.01, 0.01, 0.001]) is used to eliminate spatiotemporal deviation and ensure data synchronization of multiple sensors.

[0057] (2) Feature extraction: Extract the abrupt features of each parameter (such as displacement abruptness and angle anomaly) through db4 wavelet basis 3-level decomposition.

[0058] (3) Data fusion: The conflict data are fused using the improved DS evidence theory (the weighted average rule is enabled when the conflict coefficient k≤0.3) to finally obtain the monitoring values ​​of vertical differential settlement, horizontal contraction displacement and circumferential rotation angle.

[0059] S5: Receive the three-dimensional coupling data from step S4 and the settlement and temperature field data of the base plate from step S2, and use the feature layer fusion method to associate the strain, temperature and displacement parameters. When the monitored value exceeds the baseline threshold determined based on the dynamic baseline of the pre-cooling condition in step S3, an early warning is triggered, and the abnormal information is pushed to the following step S6.

[0060] S6: Based on the early warning results and long-term monitoring trends in step S5, optimize the sensor placement scheme of the hierarchical sensor network in step S1, adjust the maintenance cycle, and simultaneously feed back historical data to step S3 to achieve dynamic iteration of the LNG storage tank structural health monitoring device. The long-term monitoring trend refers to analyzing the parameter change patterns (such as the slope of settlement over time and the long-term distribution of the temperature field) by accumulating historical data from steps S2 to S5, and combining this with early warning records to summarize the long-term evolution trend of structural health (such as whether settlement continues to increase and whether the temperature gradient is stable).

[0061] Furthermore, in practical applications, based on the early warning results and long-term monitoring trends of step S5, the sensor deployment scheme of the hierarchical sensor network in step S1 is optimized in the following ways: (1) Density adjustment: Increase sensor density in areas where early warnings are frequently triggered (such as repeated anomalies at a certain ring monitoring point); reduce the number of points in stable areas without anomalies.

[0062] (2) Position adjustment: For structural weak points found in long-term trends (such as the rapid settlement of a certain grid on the base plate), move the sensor to the key position.

[0063] (3) Type optimization: If the failure rate of sensors in a certain area is high (such as the instability of vibrating wire strain gauges in humid environments), replace them with a more suitable type (such as fiber optic grating sensors).

[0064] (4) Backup and supplement: Add backup sensors to core monitoring points (such as the 120° node in the ring space) to improve data reliability.

[0065] In a preferred embodiment, in step S1 above, a physical quantity sensing layer can be constructed through a multimodal hierarchical sensor network. A set of three-dimensional displacement sensors (i.e., fiber optic sensors arranged in a ring on the inner wall of the LNG storage tank, with a measurement range of ±50mm and an accuracy of 0.01mm) is deployed every 120° circumferentially in the annular space. Fiber optic grating settlement sensors (range 0-100mm, resolution 0.05mm) and PT1000 temperature sensors (measurement range -200℃~85℃, accuracy ±0.5℃) are arranged in a 2m×2m grid on the bottom plate. These are then combined with vibrating wire strain gauges for complementary verification. The sampling frequency of all sensors is uniformly set to 1Hz. The operation method adopts a hierarchical deployment strategy. The outer wall sensors are integrated inside the tank's insulation layer, and the bottom plate sensors are rigidly connected to the concrete structure through pre-embedded sleeves. After deployment, a 72-hour static calibration is performed, and the sensor linearity (nonlinear error ≤0.5%FS) is verified using a standard displacement stage and a constant temperature bath. Based on this description, the location and function of each sensor are shown in Table 1.

[0066] Table 1 Sensor Installation Locations and Functions

[0067] Based on the above description, compared with the traditional single-sensor location deployment, the advantages of the layered strategy adopted in this application are at least as follows: 1. Adaptable to structural features: For different structures such as double-walled annular spaces and concrete base slabs, sensors can be arranged in layers (e.g., external wall sensors are integrated inside the insulation layer, and base slab sensors are pre-embedded with rigid connections) to improve monitoring accuracy.

[0068] 2. Stability and maintainability: The base plate sensor is rigidly connected to the concrete through a pre-embedded sleeve, making it difficult to loosen; the outer wall sensor is integrated into the insulation layer, protecting the sensor from external damage and facilitating maintenance.

[0069] 3. Enables multi-dimensional collaboration: After layering, sensors at different locations perform their respective functions, and data is collaboratively integrated, avoiding the limitations of single deployment and realizing the perception of multiple physical quantities at all times.

