Internal pressure non-contact measurement method and device for thin-walled cylindrical pressure vessel

Abstract

The invention relates to the technical field of pressure measurement, in particular to an internal pressure non-contact measurement method and device for a thin-walled cylindrical pressure vessel, and the method comprises the steps: setting axial strain measurement points in the circumferential direction of the outer wall of the pressure vessel; synchronously acquiring a real-time axial strain measurement value generated at each axial strain measurement point due to the change of the internal pressure of the container; calculating a distribution consistency index for representing the strain consistency of each current measurement point; according to the distribution consistency index, selecting a pressure calculation strategy adaptive to the current strain distribution state, and calculating to obtain an internal pressure estimation value of the pressure vessel based on the real-time axial strain measurement value; according to the method, discrete strain gauge measuring points are regarded as sparse sampling of a continuous strain field, the uniformity of the strain field is evaluated by analyzing the data consistency of the sampling points, and an optimal pressure resolving strategy is intelligently selected accordingly; and meanwhile, an environmental parameter monitoring and model self-learning mechanism is introduced to dynamically compensate interference, so that high-precision and high-reliability non-contact measurement is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pressure measurement, in particular to a method and device for non-contact measurement of internal pressure of a thin-walled cylindrical pressure vessel. BACKGROUND

[0002] For the measurement of internal pressure of a pressure vessel, the traditional method is usually to open a pressure guide hole on the vessel wall and install a pressure gauge or pressure transmitter through threaded connection; this method has obvious defects in high-pressure, high-sealing requirement or corrosive medium occasions: the pressure guide interface itself becomes a potential leakage point, and the direct contact of the pressure sensor sensitive element with the corrosive medium will cause measurement failure, shortened service life and safety risks.

[0003] Some existing technologies use a non-contact indirect measurement method based on external wall strain measurement, the basic principle of which is: under the action of internal pressure, the external wall of the thin-walled cylinder will produce elastic strain in proportion to the internal pressure in the axial and circumferential directions; by pasting resistance strain gauges on the external wall, the strain value is measured, and then the internal pressure can be calculated according to the formula of elasticity mechanics (i.e. P={4Et / [D(1-2μ)]}*ε_x); in order to improve reliability, a redundant design is usually used, in which multiple (such as four) strain gauges are arranged uniformly along the circumference.

[0004] However, in actual engineering applications, especially in long-term service, complex working conditions with material micro-uniformity, welding residual stress, non-uniform external constraints or temperature field, the above-mentioned existing non-contact measurement method still has the following technical defects: Multiple discrete strain gauges can only sense the strain of several "points" on the vessel wall; when the overall strain field of the vessel is not ideally uniform due to manufacturing, installation or use, the simple average of the measurement values of these discrete points may not accurately represent the overall average axial strain used for theoretical calculation, resulting in systematic pressure calculation errors; The output signal of the strain gauge is the result of the combined action of internal pressure, environmental temperature change, material creep / relaxation and other factors; the existing method relies on fixed material parameters (elastic modulus E, Poisson's ratio μ), and cannot dynamically distinguish and compensate the coupling interference caused by temperature change and material time-varying characteristics in long-term monitoring, resulting in difficult-to-correct drift of readings under static or slowly changing pressure; The existing multi-piece redundant design has a relatively simple data utilization mode, usually limited to simple fault alarm (such as circuit interruption of one channel) or arithmetic mean value; it fails to fully exploit the information on the overall deformation state of the vessel contained in the spatial distribution information of multiple strain gauges, thus failing to identify and correct errors caused by non-uniformity of the strain field, and also failing to realize higher-level system self-diagnosis and self-adaptation.

[0005] Therefore, aiming at the above problems, the present application provides a kind of thin-walled cylindrical pressure vessel internal pressure non-contact measurement method and device, the discrete strain gauge measuring point is regarded as the sparse sampling of continuous strain field, the data consistency of these sampling points is analyzed to evaluate the uniformity of strain field, and the optimal pressure calculation strategy is selected accordingly;Meanwhile, the environmental parameter monitoring and model self-learning mechanism are introduced, and the interference is dynamically compensated, so as to realize high-precision, high-reliability non-contact measurement. SUMMARY

[0006] In order to overcome the problems of pressure calculation error affecting the measurement accuracy and environmental interference and long-term drift leading to insufficient stability in the measurement method of the internal pressure of the pressure vessel in the prior art.

