Pneumatic zero positioning mother plate pressure self-adaptive compensation system

Abstract

The present application relates to zero point positioning mother plate pressure compensation technical field, disclose a pneumatic type zero point positioning mother plate pressure self-adapting compensation system, carry out multi-source fusion processing to initial pressure field data, cover the compensation sequence of static reference feature, the feedback sequence of dynamic deformation feature and the calibration sequence of axial offset feature. The pressure compensation module carries out dynamic fluctuation suppression to real-time deformation data stream, realizes pressure field balance identification and generates compensation instruction through multi-field coupling layer. The pressure standardization module carries out space grid division, abnormal filtering and data calibration, and the pressure field analysis layer, zero point adaptive layer and compensation trigger layer of field strength reconstruction module complete feature fusion, dynamic correlation modeling and multi-dimensional field strength fusion in turn. Through deformation segmentation, virtual interpolation, dynamic clustering and other algorithms, the system improves the pressure field balance ability and positioning accuracy of the mother plate under complex working conditions, and is suitable for high-precision positioning demand in the field of industrial automation.

Description

Technical Field

[0001] This invention relates to the field of zero-point positioning motherboard pressure compensation technology, specifically a pneumatic zero-point positioning motherboard pressure adaptive compensation system. Background Technology

[0002] In the field of industrial automation, zero-point positioning systems, as core components for achieving high-precision positioning and rapid clamping, are widely used in machine tool processing, automated production lines, and other scenarios. The mother plate, as the key carrier of the zero-point positioning system, directly affects positioning reliability and machining quality due to the stability and compensation accuracy of its pressure field distribution. Traditional zero-point positioning mother plates generally suffer from insufficient dynamic adjustment capability of the pressure field when facing complex working conditions. Specifically, this manifests in the following ways: the coupling processing of static reference features and dynamic deformation features lacks an effective fusion mechanism, resulting in pressure compensation lagging behind actual deformation requirements; the calibration process of axial offset features does not fully consider the correlation effects of historical data from multiple working conditions, leading to poor targeting and adaptability of compensation commands; and the standardization processing of the pressure field relies on fixed thresholds, making it difficult to cope with real-time fluctuating pressure data streams, easily causing the accumulation of abnormal field strength and amplification of compensation errors.

[0003] In existing technologies, some pressure compensation systems use a single sensor for data acquisition, enabling monitoring of only a single-dimensional pressure parameter. This fails to comprehensively cover multi-source characteristic data such as the load distribution of the mother plate, positioning reference offset, and dynamic deformation. Regarding pressure field equilibrium identification, traditional multi-field coupling models lack the ability to dynamically track key pressure nodes. Especially when the number of nodes in the compensation sequence is inconsistent, the lack of an effective interpolation matching mechanism leads to significant deviations in the calculation of pressure node offset coefficients under different operating conditions, severely impacting the accuracy of zero-point compensation field data. Furthermore, existing systems' fluctuation suppression processing is mostly based on fixed filtering algorithms, which struggle to adapt to the spatial lag effect of the mother plate under different load conditions. This easily leads to the accumulation of errors between the compensation command and the actual pressure distribution, thereby affecting positioning accuracy and system stability.

[0004] With the increasing demands for processing precision and flexible production in intelligent manufacturing, zero-point positioning motherboards need to maintain high-precision pressure field equilibrium capabilities under complex environments such as multi-condition switching and dynamic load fluctuations. Therefore, how to achieve deep fusion of multi-source pressure field data, real-time suppression of dynamic deformation, and adaptive generation of compensation commands has become a key problem urgently needing to be solved in the current zero-point positioning technology field. The limitations of existing technologies in multi-field coupling modeling, dynamic compensation strategies, and data standardization processing necessitate a pressure adaptive compensation system that can integrate static reference, dynamic deformation, and axial offset characteristics to meet the practical needs of industrial automation for high-precision and high-reliability positioning. Summary of the Invention

[0005] The purpose of this invention is to provide a pneumatic zero-point positioning motherboard pressure adaptive compensation system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a pneumatic zero-point positioning motherboard pressure adaptive compensation system, the system comprising:

[0007] The zero-point positioning module is used to perform multi-source pressure field fusion processing on the initial pressure field data. The initial pressure field data includes a compensation sequence corresponding to the static reference feature data, a feedback sequence corresponding to the dynamic deformation feature data, and a calibration sequence corresponding to the axial offset feature data. The static reference feature data includes a pressure label generated by the mother plate bearing distribution parameter and a zero-point label generated by the positioning reference parameter.

