Three-dimensional multi-parameter real-time monitoring system and method for floating roof of crude oil storage tank

Abstract

The invention relates to the technical field of data detection, in particular to a three-dimensional multi-parameter real-time monitoring system and method for a floating roof of a crude oil storage tank. The multi-mode sensor array is arranged on the surface of a floating roof and the inner wall of a storage tank and used for collecting displacement, deformation, pressure and environmental parameters of the floating roof in real time; the data acquisition and preprocessing module is connected with the sensor array and is used for signal conditioning, analog-to-digital conversion and noise suppression; the edge calculation unit is deployed on a storage tank site and used for processing sensor data in real time and executing initial analysis; the three-dimensional dynamic modeling module is used for constructing a floating roof three-dimensional attitude model based on multi-source sensor data; the cloud analysis platform is connected with the edge computing unit through the industrial Internet of Things and is used for deep data analysis and early warning decision making; according to the scheme, the technical problems of insufficient parameter coverage, difficult data fusion, one-sided state evaluation and the like can be solved.

Description

Technical Field

[0001] The present invention relates to the field of data detection technology, and in particular to a three-dimensional multi-parameter real-time monitoring system and method for a floating roof of a crude oil storage tank. Background Art

[0002] In the field of crude oil storage tank safety monitoring, floating roof condition monitoring is a critical component in ensuring safe tank operation. The currently widely used floating roof monitoring technology suffers from a significant technical flaw: a single monitoring parameter that fails to fully reflect the true operating status of the floating roof. Most existing monitoring systems measure only a single floating roof parameter and lack the ability to simultaneously monitor multiple parameters, including the roof's three-dimensional posture, deformation distribution, and pressure equilibrium. Traditional monitoring methods primarily rely on single sensor devices, such as magnetostrictive level gauges. These technologies have significant limitations, including limited installation space, resulting in measurement explosion-proof ratings exceeding blind spots, insufficient measurement stability under extreme operating conditions, and media compatibility issues that impact long-term reliability. More critically, single-parameter monitoring cannot establish a correlation model between the floating roof's condition and multiple physical quantities, severely limiting the system's ability to identify potential risks. For example, tilting and deformation of a floating roof are often accompanied by abnormal pressure distribution and increased localized deformation. Traditional single-parameter systems are unable to capture this multi-parameter coupling, making it difficult to detect safety hazards in a timely manner. The petrochemical industry is placing increasingly stringent requirements on floating roof condition monitoring.

[0003] Based on the above problems, there is an urgent need for a technical solution that can solve technical problems such as insufficient parameter coverage, difficulty in data fusion, and one-sided status assessment, and can simultaneously obtain multi-dimensional status parameters of the floating roof. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a three-dimensional multi-parameter real-time monitoring system for the floating roof of a crude oil storage tank, comprising: Multimodal sensor array: arranged on the surface of the floating roof and the inner wall of the tank, used to collect real-time floating roof displacement, deformation, pressure and environmental parameters; Data acquisition and preprocessing module: connected to the sensor array for signal conditioning, analog-to-digital conversion and noise suppression; Edge computing unit: deployed at the tank site to process sensor data in real time and perform initial analysis; 3D dynamic modeling module: Builds a 3D attitude model of the floating roof based on multi-source sensor data; Cloud analysis platform: connected to edge computing units through the Industrial Internet of Things for in-depth data analysis and early warning decision-making; Visual human-machine interface: used to display monitoring results and alarm information; The multimodal sensor array comprises: Distributed fiber optic strain sensors are arranged along the radial and circumferential directions of the floating roof to measure the deformation of the floating roof; a three-axis MEMS acceleration sensor array is evenly distributed on the surface of the floating roof to measure the three-dimensional displacement of the floating roof; a pressure sensor matrix is arranged on the contact surface between the floating roof and crude oil to monitor the pressure distribution; a temperature and humidity sensor group is arranged in the space above the floating roof to monitor environmental parameters.

[0005] Preferably, the distributed optical fiber strain sensor adopts a wavelength modulated optical fiber Bragg grating sensor with a sensor spacing of 3 meters. It is connected to the data acquisition module through an armored flame-retardant single-mode optical cable. The protection level of the optical cable is not less than IP66, and the explosion-proof level is not less than ExdⅡBT4; the distributed optical fiber strain sensor adopts a redundant ring topology to ensure that a single point failure does not affect the overall monitoring function; the sensor is installed with a magnetic fixed bracket for easy maintenance and replacement; the optical cable connector adopts a stainless steel waterproof connector with a protection level of IP68.

[0006] Further preferably, the edge computing unit adopts an industrial-grade explosion-proof design, including: Multi-channel high-speed data acquisition card: sampling rate of not less than 1kHz, supporting 16-bit resolution; embedded processor: equipped with a real-time operating system and a hardware watchdog circuit; local storage module: using an industrial-grade SSD with a capacity of not less than 1TB; industrial communication interface: supporting Modbus TCP / IP and OPC UA protocols; Environmental monitoring module: real-time monitoring of temperature and humidity inside the chassis; redundant power input: supports 24VDC and 220VAC dual power supply; explosion-proof housing: complies with ATEX Zone 1 standards.

[0007] Further preferably, the cloud analysis platform includes: a time series database with a distributed architecture: supporting PB-level data storage; a machine learning engine: integrating LSTM neural network and random forest algorithm; a digital twin module: realizing virtual mapping and predictive simulation of floating roof status; an alarm management subsystem: supporting multi-level warning threshold setting and intelligent push; an operation and maintenance management module: recording equipment status and maintenance history; a data visualization component: supporting multi-dimensional data analysis and report generation; and a permission management unit: realizing multi-level user access control.