[0007] The technical scheme of the present application is: a non-contact measurement method for the internal pressure of a thin-walled cylindrical pressure vessel, comprising: S1: uniformly arranging at least three axial strain measurement points on the circumference of the outer wall of the pressure vessel; S2: synchronously acquiring real-time axial strain measurement values at each axial strain measurement point due to changes in the internal pressure of the container; S3: calculating a distribution consistency index based on the at least three real-time axial strain measurement values, which represents the consistency of the current strain distribution of each measurement point; S4: selecting a pressure calculation strategy suitable for the current strain distribution state according to the distribution consistency index, and calculating the internal pressure estimate of the pressure vessel based on the real-time axial strain measurement values; Wherein, the pressure calculation strategy at least includes the first strategy based on the statistical quantity of the at least three real-time axial strain measurement values for calculation, and the second strategy for constructing an axial strain distribution model by fitting or interpolation method and calculating the equivalent axial strain based on the model when the distribution consistency index is lower than the preset threshold.

[0008] Preferably, by using a layout of at least three points, the basic morphological information of the strain distribution in the circumferential direction can be obtained, providing a basis for subsequent analysis of the uniformity of the strain field. Synchronous acquisition ensures the temporal consistency of the data, facilitating the analysis of spatial distribution at the same moment. Based on the at least three real-time axial strain measurements, a distribution consistency index is calculated to characterize the strain consistency at each measurement point. This index is a quantitative parameter used to determine whether the current deformation of the container is close to an ideal uniform state. According to the distribution consistency index, a pressure calculation strategy adapted to the current strain distribution state is selected, and the estimated internal pressure of the pressure vessel is calculated based on the real-time axial strain measurements. The pressure calculation strategy includes at least: a first strategy based on the statistics of the at least three real-time axial strain measurements; and a second strategy where, when the distribution consistency index is below a preset threshold, an axial strain distribution model is constructed through fitting or interpolation, and the equivalent axial strain is calculated based on this model before further calculation. This step achieves dynamic matching between the measurement strategy and the measured strain state.

[0009] Preferably, in step S3, the distribution consistency index is the coefficient of variation, which is calculated by dividing the standard deviation of the at least three real-time axial strain measurements by their arithmetic mean. The distribution consistency index is the difference between 1 and the calculated result.

[0010] Preferably, in step S4, the first strategy is to directly use the arithmetic mean or median of the at least three real-time axial strain measurements, combined with preset container structural parameters and material parameters, to calculate the estimated internal pressure using a pressure-strain relationship model.

[0011] Preferably, in step S4, the second strategy includes: S41: Identify and exclude outliers from the at least three real-time axial strain measurements, retaining valid measurements; S42: Based on the circumferential angle position corresponding to each effective measurement value, construct the continuous axial strain distribution function of the outer wall of the pressure vessel in the circumferential direction through function fitting or interpolation methods; S43: Integrate and average the axial strain continuous distribution function over a 360-degree circumferential range to obtain the equivalent axial strain; S44: Based on the equivalent axial strain, combined with the preset container structural parameters and material parameters, the estimated internal pressure is calculated using a pressure-strain relationship model.

[0012] Preferably, the method further includes: S5: Obtain the real-time temperature measurement value of the outer wall of the pressure vessel; S6: Based on historical data of the real-time temperature measurement and / or the internal pressure estimate, dynamically adjust at least one material parameter used in the pressure-strain relationship model, wherein the material parameter includes at least the elastic modulus; S7: Update the estimated internal pressure using the adjusted material parameters and the pressure-strain relationship model.

[0013] A non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel, used to implement the above method, includes: The strain sensing array consists of at least three axial strain sensors uniformly arranged circumferentially along the outer wall of the pressure vessel, used to synchronously measure the axial strain at each arrangement point. The signal measurement circuit is connected to each strain sensor in the strain sensing array and is used to acquire and output the real-time axial strain measurement value of each measurement point. The processing unit, which is communicatively connected to the signal measurement circuit, is configured as follows: Receive the real-time axial strain measurement value; Calculate the distribution consistency index based on the at least three real-time axial strain measurements; Select the corresponding pressure calculation strategy based on the distribution consistency index, and calculate and output the estimated internal pressure of the pressure vessel based on the real-time axial strain measurement value.

