An online monitoring system for the vacuum level of the annular space of an LNG storage tank
By constructing a hierarchical sensing area and distributed fiber optic sensing units, the vacuum degree, thermal response, and dynamic deformation data of the annular space of the LNG storage tank are acquired simultaneously, and the trend of hidden cold bridge formation is identified. This solves the problem that existing technologies cannot identify the expansion of cold bridges in advance, and improves the accuracy of monitoring and early warning capabilities.
Patent Information
- Application Number
- CN202611119662.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-27
- Publication Date
- 2026-11-10
AI Technical Summary
Existing methods for monitoring the vacuum level of the annular space in LNG storage tanks are insufficient to accurately identify the trend of hidden cold bridge formation, cannot continuously assess the cold bridge expansion process and the risk of insulation failure, and are easily affected by local operating condition fluctuations and low temperature drift.
A hierarchical sensing area is constructed, and distributed optical fiber sensing units and micro-deformation sensing points are set up along the metal heat conduction path to simultaneously acquire vacuum data, thermal response data, and dynamic deformation data of the insulation layer. Through the thermal migration feature extraction module, structural coupling analysis module, temporal correlation identification module, and cold bridge trend determination module, the trend of hidden cold bridge formation and risk assessment are realized.
It can identify local heat transfer anomalies before the vacuum level deteriorates significantly, reduce the risk of insulation performance degradation, and improve the early warning reliability of LNG storage tanks under long-term operation.
Smart Images

Figure CN123065693A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of online monitoring technology for LNG storage tanks, and specifically to an online monitoring system for the vacuum degree of the annular space of an LNG storage tank. Background Technology
[0002] LNG storage tanks typically employ a double-walled structure, forming an annular vacuum insulation space between the inner and outer tanks. This reduces the entry of external heat into the tank, ensuring stable storage of liquefied natural gas in cryogenic environments. The vacuum state within this annular space directly affects the tank's insulation performance and operational safety, thus requiring long-term online monitoring of its vacuum level. Current technologies typically involve setting up vacuum detection points, temperature detection points, and pressure acquisition structures to acquire vacuum parameters in real time within the annular space, and using changes in vacuum to determine if leaks or a decline in insulation performance are present.
[0003] However, most existing online monitoring methods rely solely on single vacuum values, local temperature values, or simple threshold comparisons for judgment. Their monitoring results are easily affected by local operating condition fluctuations, low-temperature drift, and differences in the distribution of measuring points, making it difficult to accurately identify early, latent anomalies within the annular space. Especially under multi-point online monitoring conditions, metal wires, probes, and support paths can easily form localized heat conduction channels within the vacuum insulation layer, causing non-uniform heat migration in localized areas. This leads to changes in thermal stress within the insulation layer, interlayer displacement, and localized structural disturbances. While such anomalies typically do not cause a rapid decrease in vacuum in their initial stages, they lead to a gradual deterioration of the insulation layer's performance, eventually inducing problems such as localized vacuum loss, increased heat leakage, and even low-temperature stress concentration within the inner tank.
[0004] Existing technologies can usually only identify vacuum anomalies after they have deteriorated significantly. They lack the ability to dynamically analyze the coupling relationship between thermal migration behavior and insulation layer structural response, cannot identify the trend of hidden cold bridge formation in advance, and are difficult to continuously assess the cold bridge expansion process and insulation failure risk. Summary of the Invention
[0005] The purpose of this invention is to provide an online monitoring system for the vacuum level of the annular space of an LNG storage tank, in order to address the shortcomings of the prior art.
[0006] To achieve the above objectives, the present invention provides the following technical solution: an online monitoring system for the vacuum level of the annular space of an LNG storage tank, comprising:
[0007] The layered sensing and acquisition module constructs a layered sensing area in the annular space between the inner and outer tanks of the LNG storage tank. Distributed optical fiber sensing units are set along the metal heat conduction path, and micro-deformation sensing points are set at different depths of the insulation layer to simultaneously acquire vacuum data, thermal response data, and dynamic deformation data of the insulation layer.
[0008] The thermal migration feature extraction module extracts the heat transfer hysteresis features and thermal diffusion shift features of the metal path at different time stages based on the acquired thermal response data, and generates the corresponding non-uniform thermal migration state information.
[0009] The structural coupling analysis module identifies microscale compression changes, interlayer displacement changes, and stress transmission changes caused by local thermal disturbances within the insulation layer based on the dynamic deformation data, thus forming structural coupling state information of the insulation layer.
[0010] The temporal correlation identification module performs temporal correlation analysis on the non-uniform thermal migration state information and the insulation layer structure coupling state information, establishes the synchronous evolution relationship between thermal migration behavior and insulation layer structure response, and identifies spatial clustering areas of local heat transfer anomalies.
[0011] The cold bridge trend determination module determines whether there is a hidden cold bridge formation trend in the annular space based on the vacuum change rate, thermal response continuity, and structural response stability corresponding to the spatial accumulation area.
[0012] The risk tracking output module continuously tracks the status of the corresponding area when it determines that there is a tendency for the formation of hidden cold bridges. It also outputs the cold bridge development level and vacuum insulation failure risk results by combining the characteristics of thermal migration direction change and structural response expansion characteristics.
[0013] Preferably, when constructing the layered sensing region, an inner sensing layer, a middle sensing layer, and an outer sensing layer are formed radially along the annular space, and each sensing layer is arranged correspondingly along the extension direction of the metal heat conduction path to form a spatial correspondence between vacuum data, thermal response data, and dynamic deformation data of the insulation layer.
[0014] Preferably, when setting up distributed optical fiber sensing units, the distributed optical fiber sensing units are laid parallel to both sides of the metal heat conduction path and arranged radially to correspond with the micro-deformation sensing points, so that the thermal response data and the dynamic deformation data of the insulation layer at the same spatial location can be synchronously correlated.