[0008] The pressure compensation module is used to perform dynamic fluctuation suppression processing on the real-time deformation data stream and input it into the multi-field coupling layer for pressure field equalization identification, and generate pressure compensation instructions based on the output results of the multi-field coupling layer.

[0009] The multi-field coupling layer includes a pressure standardization module and a field strength reconstruction module. The pressure standardization module is used to perform spatial grid division and abnormal field strength filtering on the original pressure data stream. The field strength reconstruction module is obtained by joint training based on historical pressure data under multiple working conditions and real-time deformation trajectory. The field strength reconstruction module includes a pressure field analysis layer, a zero-point adaptive layer and a compensation triggering layer connected in sequence.

[0010] Preferably, the pressure field analysis layer is used to perform spatial coupling processing on different compensation sequences in the original pressure data stream to generate pressure field fusion feature data; the zero-point adaptive layer is used to model the dynamic correlation between the pressure field fusion feature data corresponding to each compensation sequence to generate zero-point compensation field data; and the compensation triggering layer is used to perform multi-dimensional field strength fusion based on the zero-point compensation field data and the pressure field fusion feature data to generate pressure compensation instructions.

[0011] Preferably, the step of modeling the dynamic correlation between the pressure field fusion feature data corresponding to each compensation sequence to generate zero-point compensation field data includes:

[0012] The deformation segmentation algorithm is used to identify the pressure key nodes in the pressure field fusion feature data, and the zero-point compensation sequence corresponding to each compensation sequence is determined based on the working condition type corresponding to each pressure key node.

[0013] Calculate the offset coefficient between pressure nodes under the same working condition in the zero-point compensation sequence corresponding to any two compensation sequences, and generate zero-point compensation field data between the two compensation sequences based on the offset coefficient.

[0014] Preferably, the calculation of the offset coefficient between pressure nodes under the same working condition in the zero-point compensation sequences corresponding to any two compensation sequences includes:

[0015] When the number of pressure nodes in any two compensation sequences is inconsistent, virtual field strength point interpolation is performed based on the working condition parameters corresponding to the end pressure node in the sequence with fewer pressure nodes, and the offset coefficient between pressure nodes under the same working condition is calculated based on the interpolated data.

[0016] Preferably, the pressure standardization module is specifically used for:

[0017] The compensation sequence, feedback sequence, and calibration sequence are divided into equal field strengths according to a preset spatial grid to generate standardized pressure data, standardized deformation data, and standardized calibration data.

[0018] A dynamic field strength clustering method is used to match the standardized pressure data and standardized deformation data in real time, and a fixed space sliding mechanism is used to perform steady-state optimization on the standardized calibration data, outputting a first reference sequence, a second reference sequence, and a third reference sequence; wherein, the first reference sequence includes calibrated pressure labels and calibrated zero-point labels.

[0019] Preferably, the pressure standardization module is further used for:

[0020] Calculate the deformation offset coefficient between the calibrated pressure tag and the calibrated zero-point tag during the historical monitoring period;

[0021] Based on the deformation offset coefficient and the operating parameters of the calibrated pressure tag during the real-time monitoring period, predict the expected distribution value of the calibrated zero-point tag during the real-time monitoring period.

[0022] Target zero-point compensation data is generated based on the calibrated zero-point labels and their expected distribution values, and the compensation sequence corresponding to the target zero-point compensation data is used as the first reference sequence.

[0023] Preferably, the zero-point adaptive layer specifically includes:

[0024] The field strength tracing unit is used to perform pressure field tracing on each compensation sequence in the pressure field fusion feature data, so as to extract the corresponding pressure conduction chain from each compensation sequence.

[0025] The working condition matching unit is used to spatially superimpose the pressure transmission chain extracted from each compensation sequence with the corresponding pressure field fusion feature data to generate zero-point compensation field data.

[0026] Preferably, the zero-point adaptive layer further includes:

[0027] The fluctuation suppression unit is used to perform spatial lag effect elimination processing on the zero-point compensation field data.

[0028] Preferably, the compensation triggering layer specifically includes:

[0029] The multi-dimensional field strength coordination unit contains multiple compensation decision nodes. Each compensation decision node is connected to each compensation sequence in the zero-point compensation field data and the pressure field fusion feature data through parameter configuration.

[0030] The dynamic field strength optimization unit is used to iteratively optimize the parameter configuration through a dynamic field strength adjustment algorithm to minimize the error between the pressure compensation command and the actual pressure distribution.

[0031] The pressure equalization unit is used to locate high-pressure operating conditions based on the zero-point compensation field data and the pressure field fusion feature data, and to generate pressure compensation commands.