[0014] Preferably, the strain sensing array includes four axial strain sensors, which are evenly spaced at 90-degree intervals along the circumferential direction of the outer wall of the pressure vessel.

[0015] Preferably, the axial strain sensor is a metal strain gauge, and the axis of its sensing grid is parallel to the axis of the pressure vessel.

[0016] Preferably, the device further includes a temperature sensor configured to measure the temperature of the outer wall of the container in the strain sensing array region and to communicate with the processing unit; the processing unit is further configured to receive the temperature of the outer wall of the container and, based on historical data of the temperature value and / or estimated internal pressure, dynamically correct the material parameters used for pressure calculation.

[0017] Preferably, the processing unit is configured to: select a pressure calculation strategy as follows: When the distribution consistency index is higher than the first preset threshold, the first calculation module is used, which calculates based on the statistics of the at least three real-time axial strain measurements. When the distribution consistency index is lower than the second preset threshold, the second calculation module is used. This module performs the following operations: identifying and eliminating outliers in the measured values, constructing an axial strain distribution model based on the remaining valid measured values ​​and their spatial locations, calculating the equivalent axial strain through the model, and calculating the estimated internal pressure based on this model. Wherein, the second preset threshold is less than or equal to the first preset threshold.

[0018] The beneficial effects of this invention are: 1. By introducing a distribution consistency index and strain field modeling and equivalent strain calculation, this invention can automatically identify and effectively compensate for systematic errors caused by strain field non-uniformity due to uneven container manufacturing, installation or stress. Compared with the traditional simple averaging method, the "equivalent axial strain" calculated by this invention under non-ideal conditions is more in line with mechanical principles, so that the calculated internal pressure value is closer to the true value, which significantly improves the measurement accuracy under complex working conditions. 2. By integrating temperature monitoring and dynamic self-calibration mechanism for material parameters, this invention can correct key parameters such as elastic modulus in real time or periodically, actively compensate for the coupling effect of time-varying factors such as environmental temperature changes and material creep on measurement results, fundamentally reduce the reading drift problem in long-term monitoring, and enhance the stability and anti-interference ability of long-term measurement. 3. The change in the distribution consistency index output by this invention can indirectly reflect the structural health status of the pressure vessel. For example, when the vessel becomes out of round, bulges locally, or is subjected to abnormal external constraints, the index will drop significantly. Therefore, while performing pressure measurement, this device can also provide early warning signals for the safety status assessment of the vessel, thus expanding its functionality. Attached Figure Description

[0019] Figure 1 The diagram shows a three-dimensional assembly structure of the internal pressure non-contact measuring device and the container of the thin-walled cylindrical pressure vessel of the present invention. Figure 2 The diagram shown is a cross-sectional view of the internal pressure non-contact measuring device and the container assembly of the thin-walled cylindrical pressure vessel of the present invention. Figure 3 The diagram shows a partial structural schematic of the non-contact internal pressure measuring device for the thin-walled cylindrical pressure vessel of the present invention. Figure 4 The diagram shows the steps of the non-contact internal pressure measurement method for the thin-walled cylindrical pressure vessel of the present invention. Explanation of reference numerals in the attached figures: 1. Strain sensor array; 2. Temperature sensor. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] Example 1 Please see Figure 1 , Figure 2 , Figure 3 and Figure 4 This invention provides an embodiment of a non-contact method for measuring the internal pressure of a thin-walled cylindrical pressure vessel, comprising: S1: At least three axial strain measurement points are uniformly set along the circumference of the outer wall of the pressure vessel; S2: Synchronously acquire the real-time axial strain measurement values ​​generated by the pressure change inside the container at each axial strain measurement point; S3: Based on at least three real-time axial strain measurements, calculate the distribution consistency index to characterize the strain consistency at each measurement point. S4: Based on the distribution consistency index, select a pressure calculation strategy that is compatible with the current strain distribution state, and calculate the estimated internal pressure of the pressure vessel based on the real-time axial strain measurement value. The pressure calculation strategy includes at least a first strategy that calculates based on statistics of at least three real-time axial strain measurements, and a second strategy that, when the distribution consistency index is lower than a preset threshold, constructs an axial strain distribution model by fitting or interpolation and calculates the equivalent axial strain based on the model before performing the calculation.

