An intelligent monitoring system for LNG storage tank structure

Abstract

The application belongs to the field of LNG storage tank monitoring and evaluation, and relates to an intelligent LNG storage tank structure monitoring system, comprising: a state monitoring module for collecting and monitoring monitoring data of the LNG storage tank structure; a digital information module for establishing a visual model according to the monitoring data collected by the state monitoring module, and for visual display in combination with user input information according to the visual model; and an intelligent post-processing module for calculating the monitoring data collected by the state monitoring module through a finite element model, evaluating the health state of the LNG storage tank structure according to the calculation results, and displaying the calculation results and the structure health state evaluation in the digital information module. The application has reasonable design, high intelligence, safe and stable operation, and high efficiency, and solves the problems of limited measuring points, high cost, limited data processing capacity, low safety, poor corrosion resistance and insufficient intelligence of traditional structure monitoring and state evaluation systems.

Description

TECHNICAL FIELD

[0001] The application relates to an LNG storage tank structure intelligent monitoring system and belongs to the technical field of LNG storage tank monitoring and evaluation. BACKGROUND

[0002] The LNG storage tank is the most important and largest single structure in the receiving station, stores a large amount of ultra-low temperature liquefied natural gas, and its structural stability is crucial in the construction and operation stages, directly affecting the safety of the receiving station and the surrounding environment. Since the Dapeng LNG receiving station was put into production in 2006, the longest operation time of the LNG storage tank in China has reached 16 years, which is more than 60% of the designed life. At present, more than 100 storage tanks have been built in China, and more than 50 storage tanks are under construction and planning. With the construction and operation of a large number of storage tanks, more storage tanks will approach and exceed the designed life in the future.

[0003] At present, there are few researches and engineering applications on the structural health state of the LNG storage tank, and the traditional monitoring and evaluation system cannot fully judge the safety and health state of the structure, so it is urgent to establish a structural state evaluation system based on monitoring data and artificial intelligence algorithms. SUMMARY

[0004] In view of the above problems, the application aims to provide an LNG storage tank structure intelligent monitoring system, which has reasonable design, high intelligence, safe and stable operation, and solves the problems of limited measuring points, high cost, limited data processing capacity, low safety, poor corrosion resistance and insufficient intelligence of the traditional structure monitoring and state evaluation system, and realizes efficient and feasible structure state monitoring and evaluation of the LNG storage tank structure in the whole life cycle.

[0005] To achieve the above-mentioned purpose, the application provides the following technical scheme: an LNG storage tank structure intelligent monitoring system, comprising a state monitoring module, a digital information module and an intelligent post-processing module; the state monitoring module is used for collecting and monitoring the monitoring data of the LNG storage tank structure; the digital information module establishes a visual model according to the monitoring data collected by the state monitoring module, and performs visual display according to the visual model combined with user input information; the intelligent post-processing module is used for calculating the monitoring data collected by the state monitoring module through a finite element model, evaluating the health state of the LNG storage tank structure according to the calculation result, and displaying the calculation result and the structural health state evaluation in the digital information module.

[0006] Further, the state monitoring module comprises a sensor unit, a data acquisition processing unit, a data transmission unit and a data storage management unit; the sensor unit is arranged inside or on the surface of the LNG storage tank structure for monitoring the monitoring data of the structure; the data acquisition processing unit is used for dividing the sensor unit into several areas according to the types and monitoring point positions, integrating the sensors in different areas into a plurality of integrated points, and transmitting the monitoring data of each integrated point to a modulation device through the data transmission unit; the data transmission unit is used for transmitting the monitoring data in the sensor unit and the data acquisition processing unit; and the data storage management unit is used for storing the monitoring data output by the modulation device.

[0007] Further, the sensor unit is one or more of a distributed optical fiber based on Brillouin scattering light frequency domain analysis, a dense distributed fiber grating, a weak fiber grating, a MEMS acceleration vibration sensor or a three-component force balance acceleration sensor.

[0008] Further, the digital information module comprises a storage tank digital information model unit, a storage tank structure twin operation visualization unit and a storage tank information management unit; the storage tank digital information model unit is used for establishing a storage tank structure digital information model according to BIM standards, LNG storage tank structure and equipment information; the storage tank structure twin operation visualization unit is used for realizing lightweight flow of the model according to the storage tank structure digital information model and three-dimensional graphics; and the storage tank information management unit is used for human-computer interaction, realization of information inquiry and calling window, real-time push of storage tank related early warning information and mobile inspection.

[0009] Further, the storage tank structure twin operation visualization unit can realize three-dimensional display of storage tank pile foundation stress monitoring system data, three-dimensional display of prestressed cable force monitoring system data, three-dimensional display of storage tank prestress monitoring system data, three-dimensional display of storage tank outer tank stress monitoring system data, three-dimensional display of storage tank top beam frame stress monitoring system data, three-dimensional display of ground temperature field monitoring system data, three-dimensional display of storage tank structure settlement monitoring system data, three-dimensional display of storage tank vibration monitoring system data, three-dimensional display of storage tank wind speed and direction monitoring system data, three-dimensional visualization model and surrounding display of the storage tank, display of structure component data information, display and visualization model positioning of related threshold value and early warning information, three-dimensional cloud map display of intelligent twin analysis results and display of service state evaluation results.

