A method of fluid pressure monitoring for a heavy duty workbench

Abstract

The application relates to the technical field of static pressure support monitoring, and discloses a fluid pressure monitoring method for a heavy-duty workbench, which comprises the following steps: a spatial coordinate model is established for a multi-oil-cavity system; pressure signals and real-time temperature values of each monitoring node are synchronously collected; a reference oil pressure is corrected by using a temperature parameter; a high-frequency pressure pulsation component is separated by using band-pass filtering; a signal response lag time of the component relative to the reference oil pressure is determined; a spatial pressure distribution field is decoupled in combination with a spatial coordinate distribution of the reference oil pressure; false pressure interference components induced by elastic deformation of the workbench and a base thereof are identified and removed; the application uses the propagation characteristics of high-frequency perturbation to strip the influence of structural deformation, solves the nonlinear distortion problem of pressure sensing under heavy-load and partial-load working conditions, obtains an oil film gap characteristic parameter, and improves the operation stability of the support system.

Description

Technical Field

[0001] This invention belongs to the field of hydrostatic support monitoring technology, and particularly relates to a fluid pressure monitoring method for a heavy-duty workbench. Background Technology

[0002] Currently, heavy-duty worktables typically use hydrostatic support systems for load bearing. This system provides fluid pressure through multiple independent oil chambers, which distributes the load more evenly across the support positions and maintains the stability of the worktable. For equipment that bears tens of tons and requires motion precision down to the micrometer level, the formation of a support oil film using fluid lubrication is an important means of ensuring the smooth operation of the worktable. When the machining specifications reach 3 meters or more, existing solutions often use independent control of multiple oil chambers and monitor the pressure values ​​at each support position in real time. The oil film thickness is adjusted based on the monitoring results to maintain the flatness of the table surface and cutting stability.

[0003] According to the Reynolds equations for lubrication and fluid-structure interaction, there is a significant nonlinear relationship between the oil film stiffness and the oil film gap in a hydrostatic support system. In actual operation, once extreme off-center loading occurs, the large mechanical load will cause the heavy-duty workbench and its base to undergo micron-level elastic bending. This structural deformation will change the stress state at the oil film boundary and be superimposed on the measurement results of the pressure sensor. This will cause the collected quasi-static pressure value to contain not only the fluid bearing pressure but also a large amount of additional stress caused by structural bending. The pseudo-pressure interference component caused by structural deformation mainly manifests as low-frequency, quasi-static changes in the time domain. When relying solely on pressure values ​​for conventional monitoring, it is difficult to distinguish between the pressure actually borne by the fluid and the structural elastic stress. Consequently, the oil film thickness calculated from the pressure gradient will exhibit nonlinear distortion. Under extreme off-center loading conditions of 150%, the control system may even make incorrect leveling compensations based on this. Although existing technologies attempt to increase pressure sampling points or use static compensation matrices for correction, the dynamic viscosity of the fluid changes with temperature, and the hydraulic pipeline itself has a cavity effect. It is still difficult to avoid the structural elastic deformation masking the actual changes in fluid pressure.

[0004] Therefore, how to utilize the high-frequency dynamic changes in fluid pressure signals to identify and eliminate pseudo-pressure interference components caused by mechanical structural deformation, and accurately measure the thickness of minute oil films under extreme off-center load conditions, has become a technical problem that needs to be solved in this field. Summary of the Invention

[0005] This invention provides a method for monitoring fluid pressure on a heavy-duty workbench, comprising the following steps: Step 101: Establish a spatial coordinate model for the sixteen independent oil chambers of the hydrostatic support system, and define each independent oil chamber as a pressure monitoring node. Step 102: Obtain the raw fluid pressure signal of each pressure monitoring node, and simultaneously collect the real-time temperature value characterizing the fluid medium environment within the hydrostatic support system. Step 103: Use real-time temperature values ​​to correct temperature drift of each original fluid pressure signal and determine the reference oil pressure of each pressure monitoring node. Step 104: Extract high-frequency pressure pulsation components from each original fluid pressure signal through frequency filtering. The high-frequency pressure pulsation components are used to characterize the dynamic fluctuation characteristics inside the fluid field. Step 105: Calculate the signal response lag time of the high-frequency pressure pulsation component relative to the reference oil pressure, and construct a spatial pressure distribution field by combining the distribution of each reference oil pressure in the spatial coordinate model; Step 106: Based on the mapping relationship between signal response lag time and spatial pressure distribution field, identify the pseudo-pressure interference components induced by the structural deformation of the heavy-duty worktable and its base, remove the pseudo-pressure interference components from the original fluid pressure signal, and output characteristic parameters characterizing the actual oil film gap state.

[0006] Preferably, step 105 specifically includes: step 1051, analyzing the propagation waveform differences of high-frequency pressure pulsation components between adjacent pressure monitoring nodes, and converting the propagation waveform differences into signal response lag time; step 1052, performing interpolation calculations based on the coordinates of each reference oil pressure in the spatial coordinate model to generate a pressure distribution vector characterizing the load center of gravity offset trend; step 1053, correcting the pressure distribution vector through the signal response lag time to determine the coupling ratio between pressure fluctuations caused by mechanical load changes and pressure offsets caused by structural deformation.

