An automatic control system for a gas cabinet and a control method thereof

Abstract

The application relates to the field of automation control technology and discloses an automatic control system for a gas cabinet and a control method thereof, which comprises the following steps: acquiring an absolute temperature field by using a sensor array to reconstruct a local transient sealing gap; combining a piston transient speed and an absolute oil temperature to calculate micro non-uniform shearing stress, and integrating to extract a global eccentric friction torque; collecting gas chamber parameters to extract a real gas volume change rate; combining inherent quality and real-time sound velocity to comprehensively generate a dimensionless jam self-excitation intensity index. Accordingly, an endogenous particle swarm optimization is driven to deduce an asymmetric volume flow vector of an array valve, and the valve angle acceleration is inversely analyzed by taking a fluid relaxation time as a limit and is issued to a frequency converter; mechanical friction and conduction parameters are synchronously calculated to control an induced heat tracing array to execute compensation. The application reduces the jam risk of the gas cabinet under complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, and more specifically, to an automated control system and control method for a gas cabinet. Background Technology

[0002] Dry-type piston gas holders mainly consist of a thin-walled steel cylindrical shell and a rigid piston located within it. Dynamic sealing and lubrication between the piston edge and the outer shell rely on a gap filled with high-molecular-weight sealing oil. Under long-term outdoor operation, unilateral sunlight causes thermal expansion of the sun-facing steel shell, resulting in significant unevenness in the circumferential sealing gap. Simultaneously, ambient temperature differences cause variations in the dynamic viscosity of the sealing oil in different locations, creating an unbalanced distribution of frictional resistance around the piston. When a large flow of gas is pumped through the pipeline, forcing the piston to rise and fall, this unbalanced frictional force generates an eccentric overturning moment, causing the piston to slightly deflect. After this slight deflection, the high-speed influx of gas creates a pneumatic lifting effect in extremely narrow areas, further increasing the tendency to tilt. Current automated control systems typically rely on preset static parameters such as global total gas pressure and displacement alarm thresholds for adjustment, making it difficult to detect and calculate the underlying microscopic asymmetric eccentric moment in real time. This results in significant lag in the control of the existing system when faced with multi-field coupled changes, making it unable to promptly interrupt the negative cycle of frictional imbalance and pneumatic lifting, thus increasing the safety hazards of piston jamming and gas leakage. Summary of the Invention

[0003] This invention provides an automated control system and control method for gas cabinets, which solves the technical problems mentioned in the background art.

[0004] In the first aspect, an automated control system for a gas cabinet is applied to a gas cabinet system comprising a distributed fiber Bragg grating sensor array, an array pneumatic valve, a bottom-level frequency converter, and a localized electromagnetic induction heat tracing array. The system executes the following sequentially: Obtain the instantaneous piston speed and absolute oil temperature; The absolute temperature field around the circumference is obtained by using a distributed fiber Bragg grating sensor array, and the local instantaneous sealing gap is reconstructed. Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The piston's inherent mass and real-time sound velocity are obtained. Using the piston's inherent mass and real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless sticking self-excitation strength index. Based on the endogenous particle swarm optimization driven by the self-excitation intensity index, the asymmetric array valve body volume flow vector is deduced. With the natural relaxation time of the fluid as a constraint, the volumetric flow vector of the array valve body is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, microscopic non-uniform shear stress, and absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

[0005] Secondly, an automated control method for a gas cabinet, applied in any of the automated control systems for a gas cabinet described in any one of the claims, includes: Obtain the instantaneous piston speed and absolute oil temperature; The distributed fiber optic grating sensor array is used to acquire the circumferential absolute temperature field and reconstruct the local instantaneous sealing gap; Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and the absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The inherent mass of the piston and the real-time sound velocity are obtained. Using the inherent mass of the piston and the real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless jamming self-excitation strength index. Based on the self-excitation intensity index of the hysteresis, the endogenous particle swarm optimization is driven, and the asymmetric array valve body volume flow vector is deduced. With the fluid's natural relaxation time as a constraint, the array valve body volume flow vector is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, the microscopic non-uniform shear stress, and the absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

[0006] The beneficial effects of this invention are as follows: By real-time acquisition of thermodynamic and hydrodynamic parameters, the local sealing gap is dynamically reconstructed and the global eccentric friction torque is extracted. Then, combined with the real gas volume change rate, a dimensionless self-excited intensity index of hysteresis is generated, driving an endogenous algorithm to perform asymmetric flow regulation on the array valves, while simultaneously executing localized heat tracing compensation. The entire process avoids dependence on preset empirical parameters and static thresholds, effectively breaking the positive feedback loop caused by the superposition of temperature distortion, frictional imbalance, and pneumatic lifting effects. This improves the system's dynamic response capability under multi-field coupled changing conditions, reduces the probability of skew jamming, and enhances the stability of the gas holder's operation. Attached Figure Description

[0007] Figure 1 This is a flowchart of an automated control system for a gas cabinet according to the present invention. Detailed Implementation

[0008] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0009] Example 1: As Figure 1 As shown, an automated control system for a gas cabinet is applied to a gas cabinet system comprising a distributed fiber Bragg grating sensor array, an array pneumatic valve, a bottom-level frequency converter, and a localized electromagnetic induction heat tracing array. The system executes the following sequentially: Obtain the instantaneous piston speed and absolute oil temperature; The absolute temperature field around the circumference is obtained by using a distributed fiber Bragg grating sensor array, and the local instantaneous sealing gap is reconstructed. Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The piston's inherent mass and real-time sound velocity are obtained. Using the piston's inherent mass and real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless sticking self-excitation strength index. Based on the endogenous particle swarm optimization driven by the self-excitation intensity index, the asymmetric array valve body volume flow vector is deduced. With the natural relaxation time of the fluid as a constraint, the volumetric flow vector of the array valve body is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, microscopic non-uniform shear stress, and absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

[0010] Preferably, the distributed fiber optic grating sensor array is used to acquire the circumferential absolute temperature field and reconstruct the local instantaneous sealing gap, including: Obtain the inner diameter of the cylindrical cabinet wall and the inherent linear expansion coefficient of the steel, as well as the initial nominal mechanical clearance; The integral circumferential temperature field is obtained by integrating the absolute temperature field over the entire circumference and dividing it by twice the value of pi. The differential mode temperature is obtained by subtracting the integral circumferential temperature from the absolute circumferential temperature field. By continuously multiplying the inner diameter of the cylindrical cabinet wall, the inherent linear expansion coefficient of the steel, and the differential temperature, the transient thermo-elastic-plastic distortion variable is obtained. Subtracting the transient thermo-elastic-plastic distortion from the initial nominal mechanical clearance yields the local instantaneous sealing clearance. It satisfies the following relationship: in, The local instantaneous sealing gap, The initial nominal clearance of the machine. The inner diameter of the cylindrical wall of the cabinet is given. The coefficient of linear expansion of the steel is denoted as . Polar angle variable The circumferential absolute temperature field at that location, For time variables, For polar coordinate integration variables, Let be the absolute temperature field of the circumference corresponding to the integral variable, and The overall integral circumferential temperature is characterized.

[0011] The circumferential absolute temperature field is the absolute temperature distribution at various locations along the complete circumference of the cabinet wall at the current moment. It can be obtained in real time by a distributed fiber Bragg grating sensor array uniformly arranged at the corresponding height along the sealing strip of the cabinet wall.

[0012] The inner diameter of the cylindrical casing is the actual inner diameter of the cylindrical shell of the gas cabinet. It can be obtained through laser ranging, total station re-measurement, or actual measurement using inner diameter measuring tools during the installation and acceptance phase.