[0010] Further, the storage tank information management unit is a mobile terminal APP monitoring alarm device, which acquires the monitoring data obtained in the sensor unit in real time, provides a chart visualization display of the monitoring data, adjusts the threshold value of the monitoring data, and receives an abnormal alarm of the monitoring data in real time.

[0011] Further, the intelligent post-processing module comprises a parameterized finite element model unit, a tank finite element simulation real-time calculation unit and a tank structure health state evaluation unit; the parameterized finite element model unit is configured to automatically adjust the finite element model according to different analysis requirements of global structure analysis and local component analysis; the tank finite element simulation real-time calculation unit is configured to establish a simulation result dataset according to the finite element model; and the tank structure health state evaluation unit is configured to perform online evaluation on the structure health state of the tank according to the monitoring data and the simulation result dataset and give a structure health state grade of the tank.

[0012] Further, the finite element model adjustment method comprises defining material properties and element types, then establishing a geometric model according to drawings and simplified theories, and performing mesh division; inputting loads into the geometric model, and performing calculation through multi-source heterogeneous data conversion and parameterized pre-processing and post-processing algorithms; establishing data representation and exchange mechanism of the multi-source heterogeneous data, and completing multi-source heterogeneous data fusion conversion and seamless application under different design scenarios; creating split surfaces, and loading and applying constraint conditions on a part of the surfaces; and performing statistical analysis on working conditions and load combinations according to different consideration principles.

[0013] Further, the method for establishing the data representation and exchange mechanism of the multi-source heterogeneous data comprises: quickly establishing tree node corresponding relationship through a nearest search algorithm; reading tree structures in different design scenarios by using relational database technology, mapping corresponding structures through a visual interface of left and right layout frames, and representing mapping states through visual connection lines; quickly positioning corresponding structure tree nodes according to tree structure characteristics through the nearest search algorithm when refreshing states, and realizing standardized processing of heterogeneous data, including unified arrangement of data types of multi-source heterogeneous data and matching conversion of heterogeneous data contents through conversion relationships; the method for loading and applying constraint conditions on a part of the surfaces comprises: creating a beam element model by selecting a cross section from a template; building a shared node model, establishing an assembly shared node model; setting a surrounding area of a liquid generated during fluid simulation; repairing damaged and defective surfaces in the model; establishing mesh division size, setting a division method, and performing mesh division on the established LNG tank model; automatically generating a fluid region mesh by specifying a fluid region in a closed space; and creating conversion of the finite element model.

[0014] Further, the method for performing online evaluation on the structure health state of the tank according to the monitoring data and the simulation result dataset and giving a structure health state grade of the tank comprises: taking detection data of each index as input of a fuzzy neural network, training the fuzzy neural network, inputting detection data to be tested into the trained neural network to obtain evaluation results representing experts, and completing evaluation on the health state of the LNG tank structure.

[0015] The present application has the following advantages due to the above technical solutions:

[0016] 1. The LNG storage tank structure intelligent monitoring system is safe and stable, has good durability, and is low in cost, and can realize intelligent monitoring and evaluation of the LNG storage tank structure throughout the life cycle; the LNG storage tank structure intelligent monitoring system has stronger data processing and analysis capabilities; the LNG storage tank structure intelligent monitoring system has strong environmental adaptability and can adapt to low-temperature and corrosive environmental conditions;

[0017] 2. The LNG storage tank structure intelligent monitoring system can realize evaluation requirements under multi-field coupling conditions;

[0018] 3. The LNG storage tank structure intelligent monitoring system can realize LNG storage tank structure digital twinning and physical state prediction;

[0019] 4. The LNG storage tank structure intelligent monitoring system can establish an LNG storage tank structure health state evaluation system based on measured data. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a structure schematic diagram of the LNG storage tank structure intelligent monitoring system in an embodiment of the present application;

[0021] Figure 2 is a structure schematic diagram of the state monitoring module in an embodiment of the present application;

[0022] Figure 3 is a structure schematic diagram of the digital informationization module and the intelligent post-processing module in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order for those skilled in the art to better understand the technical solutions of the present application, the present application is described in detail through specific embodiments. However, it should be understood that the specific embodiments are provided only to better understand the present application, and they should not be understood as limiting the present application. In the description of the present application, it should be understood that the terms used are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0024] The LNG storage tank stores a large amount of ultra-low temperature liquefied natural gas, and has high requirements for air tightness. On the other hand, the development trend of large-scale storage tank structure is obvious. The surface area of the world's largest 270,000 LNG storage tank exceeds 30,000 square meters, and the areas that need to be focused on are widely distributed and large in area. If damage positioning is to be realized, hundreds of thousands of measuring points are required. The traditional monitoring system is difficult to meet the needs of engineering application under this condition, and it is almost impossible to have long-term monitoring possibility, while the system cost is extremely high, and it does not have economic implementation.