[0007] Preferably, step 103 specifically includes: step 1031, determining the hydrodynamic viscosity of the hydrostatic support system under the current environment based on the preset nonlinear correlation curve of fluid viscosity, pressure and temperature; step 1032, using the hydrodynamic viscosity as a compensation factor, performing zero-point correction and proportional gain adjustment on the original fluid pressure signal to obtain the reference oil pressure.

[0008] Preferably, in step 104, an AC signal with a frequency between 10Hz and 500Hz is separated from the original fluid pressure signal using a bandpass filtering algorithm, and used as a high-frequency pressure pulsation component.

[0009] Preferably, in step 105, the spatial coordinate model is divided into multiple local sampling clusters containing four adjacent pressure monitoring nodes, and the pressure change rate within each local sampling cluster is calculated to synthesize the spatial pressure distribution field.

[0010] Preferably, step 106 specifically includes: step 1061, monitoring the rate of change of the signal response lag time, and when the rate of change exceeds a preset abrupt change threshold, determining that there is a pseudo-pressure interference component caused by structural deformation in the original fluid pressure signal; step 1062, dynamically decoupling the spatial pressure distribution field based on the signal response lag time, and separating the real pressure signal caused by fluid static load.

[0011] Preferably, after step 1061, the method further includes: step 1063, calculating the bending deflection of the heavy-duty worktable body and base based on the ratio of the real pressure signal to the pseudo pressure interference component, and outputting the machine tool geometric accuracy compensation command.

[0012] Preferably, in step 102, a multi-channel synchronous sampling controller is used to acquire sixteen raw fluid pressure signals, with a sampling frequency of not less than 2kHz.

[0013] Preferably, the characteristic parameters include the real-time value of the oil film thickness and the load distribution balance, which are used to determine the support stability of the hydrostatic support system.

[0014] Compared with existing technologies, the fluid pressure monitoring method for a heavy-duty workbench of the present invention has the following advantages: 1. This invention utilizes the difference in response speed between the high-frequency pressure pulsation component and the low-frequency elastic deformation of the heavy-duty worktable and its base to distinguish the real pressure signal formed by hydrostatic load-bearing from the pseudo-pressure interference component caused by structural deformation. During the propagation of high-frequency pressure pulsation, the heavy-duty worktable and its base are difficult to change synchronously with the pressure pulsation, and the oil film boundary can be approximated as a quasi-static boundary. The propagation state of pressure pulsation is mainly affected by the fluid volume elasticity and local compression damping. Therefore, it can reduce the interference of structural bending on stress measurement results, reduce the nonlinear distortion of the original fluid pressure signal under heavy load and off-center load conditions, and improve the accuracy of oil film gap state monitoring.

[0015] 2. This invention reflects the propagation state of high-frequency pressure pulsation components between adjacent pressure monitoring nodes by the signal response lag time, so that the monitoring results no longer rely solely on the absolute pressure reading of the pressure sensor. This reduces the impact of pressure reference drift and absolute reading error on the measurement results. Even when extreme off-center loads cause the oil film thickness to decrease rapidly, characteristic parameters reflecting the actual oil film gap state can still be obtained, avoiding delays in feedback adjustment due to pressure reference changes.

[0016] 3. This invention combines reference oil pressure with high-frequency pressure pulsation components for monitoring, which can simultaneously reflect the overall load distribution of the hydrostatic support system and the instantaneous changes in the oil film gap. The reference oil pressure is used to determine the load distribution state, and the signal response lag time is used to reflect the local oil film state changes. This makes up for the insufficiency of relying solely on static pressure amplitude to detect oil film anomalies in a timely manner. Even before the static pressure shows significant changes, the risk of oil film collapse can be identified based on the changes in the pressure pulsation propagation state, thus improving the early warning capability under complex working conditions. Attached Figure Description

[0017] Figure 1 This is a flowchart of the fluid pressure dynamic decoupling process of the present invention; Figure 2 This is a schematic diagram of the oil chamber pressure data processing unit of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0019] It should be noted that all directional and positional terms used in this invention, such as: up, down, left, right, front, back, vertical, horizontal, inner, outer, top, bottom, transverse, longitudinal, center, etc., are only used to explain the relative positional relationship and connection between components in a specific state (as shown in the accompanying drawings). They are only for the convenience of describing this invention and do not require that this invention be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention. In addition, the descriptions of "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated.