[0015] Preferably, when extracting heat transfer hysteresis features, the continuously acquired thermal response data are arranged according to time stages to form a thermal response change sequence, and the thermal response delay change state between adjacent regions is identified based on the order in which adjacent regions reach the effective thermal response state.
[0016] Preferably, when extracting thermal diffusion offset features, the corresponding areas on both sides of the metal path are divided into continuous thermal diffusion segments, and the center line of the metal thermal conduction path is used as a comparison benchmark to perform corresponding analysis on the order of thermal response propagation on both sides in order to determine the deviation state of the thermal diffusion direction.
[0017] Preferably, based on the thermal response delay change state, the degree of offset of the thermal diffusion direction on both sides of the metal path is compared in segments to determine the asymmetric diffusion region existing in the local heat transfer process; the thermal response continuity of the asymmetric diffusion region is tracked in the continuous time stage to identify the dynamic offset state of the thermal diffusion boundary and the heat accumulation change state; based on the dynamic offset state and the heat accumulation change state, corresponding non-uniform thermal migration state information is generated.
[0018] Preferably, when forming the coupling state information of the insulation layer structure, the local deformation propagation path is determined based on the dynamic deformation data at different depths of the insulation layer, and the interlayer displacement offset region is identified based on the order of deformation response between adjacent depth regions.
[0019] Preferably, identifying spatially concentrated areas of localized heat transfer anomalies includes:
[0020] The non-uniform thermal migration state information and the insulation layer structure coupling state information are synchronously sorted according to the corresponding time stages to form a corresponding change sequence of thermal response and structural response.
[0021] Based on the corresponding change sequence, the continuous correspondence between the thermal migration direction change state and the structural response expansion state is compared stage by stage to identify the synchronous offset region between the thermal response change and the structural response change.
[0022] The spatial distribution changes of the synchronous offset region in a continuous time period are correlated and tracked to determine the extension continuity and aggregation continuity of local heat transfer anomalies inside the insulation layer.
[0023] Based on the aforementioned extensional continuity and aggregation continuity, a synchronous evolution relationship between thermal migration behavior and insulation layer structural response is established, and spatial aggregation regions of local heat transfer anomalies are identified.
[0024] Preferably, determining whether there is a tendency for the formation of hidden cold bridges within the annular space includes:
[0025] The vacuum change state of the spatial cluster region is tracked in a continuous time period to identify the slow-changing continuous region and the intermittent recovery region in the local vacuum fluctuation process.
[0026] Based on the gradual change continuous region and the intermittent recovery region, the continuity of the thermal response in the corresponding region is synchronously compared to determine the thermal response continuous region and the thermal response repeated region that exist in the process of heat transfer change.
[0027] By combining the thermal response continuation region and the thermal response repetition region, the stability of the insulation layer structure response is continuously verified, and the deformation accumulation offset region existing in the process of local structural changes is identified.
[0028] Based on the expansion state of the deformation accumulation offset region over a continuous time period, it is determined whether there is a tendency for the formation of hidden cold bridges within the annular space.
[0029] Preferably, the output results for cold bridge development level and vacuum insulation failure risk include:
[0030] Continuous state tracking is performed on the corresponding regions where a hidden cold bridge formation trend is identified, and the changes in the direction of thermal migration and the expansion of the structural response are obtained over a continuous time period.
[0031] Based on the change in the direction of thermal migration, the continuity of the local heat diffusion boundary is analyzed in stages to identify the expansion enhancement region and expansion stagnation region that exist during the thermal migration process.
[0032] By combining the extended enhancement region and the extended stagnation region, the spatial continuity of the structural response extension state within the corresponding region is simultaneously verified to determine the extension depth change state of the local structural disturbance.
[0033] Based on the corresponding relationship between the expansion depth change state and the heat diffusion boundary offset state, the cold bridge development level and vacuum insulation failure risk results are output.
[0034] The technical effects and advantages provided by the present invention in the above technical solution are as follows:
[0035] This invention differs from existing technologies that rely solely on single vacuum or local temperature values for anomaly detection. By constructing a layered sensing region distributed along the metal's heat conduction path within an annular space, and simultaneously acquiring thermal response data and dynamic deformation data of the insulation layer, it achieves synchronous correlation analysis between heat transfer behavior and changes in the insulation layer structure. Through continuous tracking of deviations in heat diffusion direction, abnormal diffusion interruption regions, and heat retention areas, it can identify localized asymmetric heat diffusion phenomena that are difficult to detect using traditional vacuum monitoring methods. This solves the problem of existing technologies being unable to identify the formation of hidden cold bridges in advance. Therefore, localized heat transfer anomalies can be detected before the vacuum level significantly deteriorates, reducing the risk of insulation performance degradation due to localized heat leakage.
[0036] This invention further establishes a synchronous evolutionary relationship between thermal migration behavior and structural response expansion by continuously superimposing and propagating dynamic deformation data from different depth regions, and continuously tracks spatial accumulation areas in conjunction with vacuum changes. Unlike existing technologies that treat temperature and structural changes independently, this invention can identify interlayer displacement shifts, stress transfer retention, and expansion depth changes caused by local thermal disturbances within the insulation layer, thereby achieving phased identification of the cold bridge development process. This not only improves the accuracy of locating hidden cold bridges but also outputs corresponding cold bridge development levels and vacuum insulation failure risk results, enhancing the reliability of early warning systems during long-term operation of LNG storage tanks. Attached Figure Description
[0037] 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 recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.
[0038] Figure 1 This is a flowchart of a module for an online monitoring system for the vacuum level of an annular space in an LNG storage tank according to the present invention. Detailed Implementation
[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.
[0040] For examples, please refer to Figure 1 As shown in this embodiment, an online monitoring system for the vacuum level of the annular space of an LNG storage tank includes:
[0041] The layered sensing and acquisition module constructs a layered sensing area in the annular space between the inner and outer tanks of the LNG storage tank. Distributed optical fiber sensing units are set along the metal heat conduction path, and micro-deformation sensing points are set at different depths of the insulation layer to simultaneously acquire vacuum data, thermal response data, and dynamic deformation data of the insulation layer.