[0025] The LNG storage tank structure has high safety level, and has high requirements for the timeliness of the monitoring and evaluation system. Hundreds of thousands of measuring points will generate massive monitoring data within a monitoring period, and the traditional monitoring and structure evaluation system needs to be manually inspected, collected and analyzed regularly, which cannot complete the monitoring data analysis and processing work with timeliness requirements.

[0026] The LNG storage tank stores a large amount of ultra-low temperature liquefied natural gas, and has extremely high requirements for the explosion-proof level of the facilities and equipment in the structure and the surrounding area. The traditional monitoring system mostly uses sensing technology to transmit information through electrical signals, which has safety risks when applied to LNG storage tank structure monitoring.

[0027] The LNG storage tank stores a large amount of ultra-low temperature liquefied natural gas, and the internal structure has a working condition as low as -165℃. The influence of the low-temperature environment on the structure state cannot be ignored, and the disturbance factors of the structure state are more diverse and complex. The minimum temperature of the application scene of the traditional structure monitoring and evaluation system is much higher than -165℃, which is insufficient to cope with the structure state monitoring and evaluation under the whole life cycle operation scene of the LNG storage tank.

[0028] For safety and convenience, LNG storage tanks are usually located on the sea away from personnel gathering areas, and are in a salt spray environment for a long time, which puts higher requirements on the corrosion resistance of the monitoring system. The traditional monitoring system mostly uses metal sensors and data transmission lines, which are difficult to meet the monitoring and evaluation requirements of the whole life cycle of the LNG storage tank structure.

[0029] The LNG storage tank structure is complex, has high safety level, has various environmental impact factors, includes ultra-low temperature scene, and the structure influence mechanism under coupling is not clear. The traditional structure state evaluation system mostly targets a single impact factor, which is difficult to meet the evaluation requirements under multiple field coupling conditions. On the other hand, the traditional structure state evaluation system is difficult to process massive monitoring data, and cannot meet the timeliness requirements of LNG storage tank structure state evaluation.

[0030] In order to solve the above problems existing in the prior art, the present application provides an LNG storage tank structure intelligent monitoring system, comprising: a state monitoring module, a digital informatization module and an intelligent post-processing module; the state monitoring module is used for collecting and monitoring the monitoring data of the LNG storage tank structure; the digital informatization module establishes a visual model according to the monitoring data collected by the state monitoring module, and performs visual display in combination with user input information according to the visual model; the intelligent post-processing module is used for calculating the monitoring data collected by the state monitoring module through a finite element model, and performing LNG storage tank structure health state evaluation according to the calculation result. The present application has the advantages of reasonable design, high intelligent degree, safe and stable and efficient operation, and solves the problems of limited measuring points, high cost, limited data processing capacity, low safety, poor corrosion resistance and insufficient intelligence of the traditional structure monitoring and state evaluation system, and realizes efficient and feasible structure state monitoring and evaluation of the LNG storage tank structure in the whole life cycle. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

[0031] Embodiment

[0032] The present embodiment provides an LNG storage tank structure intelligent monitoring system, as shown in Figure 1 which is an LNG storage tank structure intelligent monitoring system based on advanced structure state sensing technology, three-dimensional finite element structure simulation and artificial intelligence algorithm, comprising: a state monitoring module, a digital informatization module and an intelligent post-processing module.

[0033] The state monitoring module is used for collecting and monitoring the monitoring data of the LNG storage tank structure.

[0034] The digital informatization module establishes a visual model according to the monitoring data collected by the state monitoring module, and performs visual display in combination with user input information according to the visual model.

[0035] The intelligent post-processing module is used for calculating the monitoring data collected by the state monitoring module through a finite element model, and performing LNG storage tank structure health state evaluation according to the calculation result, and displaying the calculation result and the structure health state evaluation in the digital informatization module.

[0036] As shown in Figure 2 , the state monitoring module comprises a sensor unit, a data collection and processing unit, a data transmission unit and a data storage and management unit.

[0037] The sensor unit is installed in the interior or surface of the LNG storage tank structure during the construction stage of the storage tank structure, and is used for monitoring the monitoring data of the structure; the specific form is a sensor fixed on the steel bar and simultaneously cast in the concrete interior and permanently fixed on the structure surface.

