Online fault early warning method and system based on storage tank floating disc fluid dynamics and sealing device spectrum analysis thereof

Abstract

The invention provides an online fault early warning method and system based on fluid dynamics of a storage tank floating disc and spectral analysis of a sealing device of the storage tank floating disc, and the method comprises the steps: collecting structure parameters and operation parameters of the storage tank floating disc, and calculating fluid dynamics parameters for the quantitative evaluation of an inclination state, a buoyancy state and a friction state of the storage tank floating disc; collecting a plurality of vibration time domain signals of the sealing device on the floating disc, and converting the vibration time domain signals into frequency domain signals to obtain a frequency spectrum curve; constructing a frequency spectrum feature library based on the frequency spectrum curves in different operation states, comparing the actually monitored frequency spectrum curves with the frequency spectrum feature library, and identifying differentiated frequency spectrum features and corresponding fault types; and performing comprehensive evaluation by combining a quantitative evaluation result of the hydrodynamic parameters and a spectral analysis identification result, thereby realizing intelligent fault early warning of the storage tank floating tray system. The method solves the technical problems that the existing storage tank floating disc operation state monitoring cannot accurately find the early-stage potential fault hidden danger of the floating disc in real time, and floating disc faults and even safety accidents are easily caused.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of storage tank safety monitoring, and particularly relates to an online fault early warning method and system based on fluid dynamics of a storage tank floating disc and spectrum analysis of a sealing device thereof. BACKGROUND

[0002] Large storage tanks are key facilities for storing liquid media in the petroleum and chemical industries, and usually have a floating disc and a sealing device installed therein to reduce media volatilization and ensure storage safety. The floating disc floats up and down with the liquid level in the tank, and its running stability is directly related to the overall safety of the storage tank. However, in actual operation, due to factors such as media disturbance, tank deformation, sealing wear, and guide mechanism failure, the floating disc may tilt, be blocked, or even sink, and other faults. These faults not only cause economic losses, but also may cause major safety accidents such as fire and explosion due to media leakage, posing a serious threat to personnel safety and the environment.

[0003] At present, the monitoring means for the running state of the storage tank floating disc mainly relies on manual inspection or simple liquid level and pressure monitoring, and lacks real-time perception and quantitative evaluation of the dynamic behavior of the floating disc and the state of the sealing device. The traditional monitoring method cannot capture potential hazards such as small tilting changes of the floating disc, gradual wear of the sealing, and early failure of the guide roller bearing, and often can only be discovered after the fault occurs, and cannot achieve early warning.

[0004] Therefore, how to realize online, accurate, and early warning of the running state of the storage tank floating disc, so as to effectively improve the safety and reliability of the storage tank operation and prevent major accidents, is a technical problem to be solved in the field. SUMMARY

[0005] The purpose of the present application is to provide an online fault early warning method and system based on fluid dynamics of a storage tank floating disc and spectrum analysis of a sealing device thereof, which solves the technical problem that the existing storage tank floating disc running state monitoring cannot discover early potential fault hazards of the floating disc in real time and accurately, and easily causes floating disc faults and even safety accidents.

[0006] To solve the above technical problems, the technical scheme of the present application is as follows: In a first aspect, the present application provides an online fault early warning method based on fluid dynamics of a storage tank floating disc and spectrum analysis of a sealing device thereof, comprising the following steps: Collecting structure parameters and running parameters of the floating disc of the storage tank, and calculating relevant fluid dynamics parameters based on the structure parameters and running parameters, for quantitatively evaluating the tilting state, buoyancy state, and friction state of the floating disc; Collecting vibration time domain signals of the sealing device at multiple positions on the floating disc, and converting the time domain signals into frequency domain signals to obtain corresponding frequency spectrum curves; constructing a spectrum feature library based on spectrum curves under different operating states, the spectrum feature library at least including spectrum curves of a healthy state and a plurality of fault states, comparing an actually monitored spectrum curve with the spectrum feature library, identifying a differential spectrum feature and a corresponding fault type; comprehensively evaluating the quantitative evaluation result of the fluid dynamics parameter and the spectrum analysis identification result to realize intelligent fault early warning of the floating roof system of the storage tank.

[0007] Further, the fluid dynamics parameters include a plurality of parameters of a floating roof inclination angle, an inclination direction, a floating center offset distance, a restoring torque, total floating force, effective weight, total positive pressure and total friction.

[0008] Further, the floating roof inclination angle is calculated based on oil immersion depths of different positions of the floating roof, and the oil immersion depths of the different positions are continuously used to represent the inclination direction of the floating roof.

[0009] Further, before collecting the vibration time domain signals of the sealing device at a plurality of positions of the floating roof, the method further includes estimating a fault characteristic frequency range caused by a tank wall defect or a mechanical structure of the floating roof based on an operating speed of the floating roof, and setting a sampling frequency of the vibration signals based on the frequency range.

[0010] Further, collecting the vibration time domain signals of the sealing device at a plurality of positions of the floating roof specifically includes: installing vibration sensors on the sealing device to collect vibration signals of contact parts of the sealing device with the tank wall and obtain the time domain signals.

[0011] Further, the number of the vibration sensors is a plurality, which are installed on secondary sealing sliding sheets of the sealing device of the floating roof in contact with the tank wall, are arranged in a circumferential direction, and collect discrete time signals.

[0012] Further, the spectrum curves of the plurality of fault states at least include bearing fault spectrum, sealing local leakage spectrum and sealing resonance spectrum, and different fault types correspond to different spectrum curve characteristics.

[0013] Further, the fault early warning of the floating roof of the storage tank further includes combining a numerical range of the fluid dynamics parameter, a vibration signal collection position and a fault type identified by spectrum identification to realize hierarchical and positioning early warning.