[0032] Preferably, the dynamic field strength clustering method specifically includes:

[0033] Axial compensation parameters are generated based on the pressure field distribution characteristics of real-time operating parameters.

[0034] The standardized pressure data is processed using a spatial grid segmentation mechanism for multi-scale matching.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] The system utilizes a zero-point positioning module to perform multi-source fusion processing on the initial pressure field data, transforming static reference feature data, dynamic deformation feature data, and axial offset feature data into compensation sequences, feedback sequences, and calibration sequences, respectively. This enables comprehensive monitoring of the motherboard's load distribution, positioning reference, and dynamic deformation. The pressure compensation module, through a multi-field coupling layer's pressure standardization module and field strength reconstruction module, dynamically suppresses fluctuations and identifies pressure field equilibrium in the real-time deformation data stream, effectively solving the problem of pressure compensation lag in traditional systems. The pressure standardization module, through spatial grid partitioning and dynamic field strength clustering methods, standardizes and matches multi-source data in real time, ensuring the calibration accuracy of pressure and zero-point labels. Simultaneously, it predicts the expected distribution value of the zero-point label using the deformation offset coefficient, improving the foresight and adaptability of the compensation sequence.

[0037] The pressure field analysis layer, zero-point adaptive layer, and compensation triggering layer in the field strength reconstruction module form a progressive processing link. The pressure field analysis layer spatially couples different compensation sequences to generate fused feature data. The zero-point adaptive layer identifies key pressure nodes through deformation segmentation algorithms, establishes a dynamic correlation model between compensation sequences, and generates accurate zero-point compensation field data. The compensation triggering layer, through multi-dimensional field strength coordination units and dynamic field strength optimization units, realizes dynamic optimization of compensation commands and high-pressure condition positioning, significantly reducing the error between compensation commands and actual pressure distribution. When dealing with the problem of inconsistent number of nodes in compensation sequences, the system uses virtual field strength point interpolation technology to ensure the accuracy of offset coefficient calculation and avoid compensation deviations caused by data mismatch. The fluctuation suppression unit in the zero-point adaptive layer further improves the stability of zero-point compensation field data and enhances the system's adaptability to dynamic load fluctuations by eliminating spatial lag effects. The dynamic field strength clustering method combines real-time operating parameters to generate axial compensation parameters and achieves multi-scale matching through a spatial grid segmentation mechanism, enabling the system to automatically adjust the compensation strategy according to different operating conditions, improving the dynamic response speed and uniformity of the pressure field distribution. Attached Figure Description

[0038] Figure 1 This is a schematic diagram illustrating the working principle of the pneumatic zero-point positioning motherboard pressure adaptive compensation system described in this invention.

[0039] Figure 2 A schematic diagram illustrating the working principle of generating zero-point compensation field data in the zero-point adaptive layer;

[0040] Figure 3 This is a schematic diagram of the working principle of the pressure standardization module;

[0041] Figure 4 This is a schematic diagram of the working principle of the field strong tracing unit and the working condition matching unit in the zero-point adaptive layer. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figures 1-4 The present invention relates to a pneumatic zero-point positioning motherboard pressure adaptive compensation system, the system comprising:

[0044] The zero-point positioning module is used to perform multi-source pressure field fusion processing on the initial pressure field data. The initial pressure field data includes a compensation sequence corresponding to the static reference feature data, a feedback sequence corresponding to the dynamic deformation feature data, and a calibration sequence corresponding to the axial offset feature data. The static reference feature data includes a pressure label generated by the mother plate bearing distribution parameter and a zero-point label generated by the positioning reference parameter.

[0045] The pressure compensation module is used to perform dynamic fluctuation suppression processing on the real-time deformation data stream and input it into the multi-field coupling layer for pressure field equalization identification, and generate pressure compensation instructions based on the output results of the multi-field coupling layer.

[0046] The multi-field coupling layer includes a pressure standardization module and a field strength reconstruction module. The pressure standardization module is used to perform spatial grid division and abnormal field strength filtering on the original pressure data stream. The field strength reconstruction module is obtained by joint training based on historical pressure data under multiple working conditions and real-time deformation trajectory. The field strength reconstruction module includes a pressure field analysis layer, a zero-point adaptive layer and a compensation triggering layer connected in sequence.