[0038] The sensor unit in the embodiment includes, but is not limited to, a prestressed cable force monitoring system, a pile foundation structure response monitoring system, a pile cap structure response monitoring system, a wall structure response monitoring system, a dome structure response monitoring system, a roof beam frame structure response monitoring system, a storage tank structure settlement monitoring system, and a storage tank earthquake monitoring system. In the construction stage of the LNG storage tank structure pile foundation, the pile foundation structure response monitoring system is bound to the reinforcement cage and installed and laid during the concrete pouring process. In the construction stage of the LNG storage tank structure pile cap, the pile cap structure response monitoring system is bound to the steel mesh according to the design scheme and installed and laid during the concrete pouring process. In the construction stage of the LNG storage tank structure wall, the wall structure response monitoring system is bound to the steel mesh according to the design scheme and installed and laid during the concrete pouring process. In the steel strand threading stage, the prestressed cable force monitoring system is installed and laid together with the prestressed steel strand. In the construction process of the LNG storage tank structure roof beam frame, the roof beam frame structure response monitoring system is pasted on the roof beam frame. In the construction stage of the LNG storage tank structure dome, the dome structure response monitoring system is bound to the steel mesh and installed and laid during the concrete pouring process. After the construction of the LNG storage tank structure pile cap is completed, the storage tank structure settlement monitoring system is installed outside the pile cap structure and uniformly laid. After the construction of the LNG storage tank structure outer tank is completed, the storage tank earthquake monitoring system is installed on the pile cap, wall, and dome structure.

[0039] The sensor unit in the embodiment is one or several of a distributed optical fiber BOFDA (Brillouin Optical Frequency Domain Analysis) based on Brillouin scattering light frequency domain analysis, a dense distributed fiber grating, a weak fiber grating, a MEMS acceleration vibration sensor, or a three-component force balance acceleration sensor.

[0040] The data acquisition and processing unit is used for dividing the sensor unit into several areas according to the categories and the positions of the monitoring points, integrating the sensors in different areas into a plurality of integrated points, gathering the monitoring data of each integrated point to the main optical cable through the data transmission unit, transmitting the data to the health station through the main optical cable, and finally transmitting the data to the modulation equipment, which automatically collects various sensor signal of the vibrating string, the differential resistance, the resistance, the current, the voltage, the switch and the digital output. The sensor signal mentioned here is only illustrative, but not limited, and the specific sensor signal can be selected according to the actual needs. The data acquisition and processing unit has the ability of waterproof, anti-electricity and anti-electromagnetic interference, and can be applied to different engineering site environments. The data acquisition and processing unit can include a waterproof case, a main control module, a power module, a full-function measurement module, a temperature and humidity measurement module and a lightning protection module, etc. The specific position and connection mode of the above components are not unique, which has been fully explained in the prior art, and does not belong to the invention content of the embodiment, so it is not described in detail here, and it can be realized by using the existing method. The data acquisition and processing unit can also be provided with a wireless communication module, a Bluetooth communication module, a network port module and a WIFI network module for internal or external data transmission. The data protocol of the acquisition instrument for collecting sensor data can be opened to the outside.

[0041] The data transmission unit is used for transmitting the monitoring data in the sensor unit and the data acquisition and processing unit. The data transmission unit includes line integration in the sensor unit, which includes lead integration, line protection, device integration, device debugging, etc.

[0042] The data storage management unit is used for storing the monitoring data output by the modulation equipment.

[0043] As shown in Figure 3 The digital informationization module includes a storage tank digital informationization model unit, a storage tank structure twin operation visualization unit and a storage tank informationization management unit.

[0044] The tank digital information model unit is used for compiling a tank digital information model and BIM standards related to data import requirements of the digital information module according to relevant BIM standards published by the state and the industry, forming standardized files, stipulating tank structure digital information model classification principles, modeling ranges and detail levels of the as-built model, original attribute definitions of model graphs, model engineering attribute definitions and detail levels, tank digital information model family library management and the like, stipulating data ranges, data contents, delivery requirements, delivery quality requirements, platform delivery modes of the tank structure digital information model for implementing import of the digital information module, and respectively defining standard templates of various data. The tank structure digital information model is established according to LNG tank structure, i.e. relevant as-built drawings and equipment information, according to modeling principles. After completion of the model, its integrity, standardization and coordination are tracked and maintained. As-built equipment information is supplemented according to equipment procurement and installation, and specified operation and maintenance information is supplemented according to operation and maintenance scheme requirements. After completion of the tank structure digital information model, lightweight operation is performed according to requirements of the access digital information module, and relevant format files are formed, satisfying import requirements.

[0045] The tank digital information model unit in the embodiment includes but is not limited to LNG tank structure bodies, auxiliary facilities and various sensors, and can be used as a carrier of the platform, structure, equipment, sensors and related data to display various data and graphs. The tank digital information model unit can perform conversion of multiple model formats and support multiple systems.

[0046] The tank structure twin operation visualization unit is used for realizing model lightweight flow according to the tank structure digital information model and three-dimensional graphics. The tank structure twin operation visualization unit includes an LNG tank, a surrounding environment and a scene, and the display mode can include a three-dimensional visualization large screen. The tank structure twin operation visualization unit can also include data interfaces of a state monitoring module and an intelligent post-processing module. In the embodiment, the tank structure twin operation visualization unit can realize three-dimensional display of data of a tank pile foundation stress monitoring system, a pre-stressed cable force monitoring system, a tank pre-stress monitoring system, a tank outer tank stress monitoring system, a tank top beam frame stress monitoring system, a ground temperature field monitoring system, a tank structure settlement monitoring system, a tank vibration monitoring system, a tank wind speed and direction monitoring system, a tank three-dimensional visualization model and surrounding display, structure component data information display, related threshold and early warning information display and visualization model positioning, intelligent twin analysis result three-dimensional cloud map display and service state evaluation result display. Various data can be displayed in the form of graphs.