[0022] By using a layout of at least three points, the basic morphological information of strain distribution in the circumferential direction can be obtained, providing a basis for subsequent analysis of strain field uniformity. Synchronous acquisition ensures data temporal consistency, facilitating analysis of spatial distribution at the same time. Based on at least three real-time axial strain measurements, a distribution consistency index is calculated to characterize the strain uniformity at each measurement point. This index is a quantitative parameter used to determine whether the current deformation of the container is close to an ideal uniform state. According to the distribution consistency index, a pressure calculation strategy adapted to the current strain distribution state is selected, and the estimated internal pressure of the pressure vessel is calculated based on the real-time axial strain measurements. The pressure calculation strategy includes at least: a first strategy based on statistics from at least three real-time axial strain measurements; and a second strategy where, when the distribution consistency index is below a preset threshold, an axial strain distribution model is constructed through fitting or interpolation, and the equivalent axial strain is calculated based on this model before further calculation. This step achieves dynamic matching between the measurement strategy and the measured strain state.

[0023] Furthermore, in step S3, the distribution consistency index is the coefficient of variation, which is calculated by dividing the standard deviation of at least three real-time axial strain measurements by their arithmetic mean. The distribution consistency index is the difference between 1 and the calculated result. The closer the value is to 1, the more consistent the measurements at each point are and the more uniform the strain field is; conversely, the less uniform the value is, the worse the uniformity is.

[0024] Furthermore, in step S4, the first strategy is to directly use the arithmetic mean or median of at least three real-time axial strain measurements, combined with preset container structural parameters and material parameters, to calculate the estimated internal pressure using a pressure-strain relationship model; this strategy is highly efficient and has the best accuracy under ideal or near-ideal conditions.

[0025] Furthermore, in step S4, the second strategy includes: S41: First, statistical methods are used to identify and eliminate possible outliers from at least three real-time axial strain measurements. These outliers may originate from an abnormal strain gauge adhesion, local material defects, or abnormal external disturbances. Eliminating them can improve the robustness of subsequent modeling. S42: Based on the circumferential angular positions of each effective measurement value on the outer wall of the container, a continuous axial strain distribution function that can describe the entire circumference of the outer wall of the pressure vessel is constructed through function fitting or interpolation methods; this step expands the discrete point information into continuous field information. S43: Integrate the axial strain continuous distribution function obtained in step S42 within the circumferential range of 0 to 360 degrees and calculate the average value to obtain the "equivalent axial strain". This equivalent value is the result of weighted averaging of the entire circumferential strain distribution through a mathematical model. Theoretically, it is more representative of the "average axial strain" used for theoretical formula calculation than the simple arithmetic mean. S44: Based on equivalent axial strain, combined with preset container structural parameters and material parameters, the internal pressure is estimated using a pressure-strain relationship model.

[0026] Furthermore, the methods also include: S5: Obtain the real-time temperature measurement value of the outer wall of the pressure vessel; S6: Based on historical data of real-time temperature measurements and / or internal pressure estimates, dynamically adjust at least one material parameter used in the pressure-strain relationship model, where the material parameter includes at least the elastic modulus; S7: Update the internal pressure estimate using the adjusted material parameters and pressure-strain relationship model.

[0027] A non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel, used to implement the above method, includes: The strain sensing array 1 consists of at least three axial strain sensors uniformly arranged along the circumference of the outer wall of the pressure vessel, used to synchronously measure the axial strain at each arrangement point. The signal measurement circuit is connected to each strain sensor in the strain sensing array 1 and is used to acquire and output the real-time axial strain measurement value of each measurement point. The processing unit, which is communicatively connected to the signal measurement circuit, is configured as follows: Receives real-time axial strain measurements; Calculate the distribution consistency index based on at least three real-time axial strain measurements; Select the corresponding pressure calculation strategy based on the distribution consistency index, and calculate and output the estimated internal pressure of the pressure vessel based on the real-time axial strain measurement value.