[0047] The present invention will be further described below with reference to Examples 1 to 5:

[0048] Example 1:

[0049] The pressure field analysis layer, zero-point adaptive layer, and compensation triggering layer are sequentially connected to form the field strength reconstruction module. The pressure field analysis layer performs spatial coupling processing on different compensation sequences in the original pressure data stream to generate pressure field fusion feature data. The original pressure data stream includes compensation sequences corresponding to static reference feature data output by the zero-point positioning module, feedback sequences corresponding to dynamic deformation feature data, and calibration sequences corresponding to axial offset feature data. This data includes pressure labels generated by the mother plate load distribution parameters, zero-point labels generated by the positioning reference parameters, and real-time monitored pressure changes and deformation information. During processing, the pressure field analysis layer first preprocesses various sequences to remove noise and outliers, ensuring data accuracy. Then, through a spatial coupling algorithm, it analyzes the spatial distribution patterns of different compensation sequences, such as the distribution of pressure values ​​at different locations on the mother plate and the spatial correlation between different sequences. Meanwhile, considering the conduction characteristics of pressure in the base plate material, such as conduction direction and attenuation, these spatial dimensions of information are fused to form pressure field fusion characteristic data that can comprehensively reflect the spatial distribution and energy transfer characteristics of the pressure field. This data includes information such as pressure intensity at different locations, pressure change trends, and spatial correlation between sequences.

[0050] The zero-point adaptive layer models the dynamic correlation between the pressure field fusion feature data corresponding to each compensation sequence, generating zero-point compensation field data. During this process, a deformation segmentation algorithm is used to identify pressure-critical nodes in the pressure field fusion feature data. The deformation segmentation algorithm determines which nodes have a significant impact on the entire pressure field based on the magnitude and gradient of pressure value changes in the pressure field fusion feature data. These nodes are typically located in stress-concentrated areas or structurally weak parts of the motherboard. For example, when a certain area of ​​the motherboard bears a large load, that area and its nearby nodes may be identified as pressure-critical nodes. Then, based on the operating condition type corresponding to each pressure-critical node, the zero-point compensation sequence corresponding to each compensation sequence is determined. The operating condition type includes the operating state of the motherboard under different load conditions and working environments, such as light load, heavy load, high temperature, and normal temperature. Each pressure-critical node exhibits different pressure characteristics under different operating conditions, thus dividing the compensation sequence into different zero-point compensation sequences.

[0051] Calculate the offset coefficient between pressure nodes under the same operating condition in the zero-point compensation sequences corresponding to any two compensation sequences. When the number of pressure nodes in any two compensation sequences is inconsistent, virtual field strength point interpolation is required. Specifically, taking the compensation sequence with fewer pressure nodes as the benchmark, analyze the operating parameters corresponding to its terminal pressure nodes, such as pressure value, temperature, and load type. Then, based on the positions of missing nodes in the compensation sequence with more pressure nodes, insert virtual field strength points to make the number of pressure nodes in the two compensation sequences consistent. During the interpolation process, the pressure change trend and spatial distance of adjacent nodes are considered to ensure that the inserted virtual field strength point data conforms to the actual distribution law of the pressure field. After interpolation, calculate the pressure value difference of corresponding pressure nodes in the two compensation sequences under the same operating condition. Compare this difference with the standard pressure value or reference pressure value to obtain the offset coefficient, which reflects the degree of pressure deviation between the two compensation sequences under the same operating condition. Finally, based on these offset coefficients, generate zero-point compensation field data between any two compensation sequences. This data describes the pressure correlation and deviation of different compensation sequences under different operating conditions, providing a basis for subsequent zero-point compensation.

[0052] The compensation trigger layer performs multi-dimensional field strength fusion based on zero-point compensation field data and pressure field fusion feature data to generate pressure compensation commands. During multi-dimensional field strength fusion, it comprehensively considers the pressure deviation information between different compensation sequences in the zero-point compensation field data, as well as the spatial distribution and trend of pressure in the pressure field fusion feature data. For example, when the zero-point compensation field data shows a large pressure offset between two compensation sequences under heavy load conditions, and the pressure field fusion feature data indicates that this offset causes pressure concentration in a certain area of ​​the motherboard, the compensation trigger layer determines the location and amount of pressure compensation based on this information. Specifically, through a specific fusion algorithm, the information in each dimension of the zero-point compensation field data and pressure field fusion feature data is weighted and combined to obtain a comprehensive field strength distribution. Then, based on preset compensation rules and the structural characteristics of the motherboard, a specific pressure compensation command is generated. This command includes information such as the compensation location, the compensation pressure value, and the compensation method, so that the pressure compensation module can perform the corresponding compensation operation, achieving adaptive adjustment of the motherboard pressure field and ensuring the zero-point positioning accuracy and operational stability of the motherboard. Throughout the process, data transmission and processing between layers are continuous and dynamic, enabling real-time response to changes in the motherboard pressure field and ensuring the system's compensation effectiveness and adaptability.