[0014] In a second aspect, the application further provides an online fault early warning system based on fluid dynamics of a floating roof of a storage tank and spectrum analysis of a sealing device of the floating roof, which is used to implement the method of any one of the above aspects, and includes: A parameter acquisition and calculation module is configured to acquire structure parameters and operating parameters of the floating roof of the storage tank, and calculate relevant fluid dynamics parameters based on the structure parameters and the operating parameters, so as to quantitatively evaluate an inclination state, a floating force state and a friction state of the floating roof of the storage tank. The signal acquisition and processing module acquires vibration time domain signals of the contact part between the sealing device and the tank wall on the floating disc through a plurality of vibration sensors evenly arranged in the circumferential direction of the floating disc sealing device, and converts the time domain signals into frequency domain signals to obtain the frequency spectrum curves of the sensors.

[0015] The spectrum comparison and analysis module is used for comparing the actual monitored frequency spectrum curves with the preset spectrum feature library to identify the differential spectrum features and the corresponding fault types. The comprehensive evaluation and early warning module is used for comprehensively evaluating the quantitative evaluation results of the fluid dynamics parameters and the spectrum analysis identification results to realize intelligent fault early warning of the floating disc system of the storage tank.

[0016] Further, the fault early warning of the floating disc of the storage tank specifically includes the numerical range of the fluid dynamics parameters, the vibration signal acquisition position and the fault type identified by the spectrum identification, and realizes hierarchical and positioning early warning.

[0017] Compared with the prior art, the present application at least has the following beneficial effects: The present application calculates the fluid dynamics parameters such as the inclination angle of the floating disc, the restoring torque and the friction by acquiring the structure and operation parameters of the floating disc of the storage tank, realizes quantitative evaluation of the inclination, buoyancy and friction state of the floating disc, converts the operation state of the floating disc into digital indicators that can be accurately judged, can timely identify potential risks such as instability and high friction of the floating disc from the perspective of mechanics, provides a scientific quantitative basis for fault early warning, and makes up for the defect that the traditional monitoring method cannot accurately grasp the stress and motion state of the floating disc.

[0018] The present application evenly arranges vibration sensors in the circumferential direction of the floating disc sealing device, converts vibration time domain signals into frequency domain spectrum curves through discrete Fourier transform, constructs a spectrum feature library containing healthy and various fault states, realizes accurate identification of fault types such as bearing failure, sealing leakage and resonance, and can locate the fault occurrence direction by combining the sensor installation position, so that early faults of the floating disc sealing device can be captured in real time and accurately, and the pain points that the traditional monitoring method cannot find subtle problems such as sealing wear and roller bearing failure are solved.

[0019] The present application comprehensively evaluates the quantitative evaluation results of the fluid dynamics parameters and the fault identification results of the spectrum analysis, establishes a hierarchical and positioning fault early warning mechanism, can not only divide the risk level according to parameters such as the inclination angle, but also determine the fault type and position according to the spectrum features, realizes online, accurate and early warning of the floating disc system of the storage tank, can eliminate hidden dangers such as chuck, sink disc and sealing leakage in advance, effectively avoids safety accidents such as fire and explosion caused by medium leakage, greatly improves the safety and reliability of the operation of the large storage tank, and provides data support for the maintenance and design optimization of the floating disc of the storage tank. BRIEF DESCRIPTION OF DRAWINGS

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0021] Figure 1 The diagram shows the dynamic analysis of the influence of the inclination and oil immersion depth of the tank floating roof on the distance d between the center of gravity and the center of buoyancy and the restoring torque provided by the present invention. Figure 2 A mechanical analysis diagram for calculating the inclination of the tank floating roof and the frictional force between the seal and the tank wall provided by the present invention; Figure 3 The present invention provides time-domain waveforms and spectral curves of the health status of tank floating roofs; Figure 4 The present invention provides time-domain waveform and spectrum curve of tank floating roof bearing failure; Figure 5 The present invention provides time-domain waveform and spectrum curve of local leakage of the floating roof seal of the storage tank; Figure 6 The time-domain waveform and spectrum curve of the tank floating roof seal resonance provided by the present invention; Figure 7 The installation diagrams of the tank floating roof, sealing device and vibration sensor provided by the present invention are shown from different perspectives. Figure 8 for Figure 7 A schematic diagram of the sealing structure of the floating disc sealing device and the installation of the vibration sensor on it.

[0022] Figure label: 1-Column; 2-Sealing device; 3-Tank wall; 4-Top of the pan; 5-Sealed chamber; 6-Bottom of the pan; 7-L-shaped sliding plate; 8-Vibration sensor; 9-Elastic element; 10-Rubber toothed belt; 11-Pressure-bearing diaphragm assembly. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0025] Contents of each embodiment This embodiment provides an online fault early warning method based on the fluid dynamics of a storage tank floating roof and the spectrum analysis of its sealing device. Please refer to [reference needed]. Figures 1-8 As shown, this method first collects the structural parameters (such as diameter and mass) and operational parameters (such as oil immersion depth) of the floating roof of the storage tank, and calculates the hydrodynamic parameters such as the floating roof tilt angle, effective weight, normal pressure, and friction force, thereby quantitatively evaluating the tilt state, buoyancy state, and friction state of the floating roof. Simultaneously, vibration sensors are installed at multiple locations on the floating roof sealing device to collect the vibration time-domain signals of the contact points with the tank wall. The time-domain signals are converted into frequency-domain spectrum curves using Fourier transform. Then, a spectral feature library containing healthy states and various fault states (such as bearing failure, seal leakage, and resonance) is constructed. The measured spectrum is compared with the features in the library to identify differentiated spectral features and corresponding fault types. Finally, the hydrodynamic quantitative evaluation results are integrated with the spectral analysis results to achieve intelligent fault early warning for the storage tank floating roof system. This method, through the fusion analysis of mechanical and vibration signals, can detect potential problems such as jamming, sinking, and seal failure in advance, improving the accuracy and reliability of early warning.