[0020] In the description of this invention, unless otherwise explicitly specified and limited, the terms installation, connection, and linking should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections; they can refer to direct connections or indirect connections through an intermediate medium; they can refer to the internal connection of two components. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0021] In the description of this specification, references to the terms "an embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0022] A method for monitoring fluid pressure on a heavy-duty workbench includes the following steps: Step 101: Establish a spatial coordinate model for the sixteen independent oil chambers of the hydrostatic support system, and define each independent oil chamber as a pressure monitoring node. Step 102: Obtain the raw fluid pressure signal of each pressure monitoring node, and simultaneously collect the real-time temperature value characterizing the fluid medium environment within the hydrostatic support system. Step 103: Use real-time temperature values ​​to correct temperature drift of each original fluid pressure signal and determine the reference oil pressure of each pressure monitoring node. Step 104: Extract high-frequency pressure pulsation components from each original fluid pressure signal through frequency filtering. The high-frequency pressure pulsation components are used to characterize the dynamic fluctuation characteristics inside the fluid field. Step 105: Calculate the signal response lag time of the high-frequency pressure pulsation component relative to the reference oil pressure, and construct a spatial pressure distribution field by combining the distribution of each reference oil pressure in the spatial coordinate model; Step 106: Based on the mapping relationship between signal response lag time and spatial pressure distribution field, identify the pseudo-pressure interference components induced by the structural deformation of the heavy-duty worktable and its base, remove the pseudo-pressure interference components from the original fluid pressure signal, and output characteristic parameters characterizing the actual oil film gap state.

[0023] Preferably, step 105 specifically includes: step 1051, analyzing the propagation waveform differences of high-frequency pressure pulsation components between adjacent pressure monitoring nodes, and converting the propagation waveform differences into signal response lag time; step 1052, performing interpolation calculations based on the coordinates of each reference oil pressure in the spatial coordinate model to generate a pressure distribution vector characterizing the load center of gravity offset trend; step 1053, correcting the pressure distribution vector through the signal response lag time to determine the coupling ratio between pressure fluctuations caused by mechanical load changes and pressure offsets caused by structural deformation.

[0024] Preferably, step 103 specifically includes: step 1031, determining the hydrodynamic viscosity of the hydrostatic support system under the current environment based on the preset nonlinear correlation curve of fluid viscosity, pressure and temperature; step 1032, using the hydrodynamic viscosity as a compensation factor, performing zero-point correction and proportional gain adjustment on the original fluid pressure signal to obtain the reference oil pressure.

[0025] Preferably, in step 104, an AC signal with a frequency between 10Hz and 500Hz is separated from the original fluid pressure signal using a bandpass filtering algorithm, and used as a high-frequency pressure pulsation component.

[0026] Preferably, in step 1032, the normalized corrected reference oil pressure Satisfy the following formula: ,in, This represents the average static pressure in the original fluid pressure signal. This is the real-time temperature value. β is the preset reference temperature, and β is the viscosity-temperature constant of the fluid medium.

[0027] Preferably, in step 105, the spatial coordinate model is divided into multiple local sampling clusters containing four adjacent pressure monitoring nodes, and the pressure change rate within each local sampling cluster is calculated to synthesize the spatial pressure distribution field.

[0028] Preferably, step 106 specifically includes: step 1061, monitoring the rate of change of the signal response lag time, and when the rate of change exceeds a preset abrupt change threshold, determining that there is a pseudo-pressure interference component caused by structural deformation in the original fluid pressure signal; step 1062, dynamically decoupling the spatial pressure distribution field based on the signal response lag time, and separating the real pressure signal caused by fluid static load.

[0029] Preferably, after step 1061, the method further includes: step 1063, calculating the bending deflection of the heavy-duty worktable body and base based on the ratio of the real pressure signal to the pseudo pressure interference component, and outputting the machine tool geometric accuracy compensation command.

[0030] Preferably, in step 102, a multi-channel synchronous sampling controller is used to acquire sixteen raw fluid pressure signals, with a sampling frequency of not less than 2kHz.

[0031] Preferably, the characteristic parameters include the real-time value of the oil film thickness and the load distribution balance, which are used to determine the support stability of the hydrostatic support system.

[0032] Example 1: A fluid pressure monitoring method for a heavy-duty workbench was applied to a hydrostatic support system of a precision machine tool with a diameter of 3.5m and a rated load of 120t. Under 150% rated off-center load, the heavy-duty workbench body and base underwent elastic bending. The original fluid pressure signal collected by the pressure sensor contained pseudo-pressure interference components induced by structural deformation, resulting in nonlinear distortion of the oil film thickness calculated based on the pressure gradient. To establish the positional relationship between the 16 pressure monitoring nodes, a spatial coordinate model was established on the guide rail surface of the hydrostatic support system. The center of the workbench was set as the origin of the coordinate system. The 16 independent oil chambers were symmetrically distributed in a double-ring concentric circle. The inner ring contained 8 independent oil chambers with a radius of 1.2m, and the polar angles corresponding to the center lines of each independent oil chamber were equally distributed at 45° intervals. The outer ring contained 8 independent oil chambers with a radius of 1.6m, and the polar angles corresponding to the center lines of each independent oil chamber were also equally distributed at 45° intervals, and were radially aligned with the independent oil chambers of the inner ring.