[0013] The inherent coefficient of linear expansion of steel is a proportional constant that represents the linear dimensional change of the steel in the cabinet wall under a unit temperature change. A preferred value is 0.000011 to 0.000013 per degree Celsius, a range that aligns with the linear expansion characteristics of commonly used carbon structural steel and low-alloy steel within the operating temperature range of gas cabinets.

[0014] The initial nominal mechanical clearance is the design reference clearance between the piston edge and the cabinet wall when there is no thermal distortion or operational misalignment. A preferred value is 10 to 25 mm, a range that simultaneously meets the requirements for sealing oil film establishment, installation deviation absorption, and thermal expansion compensation.

[0015] The integral circumferential temperature is the average temperature characterization value of the absolute temperature field of a circle over the entire circumference.

[0016] Differential temperature is the deviation of the temperature at a local corner point from the temperature of the integral circle.

[0017] The transient thermo-elastic-plastic distortion is the equivalent radial thermal distortion value driven by local temperature difference.

[0018] The local instantaneous sealing gap is the actual width of the sealing oil gap at a certain polar angle position at the current moment.

[0019] Polar angle variables are angular coordinates used to identify the position of a circle.

[0020] The time variable is the time coordinate used to identify the current sampling moment.

[0021] Polar coordinate integral variables are auxiliary angle variables that traverse the circumference when performing full circle integrals.

[0022] In practical implementation, for the arrangement and calibration of the distributed fiber Bragg grating sensor array, it is necessary to arrange measuring points at equal angular intervals along the circumference corresponding to the center height of the cabinet wall sealing strip. The number of measuring points is preferably 12 to 48, and the preferred angular interval is 7.5 degrees to 30 degrees. Each measuring point is encapsulated with a temperature-compensated fiber Bragg grating and installed in a position that is in close contact with the cabinet wall to reduce surface temperature drift caused by wind and sunlight scattering. After installation, zero-point calibration is first performed under the condition of shutdown and uniform temperature, and then temperature calibration is performed at no less than 3 points using a constant temperature source. If there is significant mechanical strain interference on site, a dual-grating compensation structure is adopted, with one grating as a temperature measuring point and the other grating as a strain compensation measuring point.

[0023] In practical implementation, for the reconstruction of the circumferential absolute temperature field, it is necessary to first map the discrete temperature values ​​to a unified polar coordinate according to the installation angle of each measuring point on the circumference; then perform median filtering and moving average filtering to eliminate instantaneous jumps and noise; then use piecewise cubic interpolation or circumferential Fourier interpolation to reconstruct the discrete temperature points into a continuous temperature field of the complete circle; for a single missing measuring point, interpolation between two adjacent points can be used for compensation; for cases where more than two consecutive measuring points are missing, degraded control should be implemented and the temperature field at the previous moment should be fused and corrected with the current valid measuring points.

[0024] In practice, to obtain the inner diameter of the cylindrical body of the cabinet wall and the initial nominal mechanical clearance, the inner diameter of the cylindrical body needs to be remeasured circumferentially at at least three height sections using a laser total station, inner diameter fixture, or multi-point ranging device during equipment installation and acceptance, major repairs, or annual shutdown maintenance. The average value of the corresponding height sections of the sealing strip should be used as the input to the control model. The initial nominal mechanical clearance should be measured at multiple points around the entire circle using a feeler gauge, clearance sensor, or special calibration fixture under no-load, shutdown, temperature equalization, and stable and static sealing oil conditions. The average value of the entire circle should be used as the nominal value. If the local installation deviation is large, the deviation should be written into the static geometric compensation table and not included in the main value of the initial nominal mechanical clearance.

[0025] In practical implementation, the value of the inherent linear expansion coefficient of steel should be selected according to the actual material grade, manufacturing batch, and operating temperature range of the cabinet wall. The measured value in the material certificate should be used first, and the recommended value in the standard database of the same grade should be used if the measured value is unavailable. If there are dense weld areas, repair weld areas, or long-term high-temperature exposure areas in the cabinet wall, a correction table can be established according to the area. When the controller is called, the corresponding parameters should be read according to the area to which the current measuring point belongs.

[0026] In specific implementation, for the discrete integration of the integral circumferential temperature and differential mode temperature, it is necessary to discretely sample the reconstructed continuous circumferential temperature field at an equal angular step size, preferably 1 to 5 degrees; sum the temperatures of all discrete points on the whole circle and divide by the number of discrete points to obtain the integral circumferential temperature; subtract the integral circumferential temperature from the temperature of each discrete point to obtain the differential mode temperature of the corresponding corner point; the polar zero position should be fixed at the starting point of the array pneumatic valve number or the north-facing baseline of the cabinet to ensure that the subsequent friction torque calculation is consistent with the valve orientation.

[0027] In practical implementation, for the boundary protection of transient thermo-elastic-plastic distortion and local instantaneous sealing gap, at the real-time control level, the transient thermo-elastic-plastic distortion should be treated as equivalent radial thermal distortion dominated by local thermal expansion, and it is not required to solve the complete finite element elastoplastic model every time. When the equipment's historical maintenance records indicate that there is residual plastic deformation in a certain area, the residual deformation should be written into the static compensation table and used in conjunction with the real-time distortion. After calculating the local instantaneous sealing gap, a minimum allowable gap lower limit should be set, preferably 3 to 5 mm, to prevent the value from being too small and causing abnormalities in the subsequent shear stress denominator.

[0028] Preferably, the microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity with the absolute oil temperature, and the global eccentric friction torque is extracted by integration in polar coordinates, including: To obtain the fluid's infinite thermodynamic viscosity limit, activation energy, and ideal gas constant, as well as the effective sealing height and piston radius; The rheological correction coefficient is obtained by dividing the activation energy by the product of the ideal gas constant and the absolute oil temperature, and taking the natural exponent. Multiply the rheological correction factor by the fluid's infinite temperature dynamic viscosity limit to obtain the transient dynamic viscosity; Multiplying the transient dynamic viscosity by the instantaneous piston velocity and dividing by the local instantaneous sealing gap yields the microscopic non-uniform shear stress. ; Using the integral polar angle in polar coordinates as the independent variable, a directional unit vector is constructed using the sine and negative cosine values ​​of the integral polar angle; Multiply the microscopic non-uniform shear stress by the square of the effective sealing height and the piston radius, then multiply by the unit vector of the direction, and perform polar coordinate vector integration over the complete circumference to extract the global eccentric friction torque. It satisfies the following relationship: in, , For the microscopic non-uniform shear stress, The global eccentric friction torque, The infinite thermodynamic viscosity limit of the fluid. The activation energy is... Let be the ideal gas constant. The absolute oil temperature, The instantaneous velocity of the piston is... The effective height of the seal, The piston radius is... Let be the unit vector in the direction.

[0029] The instantaneous piston velocity is the instantaneous speed of the piston along the lifting direction at the current moment. It can be obtained by acquiring the displacement through a drawstring encoder, laser displacement sensor, or magnetostrictive displacement sensor, and then performing differential calculations.

[0030] Absolute oil temperature is the absolute temperature value of the sealing oil at a corresponding circumferential position. It can be obtained through an immersion platinum resistance temperature sensor, a multi-point thermal resistor in the oil tank, or an oil-immersed fiber optic grating sensor.

[0031] The infinite-temperature dynamic viscosity limit of a fluid is the asymptotic value of the dynamic viscosity of a sealing oil under the assumption of an extremely high temperature limit. A preferred value is 0.001 to 0.02 Pa·s, a range that meets the order-of-magnitude requirements for extrapolating and fitting the viscosity of common polymeric sealing oils at high temperatures.