[0026] In this embodiment, the types of fluid dynamic parameters are defined, including multiple parameters such as the floating roof tilt angle, tilt direction, center of buoyancy offset distance, restoring torque, total buoyancy, effective weight, normal force, and total friction. These parameters are clearly defined and calculated in the specific implementation details below: the tilt angle and center of buoyancy offset determine the restoring torque, reflecting the stability of the floating roof; the effective weight determines the floating roof's floating / sinking state; and the normal force and friction quantify the contact and friction between the floating roof and the tank wall. By calculating these parameters, the force and motion state of the floating roof can be comprehensively understood, providing a quantitative mechanical basis for subsequent fault early warning.

[0027] In this embodiment, the specific methods for obtaining the floating roof tilt angle and tilt direction are clarified: the tilt angle is calculated based on the oil immersion depth at different positions of the floating roof, and the tilt direction is represented by the line connecting the highest point (sinking point) and the lowest point of the oil immersion depth. Part III and Table 1 in the following detailed implementation indicate that by measuring the oil immersion depth at multiple points (e.g., h... a h b h c The tilt angle can be calculated. θ The straight line from the highest point to the lowest point is used as the direction of inclination. This digital representation transforms abstract posture into locatable fault information, making it easier for engineers to identify the specific location of problems such as tank bottom settling and localized scale buildup, thus improving maintenance efficiency.

[0028] In this embodiment, the key preprocessing step before collecting the vibration signal is emphasized: according to the running speed of the tank floating disc (such as 0.025 m / s), the fault characteristic frequency range (such as 25-50 Hz) caused by tank wall defects (such as tank wall welds and defects) or its own mechanical structure (such as bearings, column rollers) is estimated, and the sampling frequency of the vibration signal is set according to the range. The second part of the following specific implementation content calculates the defect impact frequency by example, and guides the selection of the sampling frequency according to the Shannon theorem. This step ensures that the sampling parameters cover the potential fault frequency band, avoids signal aliasing or omission, and lays the foundation for the accuracy of subsequent spectral analysis.

[0029] In this embodiment, the specific collection method of the vibration signal is described: the vibration sensor is installed on the floating disc sealing device, the vibration signal of the contact part of the sealing element and the tank wall is collected, and the time domain signal is obtained. The second part (section 5) of the following specific implementation content clearly installs the sensor on the L-shaped sliding sheet of the secondary sealing and the tank wall surface, so as to capture the vibration when the sealing element passes through the tank wall defect or the friction suddenly changes. This direct contact measurement can truly reflect the dynamic interaction between the sealing device and the tank wall, and is the core data source for identifying sealing wear, tank wall concave-convex, jamming and other faults.

[0030] In this embodiment, the layout of the sensor and the form of the signal are further limited: the vibration sensor is multiple, which is installed in a circumferential uniform manner on the secondary sealing sliding sheet, and the collected time domain signal is a discrete time signal. The following specific implementation content further points out that the number of sensors can be 8-16 sets according to the tank diameter, and the circumferential uniform distribution can comprehensively monitor the contact condition of the entire circumference; the discrete time signal is convenient for subsequent digital signal processing. Through the spectral difference of different direction sensors, the specific direction of fault occurrence (such as a certain place of tank wall protrusion) can be located, and precise fault positioning can be realized.

[0031] In this embodiment, typical fault types that should be included in the frequency spectrum feature library are listed, at least including bearing fault spectrum, sealing local leakage spectrum and sealing resonance spectrum, and different faults correspond to different spectral curve characteristics. The third section of the second part of the following specific implementation content describes the spectral morphology (curves 2-4) of the three faults in detail: bearing fault shows equidistant peaks; sealing leakage shows high-frequency wide-frequency "hump"; sealing resonance shows isolated peaks at a certain frequency. These typical spectra are included in the feature library, and when similar characteristics are found in the measured spectrum, the fault type can be quickly identified, and the diagnosis efficiency is improved.

[0032] This embodiment proposes a graded and location-based early warning mechanism: combining the numerical range of fluid dynamic parameters (such as tilt angle threshold), vibration signal acquisition location (sensor circumferential distribution), and fault type identified by spectrum, risk grading and location positioning are achieved. Table 1 in the following implementation details classifies risks into low, medium, and high risk levels based on tilt angle (e.g., θ≤0.5° is low risk), and the fault point can be determined by combining this with the sensor's orientation. This early warning method provides more accurate risk indications, guiding maintenance personnel to take targeted measures, avoiding blind troubleshooting, and effectively preventing accidents.

[0033] This embodiment also provides an online fault early warning system based on the fluid dynamics of the tank floating roof and the spectrum analysis of its sealing device. Please refer to [link / reference]. Figures 1-8 As shown, a fault early warning system for implementing any of the aforementioned fault early warning methods includes the following functional modules: a parameter acquisition and calculation module (acquiring structural / operating parameters and calculating fluid dynamic parameters), a signal acquisition and processing module (acquiring time-domain signals through circumferentially distributed vibration sensors and converting them into frequency-domain spectra), a spectrum comparison and analysis module (comparing measured spectra with a preset feature library to identify fault types), and a comprehensive evaluation and early warning module (combining fluid dynamic quantification results and spectrum identification results to achieve fault early warning). This system has a clear structure, can automatically execute the early warning process, realize real-time monitoring and intelligent diagnosis of the floating roof status of storage tanks, and improve the operational safety of storage tanks.