[0042] Within the annular space between the inner and outer tanks of the LNG storage tank, three layered sensing zones are constructed radially: an inner sensing layer near the outer wall of the inner tank, a middle sensing layer located at half the thickness of the insulation layer, and an outer sensing layer near the inner wall of the outer tank. The inner sensing layer is 50 mm from the outer wall of the inner tank, the outer sensing layer is 50 mm from the inner wall of the outer tank, and the middle sensing layer is equidistant from both. Each layered sensing zone is arranged along the direction of the metal heat conduction path, which includes a short pipe leading out from the vacuum detection interface, a metal probe, an outer protective tube for the metal wire, and a support path.
[0043] When the distributed optical fiber sensing unit is set along the metal heat conduction path, the optical fiber sensing unit is kept 15 mm apart from the outer surface of the metal heat conduction path and is laid parallel to the metal heat conduction path. The axial spacing between two adjacent fiber optic grid areas is set to 500 mm.
[0044] One fiber optic sensing unit is installed on each side of each metal heat conduction path to make the thermal response data on both sides comparable. The fiber optic sensing units are fixed to the insulation layer frame or non-metallic support through low thermal conductivity fixing points with a spacing of 1000 mm to reduce the additional impact of the fixing structure on the thermal field of the annular space.
[0045] When micro-deformation sensing points are set at different depths of the insulation layer, the micro-deformation sensing points and the corresponding fiber optic grating areas are aligned radially to form thermal response data and dynamic deformation data of the insulation layer at the same spatial location. Three micro-deformation sensing points are set at each radial measurement section, corresponding to the inner sensing layer, the middle sensing layer and the outer sensing layer, respectively, to reflect the compression changes and interlayer displacement changes generated when the insulation layer diffuses outward from the metal heat conduction path.
[0046] The micro-deformation sensing point is installed using a pre-tensioning method, with the pre-tensioning amount set at 300 micro-strains, so that it still maintains effective response after low-temperature contraction.
[0047] Vacuum data is acquired via a ring-shaped vacuum detection interface and time-stamped along with thermal response data and insulation layer dynamic deformation data, using the same acquisition period set to 10 seconds. The thermal response data is converted as follows: using the initial wavelength of the fiber optic sensing unit as a reference value, the difference between the real-time wavelength and the reference value is taken as the wavelength change, and then converted to temperature change according to the relationship that every 1 degree Celsius increase corresponds to a 10 picometer wavelength change. The insulation layer dynamic deformation data is converted as follows: first, the wavelength change caused by the temperature compensation point within the same layer is deducted, and then the deformation change is converted according to the relationship that every 1 micro-strain corresponds to a 1.2 picometer wavelength change. After being time-stamped together, the vacuum data, thermal response data, and insulation layer dynamic deformation data form synchronously acquired data within the same layered sensing area.
[0048] The thermal migration feature extraction module extracts the heat transfer hysteresis features and thermal diffusion shift features of the metal path at different time stages based on the acquired thermal response data, and generates the corresponding non-uniform thermal migration state information.
[0049] After synchronously acquiring vacuum data, thermal response data, and dynamic deformation data of the insulation layer, the thermal response data continuously acquired along the metal heat conduction path is divided into time stages. The preferred time stage is 60 seconds. Within each time stage, the thermal response change at the corresponding location is obtained by subtracting the thermal response value at the start of the stage from the thermal response value at the end of the stage. Then, according to the extension order of the metal heat conduction path, the thermal response changes at adjacent locations on the same side are arranged sequentially to construct a thermal response change sequence. To avoid misidentification of minute noise in low-temperature environments, the effective judgment value for thermal response is set to 0.8 degrees Celsius. When there is a difference in the timing of adjacent regions reaching the 0.8-degree Celsius thermal response change, the thermal response delay between adjacent regions is obtained by subtracting the earlier time from the later time reaching the judgment value. A thermal response delay greater than 20 seconds is recorded as a thermal response delay state; a delay less than or equal to 20 seconds is recorded as a normal transmission state.
[0050] After obtaining the thermal response delay change state, the thermal response data of corresponding areas on both sides of the metal path are divided into segments according to a preset time sequence. The length of each continuous thermal diffusion segment along the metal heat conduction path is preferably 1500 mm, with a 500 mm overlap between adjacent continuous thermal diffusion segments to ensure continuous recording when the thermal diffusion boundary moves across segments. Segment division is based on paired thermal response data on both sides of the metal path. The first and second sides form corresponding thermal response change sequences within the same time period, with the center line of the metal heat conduction path used as a comparison benchmark. Within each continuous thermal diffusion segment, the positional sequence of the first side reaching a thermal response change of 0.8 degrees Celsius is first determined, followed by the positional sequence of the second side reaching the same thermal response change. When the positional sequences on both sides are the same and the time difference does not exceed 20 seconds, the thermal response propagation direction is considered consistent. When the positional sequences on both sides are different, or the time difference corresponding to the same position is greater than 20 seconds, a directional deviation is considered.
[0051] When analyzing directional deviations, the direction of heat diffusion from the metal's thermal conduction path into the insulation layer is used as the basic direction. If, within a continuous heat diffusion segment, two consecutive locations on the first side reach the valid thermal response value before the second side, and the next continuous heat diffusion segment maintains the same directional relationship, then the heat diffusion direction is considered to have shifted towards the second side; conversely, if two consecutive locations on the second side reach the valid thermal response value before the first side, and the next continuous heat diffusion segment maintains the same directional relationship, then the heat diffusion direction is considered to have shifted towards the first side. The directional deviation is not judged based on a single measurement point, but rather on the simultaneous fulfillment of two consecutive locations and two consecutive time periods, thereby reducing misjudgments caused by accidental measurement fluctuations.