[0053] Example 2:

[0054] The pressure standardization module plays a crucial role in the system by standardizing the raw pressure data. Its specific implementation is as follows: The compensation sequence, feedback sequence, and calibration sequence are divided into equal-field-strength sections based on a preset spatial grid. The preset spatial grid design must comprehensively consider the geometry, material properties, and possible pressure distribution range of the parent plate. For example, the bearing surface of the parent plate is divided into several regular grid cells according to certain size specifications, with each grid cell corresponding to a specific spatial region. During the division process, the field strength distribution within each grid cell is ensured to be relatively uniform to facilitate subsequent pressure data standardization. After division, the pressure data, deformation data, and calibration data in each sequence are categorized and statistically analyzed according to the grid cells, generating standardized pressure data, standardized deformation data, and standardized calibration data, ensuring that these data accurately reflect the pressure field characteristics within each grid cell.

[0055] A dynamic field strength clustering method is employed to perform real-time matching of standardized pressure and deformation data. The implementation of this method first generates axial compensation parameters based on the pressure field distribution characteristics of real-time operating parameters. These parameters include the magnitude of the axial load on the motherboard, the rate of load change, and ambient temperature. Changes in these parameters directly affect the axial distribution of the pressure field. For example, as the axial load increases, the axial distribution of the pressure field changes. By analyzing these changes, the corresponding axial compensation parameters are calculated, which are then used to adjust the pressure data. Next, a spatial grid segmentation mechanism is used to perform multi-scale matching processing on the standardized pressure data. This mechanism divides the entire spatial grid into different scales, such as large-scale, medium-scale, and small-scale segments, and performs separate matching analysis on the pressure data within each segment. At the large scale, the overall distribution trend of the pressure field is considered; at the small scale, the focus is on the detailed changes in pressure in local areas. This multi-scale matching process allows for a more comprehensive capture of the characteristics of the pressure data, improving the accuracy of clustering.

[0056] A fixed-space sliding mechanism is employed to perform steady-state optimization on standardized calibration data. This mechanism involves setting a fixed-size sliding window within a spatial grid. This window slides gradually along a specific direction (e.g., horizontal or vertical) of the grid, analyzing and optimizing the calibration data within each slide. This method filters out random fluctuations and outliers in the calibration data, stabilizing it and resulting in more reliable calibration results. After this processing, the pressure standardization module outputs a first reference sequence, a second reference sequence, and a third reference sequence. The first reference sequence contains calibrated pressure labels and calibrated zero-point labels; these labels are crucial data that accurately reflect the pressure state and zero-point position of the motherboard after calibration.

[0057] The pressure standardization module also calculates the deformation offset coefficient of the calibrated pressure tag and the calibrated zero-point tag over the historical monitoring period. The historical monitoring period refers to the time span during which the system monitors the motherboard pressure and zero-point position. By collecting pressure tag and zero-point tag data during these periods, the module analyzes the relationship between the two and calculates the deformation offset coefficient. This coefficient reflects the degree to which pressure changes affect the zero-point position during historical operation. Then, based on the deformation offset coefficient and the operating parameters of the calibrated pressure tag in the real-time monitoring period, the expected distribution value of the calibrated zero-point tag in the real-time monitoring period is predicted. The operating parameters of the real-time monitoring period include the current load and temperature conditions. Combined with the historical deformation offset coefficient, the possible distribution of the zero-point tag under the current operating conditions can be inferred.

[0058] Target zero-point compensation data is generated based on the calibrated zero-point labels and their expected distribution values, and the compensation sequence corresponding to the target zero-point compensation data is used as the first reference sequence. Specifically, the difference between the calibrated zero-point labels and the expected distribution values ​​is compared. If a deviation exists, corresponding compensation data is generated based on the magnitude and direction of the deviation. This compensation data aims to adjust the zero-point position to return it to the expected distribution range. The compensation sequence corresponding to the generated target zero-point compensation data is used as the new first reference sequence for subsequent pressure field processing and compensation operations, ensuring that the system can accurately compensate according to real-time pressure changes and zero-point offset, maintaining the zero-point positioning accuracy of the motherboard.

[0059] Example 3:

[0060] The zero-point adaptive layer is responsible for performing in-depth analysis of the pressure field fusion feature data in the system to generate accurate zero-point compensation field data. Its specific implementation process is as follows: This layer includes a field strength tracing unit, a working condition matching unit, and a fluctuation suppression unit. Each unit works together to achieve comprehensive analysis and processing of pressure field features.