[0008] Further preferably, the visual human-computer interface includes: 3D floating roof attitude real-time display window, supporting multi-view switching and zooming; multi-parameter trend chart, supporting custom time range and curve overlay; alarm event list, displayed by priority; historical data query module, supporting multi-condition combination retrieval; system configuration interface for parameter setting and user management; Report export function supports PDF and Excel formats; mobile adaptation module supports access from mobile phones and tablets; voice alarm prompt function supports multi-language switching.

[0009] Further preferably, the system further comprises: The lightning protection device is installed on the floating roof guide column and adopts a three-level lightning protection design; The static elimination device is arranged at the edge of the floating roof and adopts ion wind static elimination technology; The backup power module uses a lithium-ion battery pack with a battery life of no less than 8 hours; Explosion-proof junction box, protection level not less than IP66, in line with IECEx certification; Emergency communication module, supporting satellite communication links; Environmental monitoring unit, real-time monitoring of wind speed and rainfall in the tank area; Equipment health monitoring system, predictive maintenance of key components.

[0010] Further preferably, the communication network of the system adopts: Field layer: Industrial Ethernet connects sensor arrays and edge computing units, supporting IEEE 802.3 standards; Transmission layer: The optical fiber ring network connects multiple tank monitoring systems and adopts a redundant dual-ring architecture; Cloud platform layer: 4G / 5G wireless communication connects to the cloud analysis platform and supports VPN encrypted transmission; Local backup channel: LoRa wireless communication is used as the emergency communication link; Network security protection: deploy industrial firewalls and intrusion detection systems; Data encryption mechanism: AES-256 encryption algorithm is used to protect transmitted data.

[0011] A three-dimensional multi-parameter real-time monitoring method for a crude oil storage tank floating roof is applied to any one of the above-mentioned three-dimensional multi-parameter real-time monitoring systems for a crude oil storage tank floating roof, comprising the following steps: S1: Real-time acquisition of floating roof displacement, deformation, pressure and environmental parameters through a multimodal sensor array; S2: Preprocessing of multimodal sensor array data, including denoising, normalization, and time alignment; S3: Fusion of multi-source sensor data based on the Kalman filter algorithm to estimate the three-dimensional posture of the floating roof; S4: Calculate the floating roof safety status index and generate early warning information; S5: Visual display of monitoring results and early warning information; The Kalman filter algorithm used by S3 is expressed as: ; Where: is the prior state estimate at time k; is the a priori estimated covariance; is the Kalman gain; Input for the system; is the posterior state estimate; is the posterior estimated covariance; A is the state transfer matrix; B is the control input matrix; H is the observation matrix; Q is the process noise covariance; R is the observation noise covariance; is the observation value at time k.

[0012] Further preferably, the floating roof safety status indicator includes a deformation index D, which is calculated as follows: ; Where: is the strain value of the i-th measurement point; is the initial strain reference value; is the maximum allowable strain of the material; N is the total number of measurement points.

[0013] Further preferably, the pressure distribution evaluation adopts the dynamic pressure equilibrium index , which is calculated as follows: ; Where: is the measurement value of the jth pressure sensor; is the average pressure value; is the weight coefficient of the jth sensor; M is the number of pressure sensors.

[0014] Technical effects: This invention proposes a three-dimensional, multi-parameter real-time monitoring system for the floating roof of crude oil storage tanks. The system consists of six core components: a multimodal sensor array, a data acquisition and preprocessing module, an edge computing unit, a three-dimensional dynamic modeling module, a cloud-based analysis platform, and a visual human-machine interface. The multimodal sensor array includes distributed fiber-optic strain sensors, a three-axis MEMS accelerometer array, a pressure sensor matrix, and a temperature and humidity sensor array, which are positioned at various locations on the floating roof surface and the inner wall of the tank, forming a comprehensive monitoring network. The data acquisition and preprocessing module is responsible for unified conditioning and conversion of various sensor signals, the edge computing unit performs field-level data processing, the three-dimensional dynamic modeling module constructs a three-dimensional attitude model of the floating roof, the cloud-based analysis platform performs in-depth data analysis, and the visual human-machine interface provides intuitive display of monitoring results. This system addresses the shortcomings of traditional technical solutions, including the single monitoring parameter (typically only monitoring liquid level or simple displacement, which cannot fully reflect the floating roof status); data processing lags due to reliance on central servers, resulting in delayed responses; and the lack of three-dimensional visualization, which makes it difficult to intuitively judge floating roof deformation and attitude changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is the block diagram of the three-dimensional multi-parameter real-time monitoring system for the floating roof of crude oil storage tanks in this application; Figure 2 This is a flow chart of the three-dimensional multi-parameter real-time monitoring method for the floating roof of crude oil storage tanks in this application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] See also Figure 1 Traditional technical solutions, for example, suffer from the following technical issues: Traditional crude oil tank floating roof monitoring systems suffer from three major technical bottlenecks: First, the monitoring parameters are limited, typically only monitoring liquid level or simple displacement, which cannot fully reflect the floating roof status; second, data processing lags, relying on central servers for processing, resulting in response delays; and third, the lack of three-dimensional visualization makes it difficult to intuitively determine floating roof deformation and posture changes. These issues make it difficult to detect abnormal floating roof conditions in a timely manner, increasing the risk of tank operations.