[0052] After identifying the directional deviation, the consistency of the thermal diffusion transition between adjacent continuous thermal diffusion sections is checked segment by segment. The method for judging the consistency of the thermal diffusion transition is as follows: the direction of thermal response propagation at the end of the preceding continuous thermal diffusion section should be consistent with the direction of thermal response propagation at the beginning of the following continuous thermal diffusion section; if the two directions are opposite, and this state lasts for two time stages, it is identified as an abnormal diffusion interruption region. If there is no directional reversal between adjacent continuous thermal diffusion sections, but the change in thermal response at the same location continues to increase for three consecutive time stages, and the change in thermal response transmitted to adjacent locations does not reach 0.8 degrees Celsius, it is identified as a heat retention region. Abnormal diffusion interruption regions reflect local blockage of the thermal diffusion path, while heat retention regions reflect local accumulation of heat around the metal path; both are used to subsequently determine asymmetric diffusion regions.
[0053] Based on the spatial distribution of abnormal diffusion interruption regions and heat retention regions, asymmetric diffusion regions existing in the local heat transfer process are identified. The determination method is as follows: when an abnormal diffusion interruption region exists on one side of the same metal heat conduction path, and a heat retention region exists on the other side, and the axial distance between the two does not exceed 500 mm, this location is recorded as a candidate region for asymmetric diffusion; when a candidate region for asymmetric diffusion repeatedly appears in two consecutive time stages, it is confirmed as an asymmetric diffusion region. Through this method, asymmetric diffusion regions possess both spatial correspondence and temporal continuity, distinguishing them from single-point temperature fluctuations.
[0054] After determining the asymmetric diffusion region, the continuity of the thermal response of the asymmetric diffusion region in the continuous time stage is tracked.
[0055] The thermal diffusion boundary is determined by the location where the change in thermal response first reaches 0.8 degrees Celsius. When the location of the thermal diffusion boundary shifts relative to the previous time stage in a subsequent time stage, it is recorded as a thermal diffusion boundary shift. A dynamic shift is considered to exist when the thermal diffusion boundary shifts to the same side for three consecutive time stages. The heat accumulation change state is determined by the thermal response holding time; that is, if the temperature at the same location reaches 0.8 degrees Celsius and remains at or above that value for 180 seconds, and adjacent locations do not reach that value synchronously, a heat accumulation change state is considered to exist at that location.
[0056] Non-uniform thermal migration state information is formed by dynamic offset state, heat accumulation change state, and abnormal diffusion interruption region.
[0057] The degree of local heat transfer anomaly is represented using a weighted method, where the weight for dynamic offset state is set to 0.45, the weight for heat accumulation change state is set to 0.35, and the weight for abnormal diffusion interruption region is set to 0.20. During calculation, the value corresponding to dynamic offset state is determined by the number of consecutive unidirectional offset time stages, the value corresponding to heat accumulation change state is determined by the continuous holding time, and the value corresponding to abnormal diffusion interruption region is determined by the number of consecutive occurrences. These three values are multiplied by their respective weights and then summed to obtain the anomaly degree representation value in the non-uniform heat migration state information. An anomaly degree representation value reaching 0.60 is recorded as a local heat transfer anomaly; a value below 0.60 is recorded as heat transfer fluctuation.
[0058] It should be noted that the above values were verified through a low-temperature test section with an annular interlayer. The test section was 6 meters long, with a stable heat input of 2 watts applied to the outer end of the metal heat conduction path, and the interior of the insulation layer maintained an environment of -162 degrees Celsius. Using a 60-second time stage and a valid judgment value of 0.8 degrees Celsius for the thermal response, the thermal diffusion front near the metal path could be stably identified in the third time stage. When the judgment value was set to 0.5 degrees Celsius, the number of misjudgments due to low-temperature fluctuations was 7. When the judgment value was set to 1.2 degrees Celsius, the thermal diffusion boundary identification lagged by 2 time stages. Using a 1500 mm segment length and a 500 mm overlap length, the positioning deviation of the asymmetric diffusion region was 300 mm; after removing the overlap length, the positioning deviation increased to 900 mm. Using a weight combination of 0.45, 0.35, and 0.20, all three artificially set metal thermal conductivity enhancement locations in the experiment were identified, and normal temperature fluctuation areas were not judged as local heat transfer anomalies. Therefore, this set of parameters can be used to form information on non-uniform thermal migration states.
[0059] The structural coupling analysis module identifies microscale compression changes, interlayer displacement changes, and stress transmission changes caused by local thermal disturbances within the insulation layer based on the dynamic deformation data, thus forming structural coupling state information of the insulation layer.
[0060] After obtaining dynamic deformation data at different depths of the insulation layer, the average dynamic deformation during the stable operation phase is used as the baseline deformation value. The stable operation phase is preferably defined as a period where the tank liquid level change is less than 1 percentage point, the annular space vacuum data fluctuation is less than 0.5 Pa, and this period lasts for 24 hours. The deformation change of each depth region within the current time phase is obtained by subtracting the baseline deformation value from the current dynamic deformation value; the preferred time phase is 60 seconds. The effective deformation judgment value is set at 25 microstrains; when the deformation change reaches 25 microstrains, an effective deformation response is recorded for that depth region.
[0061] When tracking the deformation changes of each depth region over a continuous time period, the dynamic deformation data are first arranged radially according to the inner sensing layer, the middle sensing layer, and the outer sensing layer. Then, the time when each depth region first reaches the effective deformation judgment value is recorded. The local deformation propagation path is determined according to the order of effective deformation responses. When the inner sensing layer responds first, followed by the middle sensing layer, and lastly responds, it is recorded as a local deformation propagation path from the inside out. When the middle sensing layer shows an effective deformation response before the inner or outer sensing layer, it is recorded as an intra-layer offset propagation path.
[0062] Based on the local deformation propagation path, the sequential states of deformation responses between adjacent depth regions are compared layer by layer. During the comparison, the time difference between the first occurrence of 25 microstrain in adjacent depth regions is used as the interlayer response difference, with a preferred value of 40 seconds. When the interlayer response difference exceeds 40 seconds and the deformation change directions of adjacent depth regions are inconsistent, an interlayer displacement shift region is identified at the corresponding location. To avoid misjudgment due to a single low-temperature disturbance, the interlayer displacement shift region needs to maintain the same shift direction for two consecutive time stages.