[0061] The field strength tracing unit performs pressure field tracing on each compensation sequence in the pressure field fusion feature data to extract the corresponding pressure transmission chain from each compensation sequence. The pressure field fusion feature data is comprehensive data generated after processing by the pressure field analysis layer, containing spatial pressure distribution and correlation information for different compensation sequences. During processing, the field strength tracing unit tracks the pressure transmission path for each compensation sequence. For example, for compensation sequences corresponding to static reference feature data, it analyzes how pressure is transmitted through the internal structure of the mother plate to various positioning reference points, starting from the bearing surface of the mother plate; for feedback sequences corresponding to dynamic deformation feature data, it tracks the transmission process of pressure changes at different parts of the mother plate. During the tracing process, the physical properties of the mother plate material, such as the elastic modulus and Poisson's ratio, as well as the attenuation law during pressure transmission, need to be considered. In this way, the pressure transmission chain corresponding to each compensation sequence is determined, which describes the specific path of pressure from input to output and the pressure distribution at each node.

[0062] The load condition matching unit spatially superimposes the pressure transmission chain extracted from each compensation sequence with the corresponding pressure field fusion feature data to generate zero-point compensation field data. The pressure transmission chain reflects the pressure transmission path within the motherboard, while the pressure field fusion feature data reflects the spatial distribution characteristics of the pressure. During spatial superposition, the load condition matching unit first maps each node on the pressure transmission chain to the spatial grid in the pressure field fusion feature data, determining the specific spatial location of each node. Then, it integrates the pressure values ​​on the pressure transmission chain with the corresponding pressure values ​​in the spatial grid, considering pressure variations under different load conditions. For example, under light load conditions, the pressure distribution on the pressure transmission chain differs from that under heavy load conditions. The load condition matching unit weights and superimposes the pressure transmission chain and pressure field fusion feature data according to the current load condition type, ensuring that the superimposed data accurately reflects the pressure distribution and zero-point offset under the current load condition, thus generating zero-point compensation field data. This data includes information such as the zero-point compensation amount and direction at different locations, providing a basis for subsequent zero-point compensation.

[0063] The fluctuation suppression unit performs spatial hysteresis elimination processing on the zero-point compensation field data. During pressure transmission, due to the inertia and damping characteristics of the base plate material, pressure changes exhibit spatial hysteresis; that is, pressure changes at one location may not immediately propagate to other locations, but rather there is a certain time delay. This hysteresis effect can cause deviations in the zero-point compensation field data, affecting the accuracy of compensation. The fluctuation suppression unit employs a specific algorithm to eliminate this hysteresis effect, which can be expressed as:

[0064]

[0065] in, This represents the zero-point compensation field data after eliminating hysteresis in spatial coordinates. Location, Time The value at time; Indicates the number of adjacent nodes involved in the calculation; Indicates the first The spatial coordinates of the adjacent nodes Location, Time The original zero-point compensation field data value at that time; Indicates the first The pressure transmission time delay between adjacent nodes is related to factors such as the distance between adjacent nodes and the properties of the base plate material. By processing the zero-point compensation field data using this formula, the spatial lag effect can be effectively eliminated, making the zero-point compensation field data more accurately reflect the current actual pressure state.

[0066] In the specific implementation process, the field strength tracing unit first performs pressure field tracing on each compensation sequence to extract the pressure transmission chain. For example, for a compensation sequence containing multiple pressure nodes, the field strength tracing unit analyzes the pressure source of each node and how the pressure is transmitted from one node to the next, thus constructing a complete pressure transmission chain. Then, the operating condition matching unit spatially superimposes the pressure transmission chain with the pressure field fusion feature data. During the superposition process, the superposition weights are adjusted according to different operating condition types to ensure that the generated zero-point compensation field data conforms to the current operating state. For example, when the mother plate is under high-temperature conditions, the elastic modulus of the material will change, thus affecting the pressure transmission. The operating condition matching unit will adjust the superposition weights accordingly to reflect the impact of this change on the zero-point compensation field data. Finally, the fluctuation suppression unit processes the zero-point compensation field data using the above formula to eliminate spatial lag effects. During the calculation process, it is necessary to determine the number of adjacent nodes. and the time delay of each node These parameters can be determined through physical modeling and experimental measurements of the mother plate. For example, by simulating the pressure transmission process in the mother plate using the finite element analysis method, the time delay between different locations can be obtained, thus providing accurate parameters for the wave suppression unit.