[0018] Based on this, the present application provides a three-dimensional multi-parameter real-time monitoring system for a crude oil storage tank floating roof, comprising: Multimodal sensor array: arranged on the surface of the floating roof and the inner wall of the tank, used to collect real-time floating roof displacement, deformation, pressure and environmental parameters; Data acquisition and preprocessing module: connected to the sensor array for signal conditioning, analog-to-digital conversion and noise suppression; Edge computing unit: deployed at the tank site to process sensor data in real time and perform initial analysis; 3D dynamic modeling module: Builds a 3D attitude model of the floating roof based on multi-source sensor data; Cloud analysis platform: connected to edge computing units through the Industrial Internet of Things for in-depth data analysis and early warning decision-making; Visual human-machine interface: used to display monitoring results and alarm information; The multimodal sensor array comprises: Distributed fiber optic strain sensors are arranged along the radial and circumferential directions of the floating roof to measure the deformation of the floating roof; a three-axis MEMS acceleration sensor array is evenly distributed on the surface of the floating roof to measure the three-dimensional displacement of the floating roof; a pressure sensor matrix is arranged on the contact surface between the floating roof and crude oil to monitor the pressure distribution; a temperature and humidity sensor group is arranged in the space above the floating roof to monitor environmental parameters.

[0019] It is worth mentioning that this embodiment proposes a three-dimensional multi-parameter real-time monitoring system for the floating roof of crude oil storage tanks. The system consists of six core parts: a multimodal sensor array, a data acquisition and preprocessing module, an edge computing unit, a three-dimensional dynamic modeling module, a cloud analysis platform, and a visual human-computer interface. The multimodal sensor array includes distributed optical fiber strain sensors, a three-axis MEMS acceleration sensor array, a pressure sensor matrix, and a temperature and humidity sensor group, which are arranged at different positions on the floating roof surface and the inner wall of the tank to form an all-round monitoring network. The data acquisition and preprocessing module is responsible for uniformly conditioning and converting the signals of various sensors, the edge computing unit realizes field-level data processing, the three-dimensional dynamic modeling module constructs a three-dimensional posture model of the floating roof, the cloud analysis platform performs in-depth data analysis, and the visual human-computer interface provides an intuitive display of monitoring results.

[0020] The technical benefits of this solution include: This solution enables comprehensive monitoring of the floating roof's status through multimodal sensor fusion, comprehensively collecting and analyzing displacement, deformation, pressure, and environmental parameters. The introduction of edge computing units significantly improves the system's real-time performance, enabling local data processing and initial warning within 50 milliseconds. Three-dimensional dynamic modeling technology enables operators to intuitively observe changes in the floating roof's three-dimensional posture, significantly improving monitoring accuracy and efficiency. Testing has shown that the system can shorten the time it takes to identify abnormal floating roof conditions and reduce false alarm rates.

[0021] Crude oil storage tanks place stringent demands on sensor systems. Traditional solutions, for example, face the following technical challenges: First, highly corrosive media can easily damage sensors; second, explosive gas environments require equipment with a high degree of explosion protection; third, the need for large-scale monitoring conflicts with the complexity of wiring; and fourth, maintenance is difficult, with traditional mounting methods making replacement difficult. These issues result in the lack of reliability and high maintenance costs of existing fiber optic sensing systems in storage tank applications.

[0022] Based on this, the distributed optical fiber strain sensor adopts a wavelength-modulated optical fiber Bragg grating sensor with a sensor spacing of 3 meters. It is connected to the data acquisition module through an armored flame-retardant single-mode optical cable. The optical cable protection level is not lower than IP66, and the explosion-proof level is not lower than ExdⅡBT4; the distributed optical fiber strain sensor adopts a redundant ring topology structure to ensure that a single point failure does not affect the overall monitoring function; the sensor is installed with a magnetic fixed bracket for easy maintenance and replacement; the optical cable connector uses a stainless steel waterproof connector with a protection level of IP68.

[0023] It is worth mentioning that this embodiment makes detailed limitations on the distributed optical fiber strain sensor of the above embodiment, adopts wavelength-modulated fiber Bragg grating (FBG) sensors, and the sensor spacing is precisely set to 3 meters, connected by specially designed armored flame-retardant single-mode optical cables. The protection level of the optical cable reaches IP66, and the explosion-proof level is not lower than ExdⅡBT4, ensuring safe use in flammable and explosive environments. The distributed optical fiber strain sensor adopts a redundant ring topology design. Even if a single sensor node fails, the system can still complete the monitoring function through the reconstruction of the adjacent node data. The installation method adopts a magnetic fixed bracket to facilitate on-site maintenance and replacement. The optical cable connector adopts a stainless steel waterproof connector with a protection level of IP68.

[0024] The technical benefits of the above solution include: This solution solves the aforementioned problems by optimizing sensor selection and layout: wavelength-modulated FBG sensors are immune to electromagnetic interference and suitable for strong electromagnetic environments; 3-meter spacing balances monitoring accuracy and cost; armored flame-retardant optical cables and IP66 / IP68 protection levels ensure long-term reliable operation; redundant ring topology improves system fault tolerance; and magnetic mounting reduces single sensor replacement time from the traditional 2 hours to 15 minutes. Practical applications have shown that this design achieves a sensor system MTBF (mean time between failures) of over 50,000 hours, reducing installation costs.

[0025] Traditional solutions face the following technical challenges: First, harsh industrial environments, such as high temperatures, high humidity, and vibration, can easily affect equipment reliability; second, high-speed, simultaneous data collection from multiple heterogeneous sensors is challenging; and third, explosion-proof requirements limit equipment heat dissipation and performance. Traditional solutions often compromise performance and reliability, making them difficult to meet real-time monitoring requirements.