[0063] After determining the inter-story displacement region, the dynamic deformation data of the inter-story displacement region at different time stages are continuously superimposed. The superposition method is as follows: for the same location within 5 consecutive time stages, when the deformation direction is the same, the deformation changes in each time stage are added together to form a cumulative deformation value; when the deformation direction is opposite, the deformation changes in the opposite direction are subtracted from the formed cumulative deformation value, thus obtaining the cumulative deformation change sequence of the corresponding region.
[0064] In this embodiment, the continuous superposition time is preferably 5 time stages, i.e. 300 seconds, to reflect whether the local compression change is continuous.
[0065] Based on the cumulative deformation sequence, the continuity of deformation propagation between adjacent regions is analyzed segment by segment. When the difference in cumulative deformation value between adjacent regions reaches 60 microstrains in the same time period, and the region with higher cumulative deformation value fails to effectively transmit to the adjacent region in the subsequent two time periods, a region where deformation propagation stagnates is identified.
[0066] In this embodiment, a deformation abrupt change region is identified when the deformation change at any location reaches 80 microstrains within a single time period, and no effective deformation response in the same direction is observed at adjacent locations. A deformation propagation stagnation region indicates that local compressional changes are impeded, while a deformation abrupt change region indicates that local compressional changes undergo discontinuous transitions.
[0067] Based on the deformation propagation stagnation region and deformation abrupt change region, the continuous change in the stress transmission direction inside the insulation layer is tracked accordingly. The tracking method is as follows: the stress transmission direction is taken from the position with high deformation accumulation value to the position with low deformation accumulation value; when the stress transmission direction reverses between two adjacent time stages, and the reversal position coincides with the deformation propagation stagnation region or deformation abrupt change region, an abnormal offset region is identified in the local stress diffusion process. When the axial movement distance of the abnormal offset region is less than 500 mm within three consecutive time stages, and the deformation accumulation value continues to increase by 90 microstrain, it is identified as a stress transmission stagnation region; when the abnormal offset region spans two adjacent depth regions within three consecutive time stages and extends axially to a distance of 1000 mm, it is identified as a stress diffusion anomaly region.
[0068] Based on the continuous spatial variation of stress transfer stagnation regions and stress diffusion anomaly regions, the coupling state information of the insulation layer structure is formed. The formation process uses a weighted representation method, where the weight for inter-layer displacement offset regions is 0.25, the weight for deformation propagation stagnation regions is 0.25, the weight for deformation abrupt change regions is 0.20, and the weight for stress diffusion anomaly regions is 0.30. During calculation, the frequency of each item within a continuous time period is normalized across five time periods, then multiplied by their corresponding weights and summed to obtain the structural coupling representation value. A structural coupling representation value reaching 0.65 is recorded as an insulation layer structural coupling anomaly; a value below 0.65 is recorded as an insulation layer structural fluctuation.
[0069] The temporal correlation identification module performs temporal correlation analysis on the non-uniform thermal migration state information and the insulation layer structure coupling state information, establishes the synchronous evolution relationship between thermal migration behavior and insulation layer structure response, and identifies spatial clustering areas of local heat transfer anomalies.
[0070] After acquiring the non-uniform thermal migration state information and the insulation layer structure coupling state information, they are synchronously sorted according to their corresponding time stages. The time stages are divided into 60-second intervals, with the end time of each time stage serving as the sorting benchmark. When the acquisition time difference between the two types of state information is no greater than 10 seconds, they are grouped into the same time stage. When the time difference is greater than 10 seconds, the average change trend of the previous and subsequent time stages is used to fill in the gaps, allowing the thermal response and structural response to be compared on the same time axis. After synchronous sorting, the non-uniform thermal migration state information, insulation layer structure coupling state information, thermal migration direction change state, and structural response expansion state within each time stage are arranged sequentially according to their spatial location, forming a corresponding change sequence of thermal response and structural response.
[0071] When comparing the corresponding change sequences of thermal response and structural response in stages, first read the abnormality level representation value in the non-uniform thermal migration state information, and then read the structural coupling representation value in the insulation layer structural coupling state information.
[0072] When the anomaly level value reaches 0.60 and the structural coupling value reaches 0.65 at the same spatial location, the location is recorded as the corresponding location of the thermal structure. When the axial distance between the spatial location pointed to by the thermal migration direction change state and the spatial location pointed to by the structural response expansion state is no greater than 500 mm and the radial depth difference is no more than one layered sensing area, the corresponding location of the thermal structure is recorded as a candidate location for synchronous offset.
[0073] When a candidate location for synchronous migration repeatedly appears within two consecutive time phases, and the change in thermal migration direction is in the same or adjacent direction as the expansion of structural response, it is identified as a synchronous migration region.
[0074] The identification of the synchronous offset region also includes time lag verification. Since the transmission of local thermal disturbances to the insulation structure response is typically delayed, a one-time-stage lag between the structural response extension state and the thermal migration direction change state is allowed. If the thermal migration direction change state occurs first in the current time stage, and the structural response extension state occurs in the next time stage at the same axial position or adjacent radial position, and the structural coupling representation value increases by 0.10 compared to the previous time stage, then a continuous correspondence is still considered. If the structural response extension state lags by more than two time stages, or the spatial position deviation exceeds 500 mm, it is not included in the synchronous offset region.
[0075] When tracking the spatial distribution changes of the synchronous offset region over consecutive time periods, the geometric center of the synchronous offset region is used as the tracking position. If the distance between the geometric centers in adjacent time periods is no more than 500 mm, and the direction of movement remains consistent with the previous time period, it is recorded as continuous expansion. When continuous expansion reaches three time periods, it is determined that the local heat transfer anomaly has continuous expansion within the insulation layer. If the distance between the geometric centers of the synchronous offset region is less than 200 mm, but the anomaly degree indicator and the structural coupling indicator both increase over three consecutive time periods, it is recorded as local aggregation. When local aggregation reaches three time periods, it is determined that the local heat transfer anomaly has aggregation continuity.