[0032] Activation energy is a characteristic energy parameter that characterizes the sensitivity of sealing oil viscosity to temperature changes. A preferred value is 12,000 to 35,000 joules per mole, a range that conforms to the conventional viscosity-temperature fitting range for both mineral-based and synthetic sealing oils.

[0033] The ideal gas constant is a standard physical constant used for normalization in temperature exponential models. A preferred value is 8.314 joules per mole per Kelvin, a fixed value for the standard physical constant.

[0034] The effective sealing height is the axial effective height at which the sealing oil actually participates in shearing and heat transfer. It can be obtained through re-measuring the sealing structure dimensions, measurement during shutdown maintenance, or verification of the equipment's as-built dimensions.

[0035] The piston radius is the geometric radius from the outer edge of the piston to its center. It can be obtained through actual measurement of the piston structure dimensions or verification of the equipment's as-built dimensions.

[0036] The rheological correction factor is a dimensionless correction factor used to map changes in absolute oil temperature to changes in viscosity.

[0037] Transient dynamic viscosity is the real-time dynamic viscosity of sealing oil under the current oil temperature conditions.

[0038] Microscopic non-uniform shear stress is the local tangential frictional stress formed by sealing oil under local gap and local temperature conditions.

[0039] The directional unit vector is a unit vector description of the direction of contribution of local shear stress to the global eccentric frictional torque.

[0040] The global eccentric frictional torque is the overall eccentric overturning torque formed by integrating the local shear stress over the entire circumference.

[0041] In practical implementation, for the acquisition and spatiotemporal alignment of piston instantaneous velocity and absolute oil temperature, the piston instantaneous velocity needs to be obtained by the displacement measurement link, with a preferred sampling period of 0.05 to 0.5 seconds. The displacement signal is obtained after center differential and low-pass filtering. The absolute oil temperature should be distributed at multiple points along the circumference at the height of the sealing oil, and a unified polar angle numbering system should be used with the circumferential absolute temperature field. To ensure the accuracy of shear stress calculation, the timestamp alignment error between velocity data and oil temperature data is preferably less than 0.05 seconds. When the velocity sampling frequency is higher than the temperature sampling frequency, the temperature can be synchronized by the nearest time hold method or the linear interpolation method.

[0042] In practical implementation, the values ​​of the infinite temperature dynamic viscosity limit, activation energy, and ideal gas constant of the fluid should be determined by experimental fitting of the fluid's infinite temperature dynamic viscosity limit and activation energy for the actual sealing oil used, and should not be directly applied to other oil parameters. During the test, the dynamic viscosity should be measured at no less than 5 temperature points, preferably covering a temperature range of 20 degrees Celsius to 80 degrees Celsius. Then, the fluid's infinite temperature dynamic viscosity limit and activation energy should be fitted according to the viscosity-temperature relationship adopted in this application. The ideal gas constant should use a fixed standard value.

[0043] In practical implementation, for the applicability of rheological correction coefficients and transient dynamic viscosity, the absolute oil temperature must be input using an absolute temperature scale. If the output of the field temperature sensor is in Celsius, then 273.15 should be added before participating in the calculation. This model is applicable to the same sealing oil, with stable composition and no obvious emulsification or contamination. When the sealing oil is severely aged, has a significantly increased water content, or is mixed with particulate impurities, the model parameters should be recalibrated or switched to an empirical lookup table model.

[0044] In practical implementation, considering the engineering assumptions of microscopic non-uniform shear stress, this application requires the use of a real-time control model under the conditions of local parallel plate approximation and no-slip boundary conditions, treating the local oil film of the sealing strip as a shear layer with a thickness equal to the local instantaneous sealing gap; this model is suitable for working conditions where the gap is relatively small, the flow is dominated by shear and the local Reynolds number is not high; when severe disturbances, cavitation, oil film rupture or mixed-phase erosion occur locally, the calculation results should be overridden by protection logic.

[0045] In practical implementation, for the coordinate definition and sign convention of the directional unit vector, the geometric center of the piston should be used as the origin of torque calculation, the fixed reference direction of the cabinet should be used as the zero-degree direction, and counterclockwise should be defined as the positive polar angle direction; the contribution direction of each local shear stress to the global eccentric friction torque should be determined by the tangential force direction and lever arm direction corresponding to the local polar angle; a unified directional convention should be solidified in the controller to avoid conflicts between the valve position direction vector and the friction torque direction.

[0046] In practical implementation, for the discrete integration of the global eccentric friction torque, the entire circle needs to be divided into equiangular discrete elements, preferably with no less than 36 discrete elements. For each discrete element, the local microscopic non-uniform shear stress is calculated, and then multiplied by the effective sealing height, piston radius, and direction unit vector to obtain the vector contribution of that element to the global eccentric friction torque. Finally, the vector contributions of all elements are summed to obtain the global eccentric friction torque.

[0047] Preferably, the measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on a small rate of change over time, including: Obtain the cross-sectional area of ​​the piston, multiply the cross-sectional area of ​​the piston by the instantaneous velocity of the piston, and obtain the mechanical sweep volume change rate; Divide the measured volume of the air chamber by the absolute pressure, and multiply by the first derivative of the absolute pressure with respect to time to obtain the pseudo pressure fluctuation rate. Divide the measured volume of the air chamber by the absolute temperature, and multiply by the first derivative of the absolute temperature with respect to time to obtain the pseudo-thermal thermal expansion coefficient; The true gas volume change rate is obtained by subtracting the pseudo pressure fluctuation rate from the mechanical sweep volume change rate and adding the pseudo thermal expansion rate. It satisfies the following relationship: in, The actual gas volume change rate, The cross-sectional area of ​​the piston is... The instantaneous velocity of the piston is... The measured volume of the air chamber is... The absolute pressure is... The absolute temperature, Let be the first derivative of the absolute pressure with respect to time. Let be the first derivative of the absolute temperature with respect to time.

[0048] The measured volume of the gas chamber is the actual volume of gas that the chamber can hold at the current moment. It can be obtained in real time using a piston position sensor combined with a volume correspondence table calibrated on-site.

[0049] Absolute pressure is the actual pressure value inside the gas chamber relative to a vacuum reference. It can be obtained through an absolute pressure sensor.

[0050] Absolute temperature is the absolute temperature value of the gas inside the chamber at the current moment. It can be obtained through a pressure-resistant temperature sensor, a multi-point thermal resistor in the chamber, or a fiber optic temperature measuring point.

[0051] The piston cross-sectional area is the effective projected area of ​​the piston in the lifting direction. It can be obtained by re-measuring the piston's geometric dimensions and verifying the as-built dimensions of the equipment.

[0052] The rate of change of mechanical sweep volume is the rate of change of mechanical volume determined by both the piston cross-sectional area and the piston instantaneous velocity.

[0053] The first derivative of absolute pressure with respect to time is the instantaneous rate of change of absolute pressure over time.

[0054] The pseudo-pressure volatility is an equivalent volume fluctuation term caused by pressure changes, but not by actual throughput.

[0055] The first derivative of absolute temperature with respect to time is the instantaneous rate of change of absolute temperature over time.

[0056] The pseudo-thermal thermal expansion is caused by changes in gas temperature, but is not an equivalent volumetric thermal expansion term caused by actual gas flow.

[0057] The real gas volume change rate is the effective gas volume change rate obtained after deducting the pseudo-contributions of pressure fluctuations and thermal expansion.