[0026] Based on this, the edge computing unit adopts an industrial-grade explosion-proof design, including: Multi-channel high-speed data acquisition card: sampling rate of not less than 1kHz, supporting 16-bit resolution; embedded processor: equipped with a real-time operating system and a hardware watchdog circuit; local storage module: using an industrial-grade SSD with a capacity of not less than 1TB; industrial communication interface: supporting Modbus TCP / IP and OPC UA protocols; Environmental monitoring module: real-time monitoring of temperature and humidity inside the chassis; redundant power input: supports 24VDC and 220VAC dual power supply; explosion-proof housing: complies with ATEX Zone 1 standards.

[0027] It's worth noting that this embodiment defines the specific implementation of the edge computing unit, which utilizes an industrial-grade explosion-proof design. Its core components include a multi-channel high-speed data acquisition card, an embedded processor, a local storage module, and an industrial communication interface. The data acquisition card supports a sampling rate of at least 1kHz and 16-bit resolution to ensure data acquisition accuracy. The embedded processor is equipped with a real-time operating system and a hardware watchdog circuit to prevent system crashes. Local storage utilizes an industrial-grade SSD with a capacity of at least 1TB. The communication interface supports Modbus TCP / IP and OPC UA protocols, enabling interconnection with various industrial equipment. The unit also includes an environmental monitoring module, redundant power inputs, and an explosion-proof enclosure, conforming to ATEX Zone 1 standards.

[0028] The technical benefits of this solution include: This solution achieves a balance of performance and reliability through an optimized edge computing unit design. Industrial-grade components ensure stable operation in temperatures ranging from -40°C to 70°C and at 95% humidity. The high-speed data acquisition card supports 32-channel simultaneous sampling with a time synchronization accuracy of 1μs. The explosion-proof design is ATEX-certified, ensuring safe operation in explosive gas environments. Test data shows that the unit achieves a 98% data acquisition completeness rate and a processing latency of less than 10ms, significantly improving the 200ms latency of traditional solutions.

[0029] For example, traditional technical solutions have the following technical problems: The floating roof monitoring cloud platform faces challenges such as large data volume, high analysis complexity, and strict security requirements: First, the massive monitoring data, the annual data volume of a single tank can reach 10TB, which will cause high storage and processing pressure; second, the problem of intelligent identification of abnormal conditions under complex working conditions; third, the need for permission management for multi-user collaborative work; fourth, insufficient value mining of long-term operating data.

[0030] The cloud-based analysis platform includes: a time series database with a distributed architecture that supports PB-level data storage; a machine learning engine that integrates LSTM neural networks and random forest algorithms; a digital twin module that implements virtual mapping and predictive simulation of floating roof status; an alarm management subsystem that supports multi-level warning threshold setting and intelligent push notifications; an operation and maintenance management module that records equipment status and maintenance history; a data visualization component that supports multi-dimensional data analysis and report generation; and a permission management unit that implements multi-level user access control.

[0031] It is worth mentioning that this embodiment describes in detail the technical implementation of the cloud-based analysis platform, including a time series database, a machine learning engine, a digital twin module, an alarm management subsystem, an operations and maintenance management module, a data visualization component, and a permissions management unit. The time series database adopts a distributed architecture, supporting petabyte-level data storage and high-speed queries; the machine learning engine integrates LSTM neural networks and random forest algorithms for anomaly detection and predictive maintenance; the digital twin module implements virtual mapping and predictive simulation of the floating roof state; the alarm management subsystem supports multi-level warning threshold setting and intelligent push notification; the operations and maintenance management module records equipment data throughout its lifecycle; the visualization component supports multi-dimensional data analysis; and the permissions management unit implements fine-grained access control.

[0032] The technical effects of the above solution include: this solution builds a complete cloud-based analysis system: the distributed time series database supports the writing of millions of data points per second; it can improve the accuracy of anomaly detection by machine learning algorithms; digital twin technology can realize the prediction of floating roof status within 72 hours; fine-grained permission management supports up to 6 levels of permission division.

[0033] Traditional technical solutions suffer from the following technical issues: 1. Unintuitive 3D posture display; 2. Difficulty analyzing multi-parameter correlations; 3. Alarm information overload; 4. Unable to meet mobile office needs; 5. Inadequate support for multilingual environments. These issues reduce the usability and practicality of monitoring systems.

[0034] Based on this, the visual human-computer interface includes: 3D floating roof attitude real-time display window, supporting multi-view switching and zooming; multi-parameter trend chart, supporting custom time range and curve overlay; alarm event list, displayed by priority; historical data query module, supporting multi-condition combination retrieval; system configuration interface for parameter setting and user management; Report export function supports PDF and Excel formats; mobile adaptation module supports access from mobile phones and tablets; voice alarm prompt function supports multi-language switching.

[0035] It's worth noting that this embodiment details the functional components of the visual human-machine interface, including a real-time 3D floating roof attitude display window, multi-parameter trend charts, an alarm event list, a historical data query module, a system configuration interface, a report export function, a mobile terminal adaptation module, and a voice alarm prompt function. The 3D display window supports multiple perspectives and zooming; the trend chart allows for customizable time ranges and curve overlay analysis; the alarm list is displayed by priority; historical queries support multi-condition combination retrieval; report exports are available in PDF and Excel formats; mobile adaptation ensures a smooth access experience on mobile phones and tablets; and the voice alarm supports multi-language switching.