[0067] After sequential processing by the field strength tracing unit, operating condition matching unit, and fluctuation suppression unit, the zero-point adaptive layer generates accurate zero-point compensation field data. This data is then transmitted to the compensation triggering layer to generate pressure compensation commands. Throughout the process, data transmission and processing between units are continuous and dynamic, enabling real-time response to changes in the motherboard pressure field and ensuring the system's compensation effectiveness and adaptability.

[0068] Example 4:

[0069] The compensation triggering layer, as a key component in the system for generating pressure compensation commands, comprises a multi-dimensional field strength coordination unit, a dynamic field strength optimization unit, and a pressure equalization unit. Each unit achieves precise control of the pressure field through data interaction and algorithm processing. The implementation method is detailed below with reference to a specific application scenario:

[0070] In an industrial automated production line, when a mother plate supports precision machining equipment, multiple compensation decision nodes within a multi-dimensional field strength coordination unit receive real-time zero-point compensation field data and pressure field fusion characteristic data. For example, when the right side of the mother plate is subjected to a 300N axial load, compensation decision node A obtains a pressure value of 1.2MPa from the pressure field fusion characteristic data of that area, while the zero-point compensation field data shows a pressure offset coefficient of 0.35 between that area and the left reference point. Each compensation decision node is connected to various compensation sequences through preset parameter configurations, such as node B corresponding to the feedback sequence of dynamic deformation characteristic data and node C corresponding to the calibration sequence of axial offset characteristic data. These nodes are distributed in a matrix in key stress areas of the mother plate, such as around the positioning pin holes and at the connection points of the support ribs, ensuring coverage of all possible areas of abnormal pressure.

[0071] The dynamic field strength optimization unit iteratively optimizes the parameter configuration through a dynamic field strength adjustment algorithm. Assuming the initial parameter configuration of compensation decision node D is a weight coefficient of [0.4, 0.3, 0.3] (corresponding to pressure label, zero-point label, and deformation data, respectively), when the system detects a 0.5mm deformation in the center of the motherboard due to a sudden load change, the dynamic field strength adjustment algorithm calculates the error between the current pressure compensation command and the actual pressure distribution. For example, if the actual pressure distribution shows a pressure of 1.8MPa in the central region, while the expected compensation command is 1.5MPa, the error is 0.3MPa. The algorithm automatically adjusts the weight coefficient of node D to [0.5, 0.2, 0.3], gradually reducing the error through multiple iterations. During this process, the algorithm references historical pressure data under multiple operating conditions. For instance, under similar load conditions, adjusting the weight coefficient to [0.6, 0.1, 0.3] resulted in better compensation, thus accelerating the optimization speed.

[0072] The pressure equalization unit performs high-pressure condition positioning based on zero-point compensation field data and pressure field fusion characteristic data. When a local pressure surges suddenly in a certain area of ​​the motherboard due to workpiece installation eccentricity, such as a sudden increase in pressure from 0.8MPa to 2.1MPa near the positioning pin hole, the pressure equalization unit determines the specific location and influence range of the high-pressure area by analyzing the pressure deviation between this area and adjacent areas in the zero-point compensation field data and the pressure gradient change in the pressure field fusion characteristic data. For example, if the influence radius of the high-pressure area is found to be 50mm, and it causes an offset of 0.03mm in the adjacent positioning reference point, a pressure compensation command is generated. The command includes: applying a reverse pressure of 0.5MPa at the pneumatic compensation module corresponding to the high-pressure area for a compensation time of 200ms, and adjusting the pressure distribution of adjacent areas to balance the overall field strength.

[0073] In practical implementation, the compensation decision nodes of the multi-dimensional field strength collaborative unit are deployed according to the partition of the mother plate structure. Taking a rectangular mother plate as an example, one node is set every 200mm along the X and Y axes, for a total of 9 nodes. Each node collects pressure and deformation data of the corresponding area in real time. When a node detects that the pressure data fluctuation exceeds a preset threshold (e.g., ±0.2MPa), it immediately transmits the data to the dynamic field strength optimization unit. The dynamic field strength optimization unit uses a gradient descent algorithm for parameter iteration. The step size of each iteration is automatically adjusted according to the error magnitude. For example, the step size is set to 0.1 when the error is greater than 0.5MPa, and 0.01 when the error is less than 0.1MPa, ensuring the stability and efficiency of the optimization process. The pressure equalization unit assists in positioning by establishing a high-pressure operating condition feature library. The feature library stores pressure mutation patterns under different load types, such as edge high pressure caused by radial load and center high pressure caused by axial load. When real-time data matches a certain pattern, the compensation strategy is quickly determined.