[0028] Furthermore, the strain sensing array 1 includes four axial strain sensors, which are evenly spaced at 90-degree intervals along the circumference of the outer wall of the pressure vessel. The four-point arrangement is the best balance between cost and the ability to acquire circumferential distribution information, and can effectively capture low-order non-uniform modes such as possible elliptic deformation.

[0029] Furthermore, the axial strain sensor is a metal strain gauge, with the axis of its sensitive grid parallel to the axis of the pressure vessel to ensure accurate measurement of axial strain.

[0030] Furthermore, the device also includes a temperature sensor 2, which is configured to measure the temperature of the outer wall of the container in the region of the strain sensor array 1 and is communicatively connected to the processing unit; the processing unit is further configured to receive the temperature of the outer wall of the container and dynamically correct the material parameters used for pressure calculation based on historical data of the temperature value and / or the internal pressure estimate; thereby realizing temperature compensation and model self-calibration.

[0031] Furthermore, when selecting a stress calculation strategy, the processing unit is specifically configured to execute a judgment logic: When the distribution consistency index is higher than the first preset threshold, the strain field is determined to be highly uniform, and the first calculation module is called to execute the first strategy for calculation.

[0032] When the distribution uniformity index is lower than the second preset threshold (the second preset threshold ≤ the first preset threshold), it is determined that there is obvious non-uniformity in the strain field, and the second calculation module is called to execute the second strategy for calculation.

[0033] If the metric falls between the two, a weighted fusion or a default strategy can be used for processing.

[0034] Through the above steps, this invention treats discrete strain gauge measurement points as sparse sampling of a continuous strain field, evaluates the uniformity of the strain field by analyzing the data consistency of these sampling points, and intelligently selects the optimal pressure calculation strategy accordingly. At the same time, it introduces environmental parameter monitoring and model self-learning mechanisms to dynamically compensate for interference, thereby achieving high-precision and high-reliability non-contact measurement.

[0035] Example 2 Optionally, the present invention also provides an embodiment.

[0036] In this embodiment, the measuring device mainly consists of the following parts: Strain sensing array 1: Four identical metal resistance strain gauges (e.g., nominal resistance 350Ω, sensitivity coefficient K=2.0) are used. These four strain gauges (marked as R1, R2, R3, R4) are uniformly pasted circumferentially along the same circumferential cross-section of the outer wall of the thin-walled cylindrical pressure vessel being measured. During pasting, it is ensured that the sensitive grid axis of each strain gauge is strictly parallel to the axial direction of the vessel to sense axial strain. The circumferential angle between the four strain gauges is 90 degrees.

[0037] Signal measurement circuit: A full-bridge or half-bridge measurement circuit is used; four strain gauges (R1-R4) serve as the four arms of the bridge (or two of them are measurement arms and the other two are temperature-compensated fixed resistors), connected to the signal conditioning module; this module includes an excitation power supply, an amplifier, a filter, and an analog-to-digital converter (ADC) to convert the strain gauge resistance change into a digital voltage signal, and finally calculate four independent real-time axial strain measurements (ε1, ε2, ε3, ε4); the signal conditioning module acquires data from the four channels in a synchronous sampling manner to ensure the time consistency of the data.

[0038] Processing unit: can be a microcontroller (MCU), digital signal processor (DSP) or industrial computer; it is connected to the ADC output of the signal measurement circuit to receive and process strain digital signals from four channels.

[0039] The processing unit contains a pre-stored program for performing the following steps: Step S1: Initial calibration and parameter setting In the initial state of the container without pressure (internal pressure P=0), the initial readings of four strain gauges are recorded as reference values ​​(ε1_0, ε2_0, ε3_0, ε4_0). At the same time, the structural and material parameters of the container are input into the processing unit, including: average diameter of the cylinder D (D=(D_inner+D_outer) / 2, unit: mm), wall thickness t (unit: mm), material elastic modulus E (unit: MPa), and Poisson's ratio μ.

[0040] Step S2: Real-time data synchronization acquisition When the pressure inside the container changes, the processing unit controls the signal measurement circuit to synchronously acquire the real-time output of the four strain gauges, and obtain a set of real-time strain measurement values ​​(ε1, ε2, ε3, ε4).