[0033] The shortest circumferential geometric boundary span between adjacent independent oil chambers is 0.35m. Each independent oil chamber is equipped with a cylindrical pressure sensor acquisition port. The three-dimensional coordinates of the center of each acquisition port are input and fixed in the topology mapping matrix of the controller, thus forming a spatial coordinate model of the pressure monitoring node. The multi-channel synchronous sampling controller synchronously acquires 16 channels of raw fluid pressure signals at a sampling frequency of 2kHz, and simultaneously acquires the real-time temperature value of the fluid medium environment within the hydrostatic support system. The controller uses the current pressure and real-time temperature values ​​based on a preset nonlinear correlation curve of fluid viscosity, pressure, and temperature. Determine the fluid dynamic viscosity, and then adjust the zero point and proportional gain of each pressure acquisition channel according to the calibration parameters corresponding to the fluid dynamic viscosity.

[0034] After zero-point correction, the static pressure average value is calculated from the original fluid pressure signal. The reference oil pressure for each pressure monitoring node is determined according to the following formula. : ,in, Reference oil pressure; The static pressure average value after zero-point correction of the original fluid pressure signal; This is the real-time temperature value; This is the preset reference temperature; The viscosity-temperature characteristic constant of the fluid medium is used; the exponential term is used to adjust the proportional gain caused by temperature changes, so that the reference oil pressure obtained at different temperatures has the same pressure reference. Subsequently, the AC signal with a frequency between 10Hz and 500Hz is separated from the original fluid pressure signal by a bandpass filtering algorithm as the high-frequency pressure pulsation component. ,in, The number indicates the pressure monitoring node. The pressure pulsations in this frequency band are mainly generated by machining and cutting vibrations. The elastic bending changes of the heavy-duty worktable body and base are concentrated in the low-frequency range, thus separating two types of signals in the frequency domain. The reference oil pressure of each pressure monitoring node is used as the steady-state reference for the high-frequency pressure pulsation component. The high-frequency pressure pulsation components of adjacent pressure monitoring nodes are analyzed within the sliding time window. and The difference in propagation waveforms between them.

[0035] Cross-correlation integration is performed on the two high-frequency pressure pulsation components, and the time offset corresponding to the peak value in the cross-correlation result is taken as the signal response lag time of the high-frequency pressure pulsation component relative to the reference oil pressure. ,in, and This indicates the number of two adjacent pressure monitoring nodes. This indicates that high-frequency pressure pulsation is caused by nodes. propagation to nodes Based on the signal response lag time, according to the adjacency relationship of pressure monitoring nodes in the spatial coordinate model, two circumferentially adjacent inner ring pressure monitoring nodes and their radially corresponding two outer ring pressure monitoring nodes are divided into a local sampling cluster.

[0036] Based on the reference oil pressure difference and sampling interval between adjacent sampling times at each pressure monitoring node, the pressure change rate within the local sampling cluster is calculated. Using the reference oil pressure as the field value and the pressure change rate as the local update value, interpolation is performed using the coordinates of each pressure monitoring node to synthesize a spatial pressure distribution field. A pressure distribution vector is generated based on the pressure change along the spatial coordinate axes according to the interpolation results. The direction of the pressure distribution vector represents the offset direction of the load's center of gravity, and its amplitude represents the degree of change in the load's center of gravity offset. The signal response lag time is then used to... The components of the pressure distribution vector are corrected. Components synchronized with changes in the reference oil pressure are identified as pressure fluctuations caused by changes in mechanical load, while components exhibiting a response lag relative to the reference oil pressure are identified as pressure offsets caused by structural deformation. The coupling ratio is determined based on the amplitude ratio of the two types of components. When the signal response lag time... When the rate of change exceeds the preset abrupt change threshold, it is determined that there are pseudo-pressure interference components in the original fluid pressure signal. The controller dynamically decouples the spatial pressure distribution field according to the coupling ratio, maps the pressure offset caused by structural deformation to the corresponding pressure monitoring node, and removes the corresponding pseudo-pressure interference components from the original fluid pressure signal to obtain the real pressure signal caused by hydrostatic bearing.

[0037] Signal response lag time It is also used to calculate the real-time value of oil film thickness and to determine the geometric distance between adjacent pressure monitoring nodes based on the spatial coordinate model. Geometric distance Divide by signal response lag time The actual propagation phase velocity of the high-frequency pressure pulsation wave within the oil film is obtained. The theoretical sound velocity is calculated based on the equivalent bulk elastic modulus and fluid density of the lubricating fluid. The flow limiting factor is then determined based on the ratio of the actual propagation phase velocity to the theoretical sound velocity. This flow limiting factor is substituted into a pre-calibrated gap damping power function model to obtain the real-time oil film thickness between the two pressure monitoring nodes. The gap damping power function model is calibrated based on the known oil film gap state and its corresponding signal response lag time. During calculation, the absolute dynamic oil film thickness value is determined based on the corresponding interval of the current flow limiting factor in the calibration data. After eliminating pseudo-pressure interference components, the controller calculates the load distribution balance based on the real pressure signals from the 16 pressure monitoring nodes.