[0076] Extended continuity and aggregated continuity are weighted to form a synchronous evolution representation value. In the calculation, the weights for extended continuity and aggregated continuity are set to 0.35, for thermal structure time lag compliance to 0.20, and for spatial overlap to 0.10. Extended continuity is obtained by dividing the number of continuous extended time stages by 3, with an upper limit of 1; aggregated continuity is obtained by dividing the number of local aggregated time stages by 3, with an upper limit of 1; time lag compliance is 1 when lagging by 0 or 1 time stages, and 0.5 when lagging by 2 time stages; spatial overlap is 1 when the axial distance is no greater than 500 mm, and 0 when it is greater than 500 mm. Each value is multiplied by its corresponding weight and then summed to obtain the synchronous evolution representation value. When the synchronous evolution representation value reaches 0.70, a synchronous evolution relationship between thermal migration behavior and the insulation layer structural response is established; when it does not reach 0.70, it is recorded as a local asynchronous fluctuation of the thermal structure.
[0077] After the synchronous evolution relationship is established, spatial clusters of local heat transfer anomalies are identified based on the continuous spatial distribution of the synchronous offset regions. The identification method is as follows: around the same metal heat conduction path, at least two adjacent synchronous offset regions appear continuously along the axial direction, and the interval between adjacent synchronous offset regions is no greater than 500 mm. Simultaneously, at least one of the following conditions must be met: extended continuity or clustered continuity. This continuous region is then identified as a spatial cluster. If a spatial cluster spans two layered sensing regions radially and continuously covers a length of 1500 mm axially, it is recorded as a valid spatial cluster; if it appears only at a single point, it is not output as a spatial cluster.
[0078] The cold bridge trend determination module determines whether there is a hidden cold bridge formation trend in the annular space based on the vacuum change rate, thermal response continuity, and structural response stability corresponding to the spatial accumulation area.
[0079] After identifying spatial clusters of localized heat transfer anomalies, the vacuum change state of these clusters is continuously tracked over corresponding time periods. The expression for calculating the vacuum change rate is: In the formula, This indicates the rate of change of vacuum, measured in Pascals per minute. This represents the average pressure value of the vacuum data within the current time period; The average pressure value of the vacuum data within the previous time period is represented by Δt; Δt represents the time interval between two adjacent time periods, which is taken as 1 minute in this embodiment. The average pressure value of the vacuum data within the current time period is obtained by averaging all vacuum pressure data collected within that time period; the average pressure value of the vacuum data within the previous time period is obtained by averaging all vacuum pressure data collected within the previous time period. When the calculation result is positive, it indicates that the pressure in the annular space is increasing and the vacuum state is decreasing; when the calculation result is close to zero, it indicates that the vacuum state remains stable. An increase in pressure value represents a decrease in the vacuum state of the annular space. The preferred value for the vacuum change rate is 0.04 Pa per minute. When the vacuum change rate corresponding to the spatial accumulation region reaches 0.04 Pa per minute for four consecutive time periods, and the pressure change direction in each time period is increasing, this region is recorded as a slowly changing, continuous region.
[0080] If the vacuum rate of change corresponding to a spatial accumulation region increases and then decreases over four consecutive time periods, only to increase again over the next two time periods, and the pressure value after the decrease is not lower than the pressure value of the initial time period, then this region is recorded as an intermittent recovery region. The intermittent recovery region is not directly used as the result of cold bridge formation judgment, but rather as a correction basis for vacuum instability. The vacuum rate of change is obtained by dividing the number of consecutive increase time periods by 4, with a maximum value of 1. For each intermittent recovery, 0.20 is subtracted from the vacuum rate of change; if the result is less than 0, it is counted as 0.
[0081] After obtaining the gradually changing continuous region and the intermittent recovery region, the continuity of the thermal response within the corresponding regions is compared synchronously. During the comparison, the anomaly degree representation value in the non-uniform thermal migration state information at the same time stage is read. When the anomaly degree representation value is not lower than 0.60 for four consecutive time stages, and the direction of movement of the thermal diffusion boundary remains consistent between adjacent time stages, the region is determined as a continuous thermal response region. If the anomaly degree representation value repeatedly changes between above and below 0.60, and the axial distance between each newly reached position above 0.60 and the previous position is not greater than 500 mm, the region is determined as a repetitive thermal response region. The continuous thermal response representation value is obtained by dividing the number of consecutively satisfied time stages by 4, with a maximum value of 1; the repetitive thermal response representation value is obtained by dividing the number of repetitions by 2, with a maximum value of 1.
[0082] Subsequently, the structural stability of the insulation layer is continuously verified by combining the thermal response continuation region and the thermal response repetition region. During verification, the structural coupling representation value and the cumulative deformation change sequence in the insulation layer's structural coupling state information are read. When the structural coupling representation value is not lower than 0.65 for three consecutive time stages, and the direction of deformation increase is consistent in adjacent time stages in the cumulative deformation change sequence, it is recorded as a stable offset state of the structural response. If the axial distance between the location corresponding to the stable offset state of the structural response and the thermal response continuation region or the thermal response repetition region is not greater than 500 mm, and the radial position spans at least two layered sensing regions, it is identified as a cumulative deformation offset region.
[0083] When judging the expansion state of the deformation accumulation offset region over continuous time stages, the geometric center of the deformation accumulation offset region is used as the tracking position. When the geometric center moves 500 mm along the metal heat conduction path axially within adjacent time stages, and the direction of movement is consistent with the change in the direction of heat migration, it is recorded as an effective expansion. If three consecutive effective expansions occur, or if the deformation accumulation offset region remains in the same position for five time stages and the deformation accumulation value continues to increase, it is judged that there is a tendency for the formation of a hidden cold bridge in the annular space.