[0047] The three-dimensional graphics in the embodiment are realized by a three-dimensional visualization graphics engine, and a development and running environment thereof includes a mainstream development system such as JAVA,.NET, GO and the like, and a running environment thereof includes Linux and Windows systems to develop a storage tank structure twin running visualization unit. The storage tank structure twin running visualization unit supports distributed deployment of server-side three-dimensional data and three-dimensional model information service and networked access of a client, and the client is not limited in number. The storage tank structure twin running visualization unit includes an LNG storage tank structure, a surrounding environment and a scene, creates a system interface, ensures real-time display of monitoring data, takes the storage tank structure digital information model as a display base of data, and assists in monitoring a situation of the LNG storage tank structure.

[0048] The storage tank informatization management unit is used for human-computer interaction, realizes an information inquiry and calling window, real-time push of storage tank related early warning information, and mobile inspection. In the embodiment, the storage tank informatization management unit is a mobile terminal APP monitoring and alarming device. The mobile terminal APP can realize storage tank digital information model browsing, including: browsing of the storage tank three-dimensional model on the mobile terminal (loading the whole when looking from a distance, and displaying details and partials when zooming in the visual angle); panning and rotating of the storage tank model on the mobile terminal (rotating around an arbitrary component as a center point); panoramic display and other visual angle display of the storage tank model on the mobile terminal, realizing selection (including point selection and frame selection), display, hiding and display of only the selected components (and being capable of displaying information of the corresponding components). The mobile terminal APP monitoring and alarming device acquires monitoring data obtained from sensor units, that is, a prestress monitoring system, an outer tank mechanical response monitoring system, a pile foundation mechanical response monitoring system, a storage tank structure settlement monitoring system, a wind speed monitoring system, a ground temperature monitoring system and the like, and provides monitoring data chart visualization display. The threshold of the monitoring data can be adjusted, and an abnormal alarm of the monitoring data can be received in real time. The mobile terminal can edit, dispatch and receive maintenance work orders at any time, the mobile terminal APP can process and forward work orders, and the work order processing flow and message notification can be acquired in real time.

[0049] In the embodiment, the mobile terminal APP is linked with a PC terminal, can be connected with existing business and data systems, can inquire storage tank related monitoring real-time data and historical data on the PC terminal, can inquire storage tank related materials, drawings, manuals and the like on the PC terminal, can push storage tank monitoring system early warning information in real time on the mobile terminal, and can view real-time monitoring data of the storage tank on the mobile terminal.

[0050] The intelligent post-processing module includes a parameterized finite element model unit, a storage tank finite element simulation real-time calculation unit and a storage tank structure health state evaluation unit.

[0051] The parameterized finite element model unit is used for specifying a finite element calculation software, implementing modeling and calculation in the relevant calculation software, calling LNG storage tank monitoring data, automatically adjusting the finite element model according to different analysis requirements of global structure analysis and local component analysis, developing an LNG storage tank parameterized finite element modeling and calculation program, proposing a multi-level multi-component finite element adaptive model, considering multiple load combinations and stress working conditions to carry out calculation and result analysis, summarizing and inducing deformation and stress characteristics of the LNG storage tank at different stages, and realizing automatic establishment and updating of a mechanical simulation model based on design parameters and monitoring data and adaptive adjustment.

[0052] The finite element model adjustment method is:

[0053] The material properties and element types are defined, then the geometric model is established according to the drawings and simplified theory, and the mesh is divided; the load is input into the geometric model, and the calculation is carried out through multi-source heterogeneous data conversion and parameterized pre-processing algorithm for the finite element calculation software, and the modeling and calculation modules in the relevant calculation software, the calling of LNG storage tank structure monitoring data are realized through programming. In the path planning design of establishing the finite element model, the key modeling technologies such as multi-point constraint technology (MPC), matrix interpolation technology, pile-soil interaction simulation technology, steel reinforcement modeling technology and interface coupling technology are used.

[0054] The multi-source heterogeneous data service mode of different specialties is analyzed, the description method of multi-platform data is researched, the data resource management problems such as discovery and matching mechanism are found out, the data representation and exchange mechanism of multi-source heterogeneous data is established, and the multi-source heterogeneous data fusion conversion and seamless application under specific design scenarios are completed.