[0034] In this embodiment, the function of the early warning system is further defined, namely, the system can achieve graded and location-based early warning: Risk levels are classified based on the numerical range of fluid dynamic parameters (such as tilt angle thresholds; other thresholds, such as buoyancy and friction thresholds, can be determined according to actual structural and operational parameters), and combined with the vibration signal acquisition location (sensor orientation) and the fault type identified by the spectrum, the system outputs early warning information with specific location and risk level. The tilt angle classification, tilt direction positioning, and spectrum fault type identification described in the following specific implementation details provide the foundation for this function. This function makes the early warning results more operable, helping on-site personnel quickly locate problems and take corresponding measures, effectively preventing the escalation of accidents.

[0035] Specific implementation details The online fault early warning method and system based on the fluid dynamics of the floating roof of a storage tank and the spectral analysis of its sealing device are implemented through the following core technical solutions, as detailed below: I. Dynamic Analysis and Calculation of Floating Roof for Storage Tanks (I) Explanation of the names, symbols, algorithms, and physical meanings of the fluid dynamics of the floating roof of the storage tank Combination Figure 1 (The dynamic analysis diagram of the floating roof's inclination and oil immersion depth relative to the distance d between the center of gravity and the center of buoyancy, and the buoyancy restoring moment is shown below:) Among them, floating roof tilt =2.3° (example); Floating roof tilt angle The larger the diameter, the greater the degree of buoyancy center offset; the larger the diameter D of the floating roof, the greater the maximum distance the buoyancy center may offset under the same tilt angle; the greater the oil immersion depth, the greater the "weight" of the point, which will pull the buoyancy towards itself (see arrow ← in the figure); YM is the liquid surface of the storage tank; DM is the bottom surface of the floating roof; the buoyancy center CB is the geometric center of the submerged volume below the tilt plane; The specific content involved also includes: 1. CG: Center of Gravity of the floating platform, which is the center of mass of the floating platform. It is manufactured according to standards and the materials are installed evenly. Generally, the center of gravity of the floating platform is located at the center of the floating platform.

[0036] 2. CB: The center of buoyancy of the floating roof (abbreviated as buoyancy center), which is the center of the volume of liquid displaced by the floating roof.

[0037] 3.d: The straight-line distance between the center of buoyancy CB and the center of gravity CG of the floating platform projected vertically onto the horizontal plane.

[0038] Calculation formula: d = , where: k: buoyancy center offset coefficient (0<k≤1); It is a coefficient related to the bottom shape and oil immersion depth of the floating roof, and can be approximated as 1 for large-area flat-plate floating roofs. The closer the k value is to 1, the more likely the center of buoyancy is to deviate from the edge.

[0039] 4.D: is the diameter of the floating roof.

[0040] 5.M: Restoring torque is the torque generated by buoyancy around the center of gravity of the disk, used to resist tilting.

[0041] Right now Its physical meaning is as follows: M > 0: Stable state, there exists a torque that restores the floating roof to horizontal position. The larger the M value, the better the stability and the stronger the resistance to wind and oil flow.

[0042] M→0: In a state of equilibrium (dangerous state), the floating roof loses its ability to recover and is at the critical point of instability, which is a direct precursor to jamming or sinking.

[0043] M < 0: The floating roof is in a capsized state and cannot recover on its own, so an accident is inevitable.

[0044] 6. Fb: Total buoyancy Calculation formula: Total buoyancy ; in, Density (kg / ); ρ is the acceleration; h is the average oil immersion depth; A is the bottom area of ​​the disk, A= .

[0045] 7. Weff= Mo - W , where Weff is the actual weight of the floating disc in operation, where Mo is the mass of the floating disc itself (kg); the physical meaning is as follows: Weff>0: the floating disc sinks, the weight is supported by the tank bottom, and the lateral pressure on the tank wall mainly comes from the inclination of the floating disc.

[0046] Weff<0: the floating disc floats, the weight is supported by the buoyancy, and contact with the tank top structure may occur.

[0047] Weff=0: the floating disc is suspended, ideal state.

[0048] 8. N: positive pressure; Calculation formula: positive pressure N=Ntilt+Nmech= +Nmech; Where the positive pressure N is composed of two parts: (1) Lateral force Ntilt caused by inclination: after the floating disc is inclined, its effective weight is no longer vertical, and a lateral force parallel to the inclined surface will be generated, which will squeeze the tank wall or guide column.

[0049] Ntilt= , where is the inclination angle.

[0050] (2) Mechanical contact force Nmech: even if the floating disc is horizontal, initial contact force with the tank wall may occur due to installation, thermal deformation, tank body out-of-roundness, etc., which needs to be estimated by experience.

[0051] (2) Calculation of the friction between the floating disc of the storage tank and the tank wall Combined Figure 2 (Analysis of the inclination of the floating disc of the storage tank and the calculation of the friction between the floating disc and the tank wall) as follows: Where the inclination of the floating disc is 1.5° (example); two columns (symmetric); sealing device 2 (combination of butyronitrile rubber and polyurethane stainless steel sheet); Ntilt (N1) = , where is the inclination angle, and N1 is the force (N1) squeezing the tank wall or guide column, as shown by the arrow←in the figure; Where the involved content specifically includes: 1. Ff: total friction Calculation formula: Ff= = (Ntilt+Nmech); Total normal force: N = Ntilt (N1) + Nmech (N2); Where Ntilt (N1) = N1 * cos (a) Where Nmech (N2) = N2 * sin (a) N1 is the force (N) of the extruded tank wall or guide column at the tilt angle a; where Weff is the actual weight of the floating disc, which is always changing, related to the buoyancy and leakage in the floating disc, and also related to the upward and downward directions (+, -), here Weff is considered as an absolute value.