[0038] The load distribution balance is determined by the dispersion of the actual pressure signal of each pressure monitoring node relative to the average pressure of all pressure monitoring nodes. The smaller the dispersion, the higher the load distribution balance. The controller compares the real-time oil film thickness and load distribution balance with the preset stability range of the hydrostatic support system to determine the support stability of the hydrostatic support system. Under 150% rated off-center load conditions, this method identifies the 0.52MPa pseudo-pressure interference component induced by base bending and determines the load distribution balance based on the signal response lag time. The system obtains the real-time value of the oil film thickness and triggers graded pressure adjustment compensation accordingly, so that the actual oil film fluctuation is stabilized at 0.72μm and the table tilt is maintained at 0.28μm / m. By the correspondence between the spatial distribution of the reference oil pressure, the difference in the propagation waveform of the high-frequency pressure pulsation component and the signal response lag time, the pressure interference caused by structural deformation can be separated from the original fluid pressure signal, and the real-time value of the oil film thickness and the load distribution balance can be output.

[0039] Example 2: A hydrostatic support system for a precision machine tool with 16 independent oil chambers, a diameter of 3.5m, and a rated load of 120t was used to verify the monitoring results of the fluid pressure monitoring method under stepped off-center load conditions. Each independent oil chamber served as a pressure monitoring node. The pressure sensor had a range of 0 to 25MPa and a resolution of 0.01MPa. A multi-channel synchronous sampling controller synchronously acquired the original fluid pressure signals of each pressure monitoring node at a sampling frequency of 2kHz. This sampling frequency was determined based on the 500Hz maximum bandwidth of machining vibration and its corresponding signal period. To simulate electromagnetic interference at the machining site, 20dB Gaussian white noise and 50Hz power frequency interference were superimposed on the original fluid pressure signals. The experiment used four stepped off-center load conditions: 80t, 100t, rated load, and 150t. Under each condition, the original fluid pressure signal and real-time temperature value were synchronously acquired. The reference oil pressure of each pressure monitoring node was calculated according to the temperature correction relationship composed of the average static pressure, real-time temperature value, reference temperature, and viscosity-temperature constant. The reference temperature was taken as 20℃.

[0040] Based on this, the original fluid pressure signal was frequency filtered to extract the high-frequency pressure pulsation component. The steady-state pressure and oil film gap state corresponding to the reference oil pressure were used as the propagation background to calculate the signal response lag time between adjacent pressure monitoring nodes. The control group only corrected the temperature drift of the original fluid pressure signal and predicted the load offset based on the pressure amplitude. It did not use the signal response lag time to identify and eliminate pseudo-pressure interference components. When the pressure reading deviation of the control group increased from 0.25MPa to 0.88MPa when the load increased from the first step condition to the rated load, the experimental group performed cross-correlation calculation on the high-frequency pressure pulsation components of adjacent pressure monitoring nodes. The correlation peak value used to determine the signal response lag time could still be obtained in the original fluid pressure signal with superimposed noise. Based on this, the pseudo-pressure interference component caused by structural deformation was identified. This pseudo-pressure interference component accounted for 12.5% ​​of the original fluid pressure signal. After removing it, the equivalent pressure error between the output oil film gap characteristic parameter and the oil film thickness value measured by the laser interferometer remained within 0.08MPa.

[0041] In the above calculation process, the reference oil pressure is used to determine the steady-state pressure background of each pressure monitoring node under the current temperature and load conditions, and corresponds to the nominal oil film gap of the hydrostatic oil chamber. When the high-frequency pressure pulsation component propagates under this steady-state background, its signal response lag time varies with the local oil film gap and fluid damping. Therefore, the signal response lag time is not obtained by directly comparing the high-frequency pressure pulsation component with the reference oil pressure in the time domain, but by establishing a propagation reference through the oil film gap state corresponding to the reference oil pressure, and then determining it based on the difference in propagation waveform between adjacent pressure monitoring nodes. In order to verify the identification effect of the signal response lag time on structural deformation interference, a partial missing control group that does not perform this time determination is set up.

[0042] Under the second stage of the operating condition, the structural deformation of the heavy-duty workbench body and base causes low-frequency elastic stress to be superimposed on the oil film pressure. The control group could not distinguish between the pressure shift caused by changes in mechanical load and structural deformation, and the resulting oil film thickness deviation reached 4.8 μm. The signal response lag times obtained by the test group under the first three stages of the operating condition were 0.15 ms, 0.18 ms, and 0.21 ms, respectively. As the load increased, the oil film gap decreased, the local fluid damping increased, and the signal response lag time increased accordingly, thus forming a monotonic correspondence between the signal response lag time and the oil film gap state. Under the final extreme operating condition, the oil film gap decreased to below 5 μm, and the signal response lag time jumped from the value corresponding to the rated load condition to 0.45 ms.

[0043] The changes indicate that the fluid field has entered the boundary lubrication state, the rate of change of fluid squeezing film damping has increased, and the original approximately linear relationship between the signal response lag time and the oil film gap has an inflection point. Under the aforementioned pressure sensor range and sampling conditions, the phase characteristics of the first three step conditions can be used to identify and eliminate pseudo-pressure interference components. After entering this extreme condition, the signal response lag time can still reflect the sudden change in oil film state, but the oil film gap is no longer directly calculated according to the original approximately linear relationship. The above experimental data show that, by combining the reference oil pressure and the propagation delay of the high-frequency pressure pulsation component, the pressure offset caused by structural deformation can be separated from the original fluid pressure signal, and the measurement error of the oil film gap characteristic parameter under the off-center load condition can be reduced.