[0070] 4. Convenient calibration: Static calibration is performed 72 hours after deployment (linearity is verified using a standard displacement stage and constant temperature bath) to ensure data reliability.

[0071] In a preferred embodiment, in step S2 above, the settlement and temperature field data acquisition stage utilizes distributed acquisition units (i.e., vibrating wire strain gauges, fiber optic settlement sensors, and distributed fiber optic temperature sensors) to achieve real-time data acquisition. The settlement data sampling cycle is conventionally set to 10 minutes / time, automatically switching to high-frequency acquisition of 1 minute / time when an anomaly is triggered. The temperature field data adopts a differential acquisition mode, with a temperature difference resolution of 0.1℃ between adjacent sensors. Data is transmitted via industrial Ethernet at a transmission rate of 100Mbps and a packet loss rate controlled within 0.1%. Furthermore, the device provided in this application can employ time-division multiplexing technology to aggregate 32 sensor signals via a 485 bus to an edge computing gateway. The gateway has a built-in digital filtering algorithm with a cutoff frequency of 0.1Hz to eliminate high-frequency noise and synchronously generates standardized data frames with timestamps, conforming to the ISO15746-2 standard. This phase requires a stable power supply voltage of DC24V±10%, a data storage capacity of no less than 1TB to support local caching of 7 days of data, a gateway operating temperature range of -30℃ to 70℃, and compliance with the electromagnetic interference immunity requirements of IEC 61000-6-2 standard.

[0072] A stable DC24V±10% voltage ensures continuous normal operation of the sensors and gateway, preventing data acquisition interruptions or errors due to voltage fluctuations and guaranteeing monitoring continuity. A 1TB local cache supports 7 days of data storage, preventing the loss of critical data due to network interruptions and preserving historical data for subsequent baseline calibration and trend analysis. An operating range of -30℃ to 70℃ adapts to the extreme environments of LNG tank pre-cooling (low temperature) and operation (normal temperature), ensuring reliable gateway operation under harsh conditions. Compliance with IEC 61000-6-2 standard (industrial environment electromagnetic interference immunity) avoids the impact of electromagnetic radiation from surrounding equipment on data transmission accuracy and reduces noise interference.

[0073] In a preferred embodiment, in step S3 above, during the pre-cooling stage, one set of baseline samples is collected every hour for 96 hours. During normal operation, the threshold is dynamically corrected three times daily. The operating method constructs segmented baseline thresholds by comparing the changes in structural parameters before and after pre-cooling: the temperature threshold during the pre-cooling stage is set to -165℃ to -160℃, and the sedimentation threshold is dynamically adjusted downwards as the temperature decreases according to a linear function, with the sedimentation threshold decreasing by 0.8 mm for every 10℃ decrease in temperature; during normal operation, the threshold is locked at 1.2 times the average value of the 24 hours after the pre-cooling stabilization period. The cooling rate during the pre-cooling stage is ≤5℃ / h, the data validity rate is ≥98%, and baseline calibration must be performed when there is no liquid inflow or outflow from the storage tank. The baseline sample includes multi-dimensional monitoring data collected hourly during the pre-cooling stage, specifically: temperature field data (such as the temperature distribution around -169℃ collected by the PT1000 temperature sensor); strain values ​​(strain of the base plate settlement deformation monitored by the vibrating wire strain gauge); settlement amount (settlement amount of the base plate monitored by the fiber optic grating settlement sensor); tilt angle (tilt angle of the outer wall monitored by the electric tilt meter); vertical / horizontal displacement and circumferential rotation angle of the annular space (data from intrinsically safe fiber optic sensors).

[0074] In a preferred embodiment, in step S4 above, the vertical differential settlement, horizontal contraction displacement, and circumferential rotation angle are simultaneously acquired during the three-dimensional coupled data fusion processing stage. The vertical differential settlement measurement accuracy is ±0.1mm, the horizontal contraction displacement range is ±100mm, and the circumferential rotation angle resolution is 0.01° with a measurement range of ±5°. The data fusion cycle is synchronized with the data acquisition stage, i.e., 10 minutes / time for normal operation and 1 minute / time for abnormal operation. The system uses Kalman filtering for spatiotemporal alignment, and the noise covariance of the state equation is set to a diagonal matrix diag[0.01, 0.01, 0.001]. A three-level decomposition using the db4 wavelet basis is performed to extract the abrupt change features of each parameter. Then, the improved DS evidence theory is used to fuse conflicting data. When the conflict coefficient k ≤ 0.3, a weighted average rule is activated. The entire fusion process must ensure that the time synchronization error between the three-dimensional sensor and the base plate sensor is ≤10ms, the data transmission delay is ≤500ms, and the processing time for a single frame of data at the edge is controlled within 2 seconds.