[0058] In practice, for the acquisition of the actual volume of the gas chamber, the piston position measurement module is needed to obtain the current piston height, and then the piston height is mapped to the actual volume of the gas chamber by combining the gas cabinet structure calibration table. For gas cabinets with regular cross-sections, the piston cross-sectional area can be multiplied by the effective height for conversion. For gas cabinets with modified cross-sections, reinforcements, or local irregular spaces, the height-volume lookup table relationship obtained after completion calibration should be used.

[0059] In practical implementation, for the arrangement and calibration of absolute pressure and absolute temperature sensors, at least two absolute pressure sensors need to be set up, one at a representative position in the air chamber and the other at a backup redundant position, using absolute pressure transmitters with the same range; at least three absolute temperature sensors need to be set up, preferably at three heights in the air chamber, and the representative temperature is obtained by weighted averaging; all sensors should be factory calibrated before installation, and the zero point and range should be checked on site after installation, preferably once every 6 months.

[0060] In practical implementation, for approximating small rates of change in time, both the first derivative of absolute pressure with respect to time and the first derivative of absolute temperature with respect to time need to be approximated by discrete difference, preferably by central difference, and by backward difference or forward difference at boundary moments; the control period is preferably 0.1 seconds to 1 second; in order to suppress noise amplification, the original pressure and temperature signals should be low-pass filtered or averaged before the difference is applied.

[0061] In practical implementation, regarding the applicable boundaries and corrections for the real gas volume change rate, the real gas volume change rate in this application refers to the effective throughput volume change rate after removing the pressure fluctuation pseudo-term and the temperature thermal expansion pseudo-term, which is based on the first-order differential approximation of the state relationship. When the gas composition is stable and the pressure variation range is small, the compressibility factor can be incorporated into the calibration coefficient for unified processing. When the gas composition fluctuates greatly or the compressibility factor changes significantly under high pressure conditions, a compressibility factor correction term should be added in conjunction with the online component analysis results.

[0062] In practical implementation, regarding the convention of flow direction sign, filtering rules, and outlier handling, it is necessary to clearly define whether the piston's upward direction or downward direction is positive, and to maintain consistency with the sign direction of the mechanical sweep volume change rate, the net flow rate of the pipeline target scheduling, and the subsequent flow fluctuation deviation; sudden peak values, short-term sensor disconnection, and obvious out-of-bounds values ​​should be excluded before calculation and should not be directly included in the volume change rate model.

[0063] Preferably, by utilizing the inherent mass of the piston and the real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless sticking self-excitation strength index, including: Obtain gravitational acceleration and target net flow rate of the pipeline network; The spatial vector magnitude of the global eccentric friction torque is calculated and divided by the product of the piston's inherent mass, gravitational acceleration, and piston radius to obtain the dimensionless torque load ratio. Subtracting the target net flow rate of the pipeline network from the actual gas volume change rate and taking the absolute value, the flow fluctuation deviation is obtained. Dividing the flow fluctuation deviation by the product of the piston cross-sectional area and the real-time sound velocity yields the dimensionless throughput limit ratio. Adding the numerical value to the dimensionless throughput limit ratio, and multiplying the whole by the dimensionless torque load ratio, yields the jamming self-excitation strength index. It satisfies the following relationship: in, The self-excitation intensity index is the self-excitation intensity index. Let be the spatial vector magnitude of the global eccentric frictional torque. The inherent mass of the piston. Let gravitational acceleration be the acceleration due to gravity. The piston radius is... The actual gas volume change rate, Schedule net flow for the target pipeline network. The cross-sectional area of ​​the piston is... The real-time speed of sound.

[0064] The inherent mass of a piston is the total mass of the piston body and its fixed accessories. It can be obtained through manufacturing record verification, hoisting weighing, or load measurement during maintenance.

[0065] Real-time sound velocity is the instantaneous speed of sound propagation in a gas under the current gas composition, pressure, and temperature conditions. It can be obtained through online gas acoustic measurement modules, ultrasonic propagation measurement modules, or sound velocity calibration devices.

[0066] Gravitational acceleration is a fixed gravitational constant used for torque normalization. A preferred value is 9.8 m / s², which is a standard value commonly used in engineering control.

[0067] The target net throughput of the pipeline network is the target net throughput flow issued to the gas storage unit by the superior dispatching system. The preferred value is 0.4 to 0.9 times the rated instantaneous throughput capacity. This range can take into account both conventional peak-shaving capacity and actuator redundancy.

[0068] The spatial vector magnitude of the global eccentric friction torque is the length of the global eccentric friction torque vector.

[0069] The dimensionless torque-load ratio is a torque-load index that normalizes the spatial vector magnitude of the global eccentric friction torque according to the piston's inherent mass, gravitational acceleration, and piston radius.

[0070] Flow fluctuation deviation is the absolute deviation between the actual gas volume change rate and the target net flow rate of the pipeline network.

[0071] The dimensionless throughput over-limit ratio is a throughput deviation index that normalizes the flow fluctuation deviation according to the piston cross-sectional area and real-time sound velocity.

[0072] The jamming self-excitation intensity index is a comprehensive jamming risk characterization quantity formed by combining the dimensionless torque load ratio and the dimensionless throughput limit ratio.

[0073] In practice, for obtaining the piston's inherent mass and real-time sound velocity, the piston's inherent mass should be verified primarily using manufacturing and assembly data. During major overhauls and hoisting, it can be verified by weighing the hook or measuring the support load. The real-time sound velocity should be measured directly using an online acoustic module. Alternatively, it can be estimated by looking up tables using absolute pressure, absolute temperature, and composition when the gas composition is stable, and then corrected by acoustic verification values.

[0074] In specific implementation, regarding the source and refresh mechanism of the target scheduling net flow of the pipeline network, this parameter needs to be issued by the superior scheduling platform, station control system or gas balance control system, and be consistent with the gas supply plan for the current period; the refresh cycle is preferably 1 to 10 seconds; if no new value is received within the refresh cycle, the previous valid value is retained; if the superior scheduling value suddenly exceeds the preset step threshold, a ramp transition should be performed to prevent instantaneous overshoot in subsequent valve allocation.

[0075] In practical implementation, to obtain the spatial vector magnitude of the global eccentric friction torque, it is necessary to first obtain the two-dimensional or three-dimensional vector expression of the global eccentric friction torque according to claim 3, and then calculate its magnitude according to a unified coordinate system. The coordinate system should use the same origin and the same rotation direction as the direction unit vector, the valve position direction vector, and the subsequent reverse jet thrust torque. If a two-dimensional planar model is used, the spatial vector magnitude can be simplified to the planar vector length.

[0076] In practical implementation, the classification and use of the self-excitation intensity index for jamming should be as follows: if the index is in the range of 0 to 0.15, it can be judged as low risk, and the system should maintain normal equilibrium control; if the index is in the range of 0.15 to 0.35, it can be judged as a warning state, and mild asymmetric flow distribution should be initiated; if the index is in the range of 0.35 to 0.60, it can be judged as a high risk state, and forced correction and heat tracing linkage should be initiated; if the index is greater than 0.60, it can be judged as a protection state, triggering speed limit, alarm and valve safety constraints.

[0077] In specific implementation, regarding the update cycle and triggering logic of the self-excitation intensity index, the update cycle of the self-excitation intensity index should be consistent with the main control cycle, preferably 0.1 seconds to 0.5 seconds; after each update, amplitude limiting and anti-jitter processing should be performed first, and then used as the input source of the inertial attenuation factor in claim 6; when the index rises for three consecutive cycles and the rise exceeds the preset threshold, a stronger valve asymmetric flow distribution should be triggered in advance.