[0052] Nmech (N2) = N2 * sin (a) Where Nmech (N2) = N2 * sin (a); where η is the coefficient for empirical estimation of mechanical contact force due to installation deviation, thermal deformation, tank body out-of-roundness, tank wall surface roughness and other factors.

[0053] (1) New tank with good quality, tank standard parameter checking meets the standard, = 10% ~ 20%.

[0054] (2) General old tank, with slight deformation, running time record and on-site light and camera data and tank standard parameter checking meet the requirements, = 20% ~ 50%.

[0055] (3) Old tank, with obvious oval or local concave and convex and rough, running time record meets the running requirements, = 50% ~ 100%.

[0056] Where, = 0.2 ~ 0.6 (friction coefficient between nitrile rubber and polyurethane and tank wall).

[0057] 2. The contact between the floating disc and the tank wall can be evaluated according to the time domain conversion frequency domain spectrum analysis of the installed vibration sensor signal, and a spectrum library of different faults and changes is established.

[0058] II. Time domain conversion frequency domain spectrum analysis of sealing device fault early warning (1) Core concept and parameter definition of floating disc sealing device spectrum analysis 1. Basic data of tank outer floating disc (1) Calculate the linear speed of the tank outer floating disc rising and falling: 1.5 meters / minute = 0.025 meters / second This is a very typical low-speed motion scenario.

[0059] (2) Estimate the potential characteristic frequency of the floating disc running: Assume that there is a local protrusion or depression (for example, a 1mm high weld or corrosion pit) on the tank wall. When the floating disc seal passes this defect, a "collision" or "friction jump" will occur.

[0060] Impact frequency (Hz) = Linear velocity (m / s) / Defect characteristic length (m); Take a 1mm (0.001m) defect as an example: Frequency = 0.025 / 0.001 = 25 Hz; If the defect is smaller (e.g. 0.5mm), the frequency can reach 50 Hz.

[0061] If there is ovality or out-of-roundness in the tank wall, the operation of the floating disc will produce periodic changes in friction force, and the frequency related to the speed of the floating disc itself will be lower (usually below 1 Hz).

[0062] Conclusion: The estimated 50Hz is completely within the reasonable fault characteristic frequency range. This means that the main energy of abnormal vibration signals caused by micro or macro defects in the tank wall is likely to be distributed in the frequency range of tens of hertz (Hz) to possibly hundreds of hertz. This frequency range is the mainstream frequency band that can be accurately captured and measured by modern vibration sensors and analysis instruments, and is technically mature.

[0063] Install vibration sensors on the elastic elements of the secondary seal of the floating disc sealing device in contact with the tank wall, and the number of vibration sensors is determined according to the size of the storage tank diameter: 8-16 sets of vibration sensors, which are installed on the secondary seal L-shaped sliding sheet (butyl rubber material) of the floating disc sealing device in contact with the tank wall.

[0064] 2. Data and collection of vibration signals (1) Time domain signal of vibration sensor: x[n] n is the sample point sequence number, n = 0, 1, 2,..., N-1; N is the total number of sampling points (i.e. data length); x[n] is the amplitude value obtained at the nth sampling time (such as the secondary seal surface of the floating disc sealing device, the floating disc anti-rotation column roller, the bearing, the acceleration g value, etc.).

[0065] (2) Sampling frequency: Fs (Hz) The number of data points collected per second. For example, Fs = 1000 Hz means 1000 samples per second.

[0066] (3) Sampling interval: Δt = 1 / Fs (seconds) The time difference between adjacent two sampling points.

[0067] (4) Total signal duration: T = N * Δt = N / Fs (seconds) 3. Discrete Fourier Transform (DFT) algorithm steps for tank floating disc seal vibration signal DFT is the bridge connecting time domain and frequency domain. Its definition is as follows: DFT forward transform formula (time domain → frequency domain) Formula: X[k] = Σ_{n=0}^{N-1} x[n]* e^{-j*(2π / N)*k*n}, where k = 0, 1, 2,..., N-1; Step-by-step interpretation: X[k]: This is the kth frequency domain component after transformation. It is a complex number containing the amplitude and phase information of that frequency component. k is also called the frequency index.

[0068] e^{-jθ}: This is the complex exponential function, according to Euler's formula e^{-jθ} = cos(θ) - j*sin(θ). It is the "basis waveform" that constitutes all frequency components.

[0069] (2π / N)*k*n: This angular part is crucial. For a fixed k, it defines a complex sinusoidal wave with a frequency of k cycles per record length. The essence of DFT is to project the time domain signal x[n] onto this series of orthogonal basis functions to get the correlation coefficients X[k].

[0070] 4. Calculation of frequency axis (horizontal coordinate) The actual physical frequency f_k corresponding to X[k] is given by: Formula: f_k = k * (Fs / N) (Hz) k = 0: Corresponds to the DC component (frequency 0 Hz), i.e. the mean value of the signal.

[0071] k = 1: Corresponds to the fundamental frequency f1 = Fs / N, which is the lowest non-zero frequency that can be distinguished in the spectrum, also known as the frequency resolution.

[0072] k = N / 2: Corresponds to the Nyquist frequency f_{Nyquist} = Fs / 2. This is the highest frequency that can be displayed in the spectrum. The part with k>N / 2 is conjugate symmetric with the first half, usually only the part with k = 0 to N / 2 is analyzed.

[0073] Frequency resolution Δf formula: Δf = Fs / N = 1 / T (Hz) Conclusion: The longer the total signal collection time T, the higher the frequency resolution Δf, and the more detailed the spectrum.

[0074] 5. Calculation of spectral amplitude (vertical coordinate) Of interest is usually the amplitude spectrum, which represents the strength of each frequency component.