[0036] The technical effects of the above scheme include: this scheme realizes through comprehensive visualization design: three-dimensional posture display makes the floating roof status clear at a glance; multi-parameter trend comparison facilitates fault tracing; hierarchical alarm management reduces the operational burden; mobile terminal support enables monitoring anytime and anywhere; multi-language interface meets international needs.

[0037] Traditional solutions, for example, face the following technical challenges: The reliable operation of tank monitoring systems faces multiple threats: lightning strikes; static electricity accumulation; power outages; communication failures; and equipment degradation. Traditional solutions often address each risk individually, lacking system-level protection.

[0038] Based on this, the system further includes: The lightning protection device is installed on the floating roof guide column and adopts a three-level lightning protection design; The static elimination device is arranged at the edge of the floating roof and adopts ion wind static elimination technology; The backup power module uses a lithium-ion battery pack with a battery life of no less than 8 hours; Explosion-proof junction box, protection level not less than IP66, in line with IECEx certification; Emergency communication module, supporting satellite communication links; Environmental monitoring unit, real-time monitoring of wind speed and rainfall in the tank area; Equipment health monitoring system, predictive maintenance of key components.

[0039] It's worth noting that this embodiment enhances the system's safety and reliability design, including lightning protection devices, an electrostatic eliminator, a backup power supply module, an explosion-proof junction box, an emergency communications module, an environmental monitoring unit, and an equipment health monitoring system. The lightning protection device utilizes a three-level protection design; the electrostatic eliminator utilizes ion wind technology; the backup power supply is a lithium-ion battery pack with an 8-hour battery life; the explosion-proof junction box complies with IECEx standards; emergency communications support satellite links; environmental monitoring covers wind speed and rainfall; and equipment health monitoring enables predictive maintenance.

[0040] The technical effects of the above scheme include: this scheme has built a comprehensive reliability assurance system: the three-level lightning protection design can withstand 100kA lightning current; ion wind static elimination keeps the surface potential below 50V; 3) the backup power supply ensures the critical 8-hour power supply; satellite communication ensures smooth communication in extreme situations; 5) equipment health monitoring reduces maintenance costs by 30%.

[0041] Traditional technical solutions present the following technical challenges: The main challenges facing industrial monitoring communication networks are: first, reliable transmission over long distances; second, multi-system interoperability; third, network security threats; and fourth, heterogeneous network convergence. Traditional solutions struggle to simultaneously meet real-time, reliability, and security requirements.

[0042] Based on this, the communication network of the system adopts: Field layer: Industrial Ethernet connects sensor arrays and edge computing units, supporting IEEE 802.3 standards; Transmission layer: The optical fiber ring network connects multiple tank monitoring systems and adopts a redundant dual-ring architecture; Cloud platform layer: 4G / 5G wireless communication connects to the cloud analysis platform and supports VPN encrypted transmission; Local backup channel: LoRa wireless communication is used as the emergency communication link; Network security protection: deploy industrial firewalls and intrusion detection systems; Data encryption mechanism: AES-256 encryption algorithm is used to protect transmitted data.

[0043] It's worth noting that this embodiment defines the system's communication network architecture, which is divided into three layers: the field layer, the transport layer, and the cloud platform layer. The field layer uses industrial Ethernet to connect sensors and edge computing units; the transport layer uses a fiber optic ring network to interconnect multiple storage tank systems; the cloud platform layer connects to the cloud via 4G / 5G wireless communication; the local backup channel uses LoRa wireless communication; network security deploys an industrial firewall and intrusion detection system; and data encryption uses the AES-256 algorithm.

[0044] The technical effects of the above solution include: This solution achieves 1ms real-time communication at the field level through a layered design; 99.999% availability of the fiber optic ring network; AES-256 encryption to ensure data security; and LoRa backup to ensure emergency communications.

[0045] See also Figure 2 Traditional solutions, for example, face the following technical challenges: The main technical challenges facing floating roof monitoring methods include: first, time synchronization issues with heterogeneous multi-source sensor data; second, insufficient state estimation accuracy in noisy environments; third, real-time processing requirements under complex working conditions; and fourth, the difficulty of accurately reconstructing 3D posture using traditional methods. These issues result in unreliable monitoring results, making it difficult to meet safety production requirements.

[0046] Based on this, this embodiment provides a three-dimensional multi-parameter real-time monitoring method for a crude oil storage tank floating roof, which is applied to any of the above-mentioned three-dimensional multi-parameter real-time monitoring systems for a crude oil storage tank floating roof, including the following steps: S1: Real-time acquisition of floating roof displacement, deformation, pressure and environmental parameters through a multimodal sensor array; S2: Preprocessing of multimodal sensor array data, including denoising, normalization, and time alignment; S3: Fusion of multi-source sensor data based on the Kalman filter algorithm to estimate the three-dimensional posture of the floating roof; S4: Calculate the floating roof safety status index and generate early warning information; S5: Visual display of monitoring results and early warning information; The Kalman filter algorithm used by S3 is expressed as: ; Where: is the prior state estimate at time k; is the a priori estimated covariance; is the Kalman gain; Input for the system; is the posterior state estimate; is the posterior estimated covariance; A is the state transfer matrix; B is the control input matrix; H is the observation matrix; Q is the process noise covariance; R is the observation noise covariance; is the observation value at time k.