[0041] Step S3: Calculate the consistency index between strain change and distribution. First, calculate the strain change at each measuring point relative to the initial state: Δε_i = ε_i - ε_i_0 (i = 1, 2, 3, 4). Then, based on this set of Δε_i values, the distribution consistency index η is calculated; in this embodiment, η is calculated using a method based on the coefficient of variation: Calculate the arithmetic mean of Δε_i: Δε_avg=(Δε_1+Δε_2+Δε_3+Δε_4) / 4 Calculate the standard deviation of Δε_i: σ=sqrt([(Δε_1-Δε_avg)^2+(Δε_2-Δε_avg)^2+(Δε_3-Δε_avg)^2+(Δε_4-Δε_avg)^2] / 4) Calculate the distribution consistency index: η = 1 - (σ / Δε_avg) In this context, the range of values ​​for η is theoretically such that a value close to 1 indicates a high degree of consistency, while a smaller value indicates a lower degree of consistency and a less uniform strain field.

[0042] Step S4: Intelligent Strategy Selection and Stress Calculation The processing unit presets a consistency threshold η_th (for example, η_th=0.93). If η ≥ η_th: It is determined that the axial strain field distribution on the outer wall of the container is uniform, and the first strategy is adopted. That is, the arithmetic mean Δε_avg of the four strain changes is directly used as the axial strain for calculation. Substituting it into the pressure-axial strain relationship formula for a thin-walled cylinder, the estimated internal pressure P_est is calculated. P_est=[4Et / (D(1-2μ))]*Δε_avg In the formula: P_est is the estimated internal pressure (MPa), E is the elastic modulus (MPa), t is the wall thickness (mm), D is the average diameter (mm), μ is Poisson's ratio, and Δε_avg is the average value of the axial strain change; If η < η_th: it is determined that there is obvious non-uniformity in the strain field, and the second strategy is adopted. In this basic embodiment, the second strategy can be simplified into a robust processing method: calculate the median value Δε_median of the four Δε_i, and then use Δε_median to replace Δε_avg and substitute it into the above formula to calculate P_est. The median value is less sensitive to outliers than the arithmetic mean, which can suppress the error caused by non-uniformity to a certain extent.

[0043] During service, changes in internal pressure cause deformation of the outer wall of the container. Four attached strain gauges synchronously sense the axial strain at their locations and convert it into electrical signals through a measurement circuit. The processing unit executes steps S2-S4 in a loop and outputs the estimated internal pressure P_est calculated based on the current strain field state using an intelligent selection strategy in real time. This scheme uses a high-precision average value when the strain field is uniform and automatically switches to the median method, which has stronger anti-interference capabilities, when non-uniformity occurs. Compared with the traditional method that uses the average value, the overall measurement accuracy and robustness are improved.

[0044] Example 3 Optionally, the present invention also provides an embodiment.

[0045] This embodiment adds the following components to the device of embodiment 2: Temperature sensor 2: A non-contact infrared temperature sensor 2 or a contact patch temperature sensor 2, installed on the outer wall of the container, for real-time measurement of the wall temperature T in the area where the strain gauge array is located.

[0046] Enhanced processing unit: It has computing power and storage space to run more complex algorithms.

[0047] The specific method steps in this embodiment differ from those in Embodiment 2 in that: Step S3: Same as in Example 2, calculate the distribution consistency index η.

[0048] Step S4: Intelligent Strategy Selection and Stress Calculation The processing unit presets two thresholds: a high threshold η_high (e.g., 0.95) and a low threshold η_low (e.g., 0.85), and η_low < η_high; If η≥η_high: adopt the first strategy, the same as in Example 2, and use Δε_avg for calculation; If η≤η_low: adopt the complete second strategy, with the following sub-steps: S41 (Outlier Removal): Tukey's Fences method is used to identify outliers; first, the interquartile range (IQR) of Δε_i is calculated as IQR = Q3 - Q1 (Q1 is the lower quartile, Q3 is the upper quartile); Δε_i values ​​less than Q1 - 1.5 * IQR or greater than Q3 + 1.5 * IQR are identified as outliers and temporarily excluded; assuming Δε_2 is excluded, the remaining valid values ​​are Δε_1, Δε_3, and Δε_4.