[0055] The data representation and exchange mechanism of the multi-source heterogeneous data is established, and the multi-source heterogeneous data fusion conversion and seamless application under different design scenarios are completed;

[0056] The method for establishing the data representation and exchange mechanism of multi-source heterogeneous data is: the tree node corresponding relationship is quickly established by using the nearest search algorithm, so that the massive data interconnection and intercommunication between any two information systems are realized, that is, the tree structure in different design scenarios is read by using the relational database technology, the corresponding structure is mapped and operated by using the visual interface of left and right layout framework, and the mapping state is represented by visual connection; when refreshing the state, the corresponding structure tree node is quickly located by using the nearest search algorithm according to the characteristics of the tree structure, the standardization processing of heterogeneous data is realized, including unified arrangement of the data types of multi-source heterogeneous data and matching conversion of the heterogeneous data content by using the conversion relationship;

[0057] The split surface is created, and the loading and constraint conditions are applied to the surface, which is extracted from the entity model. The method for loading and applying constraint conditions to the surface includes: creating a beam element model by selecting a cross section from a template; building a shared node model; building an assembly shared node model; setting a surrounding area of a liquid generated during fluid simulation, i.e., a bounding volume; repairing damaged and defective surfaces in the model; setting a mesh division size, setting a division method, and performing mesh division on the built LNG storage tank model; automatically generating a fluid region mesh without establishing a geometric model of the fluid region by specifying a fluid region in a closed space; and converting the fluid region mesh into a finite element model.

[0058] An adaptive mesh division technique is used to establish an adaptive finite element model, and energy error estimation is used to evaluate whether the grid density is sufficient. If the grid is not fine enough, the model can automatically refine the grid to reduce the error, so that a better stress distribution can be obtained. An easy-to-use h-type adaptive finite element method is established, and a quadtree algorithm is used to uniformly manage the grid and coarsen and encrypt the grid. The order of the shape function is kept unchanged, and the grid calculation precision is improved by changing the number of nodes and the grid size. A triangular element can be divided into four small triangular elements, and each edge of the triangle is also divided into two parts during the process. After each grid is refined and adjusted, a uniform triangular mesh is formed, and the grid is obtained after n times of continuous refinement, realizing the quadtree of the grid.

[0059] According to different principles, the working conditions and load combinations are statistically analyzed. The stress checking macro function and the calculation model checking stress are compiled, and the structure displacement calculation macro function is compiled to check the displacement. The formula for the reinforcement of the circular cross section is a transcendental equation, which is solved by iteration. Based on various standards, a calculation program is compiled, and a macro function is compiled for the reinforcement design of the outer tank structure. The tension side and the compression side need to call the macro function for calculation through the force judgment of the outer tank structure (such as the pile cap, outer wall, dome, and bottom pressure ring beam, etc.). A crack width calculation macro function is compiled, and the formula is:

[0060]

[0061]

[0062] wherein, ω max is the maximum crack width of the structure, α cr is the stress characteristic coefficient of the component, ψ is the strain inhomogeneity coefficient of the longitudinal tensioning steel bar between cracks, σ s is the stress of the longitudinal tensioning ordinary steel bar of the reinforced concrete component calculated according to the load quasi-permanent combination or the equivalent stress of the longitudinal tensioning steel bar of the prestressed concrete component calculated according to the standard combination, and E sis the elastic modulus of the steel bar, c s is the distance from the outer edge of the outermost longitudinal tensile steel bar to the bottom edge of the tensile steel bar area (mm), d is the equivalent diameter of the longitudinal tensile steel bar (mm), p ie is the longitudinal tensile steel bar calculated according to the effective tensile concrete section area, f tk is the standard value of the axial tensile strength of concrete.

[0063] In the solving process, dynamic adjustment is continuously carried out according to the size of the calculation error. The order of the shape function is kept unchanged, the finite element solving process is dynamically updated based on the measured data, the numerical simulation result is corrected to approximate and reflect the actual stress state of the structure, and the dynamic updating of the mechanical simulation model is realized based on the design parameters and the monitoring data.

[0064] The tank finite element simulation real-time calculation unit is used to establish a simulation result data set according to the parameterized LNG tank finite element model; contains artificial intelligence algorithm modules such as SVM and BPANN, can carry out real-time calculation intelligent model training, realize real-time intelligent analysis of the stress deformation of the LNG tank based on monitoring data driving, realize real-time prediction of the deformation stress state of the LNG tank based on data driving, the tank finite element simulation real-time calculation unit can realize modeling and calculation and other modules in the finite element calculation software, calling of the LNG tank monitoring data, self-adaptive adjustment of the finite element model, automatic establishment and updating of the mechanical simulation model based on design parameters and monitoring data. It can realize the mapping of the finite element calculation result and the digital information model of the tank, realize the data flow between the digital twin models. Among them, the spatial resolution of the three-dimensional display of the calculation result reaches the meter level, and the spatial resolution of the three-dimensional display of the calculation result of the key structural components reaches the centimeter level. The tank finite element simulation real-time calculation unit includes a digital information model environment interface, a Revit API interface program in the embodiment, which uses a one-way direct calling mode to realize lossless conversion of the model and engineering information in the Revit software into the APDL command stream format in the finite element analysis software, automatic creation of the model, meshing and the whole process of mechanical analysis.