[0044] Example 3: A hydrostatic support system for a CNC vertical boring and milling machine with 16 independent oil chambers, a diameter of 4.5m, and a rated load of 150t is used. When high dynamic pulse impact is generated by intermittent cutting with large feed rate, the multi-channel synchronous sampling controller sets a sliding time window according to the machining spindle frequency. The window length is twice the reciprocal of the lower limit of the characteristic frequency of the machining system. The original fluid pressure signal of each pressure monitoring node is collected synchronously with a sampling period of 0.5ms to form a corresponding pressure time series. Digital bandpass filtering is performed on each pressure time series to extract the high-frequency pressure pulsation component.

[0045] Using the reference oil pressure at each pressure monitoring node as a steady-state reference, a discrete cross-correlation function is calculated for the high-frequency pressure pulsation components of two adjacent pressure monitoring nodes. Locate the discrete displacement corresponding to the maximum amplitude of the cross-correlation function. The signal response lag time is determined according to the following formula: ,in, This represents the discrete displacement corresponding to the peak value of the cross-correlation function. The sampling period is 0.5ms in this embodiment; the resulting signal response lag time is... To reflect the propagation time difference of high-frequency pressure pulsation components between adjacent pressure monitoring nodes, the 16 pressure monitoring nodes are divided into multiple local sampling clusters, each containing four adjacent pressure monitoring nodes, according to the adjacency relationship in the spatial coordinate model. The pressure change rate is calculated for the reference oil pressure in each local sampling cluster, and the reference oil pressure and physical coordinates of each pressure monitoring node are used as inputs for Kriging interpolation to generate a continuous spatial pressure distribution field. The pressure distribution vector is formed according to the pressure change along the spatial coordinate axis based on the interpolation results, which is used to characterize the offset trend of the load center of gravity.

[0046] During the dynamic decoupling process, the lag time of the discrete signal response in each local sampling cluster is arranged according to the two-dimensional coordinates of the pressure monitoring node, and a first-order total differential calculation is performed to obtain the time delay spatial gradient field. Then, the time delay spatial gradient field is multiplied point by point with the gradient operator of the spatial pressure distribution field, and the pressure distribution vector is corrected using the result to determine the coupling ratio between the pressure fluctuation caused by the change of mechanical load and the pressure offset caused by structural deformation. The controller simultaneously monitors the rate of change of the signal response lag time. When the rate of change exceeds the preset abrupt change threshold, the pseudo-pressure interference component judgment process is entered. When the coupling ratio remains above 0.85 for 5 consecutive sampling cycles, the corresponding pressure offset is judged as a pseudo-pressure interference component induced by the deformation of the heavy-duty workbench body and base structure.

[0047] Subsequently, the pressure offset is integrated over the sampling time to obtain the corresponding stress increment. Based on the coordinate position of each pressure monitoring node, the corresponding stress increment is subtracted from the discrete sequence of the original fluid pressure signal to separate the true pressure signal caused by hydrostatic bearing. Under the dynamic impact condition of instantaneous 25% increase in cutting force, this method identifies and eliminates the 0.35MPa pseudo-pressure interference component. Based on the decoupled true pressure signal, the characteristic parameter representing the actual oil film gap state is output, and its error is kept within 0.05MPa equivalent pressure.

[0048] Example 4: This example combines Figures 1 to 2 A method for monitoring fluid pressure on a heavy-duty workbench is described, such as... Figure 1As shown, the process begins with spatial coordinate modeling to establish a coordinate system for sixteen independent oil cavity monitoring nodes. After spatial coordinate modeling is completed, multi-source signal acquisition is performed to synchronously obtain the original fluid pressure signal and real-time temperature value. After multi-source signal acquisition, temperature drift correction processing is performed to compensate for sensor temperature drift and determine the reference oil pressure. After temperature drift correction processing, high-frequency pulsation separation is performed to extract dynamic fluctuation characteristic components through frequency filtering. After high-frequency pulsation separation, signal hysteresis calculation is performed to determine the response time of the pulsation component relative to the reference oil pressure. After signal hysteresis calculation, spatial pressure field decoupling is performed to remove the pseudo-pressure interference component induced by structural elastic deformation. After spatial pressure field decoupling, dynamic clearance monitoring and accuracy compensation output are performed to obtain characteristic parameters and output geometric accuracy compensation commands. After dynamic clearance monitoring and accuracy compensation output are completed, the process ends.