[0075] In practical applications, Kalman filtering is used to synchronize the data (vertical differential settlement, horizontal contraction displacement, and circumferential rotation angle) collected by fiber optic sensors in a ring space in time and space. For example, the noise covariance matrix diag[0.01, 0.01, 0.001] of the state equation represents the noise variances of the three parameters as 0.01, 0.01, and 0.001, respectively. By using filtering algorithms to reduce the impact of noise on the data, the temporal and spatial alignment of data from different sensors is achieved.

[0076] For the spatiotemporally aligned data, a three-level decomposition is performed using the db4 wavelet basis—the signal is decomposed into approximate coefficients (low-frequency trends) and detail coefficients (high-frequency abrupt changes), and the abrupt change features (such as sudden increase in displacement and abnormal fluctuation of angle) in the three-level detail coefficients are extracted.

[0077] The improved DS evidence theory is used to fuse the multi-parameter features after three-level decomposition. First, the conflict coefficient k between each piece of evidence is calculated. If k ≤ 0.3, the weighted average rule is used to fuse the data. If k > 0.3, the probability of the conflicting parts is assigned to the "whole set" (i.e., the state with unclear attribution) to obtain three-dimensional coupled data to avoid distortion of the fusion result due to excessive conflict.

[0078] Furthermore, the reliability of sensor data can be re-examined (e.g., whether there are sensor malfunctions or data transmission errors), and the conflict coefficient can be recalculated after excluding abnormal data; the weight of data with large conflict coefficients can be reduced, and data with high reliability (e.g., parameter data with stable baselines) can be fused first.

[0079] In a preferred embodiment, in step S5 above, the three-dimensional coupled data (vertical differential settlement Δh, horizontal contraction displacement Δl, circumferential rotation angle θ) transmitted in S4 and the bottom plate settlement s and temperature field data T collected in S2 are first spatiotemporally aligned to eliminate sampling delays from different sensors. Then, the temporal characteristics of each parameter (such as the rate of change of the maximum value and mean within a 30-minute sliding window) are extracted, and the physical quantities are converted into dimensionless indices through feature mapping. For example, the temperature data T is normalized to the interval [-1, 1] using a piecewise linear function, where -169℃ corresponds to the minimum value -1 and the external wall ambient temperature corresponds to the maximum value 1.

[0080] In a preferred embodiment, in step S5 above, during the feature fusion stage, a weighted Euclidean distance model is used to construct a comprehensive health index, mapping strain, temperature, and displacement parameters to a unified feature space. The comprehensive health index must simultaneously meet two conditions: first, the weights of each parameter are dynamically adjusted according to the baseline threshold set in step S3; for example, the temperature weight is increased by 30% during the pre-cooling stage, and the settlement weight is increased by 50% during normal operation; second, a coupling coefficient is introduced to correct the synergistic effect of multiple parameters; for example, when the horizontal contraction displacement Δl and the circumferential rotation angle θ simultaneously exceed the threshold, the weight of their cross term is increased by 20% to amplify abnormal features.

[0081] In a preferred embodiment, in step S5 above, when the comprehensive health index exceeds the warning threshold, the system automatically generates warning information including the location of abnormal parameters (e.g., one set of fiber optic sensors is deployed every 120° around the circumference of the annular space, and the sensor number corresponding to the abnormal data (e.g., No. 3) is the No. 3 monitoring point in the annular space), the type of abnormality (e.g., the temperature field data shows low temperature (close to -169℃), and the circumferential rotation angle θ exceeds the threshold μθ=0.2°, which is determined to be excessive rotation caused by low temperature contraction), and the development trend (e.g., take the displacement monitoring values ​​of adjacent 5 minutes, calculate the difference and divide it by the set time (5 minutes) to obtain the displacement growth rate of 0.2mm / min within 5 minutes). The warning information is pushed to the maintenance decision system in S6 in real time through the Modbus protocol, and at the same time, a local audible and visual alarm is triggered. The formula for calculating the comprehensive health index is: ; In the formula, CHI represents the Comprehensive Health Index, a dimensionless index, and its threshold range is set by the baseline calibration results in step S3 (CHI ≤ 1.2 under normal operating conditions). n represents the number of physical parameter types involved in the health assessment. Based on the monitoring data from steps S4 and S2, n=5 in a typical scenario, corresponding to the following parameters: vertical differential settlement Δh, horizontal contraction displacement Δh... l Circumferential rotation angle θ, bottom plate settlement s, temperature field data T; This represents the dynamic weight of the i-th parameter, satisfying... ,in Temperature weighting increases exponentially with the degree of deviation from precooling temperature; This represents the real-time monitoring value of the i-th parameter; This represents the baseline threshold of the i-th parameter (dynamically calibrated in step S3, such as during normal operation, μ). Δh =5mm); This represents the real-time monitoring value of the j-th parameter, and... Parameters of the same type; This represents the baseline threshold for the j-th parameter; This represents the coupling coefficient, with a value range of [0, 1]. It is defined as the condition where any two parameters simultaneously exceed a threshold. =0.8 otherwise =0.2; i, j represent summation indices (i≠j), representing combinations of different parameters, such as i=Δh, j=θ, used to calculate the synergistic effect of different parameter anomalies in cross terms.