[0075] Formula: Amp[k] = |X[k]| / N (for k = 1, 2,..., N / 2 -1) Amp[0] = |X[0]| / N (DC component) Amp[N / 2] = |X[N / 2]| / N (Nyquist frequency component) |X[k]|: represents taking the magnitude of complex number X[k]. |a + jb| = sqrt(a² + b²).

[0076] N: this is to normalize the amplitude so that it has a true physical meaning (depends on sensor calibration). 2 / N can also be used to scale the one-sided spectrum amplitude in some implementations.

[0077] For vibration analysis applications, what is usually plotted is Amp[k] vs. f_k (k = 0 to N / 2), which is the "amplitude-frequency characteristic curve", i.e. the frequency spectrum.

[0078] (II) In order to convert the time domain of the tank floating seal device into the frequency spectrum algorithm process can be convenient and clear understanding, first using the conventional calculation or a manual calculation demonstration (N = 4) Use a simple example to practice DFT: 1. Time domain data (assuming the unit has been calibrated): x[0] = 0, x[1]= 1, x[2] = 0, x[3]= -1, N = 4 2. Calculate the rotation factor It is a mathematical rotation factor in DFT / FFT algorithm, which is used to generate a series of "standard ruler" (complex sine wave) of different frequencies to measure the strength of each frequency component in the signal. He is the core of the analysis tool.

[0079] The linear motion of the floating plate itself does not produce periodic vibration, but the mechanical structure (bearings, column rollers, welds and defects of the tank wall) and the running of the in-and-out tank medium flow pump and the inclination of the floating plate will introduce periodic excitation. The frequency spectrum analysis (using W (N)) detects the periodic failure rate hidden in the linear motion parameters.

[0080] W_N = e^{-j*(2π / N)} = e^{-j*(π / 2)} = cos(π / 2) - j*sin(π / 2) = -j W_N^{k*n} constitutes a table for simplified calculation:

[0081] 0 W 0 =1 W 0 =1 W 0 =1 W 0 =1 1 W 0 =1 W¹=-j W²=-1 W³=j 2 W 0 =1 W²=-1 W 0 =1 W²=-1 3 W 0 =1 W³=j W²=-1 W¹=-j 3. Calculate DFT coefficients X[k]: X[0] = 0*1 + 1*1 + 0*1 + (-1)*1 = 0 X[1] = 0*1 + 1*(-j) + 0*(-1) + (-1)*j = -j - j = -2j X[2] = 0*1 + 1*(-1) + 0*1 + (-1)*(-1) = -1 + 1 = 0 X[3] = 0*1 + 1*j + 0*(-1) + (-1)*(-j) = j + j = 2j (This is the conjugate complex number of X[1]) 4. Calculate the amplitude spectrum: Assume Fs = 4 Hz, then T = N / Fs = 1 second, Δf = Fs / N = 1 Hz, f_{Nyquist} =Fs / 2 = 2 Hz.

[0082] Amp[0] = |0| / 4 = 0 (corresponding to f=0 Hz) Amp[1] = |-2j| / 4 = 2 / 4 = 0.5 (corresponding to f=1 Hz) Amp[2] = |0| / 4 = 0 (corresponding to f=2 Hz) (Usually only take k=N / 2, that is, k=2) 5. Get the frequency spectrum data points (in table form): Frequency index k Physical frequency f_k (Hz) Complex spectrum X[k] Amplitude spectrum Amp[k] 0 0.0 0 0.0 1 1.0 -2j 0.5 2 2.0 0 0.0 (Three) Spectrum curve features and faults of tank floating roof seal device (conceptual description) The following is a conceptual spectrum curve text description based on floating roof monitoring scenarios. Illustrations can be drawn based on these descriptions.

[0083] X-axis (horizontal): Frequency (Hz), ranging from 0 to Fs / 2.

[0084] Y-axis (vertical): Amplitude (e.g., vibration acceleration g or dimensionless normalized value).

[0085] Curve 1: Healthy state (baseline) spectrum Shape description: Curve close to X-axis, appearing as a low-amplitude "noise band". There may be a small uplift in the very low frequency region (<10 Hz, corresponding to the overall movement of the floating roof), but the amplitude is close to zero in most frequency bands (>50 Hz). No obvious, sharp isolated peaks.

[0086] Text data point illustration (example): f(Hz): 10, 50, 100, 200, 500 Amp: 0.05, 0.02, 0.01, 0.05, 0.02 Curve 2: Bearing fault (e.g., floating roof anti-rotation column roller pitting) spectrum Shape description: One or more equidistant, sharp spectral peaks appear in the low frequency band. For example, a high peak appears at f_bpfo = 25 Hz (floating roof column bearing outer ring fault characteristic frequency), and amplitude-decreasing peaks appear at its harmonics 50 Hz, 75 Hz, 100 Hz... The overall background noise may be slightly higher than the baseline.

[0087] Text data point illustration (highlighting peaks): f(Hz):... 20, 25, 30, 48, 50, 52, 73, 75, 77... Amp:... 0.02, 0.5, 0.02, 0.01, 0.3, 0.01, 0.005, 0.15, 0.005... Curve 3: Local leakage of tank floating roof seal (flow-induced vibration) spectrum Shape description: A broad band of "humps" or a group of closely adjacent peaks, resembling "mountain passes", appears in the mid-high frequency region (e.g. 800 Hz - 3000 Hz). The energy in this band is significantly higher than the healthy baseline. It can be accompanied by random spikes at higher frequencies.