[0049] like Figure 2 As shown, the original fluid pressure signal and real-time temperature value are used as inputs to the temperature drift correction processing unit. The output of the temperature drift correction processing unit generates a reference oil pressure. The original fluid pressure signal is also used as input to the frequency filtering and extraction module. The output of the frequency filtering and extraction module generates a high-frequency pressure pulsation component. The reference oil pressure and the high-frequency pressure pulsation component are used together as inputs to calculate the signal response lag time. The output of the signal response lag time calculation and the reference oil pressure are connected to the spatial pressure field decoupling unit. The spatial pressure field decoupling unit decouples the input signal by internally identifying and eliminating pseudo-pressure interference components. The output of the spatial pressure field decoupling unit is connected to the oil film gap state characteristic parameter output and the machine tool geometric accuracy compensation command output, respectively.

[0050] Example 5: During offline debugging, the multi-channel synchronous sampling controller calibrates the coupling ratio using a quasi-static load gradient. Using standard counterweights, 0%, 25%, 50%, 75%, and 100% of the rated load were applied to the center and quadrant edges of the heavy-duty workbench, respectively, and measured pressure sequences from 16 pressure monitoring nodes were simultaneously collected. And calculate the first stiffness matrix based on the preset stiffness matrix in the spatial coordinate model. Theoretical fluid bearing pressure at each pressure monitoring node The residual between the measured pressure and the theoretical fluid bearing pressure is used as the pseudo-pressure interference component generated when the heavy-duty workbench body and base undergo structural deformation. The change in theoretical fluid bearing pressure between adjacent load levels is used as the pressure fluctuation caused by mechanical load changes. Under each load level, the pressure distribution vector generated by the reference oil pressure is corrected using the signal response lag time. Then, the amplitude relationship between the pressure fluctuation caused by mechanical load changes and the pressure offset caused by structural deformation is fitted using the least squares method to obtain the coupling ratio. The pseudo-pressure offset pattern of each pressure monitoring node, the real pressure signal obtained after dynamic decoupling characterizes the supporting force formed by fluid static bearing, and the pseudo-pressure interference component corresponds to the boundary displacement caused by structural deformation.

[0051] Based on the ratio of the real pressure signal to the pseudo-pressure interference component, and combined with the pre-determined elastic modulus of the component material, the large cross-section geometric moment of inertia, and the stiffness matrix, the pressure ratio of each pressure monitoring node is converted into the bending strain at the corresponding position. The integral constant is determined by the fixed support boundary conditions in the stiffness matrix, and a second spatial integration is performed along the longitudinal span of the heavy-duty workbench to obtain the bending deflection of the heavy-duty workbench body and base. According to the spatial coordinate model, the bending deflection at each position is converted into the corresponding displacement compensation amount, forming a machine tool geometric accuracy compensation command and outputting it.

[0052] When tool changes cause changes in the dynamic response characteristics of the machining system, the multi-channel synchronous sampling controller acquires the swept frequency signal under cutting conditions and determines the passband frequency range of the bandpass filter. Fast Fourier Transform analysis was performed on the power spectrum distribution of the original fluid pressure signal in the range of 0Hz to 1kHz to identify the energy peak corresponding to the rotation frequency of the machine tool spindle; in the range of 10Hz to 500Hz, the lowest frequency containing the vibration characteristic energy was determined as the lower limit of the passband frequency. The upper limit of the passband frequency is determined based on the envelope of the inherent frequency distribution of the processing system. And extract the high-frequency pressure pulsation component related to the damping of the oil film extrusion membrane.

[0053] The lower limit of the passband frequency is satisfied: ,in, This represents the lower limit of the passband frequency of the bandpass filter, expressed in Hz. This is the frequency correction factor; The frequency of the machine tool spindle is measured in Hz. The frequency correction factor is determined based on the transmission attenuation boundary of the shock wave generated by the intermittent cutting of the tool in the hydraulic pipeline. In this embodiment, it is set to 1.25. When the factor is lower than 1.00, the bandpass filter will cut off part of the high-frequency pressure pulsation component; when it is higher than 2.00, high-frequency Gaussian noise will be introduced into the filtering result.

[0054] Within a defined passband frequency range, the signal-to-noise ratio of the signal response lag time measurement is improved by 18.5 dB compared to fixed bandwidth processing, and the amplitude of the output characteristic parameter fluctuation converges to within 0.03 MPa. During dynamic monitoring of alternating load impact frequencies from 0.1 Hz to 5 Hz, the multi-channel synchronous sampling controller constructs the signal response lag time... Local pressure gradient of spatial pressure distribution field The instantaneous mapping matrix between them allows the system to extract the signal response lag time between adjacent pressure monitoring nodes in real time. Simultaneously calculate the local pressure gradient of the spatial pressure distribution field along the corresponding coordinate axis direction. The system monitors the rate of change of signal response lag time; the preset mutation threshold is determined by the standard counterweight stepped pulse load test as a delay change rate of 0.05 milliseconds per millisecond; when the change rate exceeds the threshold, the identification of pseudo-pressure interference components is initiated, and when it is below the threshold, no structural deformation is determined.