[0049] S42 (Constructing the strain distribution model): Using the remaining effective values ​​and their corresponding circumferential angles (e.g., R1: 0°, R3: 180°, R4: 270°) as data points; and fitting the data using a second-order Fourier series to construct the circumferential strain distribution function ε(θ): ε(θ)=a0+a1*cos(θ)+b1*sin(θ)+a2*cos(2θ)+b2*sin(2θ) Where θ is the circumferential angle (radians), and a0, a1, b1, a2, b2 are fitting coefficients; these coefficients are solved by the least squares method.

[0050] S43 (Calculate the equivalent average strain): Integrate the fitted continuous distribution function ε(θ) over a complete period (0 to 2π) and average it to obtain the equivalent axial strain Δε_eq. Δε_eq=(1 / 2π)*∫_0^(2π)ε(θ)dθ=a0 According to the properties of Fourier series, its average value is the constant term a0; S44 (Calculate Pressure): Substitute Δε_eq into the pressure formula: P_est=[4Et / (D(1-2μ))]*Δε_eq If η_low < η < η_high: a hybrid strategy can be adopted, such as a weighted average of the calculation results of the first strategy and the second strategy (or a simplified second strategy), with the weights proportional to η.

[0051] Step S5: Temperature Monitoring The wall temperature value T measured by temperature sensor 2 is read in real time.

[0052] Steps S6 & S7: Dynamic self-calibration of parameters A dynamic model of material parameters is established within the processing unit; taking the elastic modulus E as an example, it can be modeled as a function of temperature T and internal pressure P_est: E_corrected=E0*[1+α_E*(T-T0)+β_E*P_est] Where E0 is the initial value of the elastic modulus at the reference temperature T0, α_E is the temperature correction factor (1 / °C), and β_E is the pressure correction factor (1 / MPa); μ can also be corrected in a similar way.

[0053] The system initiates a self-learning process periodically (e.g., daily) or at a known pressure point (e.g., when the system is depressurized to atmospheric pressure): at this calibration point, the true pressure P_known is known (which may be 0 or obtained by briefly accessing a standard table). The system collects the Δε_eq (or Δε_avg) and temperature T at this time, and uses the relation P_known=[4E_correctedt / (D(1-2μ_corrected))]*Δε to back-update the adjustable parts of the parameter set (E0, α_E, β_E, ...) through an optimization algorithm (e.g., least squares) to make the model output closer to the true value.

[0054] In routine measurements, the processing unit uses the current temperature T and the latest estimated pressure P_est to calculate E_corrected and μ_corrected in real time based on the updated dynamic model. It then replaces the fixed parameters E and μ with these values ​​and substitutes them into the pressure calculation formula. Finally, it outputs the internal pressure value P_final, which has been compensated for by temperature and time-varying effects.

[0055] During operation, the device in this embodiment not only performs strain measurement and consistency judgment, but also monitors temperature and runs a background self-calibration algorithm. When the strain field is severely non-uniform, it can reconstruct the strain distribution of the entire circumference by modeling, calculate the equivalent average strain that is more in line with mechanical meaning, and reduce the error caused by non-uniform deformation. At the same time, the temperature compensation and parameter self-learning mechanism work continuously to dynamically counteract temperature drift and slow changes in material properties, ensuring the stability and accuracy of long-term measurements. In addition, the output distribution consistency index η itself can serve as an auxiliary monitoring parameter for the health status of the container structure.

[0056] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A non-contact method for measuring the internal pressure of a thin-walled cylindrical pressure vessel, characterized in that: include: S1: At least three axial strain measurement points are uniformly arranged in the circumferential direction of the outer wall of the pressure vessel; S2: Simultaneously acquire the real-time axial strain measurement values ​​generated by the pressure change inside the container at each of the axial strain measurement points; S3: Based on the at least three real-time axial strain measurements, calculate the distribution consistency index to characterize the strain consistency at each measurement point. S4: Based on the distribution consistency index, select a pressure calculation strategy that is compatible with the current strain distribution state, and calculate the estimated internal pressure of the pressure vessel based on the real-time axial strain measurement value. The pressure calculation strategy includes at least a first strategy that calculates based on statistics of the at least three real-time axial strain measurements, and a second strategy that, when the distribution consistency index is lower than a preset threshold, constructs an axial strain distribution model by fitting or interpolation and calculates the equivalent axial strain based on the model before performing the calculation.