[0065] Revit software for function-oriented secondary development, using the idea of grid to develop the required data conversion plug-in through the API interface provided; through Python and other programming languages, create intelligent algorithm information recognition program, which can recognize Revit model information and directly generate APDL command stream in corresponding intelligent algorithm to automatically execute model establishment and attribute command; compile the command stream of automatic mesh division to ensure the shape of the mesh to the greatest extent and perform calculation and analysis. At the same time, program design structural response module under the action of gravity load, structural response module under the action of wind load, structural modal analysis module and structural response module under the action of earthquake, automatically generate the command stream of corresponding functions according to the requirements of structural design. The mapping relationship between the finite element calculation results and the digital information model of the storage tank structure is established through spatial geometric relationship or component topological relationship.

[0066] The mathematical relationship of the intelligent algorithm is completely represented in the digital information model environment by using Python and other programming languages, and the model operation period is adjusted according to the monitoring data acquisition period and analysis requirements. It establishes the mapping relationship between the finite element calculation results and the digital information model through spatial geometric relationship or component topological relationship, realizes the data flow between the digital twin models. It is realized through embedded calculation in the digital information model environment, realizing the consistency of calculation and display. In this embodiment, the intelligent model result output time in the storage tank finite element simulation real-time calculation unit reaches seconds, and the error index rRMSE is less than 5%.

[0067] In this embodiment, the storage tank finite element simulation real-time calculation unit establishes a LNG storage tank numerical simulation data set for elastic-plastic and complex analysis process and object, taking deformation, cable force and modal as input, and taking the stress performance of the LNG storage tank structure as output. According to the established data set, the SVM, BPANN and other artificial intelligence algorithm modules are developed and designed, and the LSTM, CNN and other artificial intelligence algorithms are trained, and the neural network structure analysis and optimization are carried out.

[0068] Randomly initialize the weights and thresholds of the network within a certain range, and give the parameters of the network certain initial values. For each training sample, calculate the input and output of each unit layer by layer from the input layer to the output layer. For each learning sample, correct the weights and thresholds of the network in the gradient descent direction. When all the learning samples are input, if the total training error E is less than the pre-set total error e, the test total error of the validation set sample is investigated, when it is found that the test total error increases instead, the training is ended, otherwise the training is continued.

[0069] The LSTM and CNN are taken as the basic model architecture, the monitoring data of the storage tank is taken as the input data, and is divided into time series related features and time series independent features, the LSTM and CNN are used to obtain the two types of feature representations respectively, then the two are integrated, and finally the prediction result of the stress and deformation is obtained through a simple classifier. The two parts of data are processed respectively. For the time series related data, the CNN is used for pre-processing to extract the first feature, and then the LSTM which is particularly suitable for processing time series information is used for secondary processing. The time series independent features are directly processed by the CNN.

[0070] The storage tank structure health state evaluation unit is used for online evaluation of the structure health state of the storage tank according to the monitoring data and the simulation result data set, and gives the structure health state grade of the storage tank. The storage tank structure health state evaluation unit selects the prestressed cable force obtained by field monitoring, the mechanical response of the storage tank, the settlement of the storage tank structure and the stress of the storage tank calculated as the evaluation indexes. The results of numerical simulation are combined to divide the grades of the evaluation indexes, the relationship between the evaluation index system and the structure health state of the storage tank is established, and the literature research and expert questionnaire are used to carry out the division of the index grades. The results obtained by the literature research and the results obtained by the expert research are averaged as the final index grade. Finally, the structure health state of the LNG storage tank is evaluated according to the obtained evaluation indexes.

[0071] The method for online evaluation of the structure health state of the storage tank according to the monitoring data and the simulation result data set and giving the structure health state grade of the storage tank is: taking the detection data of each index as the input fuzzy neural network, training the fuzzy neural network, inputting the detection data to be tested into the trained neural network to obtain the evaluation result representing the expert, and completing the evaluation of the health state of the LNG storage tank structure.