[0047] The Kalman filter algorithm in this embodiment is the core mathematical tool for estimating the three-dimensional posture of the floating roof and is composed of five recursive equations: State prediction equation:

[0048] This equation is based on the state estimate of the previous moment and system input , predict the current state through the state transfer matrix A and the control input matrix B In floating roof monitoring, the state vector Typically, this matrix includes motion parameters such as the position, velocity, and acceleration of the floating roof. It has a 9×1 dimension and includes three degrees of freedom in three-dimensional space, corresponding to position, velocity, and acceleration. Matrix A is constructed based on the floating roof dynamics model, taking into account physical properties such as the roof's mass distribution and the effects of liquid damping. It is typically updated online every five minutes using system identification methods.

[0049] A priori covariance equation:

[0050] This equation updates the uncertainty of the state estimate, where P is the error covariance matrix and Q is the process noise covariance. The value of the Q matrix directly affects the dynamic response characteristics of the filter. In floating roof applications, it is determined statistically through long-term observation data. Its typical value is a diagonal matrix, with the diagonal elements reflecting the noise intensity of each state variable. Position noise is approximately 0.01m², and velocity noise is approximately 0.001(m / s)².

[0051] Kalman gain equation:

[0052] The Kalman gain, K, determines the weight of observations on state corrections. H is the observation matrix, mapping the state space to the observation space; R is the observation noise covariance. For multi-sensor systems, the H matrix structure is complex, considering the sensor installation location and measurement principle. The R matrix is typically obtained through static sensor testing. The R value for fiber optic sensors is approximately 1 με², while for MEMS accelerometers it is 0.01 (m / s²).

[0053] State update equation:

[0054] This equation is obtained by the actual observation Corrected forecast value, where It is called new information, which reflects the difference between observation and prediction. In floating roof monitoring, multi-source sensor data must first be time-aligned to ensure that each Time synchronization error <1ms.

[0055] Covariance update equation:

[0056] This equation updates the uncertainty of the posterior estimate, where I is the identity matrix. In practical implementations, improved algorithms such as the Joseph form or square root filtering are often used to prevent numerical instability.

[0057] The algorithm achieves the optimal estimation of the floating roof state through recursive calculation. The computational complexity is O(n³)n, where n is the state dimension. A single iteration takes less than 1ms on an embedded processor, meeting the real-time requirements.

[0058] It is worth mentioning that this embodiment proposes a three-dimensional multi-parameter real-time monitoring method for the floating roof of a crude oil storage tank, which includes five core steps: (1) multimodal sensor data acquisition; (2) data preprocessing, including denoising, normalization, and time alignment; (3) multi-source data fusion and three-dimensional posture estimation based on Kalman filtering; (4) safety status indicator calculation and warning generation; (5) visualization of monitoring results. Step (3) adopts an improved Kalman filtering algorithm to achieve optimal fusion of multi-sensor data through a state space model. The algorithm includes recursive processes such as state prediction, covariance update, Kalman gain calculation, and state correction.

[0059] The technical effects of the above solution include: the time alignment algorithm controls the synchronization error of multi-source data within 1ms; the improved Kalman filter reduces the state estimation error to 30% of the traditional method; the edge-cloud collaborative processing architecture ensures real-time performance; and the three-dimensional posture reconstruction accuracy reaches ±0.1 degrees.

[0060] Traditional solutions present the following technical challenges: Floating roof deformation assessment faces three technical difficulties: first, the difficulty in modeling the correlation between local deformation and overall safety; second, the varying contributions of deformation at different locations; and third, the lack of a quantitative method for representing deformation status. Traditional methods often rely on single-point threshold alarms, which fail to fully reflect the safety status of the floating roof.

[0061] Based on this, the floating roof safety status index includes the deformation index D, which is calculated as follows: ; Where: is the strain value of the i-th measurement point; is the initial strain reference value; is the maximum allowable strain of the material; N is the total number of measurement points.

[0062] The strain component in the above formula is:

[0063] Represents the real-time strain value of the i-th measurement point With the initial reference value Deviation of the initial benchmark It is an array of strain values recorded in the no-load stable state during the installation and commissioning phase of the floating roof. Usually, the average value of continuous 24-hour measurement is used as the benchmark. Measured by distributed fiber optic sensors with a sampling rate of 100 Hz and a resolution of 1 με, this difference reflects the deformation of the floating roof relative to its initial state, eliminating the influence of installation residual stress and temperature effects.

[0064] Normalization processing:

[0065] Normalize the strain deviation to a dimensionless quantity, where The maximum allowable strain of the floating roof material. For the commonly used floating roof steel Q235B, Take 80% of the yield strain, approximately 1200 με. Normalization makes measurement data at different locations and sensitivities comparable, while also controlling the indicator range between 0 and 1, making it easier to set a unified alarm threshold.

[0066] Square operation:

[0067] The squaring operation amplifies the contribution of significant deformations, increasing the index's sensitivity to large local deformations. Statistically, this is equivalent to calculating the mean square of the strain deviations, which better reflects the degree of deformation dispersion than a simple average. The squaring operation also ensures that all terms are positive, preventing positive and negative deformations from canceling each other out.

[0068] Average and square root:

[0069] The overall deformation index D is obtained by averaging the normalized square strains at all measurement points and taking the square root. N represents the number of valid measurement points; large floating roofs typically have 200-300 measurement points. The square root operation maintains a linear relationship between D and strain, facilitating engineering interpretation. D values range from 0 to 1, and in practice, three thresholds are set: D < 0.3 indicates safety, 0.3 ≤ D < 0.6 indicates warning, and D ≥ 0.6 indicates danger.

[0070] The index's innovation lies in combining distributed strain measurements into a single indicator that reflects both overall deformation trends and local anomalies. Field measurements have shown that when D > 0.5, the probability of failure of the floating roof sealing system increases eightfold, providing significant early warning value.