[0078] Preferably, the asymmetric array valve body volumetric flow vector is derived by driving endogenous particle swarm optimization based on the self-excitation intensity index of the hysteresis, including: The gas density, the cross-sectional area of ​​the array pneumatic valve, and the valve position direction vector are obtained, and the global highest detection temperature is obtained based on the circumferential absolute temperature field. Construct a fitness function without hyperparameters, which is composed of the sum of a torque penalty term and a flow penalty term; The single-valve volumetric flow rate is defined in the optimization process. The torque penalty term is the sum of the global eccentric friction torque and the single-phase reverse jet thrust torque of all array pneumatic valves, then the spatial vector magnitude is calculated, divided by the product of the piston's inherent mass, the gravitational acceleration, and the piston radius, and the square is obtained. The single-phase reverse jet thrust torque is the gas density multiplied by the square of the single-valve volumetric flow rate, divided by the cross-sectional area of ​​the array pneumatic valve, and then multiplied by the valve position direction vector. The flow rate penalty term is the sum of all the single-valve volumetric flow rates minus the target scheduling net flow rate of the pipeline network, divided by the product of the piston cross-sectional area and the real-time sound velocity, and the square is obtained. The negative power of the natural logarithm base, which is the exponent of the self-excitation intensity of the hysteresis, is used as the inertial decay factor. The environmental preference coordination factor is obtained by dividing the integral circumferential temperature by the global highest detected temperature. Subtracting the environmental preference coordination factor from the numerical value yields the global cognitive compliance factor. Initialize the particle velocity, individual historical optimal solution, and global historical optimal solution for the optimization iteration, and generate the first and second random numbers during the iteration; The particle velocity of the previous generation is multiplied by the inertial decay factor, and the product of the difference between the individual historical optimal solution and the current generation's single valve volume flow rate, weighted by the environmental preference coordination factor and the first random number, is added. The difference between the global historical optimal solution and the current generation's single valve volume flow rate, weighted by the global cognitive obedience factor and the second random number, is then added to iterate and obtain the particle velocity of the current generation. Based on the hyperparameter-free fitness function Based on the principle of minimization, a set of optimal single-valve volumetric flow rates is output to form the volumetric flow rate vector of the array valve body. Among them, the hyperparameter-free fitness function and the particle velocity of the current generation. The iteration satisfies the following relationship: And it satisfies: in, For the hyperparameter-free fitness function, The volumetric flow vector of the array valve body. This represents the total number of pneumatic valves. The density of the gas, The single valve volumetric flow rate, Let be the cross-sectional area of ​​a single pneumatic valve in the array. The valve position direction vector, The particle velocity of the current generation, The particle velocity described in the previous generation, The integral circumferential temperature, The highest globally detected temperature. The first random number, The second random number, This is the historical optimal solution for the stated individual. This is the globally historical optimal solution. This refers to the current generation of single-valve volumetric flow rate.

[0079] Gas density is the mass per unit volume of a gas under actual operating conditions. It can be obtained through an online density meter, a differential pressure density measuring device, or a calibrated gas composition calculation module.

[0080] The cross-sectional area of ​​a single array pneumatic valve is the effective flow cross-sectional area of ​​a single array pneumatic valve when fully open or at a specified opening degree. It can be obtained through actual measurement of valve structural dimensions, manufacturer calibration reports, or flow tests in the fully open state.

[0081] The valve position direction vector is a vector description that characterizes the spatial position of each array of pneumatic valves relative to the piston center and the direction of its torque.

[0082] The highest global detected temperature is the highest temperature value detected at the current moment in the circumferential absolute temperature field.

[0083] The hyperparameter-free fitness function is an objective function used to evaluate the quality of a set of array valve body volume flow vectors, and no additional artificial weighting coefficients are set between the torque penalty term and the flow penalty term.

[0084] The torque penalty term is a normalized penalty that characterizes the effect of the current valve distribution scheme on suppressing eccentric torque.

[0085] The flow penalty term is a normalized penalty that characterizes the degree of deviation between the current valve flow distribution scheme and the target net flow of the pipeline network.

[0086] Single-valve volumetric flow rate is the candidate volumetric flow rate allocated to a single array of pneumatic valves during the optimization process.

[0087] The single-phase reverse jet thrust torque is the single-phase torque contribution of the jet reaction of a single array of pneumatic valves to the piston.

[0088] The inertia decay factor is the search inertia decay amount obtained by mapping the hysteresis self-excitation intensity index.

[0089] The environmental preference synergy factor is a local environmental preference coefficient formed by the ratio of the integral circumferential temperature to the global highest detected temperature.

[0090] The global cognitive compliance factor is the globally optimal compliance coefficient obtained by complementing the environmental preference coordination factor.

[0091] Particle velocity is the step change of each particle in the single-valve volumetric flow search space during the particle swarm optimization process.

[0092] The individual historical optimal solution is the optimal single-valve volumetric flow distribution result obtained by a certain particle in previous iterations.

[0093] The global historical optimal solution is the optimal single-valve volumetric flow distribution result obtained by all particles in previous iterations.

[0094] The first random number is a random weighting coefficient used to adjust the traction strength of an individual's historical best solution.

[0095] The second random number is a random weighting coefficient used to adjust the traction strength of the global historical best solution.

[0096] The array valve body volumetric flow vector is the target flow distribution result formed by the combination of the volumetric flow of all individual valves.

[0097] The total number of pneumatic valves is the total number of array pneumatic valves participating in this optimization and control allocation. The preferred value is 8 to 24, which takes into account circumferential resolution, single valve actuation margin, and maintenance complexity.

[0098] In practical implementation, for obtaining gas density, the cross-sectional area of ​​the array pneumatic valve, and the valve position direction vector, the gas density should preferably be directly provided by an online density meter. If an online density meter is not available, it should be estimated from tables based on absolute pressure, absolute temperature, and gas composition, and then calibrated using an empirical correction table. The cross-sectional area of ​​the array pneumatic valve should not be directly replaced by the nominal diameter, but should be the effective flow cross-sectional area of ​​the valve. The valve position direction vector should be constructed based on the piston center as the coordinate origin and the circumferential angle of each valve installation position, and its direction should correspond to the direction of the torque formed by the jet reaction of the valve on the piston.

[0099] In practical implementation, for the engineering calculation of the thrust torque of a single reverse jet, it is necessary to first convert the volumetric flow rate of a single valve into the average flow velocity at the valve orifice, and then use the gas density and the effective flow cross-sectional area of ​​the valve orifice to calculate the jet flow rate; then, combined with the lever arm effect from the valve installation point to the piston center, the thrust torque of the single reverse jet corresponding to the valve is formed; at the real-time control level, it is permissible to use a square approximation model for rapid calculation, and it is not required to perform complex fluid simulation for each control cycle.

[0100] In practical implementation, for the constraints of the fitness function without hyperparameters, the so-called "no hyperparameters" means that no additional artificial weighting coefficients are set between the torque penalty term and the flow penalty term, but they are directly added based on a unified normalized benchmark. At the same time, a lower limit for single valve volumetric flow rate, an upper limit for single valve volumetric flow rate, a total flow conservation constraint, and a single-cycle rate of change constraint must be set. The lower limit for single valve volumetric flow rate is preferably 0, and the upper limit is preferably 1.05 to 1.20 times the rated flow rate of a single valve, so as to retain short-term correction margin.

[0101] In specific implementation, for the initialization and search boundary settings of endogenous particle swarm optimization, the number of particles is preferably 10 to 40, and the maximum number of iterations is preferably 20 to 80. The initial single-valve volumetric flow rate can be generated by fluctuating up and down by 10 to 20% of the uniform distribution value. The initial particle velocity can be set by 5 to 15% of the rated flow rate of a single valve. The individual historical optimal solution is equal to the initial position of each particle at the beginning, and the global historical optimal solution is the one with the smallest initial fitness.