[0055] After initiating identification, the coherence coefficient between the time-domain variation curve of the signal response lag time and the local pressure gradient fluctuation curve is calculated. The coupling threshold of 0.85 is obtained statistically from multiple sets of pressure time-series coherence data under no-load and various step-by-step off-center load conditions. When the coherence coefficient is in the range of 0.85 to 1.00, the two curves show consistent trends in the time domain, and the corresponding pressure offset is identified as a pseudo-pressure interference component caused by structural deformation. When the coherence coefficient is below 0.85, the pressure change is identified as a real pressure signal caused by hydrostatic load, and no interference removal is performed; the pseudo-pressure interference component is excluded. satisfy: ,in, This is a pseudo-pressure interference component, with units of MPa; This is a fluid-structure interaction adjustment factor; This represents the local pressure gradient of the spatial pressure distribution field.

[0056] The fluid-structure interaction adjustment factor is calibrated based on the ratio of the measured compliance coefficient of the heavy-duty workbench under rated operating conditions to the stiffness of the hydraulic oil cavity. In this embodiment, it is set to 0.42. When the factor is below 0.15, the compensation for rapid structural deformation is insufficient; when it is above 0.85, the hydraulic feedback adjustment system will experience self-excited oscillation and instability. The multi-channel synchronous sampling controller allocates the calculated pseudo-pressure interference components to the corresponding pressure monitoring nodes according to the preset sensitivity matrix, and subtracts them from the original fluid pressure signals of each node to obtain the real pressure signal and characteristic parameters characterizing the actual oil film gap state. Under 5Hz alternating load interference, after adopting this dynamic decoupling processing, the sensing error is reduced from 0.42MPa to within 0.05MPa.

[0057] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. A method for monitoring fluid pressure on a heavy-duty workbench, characterized in that, Includes the following steps: Step 101: Establish a spatial coordinate model for the sixteen independent oil chambers of the hydrostatic support system, and define each independent oil chamber as a pressure monitoring node. Step 102: Obtain the raw fluid pressure signal of each pressure monitoring node, and simultaneously collect the real-time temperature value characterizing the fluid medium environment within the hydrostatic support system. Step 103: Use real-time temperature values ​​to correct temperature drift of each original fluid pressure signal and determine the reference oil pressure of each pressure monitoring node. Step 104: Extract high-frequency pressure pulsation components from each original fluid pressure signal through frequency filtering. The high-frequency pressure pulsation components are used to characterize the dynamic fluctuation characteristics inside the fluid field. Step 105: Calculate the signal response lag time of the high-frequency pressure pulsation component relative to the reference oil pressure, and construct a spatial pressure distribution field by combining the distribution of each reference oil pressure in the spatial coordinate model; wherein, Step 1051: Analyze the propagation waveform difference of the high-frequency pressure pulsation component between adjacent pressure monitoring nodes, and convert the propagation waveform difference into signal response lag time; Step 1052: Perform interpolation calculation based on the coordinates of each reference oil pressure in the spatial coordinate model to generate a pressure distribution vector characterizing the load center of gravity offset trend; Step 1053: Correct the pressure distribution vector by the signal response lag time to determine the coupling ratio between pressure fluctuations caused by mechanical load changes and pressure offsets caused by structural deformation; Step 106: Based on the mapping relationship between signal response lag time and spatial pressure distribution field, identify the pseudo-pressure interference component induced by the structural deformation of the heavy-duty workbench and its base, remove the pseudo-pressure interference component from the original fluid pressure signal, and output characteristic parameters representing the actual oil film gap state; wherein, Step 1061: monitor the rate of change of signal response lag time, and when the rate of change exceeds a preset abrupt change threshold, determine that there is a pseudo-pressure interference component caused by structural deformation in the original fluid pressure signal; Step 1062: dynamically decouple the spatial pressure distribution field based on the signal response lag time to separate the real pressure signal caused by hydrostatic load.

2. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, Step 103 specifically includes: Step 1031, determining the hydrodynamic viscosity of the hydrostatic support system under the current environment based on the preset nonlinear correlation curve of fluid viscosity, pressure and temperature; Step 1032, using the hydrodynamic viscosity as a compensation factor, performing zero-point correction and proportional gain adjustment on the original fluid pressure signal to obtain the reference oil pressure.

3. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, In step 104, an AC signal with a frequency between 10Hz and 500Hz is separated from the original fluid pressure signal using a bandpass filtering algorithm, which is then used as a high-frequency pressure pulsation component.

4. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, In step 105, the spatial coordinate model is divided into multiple local sampling clusters containing four adjacent pressure monitoring nodes, and the pressure change rate within each local sampling cluster is calculated to synthesize the spatial pressure distribution field.

5. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, After step 1061, the following step is also included: Step 1063, which calculates the bending deflection of the heavy-duty worktable body and base based on the ratio of the real pressure signal to the pseudo pressure interference component, and outputs the machine tool geometric accuracy compensation command.

6. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, In step 102, a multi-channel synchronous sampling controller is used to acquire sixteen raw fluid pressure signals with a sampling frequency of not less than 2kHz.

7. The fluid pressure monitoring method for a heavy-duty workbench according to claim 1, characterized in that, The characteristic parameters include the real-time value of the oil film thickness and the load distribution balance, which are used to determine the support stability of the hydrostatic support system.

Images