[0071] It is worth mentioning that this embodiment defines the calculation method of the deformation index D in the floating roof safety status assessment. This index uses the root mean square form to comprehensively evaluate the overall deformation degree of the floating roof by measuring the relative relationship between the strain value of the measurement point and the initial reference value and the maximum allowable strain of the material. represents the real-time strain value of the i-th measurement point, is the initial reference value, is the material limit strain, and N is the total number of measurement points. This index provides a quantitative assessment of the deformation state of the floating roof, with a value range of 0-1. The closer to 1, the higher the deformation risk.

[0072] The technical effects of the above scheme include: the innovative design of the deformation index D can achieve: quantitative evaluation of the overall deformation degree, integrating data from all measurement points; normalization processing makes the results comparable and unaffected by absolute dimensions; and root mean square calculation enhances sensitivity to significant deformation.

[0073] Traditional solutions present the following technical challenges: Floating roof pressure monitoring presents the following challenges: First, the non-uniformity of pressure distribution is difficult to quantify; second, the impact of pressure in different areas on floating roof safety varies; and third, traditional methods cannot dynamically reflect the changing trends of pressure distribution. These issues make it difficult to accurately and timely identify pressure anomalies.

[0074] Based on this, the pressure distribution evaluation adopts the dynamic pressure balance index , which is calculated as follows: ; Where: is the measurement value of the jth pressure sensor; is the average pressure value; is the weight coefficient of the jth sensor; M is the number of pressure sensors.

[0075] Pressure deviation term:

[0076] Represents the jth pressure sensor measurement value and average pressure value The pressure matrix is usually arranged in the area where the lower surface of the floating roof contacts the crude oil, with a measurement range of 0-10kPa and an accuracy of 0.1%FS. The calculation uses a truncated mean, excluding the highest and lowest 10% of data to improve interference resistance. The absolute deviation reflects the degree of deviation of the pressure at each point from the average state and is a direct measure of balance.

[0077] Weight coefficient:

[0078] Weights are assigned based on the importance of sensor locations, ranging from 0.8 to 1.2. The edge of the floating roof, prone to warping, is weighted 1.2; the center is weighted 1.0; and the transition zone is weighted 0.9. Weights are based on sensitivity coefficients determined through finite element analysis and are adjusted every two years based on floating roof maintenance data. Weighted deviation calculations highlight the contributions of critical areas.

[0079] Normalized denominator:

[0080] M is the number of effective pressure sensors. Large floating roofs usually have 50-80 pressure measurement points. Divide by the average pressure To achieve dimensionless It is applicable to different liquid levels and oil conditions. At the same time, normalization makes the index insensitive to absolute pressure changes, focusing on the evaluation of pressure distribution morphology.

[0081] Dynamic characteristics The index calculation uses a sliding time window, usually 5 minutes, and the data within the window is weighted exponentially, with new data having a higher weight. It can reflect the dynamic trend of pressure distribution and is particularly sensitive to slowly developing anomalies (such as floating roof tilt). When the current value exceeds 90% of the historical percentile, an alert is triggered.

[0082] The smaller the value, the more uniform the pressure distribution. The typical value range is 0-0.5. The operating data shows that when The failure rate of floating roof machinery increased 5 times after maintenance intervention. It can be reduced to below 0.1. This index provides a quantitative basis for the evaluation of the mechanical condition of the floating roof.

[0083] It is worth mentioning that this embodiment proposes a dynamic pressure balance index The calculation method is used to evaluate the balance of the floating roof pressure distribution. The index is obtained by calculating the weighted deviation of the measured value of each pressure sensor from the average pressure and dividing it by the average pressure to obtain a dimensionless index. is the jth sensor measurement value, is the average pressure is the weight coefficient, set according to the importance of sensor location, and M is the total number of sensors. This index reflects the degree of force balance on the floating roof. A smaller value indicates a more uniform pressure distribution.

[0084] The technical effects of the above scheme include: dynamic pressure balance index The innovation is reflected in: weighted calculation takes into account the importance differences of different areas; dimensionless design facilitates setting a unified threshold; and dynamic tracking of pressure distribution trends.

[0085] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A three-dimensional multi-parameter real-time monitoring system for floating roofs of crude oil storage tanks, characterized in that: include: Multimodal sensor array: arranged on the surface of the floating roof and the inner wall of the tank, used to collect real-time floating roof displacement, deformation, pressure and environmental parameters; Data acquisition and preprocessing module: connected to the sensor array for signal conditioning, analog-to-digital conversion and noise suppression; Edge computing unit: deployed at the tank site to process sensor data in real time and perform initial analysis; 3D dynamic modeling module: Builds a 3D attitude model of the floating roof based on multi-source sensor data; Cloud analysis platform: connected to edge computing units through the Industrial Internet of Things for in-depth data analysis and early warning decision-making; Visual human-machine interface: used to display monitoring results and alarm information; The multimodal sensor array comprises: Distributed optical fiber strain sensors are arranged along the radial and circumferential directions of the floating roof to measure the deformation of the floating roof; The three-axis MEMS acceleration sensor array is evenly distributed on the surface of the floating roof to measure the three-dimensional displacement of the floating roof; the pressure sensor matrix is arranged on the contact surface between the floating roof and crude oil to monitor the pressure distribution; the temperature and humidity sensor group is arranged in the space above the floating roof to monitor environmental parameters.