2. The non-contact measurement method for the internal pressure of a thin-walled cylindrical pressure vessel according to claim 1, characterized in that: In step S3, the distribution consistency index is the coefficient of variation, which is calculated by dividing the standard deviation of the at least three real-time axial strain measurements by their arithmetic mean. The distribution consistency index is the difference between 1 and the calculated result.

3. The non-contact measurement method for the internal pressure of a thin-walled cylindrical pressure vessel according to claim 1 or 2, characterized in that: In step S4, the first strategy is to directly use the arithmetic mean or median of the at least three real-time axial strain measurements, combined with preset container structural parameters and material parameters, to calculate the estimated internal pressure using a pressure-strain relationship model.

4. The non-contact measurement method for the internal pressure of a thin-walled cylindrical pressure vessel according to claim 3, characterized in that: In step S4, the second strategy includes: S41: Identify and exclude outliers from the at least three real-time axial strain measurements, retaining valid measurements; S42: Based on the circumferential angle position corresponding to each effective measurement value, construct the continuous axial strain distribution function of the outer wall of the pressure vessel in the circumferential direction through function fitting or interpolation methods; S43: Integrate and average the axial strain continuous distribution function over a 360-degree circumferential range to obtain the equivalent axial strain; S44: Based on the equivalent axial strain, combined with the preset container structural parameters and material parameters, the estimated internal pressure is calculated using a pressure-strain relationship model.

5. The non-contact measurement method for the internal pressure of a thin-walled cylindrical pressure vessel according to claim 4, characterized in that: The method further includes: S5: Obtain the real-time temperature measurement value of the outer wall of the pressure vessel; S6: Based on historical data of the real-time temperature measurement and / or the internal pressure estimate, dynamically adjust at least one material parameter used in the pressure-strain relationship model, wherein the material parameter includes at least the elastic modulus; S7: Update the estimated internal pressure using the adjusted material parameters and the pressure-strain relationship model.

6. A non-contact measuring device for the internal pressure of a thin-walled cylindrical pressure vessel, used to implement the method according to any one of claims 1-5, characterized in that, include: The strain sensing array (1) consists of at least three axial strain sensors uniformly arranged along the circumference of the outer wall of the pressure vessel, used to synchronously measure the axial strain at each arrangement point. The signal measurement circuit is connected to each strain sensor in the strain sensing array (1) to acquire and output the real-time axial strain measurement value of each measurement point; The processing unit, which is communicatively connected to the signal measurement circuit, is configured as follows: Receive the real-time axial strain measurement value; Calculate the distribution consistency index based on the at least three real-time axial strain measurements; Select the corresponding pressure calculation strategy based on the distribution consistency index, and calculate and output the estimated internal pressure of the pressure vessel based on the real-time axial strain measurement value.

7. The non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel according to claim 6, characterized in that: The strain sensing array (1) includes four axial strain sensors, which are evenly spaced at 90-degree intervals along the circumferential direction of the outer wall of the pressure vessel.

8. The non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel according to claim 6, characterized in that: The axial strain sensor is a metal strain gauge, and the axial direction of its sensing grid is parallel to the axial direction of the pressure vessel.

9. The non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel according to claim 6, characterized in that: The device also includes a temperature sensor (2), which is configured to measure the temperature of the outer wall of the container in the region of the strain sensing array (1) and is communicatively connected to the processing unit; the processing unit is further configured to receive the temperature of the outer wall of the container and dynamically correct the material parameters used for pressure calculation based on historical data of the temperature value and / or the internal pressure estimate.

10. The non-contact internal pressure measuring device for a thin-walled cylindrical pressure vessel according to claim 6, characterized in that: When selecting a pressure calculation strategy, the processing unit is configured as follows: When the distribution consistency index is higher than the first preset threshold, the first calculation module is used, which calculates based on the statistics of the at least three real-time axial strain measurements. When the distribution consistency index is lower than the second preset threshold, the second calculation module is used. This module performs the following operations: identifying and eliminating outliers in the measured values, constructing an axial strain distribution model based on the remaining valid measured values ​​and their spatial locations, calculating the equivalent axial strain through the model, and calculating the estimated internal pressure based on this model. Wherein, the second preset threshold is less than or equal to the first preset threshold.

Images