[0084] To facilitate the output of judgment results, a cold bridge trend representation value can be generated. The weight of the vacuum gradual change representation value is 0.25, the weight of the thermal response continuity representation value is 0.25, the weight of the thermal response repetition representation value is 0.15, and the weight of the deformation accumulation offset region expansion state is 0.35. During calculation, each representation value is multiplied by its corresponding weight and then summed to obtain the cold bridge trend representation value. When the cold bridge trend representation value reaches 0.72, it is judged that there is a hidden cold bridge formation trend; when it does not reach 0.72, it is recorded that there is local disturbance in the spatial cluster region but no hidden cold bridge trend has yet formed.
[0085] The risk tracking output module continuously tracks the status of the corresponding area when it determines that there is a tendency for the formation of hidden cold bridges. It also outputs the cold bridge development level and vacuum insulation failure risk results by combining the characteristics of thermal migration direction change and structural response expansion characteristics.
[0086] After determining the presence of a hidden cold bridge formation trend within the annular space, continuous state tracking was performed on the corresponding region. Continuous state tracking was conducted in 60-second time phases, with data recorded for at least 10 time phases. Within each time phase, the non-uniform thermal migration state information and the insulation layer structural coupling state information within the corresponding region were read, obtaining the thermal migration direction change state, thermal diffusion boundary location, structural response expansion state, and deformation accumulation offset region location, respectively. The thermal diffusion boundary location was determined when the thermal response change first reached 0.8 degrees Celsius, and the structural response expansion state was determined when the structural coupling representation value reached 0.65.
[0087] When analyzing the change in the direction of heat migration in stages, the geometric center position of the heat diffusion boundary in the current time stage is first recorded, and then compared with the geometric center position in the previous time stage. If the heat diffusion boundary moves 500 mm in the same direction between two adjacent time stages, and the anomaly index increases by 0.08, it is recorded as a valid boundary shift. When valid boundary shifts occur in three consecutive time stages, the corresponding region is identified as an expanding and enhanced region. If the heat diffusion boundary moves less than 200 mm in three consecutive time stages, but the anomaly index remains above 0.60, the corresponding region is identified as an expanding and stagnant region. Expanding and enhanced regions reflect the continued extension of the cold bridge heat transfer range, while expanding and stagnant regions reflect the continuous accumulation of heat in a local location.
[0088] After obtaining the extended enhancement region and the extended stagnation region, the spatial continuity of the structural response extension state within the corresponding region is synchronously verified.
[0089] The location of the cumulative deformation offset region in the inner, middle, and outer sensing layers is recorded. If the cumulative deformation offset region extends from the inner sensing layer to the middle sensing layer, and the structural coupling representation value increases by 0.07 in adjacent time stages, it is recorded as a first-level depth extension; if it further extends from the middle sensing layer to the outer sensing layer, and the axial continuous length reaches 1500 mm, it is recorded as a second-level depth extension; if the outer sensing layer forms a stable structural response offset state for three consecutive time stages, and the vacuum change rate reaches 0.04 Pa per minute, it is recorded as a third-level depth extension.
[0090] The depth variation of local structural disturbances is determined based on the number of depth expansions and their duration. In the calculation, the value for first-level depth expansion is 0.35, for second-level depth expansion it is 0.70, and for third-level depth expansion it is 1.00. If the same depth expansion state is maintained continuously for three time stages, the corresponding value is increased by 0.10; if the increased value exceeds 1.00, it is counted as 1.00. This value is used to represent the degree to which the structural response develops from a shallow local disturbance to a deeper disturbance.
[0091] Based on the corresponding relationship between the changes in the extension depth and the shift of the heat diffusion boundary, a cold bridge development indicator value is formed. In the calculation, the weights for the continuity of the heat diffusion boundary shift are 0.30, the weight for the extension enhancement region is 0.20, the weight for the extension stagnation region is 0.10, the weight for the extension depth change state is 0.30, and the weight for the continuity of the vacuum change rate is 0.10. Each value is multiplied by its corresponding weight and then summed to obtain the cold bridge development indicator value. When the cold bridge development indicator value reaches 0.50 and is less than 0.70, a Level 1 cold bridge development level is output, and the vacuum insulation failure risk result is "concern"; when the cold bridge development indicator value reaches 0.70 and is less than 0.85, a Level 2 cold bridge development level is output, and the vacuum insulation failure risk result is "warning"; when the cold bridge development indicator value reaches 0.85, a Level 3 cold bridge development level is output, and the vacuum insulation failure risk result is "handling".
[0092] To verify the parameter values, a 1:10 scale annular sandwich low-temperature test section was used for comparison. The test section was 6 meters long, the insulation layer temperature was -162 degrees Celsius, and a stable heat input of 2 watts was applied to the outer end of the metal heat conduction path. During the test, the cold bridge development indicator value reached 0.54 at the 8th time stage, corresponding to a continuous thermal response in the inner sensing layer; it reached 0.73 at the 12th time stage, corresponding to the expansion of the deformation accumulation shift area to the middle sensing layer; and it reached 0.87 at the 16th time stage, corresponding to a continuous structural response stable shift state in the outer sensing layer. After the test, disassembly and inspection revealed that the locally compacted area of the perlite extended approximately 1600 mm along the axial direction of the metal heat conduction path, with deviations of 380 mm and 430 mm from the output positions of the second-level and third-level depth expansion, respectively. The highest cold bridge development indicator value was 0.32 under the 0-watt control condition and 0.61 under the 1-watt disturbance condition, neither of which output the third-level cold bridge development level. The above verification results show that using 500 mm effective boundary offset, 1500 mm axial continuous length, 0.50, 0.70, and 0.85 as level determination values, the cold bridge development level and vacuum insulation failure risk can be output based on the heat diffusion boundary offset state and the change state of the expansion depth.
[0093] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.