[0102] In practical implementation, the limiting of the inertial decay factor, environmental preference coordination factor, and global cognitive compliance factor requires that the inertial decay factor decrease as the hysteresis excitation intensity exponent increases, so as to accelerate convergence under high-risk conditions; the environmental preference coordination factor is obtained by the ratio of the integral circumferential temperature to the global highest detected temperature, and should be limited to the range of 0.05 to 0.95 after calculation; the global cognitive compliance factor is obtained by subtracting the environmental preference coordination factor from 1, and should also be subject to the same limiting.

[0103] In practice, the generation of the first and second random numbers requires that the first and second random numbers be generated independently in each iteration, for each particle, and for each valve position, with a value range of 0 to 1. A uniformly distributed pseudo-random sequence is preferred. To improve control repeatability, a fixed random seed can be used, or a seed can be obtained by combining the system clock and the valve position number.

[0104] In practice, for the projection correction of particle velocity and single valve volumetric flow, it is necessary to first obtain the new particle velocity according to the velocity iteration formula, and then superimpose the particle velocity onto the current single valve volumetric flow to form a candidate flow. If negative flow, flow exceeding the upper limit, or total flow deviation exceeds the limit occurs, single valve cutoff should be performed first, and then the total flow normalization projection should be performed again so that the sum of all single valve volumetric flow returns to near the target scheduling net flow of the pipeline network.

[0105] In specific implementation, the convergence criteria and output rules for the volumetric flow vector of the array valve body should be based on minimizing the fitness function without hyperparameters, and stopping when the improvement amplitude is less than the convergence threshold for 2 to 5 consecutive iterations. The convergence threshold is preferably 0.5 to 2% of the optimal fitness of the previous round. When outputting, the one-to-one correspondence between the valve number and the target flow should be retained and written to the next level execution cache.

[0106] Preferably, with the fluid's natural relaxation time as a constraint, the volumetric flow vector of the array valve body is inversely analyzed into the target angular acceleration of the valve shaft, and then sent to the underlying frequency converter, including: Obtain the average absolute height of the piston pressure chamber and the actual volumetric flow rate of the pneumatic valve; Divide the real-time sound velocity by the average absolute height of the piston pressure chamber to obtain the fluid sound wave relaxation frequency, and square the fluid sound wave relaxation frequency to obtain the angular frequency limiting term. Extract the target intake volume of a single valve from the volumetric flow vector of the array valve body, and subtract the corresponding actual volumetric flow to obtain the flow regulation gap; Divide the flow regulation gap by the product of the piston cross-sectional area and the real-time sound velocity, and calculate the arcsine function value to obtain the dimensionless driving phase angle; Multiplying the angular frequency limit term by the dimensionless drive phase angle, the target angular acceleration of the valve shaft is generated and sent to the underlying frequency converter. It satisfies the following relationship: in, The target angular acceleration of the valve shaft. The average absolute height of the piston pressure chamber. The target intake volume for the single valve is... This corresponds to the actual volumetric flow rate.

[0107] The natural relaxation time of a fluid is the characteristic time corresponding to the completion of one principal-scale equilibrium of a pressure disturbance in a pressure chamber after acoustic propagation.

[0108] The average absolute height of the piston pressure chamber is the average effective height of the piston pressure chamber in the main propagation direction at the current moment. It can be obtained by combining the piston position sensor, the top cover distance measurement module, and the height calibration table.

[0109] The actual volumetric flow rate of a pneumatic valve is the actual volumetric flow rate of gas passing through each pneumatic valve at the current moment. It can be obtained through a differential pressure flow meter downstream of the valve, an ultrasonic flow meter, or a calibrated valve flow feedback module.

[0110] The relaxation frequency of fluid sound waves is the acoustic equilibrium frequency characterized by the real-time sound velocity and the average absolute height of the piston pressure chamber.

[0111] The angular frequency limit is the upper limit constraint on the valve dynamics formed by the square of the fluid acoustic relaxation frequency.

[0112] The target intake volume of a single valve is extracted from the volumetric flow vector of the array valve body and allocated to the target volumetric flow of a specific pneumatic valve.

[0113] The flow regulation gap is the difference between the target intake volume of a single valve and the actual volumetric flow rate of that valve.

[0114] The dimensionless driving phase angle is the driving phase quantity obtained by normalizing the flow regulation gap and then mapping it with an arcsine.

[0115] The target angular acceleration of the valve shaft is the target angular acceleration of the valve shaft that is sent to the underlying frequency converter to drive the valve actuator.

[0116] In practical implementation, in view of the correspondence between the natural relaxation time of the fluid and the relaxation frequency of the fluid sound wave, this application needs to take the sound propagation time formed by the average absolute height of the piston air pressure chamber and the real-time sound speed as the engineering approximation of the natural relaxation time of the fluid; its reciprocal is taken as the relaxation frequency of the fluid sound wave, and then the square of this frequency is used to obtain the angular frequency limit term, which is used to limit the valve angular acceleration from being faster than the pressure equalization capability of the air pressure chamber itself.

[0117] In practical implementation, to obtain the average absolute height of the piston air pressure chamber and the actual volumetric flow rate of the pneumatic valve, the average absolute height of the piston air pressure chamber needs to be determined by the current height of the piston, the fixed reference height of the cabinet top, and the local structural correction value. The actual volumetric flow rate of the pneumatic valve is preferentially collected by an independent flow feedback channel for each valve. If hardware cost limitations prevent independent measurement for each valve, the valve opening degree, the pressure difference across the valve, and the valve flow characteristic curve can be used for joint estimation.

[0118] In specific implementation, for the mapping of the target air intake volume of a single valve to the pneumatic valve number, each component in the array valve body volume flow vector must be bound to a unique valve number; the valve number should be completely consistent with the valve position direction vector number used in claim 6, and reordering is not allowed; when a valve is out of maintenance, the array valve body volume flow vector length, valve position direction vector, and execution mapping table should be updated synchronously.

[0119] In practical implementation, for limiting the dimensionless driving phase angle, the flow regulation gap should be normalized by dividing it by the product of the piston cross-sectional area and the real-time sound velocity before entering the arcsine mapping. The normalization result must be limited to the range of -1 to +1, preferably within the range of -0.95 to +0.95, to prevent numerical out-of-bounds errors and control abrupt changes.

[0120] In practice, for the unit conversion and transmission interface of the target angular acceleration of the valve shaft, the target angular acceleration of the valve shaft calculated by the controller should be uniformly converted into an execution unit that can be recognized by the underlying frequency converter, and then transmitted through the fieldbus, control bus or dedicated drive interface; the transmitted content should at least include the valve number, target angular acceleration, valid timestamp and safety status bit.

[0121] In practical implementation, for the safety limit of the target angular acceleration of the valve shaft and the execution sequence, each pneumatic valve needs to be pre-calibrated with the maximum allowable angular acceleration, the maximum allowable angular velocity and the maximum allowable rate of change of opening; when the calculated target angular acceleration of the valve shaft exceeds the safety limit, it should be executed according to the safety limit; the control cycle is preferably 0.05 seconds to 0.2 seconds, and the valve execution feedback must be returned before the next control cycle.