[0088] Textual data point illustration: f (Hz): 500, 800, 1200, 1600, 2000, 2500, 3000, 3500 Amp: 0.01, 0.05, 0.15, 0.25, 0.20, 0.12, 0.04, 0.01 Curve 4: Resonant frequency spectrum of tank floating roof seal Shape description: At a certain specific frequency f_r (e.g. 45 Hz), an extremely sharp, abnormally high amplitude isolated spectral peak appears. This peak can be 1-2 orders of magnitude higher than the amplitude of the surrounding frequencies. Near f_r, the curve rises sharply and reaches a peak, then drops rapidly.

[0089] Textual data point illustration: f (Hz): 40, 42, 44, 45, 46, 48, 50 Amp: 0.1, 0.3, 0.8, 3.5, 0.7, 0.2, 0.1 (4) Key calculations in application (non-code) 1. Determine analysis parameters: According to the highest fault frequency of interest (e.g. floating roof two column roller engagement, leakage sound, seal), set the sampling frequency Fs. Must satisfy Fs>2*f_max (Shannon theorem).

[0090] According to the required frequency resolution Δf (e.g. to distinguish two fault frequencies with an interval of 1 Hz), determine the minimum sampling time T_min = 1 / Δf, and thus the minimum number of sampling points N_min = Fs / Δf.

[0091] 2. Frequency spectrum preprocessing (before FFT): Remove DC: x'[n] = x[n]- mean(x). Prevent DC component from drowning out low-frequency information.

[0092] Windowing (e.g. Hanning window): x_w[n] = x'[n]* w[n]. w[n] is the window function coefficient table, used to reduce spectral leakage. For example, Hanning window coefficients: w[n] = 0.5 * (1 - cos(2πn / (N-1))).

[0093] 3. Theoretical calculation of fault characteristic frequency: Floating disc anti-rotation column bearing fault frequency: The outer ring, inner ring, and rolling body fault frequencies need to be calculated according to the bearing model (pitch diameter, number of rollers, contact angle) and rotational speed (converted from floating disc lifting speed v and lead L: rotational speed = v / L). These formulas can be found in the bearing manual of the tank floating disc anti-rotation column.

[0094] Passing frequency: the frequency at which the floating disc passes through the tank wall defect f_pass = v / P, where v is the linear speed (m / s) and P is the defect spacing (m).

[0095] Through the above formulas, steps, and curve descriptions of the spectrum establishment and analysis process of the tank floating disc sealing device, the conversion logic from time domain signal to frequency domain spectrum can be fully understood, and the mechanical state of the secondary seal of the floating disc sealing device and the tank wall can be theoretically analyzed and fault warned according to the shape characteristics of the spectrum curve. In actual engineering, Fast Fourier Transform (FFT) is a high-efficiency algorithm for performing DFT, but its mathematical essence is exactly the same as the above DFT.

[0096] (Five) Installation of secondary seal vibration sensor of floating disc sealing device Combined with Figure 7 (Floating disc, sealing device, and vibration sensor installation schematic diagram from different perspectives) and Figure 8 (Floating disc sealing device sealing structure and vibration sensor installation schematic diagram (partial enlarged detail) in Figure 7 are as follows: Figure 7 The top view shows that there are 8 vibration sensors ZC01-ZC08; the cross-sectional view shows the tank wall 3, the disc top 4, the sealing cabin 5, and the disc bottom 6.

[0097] Figure 8 The L-shaped sliding plate 7 (as a secondary seal, made of nitrile rubber material) is movably arranged on the tank wall, the vibration sensor 8, the primary sealing structure includes the elastic element 9 (made of soft polyurethane material) and the rubber toothed belt 10 (made of nitrile rubber material) arranged integrally; the oil immersion depth h is 200mm (example).

[0098] Three, comprehensive evaluation of tank floating disc fluid mechanics and sealing device spectrum analysis fault warning 1. Inclination angle of tank floating disc The inclination direction of the floating disc is a state that indicates the up and down operation of the floating disc. It is whether the floating disc is in a normal safe range or an unsafe range, guiding the operation. And it is used as the inclination angle of the floating disc The digital dimension of the operating parameters is measured, while the inclination direction of the floating disc and the oil immersion depth 、 、 As the address and part interval of faults and hidden dangers, it is convenient for engineering personnel to handle faults.

[0099] 2. The effective weight of the floating disc The sinking, floating and suspension of the floating disc can be calculated by numbers, and the state can be explained and characterized by numbers.

[0100] 3. The positive pressure N can be calculated according to the friction coefficient between the L-shaped sealing scraper material (nitrile rubber) of the sealing device and the tank wall The detailed friction force can be calculated. The calculation process can analyze the reliability and integrity of the sealing device, find the cause of damage of the sealing device and optimize it, and provide basis for design optimization.

[0101] 4. The operation of the floating disc can be analyzed according to the time domain signal of 8-16 sets of vibration sensors installed on the circumference of the floating disc, and the time domain signal can be quickly converted into frequency domain signal by Fourier conversion formula and quickly get the frequency spectrum curve of frequency domain, and through special analysis software, the specific fault and hidden danger type of the floating disc can be found.

[0102] 5. Regarding the frequency spectrum analysis of time domain to frequency domain, at present, it is applied in rotating devices (for example: wind power generation) all over the world, and it is periodic motion. The invention is the linear motion of the floating disc rising and falling, and it is very regular linear motion. The mathematical rotation factor of DFT / FFT in the algorithm is used to generate a series of different frequency "standard ruler" (complex sine wave), so as to measure the strength of each frequency component in the signal. It is the core of analysis tool, the linear motion of the floating disc itself does not produce periodic vibration, but the mechanical structure (bearing, anti-rotation column roller of the floating disc, weld and defect of the tank wall) and the medium flow pump operation and the floating disc inclination in the movement process and the tank will introduce periodic excitation. The spectrum analysis (using W (N)) detects these implicit, periodic failure rates determined by the linear motion parameters through the floating disc fluid dynamics calculation. This periodic failure rate can be verified and digitized by the calculation formula of the floating disc fluid dynamics, get the support of big data AI, and get the mathematical theory verification of the establishment of frequency spectrum library.