[0082] The warning trigger conditions are: When CHI > 1.2 or any parameter satisfies ; An alert is triggered when this parameter reaches the standard deviation of the baseline period in step S3. For example, if Δh = 7 mm (μ) is detected at a certain moment... Δh =5mm, σ Δh =0.8mm) θ=0.3° (μ θ =0.2°, σ θ =0.05°), at this time |Δh−μ Δh |=2.5σ Δh The CHI calculation result is 1.35, which meets the dual early warning conditions. The abnormal information is immediately pushed to step S6.

[0083] In a preferred embodiment, in step S6 above, the maintenance decision system is built on an industrial control server, configured with an 8-core CPU, 32GB of memory, and redundant power supply. The decision software, developed using a B / S architecture, supports 100 concurrent users, and the early warning information push delay is controlled within 10 seconds. This system obtains the coordinates, parameter types, and trend slopes of abnormal points by parsing Modbus protocol data frames, and automatically matches them with a maintenance knowledge base covering 23 typical fault handling procedures.

[0084] In a preferred embodiment, step S6 above generates a three-level response plan: Level 1 warning (CHI 1.2~1.5) triggers planned inspections; Level 2 warning (CHI 1.5~2.0) initiates remote intervention, such as adjusting the insulation layer temperature; and Level 3 warning (CHI>2.0) automatically generates a shutdown maintenance work order. The device provided in this application must operate year-round without interruption, with a mean time between failures (MTBF) greater than or equal to 8760 hours. It can seamlessly integrate with enterprise ERP systems and supports SAP / Oracle interfaces. The maintenance decision system can generate decisions that strictly comply with the GB 50493-2019 safety standards.

[0085] In practical applications, if a certain area (such as monitoring point 3 in the annular space) frequently triggers an alarm, it indicates that the area is a structural weak point, requiring increased sensor density or adjustment of sensor positions to enhance monitoring. If certain areas remain stable and anomalies-free for a long period, the number of sensors can be reduced to lower costs. Based on the type of anomaly (such as excessive rotation caused by low-temperature contraction), the placement of temperature or displacement sensors can be optimized at corresponding structural locations (such as the outer side of the inner wall). In short, alarm information and long-term trends are the core inputs for optimizing the sensor placement scheme in step S6, enabling dynamic iteration of the monitoring device.

[0086] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0087] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0088] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0089] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (RRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0090] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0091] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0092] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for monitoring the structural health of an LNG storage tank, characterized in that, include: A hierarchical sensor network is used to collect data on the pre-cooling period and the operation period of the LNG storage tank; both the pre-cooling period data and the operation period data include temperature field data, structural deformation data and tilt angle. The pre-cooling period data and the operation period data are integrated using a unified timestamp to obtain monitoring data; Obtain the strain-displacement monitoring baseline, and based on the temperature field data, calibrate the strain-displacement monitoring baseline in real time in conjunction with the shrinkage characteristics of the inner wall of the LNG storage tank to obtain the dynamic baseline of the pre-cooling condition. Based on the dynamic baseline of the pre-cooling condition, the abnormal structural displacements of the structural deformation data are distinguished by a three-dimensional motion coupling model to obtain three-dimensional coupling data; abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank. Based on the three-dimensional coupling data, the temperature field data, and the structural deformation data, strain-temperature-displacement parameters are correlated to obtain monitored values; the strain-temperature-displacement parameters are determined based on the monitored data. Health monitoring results are generated based on the monitored values ​​and the dynamic baseline of the pre-cooling condition.