[0072] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, but not to limit it, although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced, without departing from the spirit and scope of the present application, any modification or equivalent replacement, which should be covered in the protection scope of the claims of the present application. The above content is only the specific embodiments of the present application, but the protection scope of the present application is not limited to this, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An intelligent monitoring system for LNG storage tank structure, characterized in that, The application relates to a LNG storage tank structure health state monitoring system. The system comprises a state monitoring module, a digital information module and an intelligent post-processing module. The state monitoring module is used for collecting and monitoring the monitoring data of the LNG storage tank structure. The digital information module establishes a visual model according to the monitoring data collected by the state monitoring module, and carries out visual display by combining user input information. The intelligent post-processing module is used for calculating the monitoring data collected by the state monitoring module through a finite element model, evaluating the health state of the LNG storage tank structure according to the calculation result, and displaying the calculation result and the health state evaluation of the structure in the digital information module. The intelligent post-processing module comprises a parameterized finite element model unit, a storage tank finite element simulation real-time calculation unit and a storage tank structure health state evaluation unit. The parameterized finite element model unit is used for automatically adjusting the finite element model according to different analysis requirements of global structure analysis and local component analysis. The storage tank finite element simulation real-time calculation unit is used for establishing a simulation result data set according to the finite element model. The storage tank structure health state evaluation unit is used for online evaluating the health state of the storage tank according to the monitoring data and the simulation result data set and giving the health state grade of the storage tank. The finite element model adjusting method is as follows: Define material attributes and unit types, then establish a geometric model according to drawings and simplified theory, and carry out mesh division; Input loads into the geometric model, and calculate through multi-source heterogeneous data conversion and parameterized pre-processing and post-processing algorithms; Establish data representation and exchange mechanism of the multi-source heterogeneous data, and complete multi-source heterogeneous data fusion conversion and seamless application under different design scenes; Create a split surface, and load and apply constraint conditions on a part of the surface; According to different consideration principles, statistically analyze working conditions and load combinations; The method for establishing the data representation and exchange mechanism of the multi-source heterogeneous data is as follows: Quickly establish tree node corresponding relationship through a nearest search algorithm; Read tree structures in different design scenes by using relational database technology, map corresponding structures through a visual interface of left and right layout frames, and represent mapping states through visual lines; When refreshing states, quickly locate corresponding structure tree nodes according to the characteristics of the tree structures through the nearest search algorithm, realize standardized processing of heterogeneous data, including uniformly arranging data types of the multi-source heterogeneous data and matching and converting the heterogeneous data content through conversion relations; The method for loading and applying constraint conditions on a part of the surface comprises the following steps: Create a beam element model by selecting a cross section from a template; Establish a shared node model, and establish a shared node model of an assembly; Generate a surrounding area of liquid when simulating fluid; Repair damaged and defective surfaces in the model; Establish mesh division size, set a division method, and carry out mesh division on the established LNG storage tank model; Automatically generate fluid area meshes by specifying fluid areas in a closed space; Convert the finite element model.

2. The intelligent monitoring system for LNG storage tank structure according to claim 1, wherein, The state monitoring module comprises a sensor unit, a data acquisition and processing unit, a data transmission unit and a data storage and management unit; The sensor unit is arranged inside or on the surface of the LNG storage tank structure and is used for monitoring the monitoring data of the structure; The data acquisition and processing unit is used for dividing the sensor unit into several areas according to the types and positions of the monitoring points, integrating the sensors in different areas into a plurality of integrated points, and transmitting the monitoring data of each integrated point to a modulation device through the data transmission unit; The data transmission unit is used for transmitting the monitoring data in the sensor unit and the data acquisition and processing unit; The data storage and management unit is used for storing the monitoring data output by the modulation device.

3. The intelligent monitoring system for LNG storage tank structure according to claim 2, wherein, The sensor unit is one or more of a distributed optical fiber based on Brillouin scattering light frequency domain analysis, a dense distributed fiber grating, a weak fiber grating, a MEMS acceleration vibration sensor or a three-component force balance acceleration sensor.

4. The intelligent monitoring system for LNG storage tank structure according to claim 2, wherein, The digital information module comprises a storage tank digital information model unit, a storage tank structure twin operation visualization unit and a storage tank information management unit; The storage tank digital information model unit is used for establishing a storage tank structure digital information model according to BIM standards, LNG storage tank structure and equipment information; The storage tank structure twin operation visualization unit is used for realizing lightweight flow of the model according to the storage tank structure digital information model and three-dimensional graphics; The storage tank information management unit is used for human-computer interaction, information inquiry and calling window, real-time push of storage tank related early warning information and mobile inspection.

5. The intelligent monitoring system for LNG storage tank structure according to claim 4, wherein, The storage tank structure twin operation visualization unit can realize three-dimensional display of data of a storage tank pile foundation stress monitoring system, a prestressed cable force monitoring system, a storage tank prestress monitoring system, a storage tank outer tank stress monitoring system, a storage tank top beam frame stress monitoring system, a ground temperature field monitoring system, a storage tank structure settlement monitoring system, a storage tank vibration monitoring system, a storage tank wind speed and direction monitoring system, a storage tank three-dimensional visualization model and surrounding display, structure component data information display, related threshold and early warning information display and visualization model positioning, intelligent twin analysis result three-dimensional cloud map display and service state evaluation result display.

6. The intelligent monitoring system for LNG storage tank structure according to claim 4, wherein, The storage tank information management unit is a mobile terminal APP monitoring alarm device, which obtains monitoring data obtained by the sensor unit in real time, provides monitoring data chart visualization display, adjusts the threshold of the monitoring data, and receives abnormal alarm of the monitoring data in real time.

7. The intelligent monitoring system for LNG storage tank structure according to claim 1, wherein, The method for online evaluation of the structural health state of the storage tank according to the monitoring data and the simulation result data set and giving the structural health state grade of the storage tank is: The detection data of each index is taken as input fuzzy neural network, the fuzzy neural network is trained, the detection data to be measured is input into the trained neural network to obtain the evaluation result representing the expert, and the evaluation of the health state of the LNG storage tank structure is completed.

Images