[0103] 6. The storage tank of combustible medium, the volatile gas of the medium, due to the seal damage of the floating plate sealing device, the leakage, escape, encounter lightning, static ignition can cause fire, flash explosion, which is one of the key prevention accident types, also one of the important problems for engineers and researchers to overcome. It is the effectiveness of the actual coverage of emergency plans (early warning) and on-site disposal schemes, which can prevent the leakage of flammable liquids and volatile VOCs, seal ring fire, full surface fire, large area flowing fire, floating plate tilt, floating plate sinking, medium boiling overflow, gas phase space flash explosion and other tank fire and explosion. The invention can early warning the failure, explosion, fire and other safety accidents by the calculation of the fluid mechanics platform of the floating plate and the analysis of the vibration spectrum of each important part, so as to eliminate the failure and accident before it occurs.

[0104] See Table 1 below for details: Table 1: Comprehensive evaluation of various failure early warning tables obtained by floating plate fluid mechanics algorithm and frequency spectrum analysis of sealing device between tank wall

[0105] 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 foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An online fault warning method based on the spectrum analysis of the fluid dynamics of a tank's floating plate and its sealing device, characterized in that, The method comprises the following steps: Collecting the structural parameters and operating parameters of the floating roof of the storage tank, and calculating relevant fluid dynamics parameters based on the structural parameters and operating parameters, for quantitatively evaluating the inclination state, buoyancy state and friction state of the floating roof of the storage tank; Collecting vibration time domain signals at multiple positions of the sealing device on the floating roof, and converting the time domain signals into frequency domain signals to obtain corresponding frequency spectrum curves; Building a frequency spectrum feature library based on the frequency spectrum curves under different operating states, the frequency spectrum feature library at least including frequency spectrum curves of a healthy state and multiple fault states, comparing the actually monitored frequency spectrum curves with the frequency spectrum feature library, and identifying differential frequency spectrum features and corresponding fault types; Combining the quantitative evaluation results of the fluid dynamics parameters with the identification results of the frequency spectrum analysis to comprehensively evaluate, so as to realize intelligent fault early warning of the floating roof system of the storage tank.

2. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 1, characterized in that, The fluid dynamics parameters include multiple parameters such as the inclination angle of the floating roof, the inclination direction, the offset distance of the floating center, the restoring torque, the total buoyancy, the effective weight, the total positive pressure and the total friction.

3. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 2, characterized in that, The inclination angle of the floating roof is calculated based on the oil immersion depths at different positions of the floating roof, and the oil immersion depths at different positions are continuously used to represent the inclination direction of the floating roof.

4. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 1, characterized in that, Before collecting the vibration time domain signals at multiple positions of the sealing device on the floating roof, it further comprises estimating the fault characteristic frequency range caused by the defects of the tank wall or the mechanical structure of the floating roof according to the operating speed of the floating roof of the storage tank, and setting the sampling frequency of the vibration signals based on the frequency range.

5. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 1, characterized in that, Collecting the vibration time domain signals at multiple positions of the sealing device on the floating roof specifically comprises installing vibration sensors on the sealing device to collect the vibration signals of the contact parts between the sealing device and the tank wall and to obtain the time domain signals.

6. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device, according to claim 5, characterized in that, The number of the vibration sensors is multiple, which are installed on the secondary sealing sliding sheets of the sealing device of the floating roof and in contact with the tank wall, are arranged in a circumferential uniform manner, and the collected vibration time domain signals are discrete time signals.

7. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 1, characterized in that, The frequency spectrum curves of the multiple fault states at least include bearing fault spectrum, sealing local leakage spectrum and sealing resonance spectrum, and different fault types correspond to different frequency spectrum curve characteristics.

8. The online fault warning method based on the spectrum analysis of the fluid dynamics of the tank's floating disc and its sealing device according to claim 1, characterized in that, The fault early warning of the floating roof of the storage tank further comprises combining the numerical range of the fluid dynamics parameters, the vibration signal collection positions and the fault types identified by the frequency spectrum to realize hierarchical and positioning early warning.

9. An online failure warning system based on the spectral analysis of the fluid dynamics of a tank's floating roof and its sealing devices, for implementing the method according to any one of claims 1 to 8, characterized in that, It comprises: A parameter collection and calculation module for collecting the structural parameters and operating parameters of the floating roof of the storage tank, and calculating relevant fluid dynamics parameters based on the structural parameters and operating parameters, for quantitatively evaluating the inclination state, buoyancy state and friction state of the floating roof of the storage tank; A signal collection and processing module for collecting vibration time domain signals at contact parts between the sealing device of the floating roof and the tank wall through multiple vibration sensors installed in a circumferential uniform manner on the sealing device of the floating roof, and converting the time domain signals into frequency domain signals to obtain frequency spectrum curves of each sensor; A frequency spectrum comparison and analysis module for comparing the actually monitored frequency spectrum curves with a preset frequency spectrum feature library, identifying differential frequency spectrum features and corresponding fault types; A comprehensive evaluation and early warning module for comprehensively evaluating the quantitative evaluation results of the fluid dynamics parameters and the identification results of the frequency spectrum analysis, so as to realize intelligent fault early warning of the floating roof system of the storage tank.

10. The online failure warning system based on the fluid dynamics of the tank's floating disc and the spectrum analysis of its sealing device, according to claim 9, characterized in that, The failure early warning of the tank floating disc specifically includes a failure type combined with a numerical range of fluid dynamics parameters, a vibration signal collection position and a frequency spectrum identification, thereby realizing hierarchical and positioning early warning.

Images