2. The LNG storage tank structural health monitoring method according to claim 1, characterized in that, Based on the temperature field data, and combined with the shrinkage characteristics of the LNG storage tank inner wall, the strain displacement monitoring baseline is calibrated in real time to obtain the dynamic baseline for the pre-cooling condition, including: Based on the set temperature field in the precooling period data, the strain displacement of the inner wall of the LNG storage tank caused by contraction is determined, and the structural deformation data corresponding to the strain displacement is screened out to obtain the compensated structural deformation data. Based on the compensated structural deformation data, the strain displacement monitoring baseline is calibrated in real time to obtain the dynamic baseline of the precooling condition.

3. The LNG storage tank structural health monitoring method according to claim 1, characterized in that, The construction process of the three-dimensional motion coupling model includes: Obtain the circumferential rotation angle of the LNG storage tank; Based on the structural deformation data, determine the vertical differential settlement and horizontal shrinkage displacement; Establish the coupling relationship between the vertical differential settlement, the horizontal contraction displacement, and the circumferential rotation angle and the weight load of the LNG storage tank, and use the coupling relationship as the three-dimensional motion coupling model.

4. The LNG storage tank structural health monitoring method according to claim 1, characterized in that, Based on the monitored values ​​and the dynamic baseline of the pre-cooling condition, health monitoring results are generated, including: The baseline threshold is determined based on the dynamic baseline of the precooling condition. Determine whether the monitored value is greater than the baseline threshold; When the monitored value is greater than the baseline threshold, the health monitoring result is abnormal, and abnormal monitoring information is generated based on the monitored value; Maintenance decisions are generated based on the aforementioned anomaly monitoring information; When the monitored value is less than or equal to the baseline threshold, the health monitoring result is normal.

5. The LNG storage tank structural health monitoring method according to claim 4, characterized in that, The baseline thresholds include temperature thresholds and sedimentation thresholds; Determining the baseline threshold based on the pre-cooling condition dynamic baseline includes: In the pre-cooling stage of the dynamic baseline of the pre-cooling condition, the temperature threshold is set to -165℃ to -160℃, and the sedimentation threshold is dynamically adjusted downward as the temperature decreases according to a linear function and a set rule. In the operational phase of the pre-cooling dynamic baseline, both the temperature threshold and the settlement threshold are set multiples of the average temperature and settlement values ​​over the 24 hours following the pre-cooling stabilization period.

6. An LNG storage tank structural health monitoring device, characterized in that, include: A hierarchical sensor network is used to collect data on the pre-cooling period and the operation period of LNG storage tanks; both the pre-cooling period data and the operation period data include temperature field data, structural deformation data, and tilt angle. The data synchronization and acquisition module is used to integrate the pre-cooling period data and the operation period data using a unified timestamp to obtain monitoring data. The baseline calibration module is used to acquire the strain-displacement monitoring baseline and, based on the temperature field data and the shrinkage characteristics of the inner wall of the LNG storage tank, calibrate the strain-displacement monitoring baseline in real time to obtain the dynamic baseline for the pre-cooling condition. The three-dimensional coupling module is used to distinguish abnormal structural displacements in the structural deformation data based on the dynamic baseline of the pre-cooling condition through a three-dimensional motion coupling model, thereby obtaining three-dimensional coupling data; abnormal structural displacements refer to data other than structural deformations caused by the weight of the LNG storage tank. The monitoring value determination module is used to obtain the monitoring value by associating strain-temperature-displacement parameters with the three-dimensional coupling data, the temperature field data, and the structural deformation data; the strain-temperature-displacement parameters are determined based on the monitoring data. The monitoring result generation module is used to generate health monitoring results based on the monitoring values ​​and the dynamic baseline of the pre-cooling condition.

7. The LNG storage tank structural health monitoring device according to claim 6, characterized in that, The hierarchical sensor network includes: Vibrating wire strain gauges are installed on the bottom plate of the LNG storage tank to collect the settlement deformation of the LNG storage tank in real time. Fiber optic sensors are arranged in a ring on the inner wall of the LNG storage tank to collect the vertical differential settlement level, contraction displacement and circumferential rotation angle of the LNG storage tank in real time; settlement deformation and vertical differential settlement level and contraction displacement are all part of the structural deformation data. Distributed fiber optic temperature sensors are installed on the inner wall of the LNG storage tank to collect temperature field data in real time. An electric tilt meter is installed on the outer wall of the LNG storage tank to collect the tilt angle of the LNG storage tank in real time.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the LNG storage tank structural health monitoring method according to any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the LNG storage tank structural health monitoring method according to any one of claims 1-5.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the LNG storage tank structural health monitoring method according to any one of claims 1-5.

Images