[0122] Preferably, based on the local instantaneous sealing gap, the microscopic non-uniform shear stress, and the absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power. The localized electromagnetic induction heat tracing array is then controlled to perform compensation, including: Obtain the minimum shear stress corresponding to the thermal conductivity of the sealing oil and the hottest side globally; The difference between the global highest detected temperature and the absolute oil temperature is divided by the local instantaneous sealing gap and then multiplied by the thermal conductivity of the sealing oil to obtain the rheological conduction heat flux. The mechanical friction work dissipation rate is obtained by subtracting the minimum shear stress from the microscopic non-uniform shear stress and then multiplying it by the instantaneous piston speed. The localized injected heat power is calculated by adding the rheological conduction heat flux and the mechanical friction power dissipation rate, multiplying the sum by the effective sealing height and the piston radius. It satisfies the following relationship: in, Inject thermal power into the localized area. The thermal conductivity of the sealing oil, The highest globally detected temperature. The absolute oil temperature, The local instantaneous sealing gap, For the microscopic non-uniform shear stress, The minimum shear stress, The instantaneous velocity of the piston is... The effective height of the seal, Let be the piston radius.

[0123] The thermal conductivity of a sealing oil is a parameter describing its ability to transfer heat along a temperature gradient. A preferred value is 0.12 to 0.16 W / m / °C, a range that meets the thermal conductivity requirements of common mineral-based and synthetic sealing oils in their typical operating temperature range.

[0124] The minimum shear stress corresponding to the hottest side globally is the minimum reference shear stress value selected within the circumferential region where the highest detected temperature is located globally.

[0125] The rheological heat transfer flux is a local heat transfer compensation flux determined by the local temperature difference, the local instantaneous sealing gap, and the thermal conductivity of the sealing oil.

[0126] The rate of energy dissipation due to mechanical friction is the localized rate of energy loss due to mechanical friction, which is determined by the microscopic non-uniform shear stress and the instantaneous velocity of the piston.

[0127] Localized injected heat power is the target compensation power allocated to the corresponding heat tracing unit in a certain circumferential local area.

[0128] In practical implementation, the thermal conductivity of the sealing oil should be determined based on the actual grade of the sealing oil used, through the manufacturer's property report, third-party testing report, or on-site sampling test results. If the sealing oil grade is changed, the water content changes significantly, or it has been aging for a long time, this parameter should be updated. The controller can establish a property database according to oil batches and call the corresponding thermal conductivity according to the current oil batch number.

[0129] In practice, to extract the minimum shear stress corresponding to the hottest side globally, it is necessary to first find the circumferential region where the highest detected temperature is located in the circumferential absolute temperature field, and then extract the stable minimum value as the minimum shear stress baseline in the microscopic non-uniform shear stress distribution corresponding to this region using a local search window. The search window preferably covers a range of 10 to 30 degrees to the left and right of the hottest spot. To prevent noise misjudgment, the shear stress should be averaged before extracting the minimum value.

[0130] In practical implementation, for the spatial matching of rheological conduction heat flux, the rheological conduction heat flux must use the highest global detection temperature, absolute oil temperature and local instantaneous sealing gap at the same polar angle position, and cannot be spliced ​​across positions for calculation; the circumferential angle, timestamp and local gap of the same calculation unit must be consistent; when the sampling resolution of the circumferential absolute temperature field and absolute oil temperature is different, they must be resampled to the same circumferential grid first.

[0131] In practical implementation, regarding the sign convention and local calculation of the mechanical friction work dissipation rate, the difference between the microscopic non-uniform shear stress and the minimum shear stress needs to represent the local additional friction component above the baseline; this additional friction component is then multiplied by the piston instantaneous velocity to obtain the local mechanical friction work dissipation rate; when the piston instantaneous velocity reverses, it should be treated according to the absolute energy consumption direction, and the dissipation rate should not be negative due to the sign switch.

[0132] In practice, to map the localized injected heat power to the localized electromagnetic induction heat tracing array, the sealing strip needs to be divided into 8 to 24 heat tracing zones along the circumference. Each heat tracing zone corresponds to one or more localized electromagnetic induction heat tracing units. The localized injected heat power is first mapped to the nearest heat tracing zone according to the polar angle position, and then distributed to adjacent zones according to the neighborhood weighting method to prevent single-point overheating.

[0133] In practice, for the limiting of compensation power and the handling of negative power, each heat tracing zone should be set with a maximum allowable power, a maximum heating rate, and a maximum oil temperature protection value. When the calculated result of the localized injected heat power is negative, it should be treated as 0 and no cooling action should be performed. When the oil temperature of a certain zone has reached the maximum allowable value, even if the localized injected heat power is still positive, the output of that zone should be forcibly limited or cut off.

[0134] Example 2: An automated control method for a gas cabinet, applied in any of the automated control systems for a gas cabinet described in the previous example, comprising: Obtain the instantaneous piston speed and absolute oil temperature; The distributed fiber optic grating sensor array is used to acquire the circumferential absolute temperature field and reconstruct the local instantaneous sealing gap; Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and the absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The inherent mass of the piston and the real-time sound velocity are obtained. Using the inherent mass of the piston and the real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless jamming self-excitation strength index. Based on the self-excitation intensity index of the hysteresis, the endogenous particle swarm optimization is driven, and the asymmetric array valve body volume flow vector is deduced. With the fluid's natural relaxation time as a constraint, the array valve body volume flow vector is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, the microscopic non-uniform shear stress, and the absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

[0135] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. An automated control system for a gas cabinet, applied to a gas cabinet system comprising a distributed fiber Bragg grating sensor array, an array pneumatic valve, a bottom-level frequency converter, and a localized electromagnetic induction heat tracing array, characterized in that, The system executes the following sequentially: Obtain the instantaneous piston speed and absolute oil temperature; The absolute temperature field around the circumference is obtained by using a distributed fiber Bragg grating sensor array, and the local instantaneous sealing gap is reconstructed. Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The piston's inherent mass and real-time sound velocity are obtained. Using the piston's inherent mass and real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless sticking self-excitation strength index. Based on the endogenous particle swarm optimization driven by the self-excitation intensity index, the asymmetric array valve body volume flow vector is deduced. With the natural relaxation time of the fluid as a constraint, the volumetric flow vector of the array valve body is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, microscopic non-uniform shear stress, and absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

2. The automated control system for a gas cabinet according to claim 1, characterized in that, The distributed fiber optic grating sensor array is used to acquire the circumferential absolute temperature field and reconstruct the local instantaneous sealing gap, including: Obtain the inner diameter of the cylindrical cabinet wall and the inherent linear expansion coefficient of the steel, as well as the initial nominal mechanical clearance; The integral circumferential temperature field is obtained by integrating the absolute temperature field over the entire circumference and dividing it by twice the value of pi. The differential mode temperature is obtained by subtracting the integral circumferential temperature from the absolute circumferential temperature field. By continuously multiplying the inner diameter of the cylindrical cabinet wall, the inherent linear expansion coefficient of the steel, and the differential temperature, the transient thermo-elastic-plastic distortion variable is obtained. The local instantaneous sealing gap is obtained by subtracting the transient thermo-elastic-plastic distortion from the initial nominal mechanical gap.

3. An automated control system for a gas cabinet according to claim 2, characterized in that, The microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and the absolute oil temperature, and the global eccentric friction torque is extracted by integration in polar coordinates, including: To obtain the fluid's infinite thermodynamic viscosity limit, activation energy, and ideal gas constant, as well as the effective sealing height and piston radius; The rheological correction coefficient is obtained by dividing the activation energy by the product of the ideal gas constant and the absolute oil temperature, and taking the natural exponent. Multiply the rheological correction factor by the fluid's infinite temperature dynamic viscosity limit to obtain the transient dynamic viscosity; The transient dynamic viscosity is multiplied by the instantaneous piston velocity and divided by the local instantaneous sealing gap to obtain the microscopic non-uniform shear stress. Using the integral polar angle in polar coordinates as the independent variable, a directional unit vector is constructed using the sine and negative cosine values ​​of the integral polar angle; Multiply the microscopic non-uniform shear stress by the square of the effective sealing height and the piston radius, then multiply by the directional unit vector, and perform polar coordinate vector integration on the complete circumference to extract the global eccentric friction torque.