2. The crude oil storage tank floating roof three-dimensional multi-parameter real-time monitoring system according to claim 1 is characterized in that: The distributed optical fiber strain sensor uses a wavelength-modulated fiber Bragg grating sensor, which is connected to the data acquisition module via an armored flame-retardant single-mode optical cable. The optical cable has a protection level of no less than IP66 and an explosion-proof level of no less than ExdⅡBT4. The distributed optical fiber strain sensor adopts a redundant ring topology to ensure that a single point failure does not affect the overall monitoring function. The sensor is installed using a magnetic fixing bracket for easy maintenance and replacement. The optical cable connector uses a stainless steel waterproof connector with a protection level of IP68.

3. The three-dimensional multi-parameter real-time monitoring system for floating roof of crude oil storage tank according to claim 1 is characterized in that: The edge computing unit adopts an industrial-grade explosion-proof design, including: Multi-channel high-speed data acquisition card: sampling rate of not less than 1kHz, supporting 16-bit resolution; embedded processor: equipped with a real-time operating system and a hardware watchdog circuit; local storage module: using an industrial-grade SSD with a capacity of not less than 1TB; industrial communication interface: supporting Modbus TCP / IP and OPC UA protocols; Environmental monitoring module: real-time monitoring of temperature and humidity inside the chassis; redundant power input: supports 24VDC and 220VAC dual power supply; explosion-proof housing: complies with ATEX Zone 1 standards.

4. The crude oil storage tank floating roof three-dimensional multi-parameter real-time monitoring system according to claim 1 is characterized in that: The cloud-based analysis platform includes: a time series database with a distributed architecture that supports PB-level data storage; a machine learning engine that integrates LSTM neural networks and random forest algorithms; a digital twin module that implements virtual mapping and predictive simulation of floating roof status; an alarm management subsystem that supports multi-level warning threshold setting and intelligent push notifications; an operation and maintenance management module that records equipment status and maintenance history; a data visualization component that supports multi-dimensional data analysis and report generation; and a permission management unit that implements multi-level user access control.

5. The three-dimensional multi-parameter real-time monitoring system for floating roof of crude oil storage tank according to claim 1 is characterized in that: The visual human-machine interface includes: 3D floating roof attitude real-time display window, supporting multi-view switching and zooming; multi-parameter trend chart, supporting custom time range and curve overlay; alarm event list, displayed by priority; historical data query module, supporting multi-condition combination retrieval; system configuration interface for parameter setting and user management; Report export function supports PDF and Excel formats; mobile adaptation module supports access from mobile phones and tablets; voice alarm prompt function supports multi-language switching.

6. The crude oil storage tank floating roof three-dimensional multi-parameter real-time monitoring system according to claim 1 is characterized in that: The system further comprises: The lightning protection device is installed on the floating roof guide column and adopts a three-level lightning protection design; The static elimination device is arranged at the edge of the floating roof and adopts ion wind static elimination technology; The backup power module uses a lithium-ion battery pack with a battery life of no less than 8 hours; Explosion-proof junction box, protection level not less than IP66, in line with IECEx certification; Emergency communication module, supporting satellite communication links; Environmental monitoring unit, real-time monitoring of wind speed and rainfall in the tank area; Equipment health monitoring system, predictive maintenance of key components.

7. The crude oil storage tank floating roof three-dimensional multi-parameter real-time monitoring system according to claim 1 is characterized in that: The communication network of the system adopts: Field layer: Industrial Ethernet connects sensor arrays and edge computing units, supporting IEEE 802.3 standards; Transmission layer: The optical fiber ring network connects multiple tank monitoring systems and adopts a redundant dual-ring architecture; Cloud platform layer: 4G / 5G wireless communication connects to the cloud analysis platform and supports VPN encrypted transmission; Local backup channel: LoRa wireless communication is used as the emergency communication link; Network security protection: deploy industrial firewalls and intrusion detection systems; Data encryption mechanism: AES-256 encryption algorithm is used to protect transmitted data.

8. A three-dimensional multi-parameter real-time monitoring method for a crude oil tank floating roof, applied to a three-dimensional multi-parameter real-time monitoring system for a crude oil tank floating roof according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: Real-time acquisition of floating roof displacement, deformation, pressure and environmental parameters through a multimodal sensor array; S2: Preprocessing of multimodal sensor array data, including denoising, normalization, and time alignment; S3: Fusion of multi-source sensor data based on the Kalman filter algorithm to estimate the three-dimensional posture of the floating roof; S4: Calculate the floating roof safety status index and generate early warning information; S5: Visual display of monitoring results and early warning information; The Kalman filter algorithm used by S3 is expressed as: ; Where: is the prior state estimate at time k; is the a priori estimated covariance; is the Kalman gain; Input for the system; is the posterior state estimate; is the posterior estimated covariance; A is the state transfer matrix; B is the control input matrix; H is the observation matrix; Q is the process noise covariance; R is the observation noise covariance; is the observation value at time k.

9. The method for real-time three-dimensional multi-parameter monitoring of a crude oil storage tank floating roof according to claim 8, characterized in that: The floating roof safety status index includes the deformation index D, which is calculated as follows: ; Where: is the strain value of the i-th measurement point; is the initial strain reference value; is the maximum allowable strain of the material; N is the total number of measurement points.

10. The three-dimensional multi-parameter real-time monitoring method for a crude oil storage tank floating roof according to claim 8, characterized in that: The pressure distribution is evaluated using the dynamic pressure equilibrium index , which is calculated as follows: ; Where: is the measurement value of the jth pressure sensor; is the average pressure value; is the weight coefficient of the jth sensor; M is the number of pressure sensors.

Images