Claims
1. An online monitoring system for the vacuum degree of the annular space in an LNG storage tank, characterized in that, include: The layered sensing and acquisition module constructs a layered sensing area in the annular space between the inner and outer tanks of the LNG storage tank. Distributed optical fiber sensing units are set along the metal heat conduction path, and micro-deformation sensing points are set at different depths of the insulation layer to simultaneously acquire vacuum data, thermal response data, and dynamic deformation data of the insulation layer. The thermal migration feature extraction module extracts the heat transfer hysteresis features and thermal diffusion shift features of the metal path at different time stages based on the acquired thermal response data, and generates the corresponding non-uniform thermal migration state information. The structural coupling analysis module identifies microscale compression changes, interlayer displacement changes, and stress transmission changes caused by local thermal disturbances within the insulation layer based on the dynamic deformation data, thus forming structural coupling state information of the insulation layer. The temporal correlation identification module performs temporal correlation analysis on the non-uniform thermal migration state information and the insulation layer structure coupling state information, establishes the synchronous evolution relationship between thermal migration behavior and insulation layer structure response, and identifies spatial clustering areas of local heat transfer anomalies. The cold bridge trend determination module determines whether there is a hidden cold bridge formation trend in the annular space based on the vacuum change rate, thermal response continuity, and structural response stability corresponding to the spatial accumulation area. The risk tracking output module continuously tracks the status of the corresponding area when it determines that there is a tendency for the formation of hidden cold bridges. It also outputs the cold bridge development level and vacuum insulation failure risk results by combining the characteristics of thermal migration direction change and structural response expansion characteristics.
2. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 1, characterized in that, When constructing a layered sensing region, an inner sensing layer, a middle sensing layer, and an outer sensing layer are formed radially along the annular space, and each sensing layer is arranged in a corresponding manner along the extension direction of the metal heat conduction path to form a spatial correspondence between vacuum data, thermal response data, and dynamic deformation data of the insulation layer.
3. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 1, characterized in that, When setting up distributed optical fiber sensing units, the distributed optical fiber sensing units are laid parallel to both sides of the metal heat conduction path and arranged radially to correspond with the micro-deformation sensing points, so that the thermal response data and the dynamic deformation data of the insulation layer at the same spatial location can be synchronously correlated.
4. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 1, characterized in that, When extracting heat transfer hysteresis features, the continuously acquired thermal response data are arranged according to time stages to form a thermal response change sequence, and the thermal response delay change state between adjacent regions is identified based on the order in which adjacent regions reach the effective thermal response state.
5. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 1, characterized in that, When extracting thermal diffusion offset features, the corresponding areas on both sides of the metal path are divided into continuous thermal diffusion segments. The center line of the metal thermal conduction path is used as the comparison benchmark to analyze the order of thermal response propagation on both sides in order to determine the deviation state of the thermal diffusion direction.
6. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 4, characterized in that, Based on the thermal response delay change state, the degree of offset of the thermal diffusion direction on both sides of the metal path is compared in segments to determine the asymmetric diffusion region existing in the local heat transfer process; the thermal response continuity of the asymmetric diffusion region in the continuous time stage is tracked to identify the dynamic offset state of the thermal diffusion boundary and the heat accumulation change state; based on the dynamic offset state and the heat accumulation change state, the corresponding non-uniform thermal migration state information is generated.
7. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 5, characterized in that, When generating the coupling state information of the insulation layer structure, the local deformation propagation path is determined based on the dynamic deformation data at different depths of the insulation layer, and the interlayer displacement offset region is identified based on the order of deformation response between adjacent depth regions.
8. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 1, characterized in that, Identify spatial clusters of localized heat transfer anomalies, including: The non-uniform thermal migration state information and the insulation layer structure coupling state information are synchronously sorted according to the corresponding time stages to form a corresponding change sequence of thermal response and structural response. Based on the corresponding change sequence, the continuous correspondence between the thermal migration direction change state and the structural response expansion state is compared stage by stage to identify the synchronous offset region between the thermal response change and the structural response change. The spatial distribution changes of the synchronous offset region in a continuous time period are correlated and tracked to determine the extension continuity and aggregation continuity of local heat transfer anomalies inside the insulation layer. Based on the aforementioned extensional continuity and aggregation continuity, a synchronous evolution relationship between thermal migration behavior and insulation layer structural response is established, and spatial aggregation regions of local heat transfer anomalies are identified.
9. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 8, characterized in that, Determining whether there is a tendency for hidden cold bridges to form within the annular space includes: The vacuum change state of the spatial cluster region is tracked in a continuous time period to identify the slow-changing continuous region and the intermittent recovery region in the local vacuum fluctuation process. Based on the gradual change continuous region and the intermittent recovery region, the continuity of the thermal response in the corresponding region is synchronously compared to determine the thermal response continuous region and the thermal response repeated region that exist in the process of heat transfer change. By combining the thermal response continuation region and the thermal response repetition region, the stability of the insulation layer structure response is continuously verified, and the deformation accumulation offset region existing in the process of local structural changes is identified. Based on the expansion state of the deformation accumulation offset region over a continuous time period, it is determined whether there is a tendency for the formation of hidden cold bridges within the annular space.
10. The online monitoring system for the vacuum degree of the annular space of an LNG storage tank according to claim 9, characterized in that, Output the cold bridge development level and vacuum insulation failure risk results, including: Continuous state tracking is performed on the corresponding regions where a hidden cold bridge formation trend is identified, and the changes in the direction of thermal migration and the expansion of the structural response are obtained over a continuous time period. Based on the change in the direction of thermal migration, the continuity of the local heat diffusion boundary is analyzed in stages to identify the expansion enhancement region and expansion stagnation region that exist during the thermal migration process. By combining the extended enhancement region and the extended stagnation region, the spatial continuity of the structural response extension state within the corresponding region is simultaneously verified to determine the extension depth change state of the local structural disturbance. Based on the corresponding relationship between the expansion depth change state and the heat diffusion boundary offset state, the cold bridge development level and vacuum insulation failure risk results are output.