4. An automated control system for a gas cabinet according to claim 3, characterized in that, The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on a small rate of change over time, including: Obtain the cross-sectional area of ​​the piston, multiply the cross-sectional area of ​​the piston by the instantaneous velocity of the piston, and obtain the mechanical sweep volume change rate; Divide the measured volume of the air chamber by the absolute pressure, and multiply by the first derivative of the absolute pressure with respect to time to obtain the pseudo pressure fluctuation rate. Divide the measured volume of the air chamber by the absolute temperature, and multiply by the first derivative of the absolute temperature with respect to time to obtain the pseudo-thermal thermal expansion coefficient; The real gas volume change rate is obtained by subtracting the pseudo pressure fluctuation rate from the mechanical sweep volume change rate and adding the pseudo temperature thermal expansion rate.

5. An automated control system for a gas cabinet according to claim 4, characterized in that, Using the inherent mass of the piston and the real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless sticking self-excitation strength index, including: Obtain gravitational acceleration and target net flow rate of the pipeline network; The spatial vector magnitude of the global eccentric friction torque is calculated and divided by the product of the piston's inherent mass, gravitational acceleration, and piston radius to obtain the dimensionless torque load ratio. Subtracting the target net flow rate of the pipeline network from the actual gas volume change rate and taking the absolute value, the flow fluctuation deviation is obtained. Dividing the flow fluctuation deviation by the product of the piston cross-sectional area and the real-time sound velocity yields the dimensionless throughput limit ratio. Add the numerical value to the dimensionless throughput limit ratio, multiply the whole by the dimensionless torque load ratio, and synthesize the jamming self-excitation strength index.

6. An automated control system for a gas cabinet according to claim 5, characterized in that, Based on the intrinsic particle swarm optimization driven by the self-excitation intensity index, the asymmetric array valve body volumetric flow vector is deduced, including: The gas density, the cross-sectional area of ​​the array pneumatic valve, and the valve position direction vector are obtained, and the global highest detection temperature is obtained based on the circumferential absolute temperature field. Construct a fitness function without hyperparameters, which is composed of the sum of a torque penalty term and a flow penalty term; The single-valve volumetric flow rate is defined in the optimization process. The torque penalty term is the sum of the global eccentric friction torque and the single-phase reverse jet thrust torque of all array pneumatic valves, then the spatial vector magnitude is calculated, divided by the product of the piston's inherent mass, the gravitational acceleration, and the piston radius, and the square is obtained. The single-phase reverse jet thrust torque is the gas density multiplied by the square of the single-valve volumetric flow rate, divided by the cross-sectional area of ​​the array pneumatic valve, and then multiplied by the valve position direction vector. The flow rate penalty term is the sum of all the single-valve volumetric flow rates minus the target scheduling net flow rate of the pipeline network, divided by the product of the piston cross-sectional area and the real-time sound velocity, and the square is obtained. The negative power of the natural logarithm base, which is the exponent of the self-excitation intensity of the hysteresis, is used as the inertial decay factor. The environmental preference coordination factor is obtained by dividing the integral circumferential temperature by the global highest detected temperature; the global cognitive obedience factor is obtained by subtracting the environmental preference coordination factor from the value. Initialize the particle velocity, individual historical optimal solution, and global historical optimal solution for the optimization iteration, and generate the first and second random numbers during the iteration; The particle velocity of the previous generation is multiplied by the inertial decay factor, and the product of the difference between the individual historical optimal solution and the current generation's single valve volume flow rate, weighted by the environmental preference coordination factor and the first random number, is added. The difference between the global historical optimal solution and the current generation's single valve volume flow rate, weighted by the global cognitive obedience factor and the second random number, is then added to iterate and obtain the particle velocity of the current generation. Based on the principle of minimizing the hyperparameter-free fitness function, a set of optimal single-valve volumetric flow rates is output to form the array valve body volumetric flow rate vector.

7. An automated control system for a gas cabinet according to claim 6, characterized in that, Constrained by the fluid's natural relaxation time, the volumetric flow vector of the array valve body is inversely analyzed into the target angular acceleration of the valve shaft, and then sent to the underlying frequency converter, including: Obtain the average absolute height of the piston pressure chamber and the actual volumetric flow rate of the pneumatic valve; Divide the real-time sound velocity by the average absolute height of the piston pressure chamber to obtain the fluid sound wave relaxation frequency, and square the fluid sound wave relaxation frequency to obtain the angular frequency limiting term. Extract the target intake volume of a single valve from the volumetric flow vector of the array valve body, and subtract the corresponding actual volumetric flow to obtain the flow regulation gap; Divide the flow regulation gap by the product of the piston cross-sectional area and the real-time sound velocity, and calculate the arcsine function value to obtain the dimensionless driving phase angle; Multiply the angular frequency limit term by the dimensionless drive phase angle to generate the target angular acceleration of the valve shaft, which is then sent to the underlying frequency converter.

8. An automated control system for a gas cabinet according to claim 7, characterized in that, Based on the local instantaneous sealing gap, the microscopic non-uniform shear stress, and the absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power. The localized electromagnetic induction heat tracing array is then controlled to perform compensation, including: Obtain the minimum shear stress corresponding to the thermal conductivity of the sealing oil and the hottest side globally; The difference between the global highest detected temperature and the absolute oil temperature is divided by the local instantaneous sealing gap and then multiplied by the thermal conductivity of the sealing oil to obtain the rheological conduction heat flux. The mechanical friction work dissipation rate is obtained by subtracting the minimum shear stress from the microscopic non-uniform shear stress and then multiplying it by the instantaneous piston speed. The localized injected heat power is calculated by adding the rheological conduction heat flux and the mechanical friction work dissipation rate, multiplying the total by the effective sealing height and the piston radius.

9. An automated control method for a gas cabinet, applied in an automated control system for a gas cabinet as described in any one of claims 1-8, characterized in that, include: Obtain the instantaneous piston speed and absolute oil temperature; The distributed fiber optic grating sensor array is used to acquire the circumferential absolute temperature field and reconstruct the local instantaneous sealing gap; Microscopic non-uniform shear stress is extracted by combining the instantaneous piston velocity and the absolute oil temperature, and global eccentric friction torque is extracted by integration in polar coordinates. The measured volume, absolute pressure, and absolute temperature of the gas chamber are collected, and the actual gas volume change rate is extracted based on the small change rate over time. The inherent mass of the piston and the real-time sound velocity are obtained. Using the inherent mass of the piston and the real-time sound velocity, the global eccentric friction torque and the real gas volume change rate are combined to generate a dimensionless jamming self-excitation strength index. Based on the self-excitation intensity index of the hysteresis, the endogenous particle swarm optimization is driven, and the asymmetric array valve body volume flow vector is deduced. With the fluid's natural relaxation time as a constraint, the array valve body volume flow vector is inversely analyzed into the target angular acceleration of the valve shaft and sent down to the underlying frequency converter; Based on the local instantaneous sealing gap, the microscopic non-uniform shear stress, and the absolute oil temperature, the mechanical friction work dissipation rate and rheological conduction heat flux are calculated and converted into localized localized injected heat power to control the localized electromagnetic induction heat tracing array to perform compensation.

Images