[0074] In another application scenario, the motherboard is used in automated equipment with multi-station switching. When the station switches from A to B, the load distribution changes, and the pressure in the right support area increases from 0.5MPa to 1.5MPa, accompanied by a 0.2mm deformation. Nodes E and F of the multi-dimensional field strength coordination unit detect the pressure change and deformation data, respectively, and transmit the information to the dynamic field strength optimization unit. Based on the compensation parameters of similar station switching in historical data, the optimization unit quickly adjusts the weight coefficient of node E, increasing the weight ratio of the pressure label. The pressure equalization unit determines that the high-pressure area is concentrated near the right support block and generates a compensation command: activate the three pneumatic compensation columns on the right, applying reverse pressures of 0.3MPa, 0.2MPa, and 0.3MPa respectively, forming a gradient compensation field, causing the pressure in this area to gradually drop back to 1.2MPa. At the same time, the zero-point compensation field data ensures that the offset of the positioning reference point is controlled within 0.01mm.

[0075] Throughout the entire operation of the compensation trigger layer, each unit forms a closed-loop control through real-time data interaction. The multi-dimensional field strength coordination unit continuously collects data and transmits it to the dynamic field strength optimization unit. The optimized parameters are fed back to the coordination unit to adjust the node configuration, and the pressure equalization unit generates precise compensation commands based on the optimized parameters and real-time field strength data. This mechanism ensures that the system can respond quickly and generate effective compensation strategies when facing different operating conditions, maintaining the balance of the motherboard pressure field and guaranteeing zero-point positioning accuracy. For example, under continuous alternating load conditions, the system can complete multiple parameter iterations and optimizations within each load cycle, ensuring that the deviation between the pressure compensation command and the actual pressure distribution is always controlled within a reasonable range, avoiding the accumulation of positioning errors caused by pressure imbalance.

[0076] Example 5:

[0077] The dynamic field strength clustering method is used in a pneumatic zero-point positioning motherboard pressure adaptive compensation system to perform real-time matching of standardized pressure and deformation data, thereby improving the accuracy of pressure field analysis. The implementation process is detailed below with reference to a specific production scenario:

[0078] In a machining center's motherboard application scenario, the motherboard needs to support a CNC milling machine. When the machine performs vertical cutting on the workpiece, the motherboard experiences axial loads that fluctuate with the depth of cut. In this case, the dynamic field strength clustering method first generates axial compensation parameters based on the pressure field distribution characteristics of real-time operating parameters. For example, when the cutting depth increases from 5mm to 8mm, real-time monitoring shows that the pressure field distribution in the central region of the motherboard changes from a uniform distribution of 0.8MPa to a gradient distribution with pressure increasing to 1.2MPa within a 100mm radius centered on the center, and the pressure change rate in this region is 0.15MPa / s. The system calculates the axial compensation parameters based on these characteristics. These parameters are used for subsequent adjustments to the pressure data in the axial direction to reflect the impact of load changes on the pressure field.

[0079] Next, a spatial grid segmentation mechanism was used to perform multi-scale matching processing on the standardized pressure data. It was assumed that the bearing surface of the mother plate was divided into a 200mm × 200mm spatial grid, with each grid cell corresponding to a pressure monitoring point. In large-scale matching, the entire mother plate was treated as a single region, and the influence of changes in cutting depth on the overall pressure field trend was analyzed. For example, it was found that the overall pressure field exhibited a distribution trend of high pressure at the center and low pressure at the edges. In medium-scale matching, the mother plate was divided into four quadrants, and the pressure distribution characteristics within each quadrant were analyzed. For instance, the pressure in the first quadrant increased from the upper left to the lower right corner, while the pressure in the fourth quadrant remained relatively stable. In small-scale matching, the focus was on each grid cell. For example, the pressure in the central grid cell (coordinates 100, 100) increased from 0.8MPa to 1.2MPa, and the pressure in the adjacent grid cell (coordinates 100, 150) increased from 0.7MPa to 1.0MPa. Through this multi-scale matching, the variation characteristics of the pressure data across different dimensions were comprehensively captured.

[0080] In another scenario, a motherboard is used in an automotive parts assembly line. When a robotic arm places an engine block onto the motherboard, the left side of the motherboard experiences a concentrated load. Real-time operating parameters include the weight of the engine block (approximately 500 kg), the placement speed (0.5 m / s), and the contact area (300 mm × 200 mm). Dynamic field strength clustering, based on these parameters, generates a pressure field distribution characteristic showing that the pressure in the left side region spikes to 1.8 MPa at the moment of contact, then gradually stabilizes to 1.5 MPa within 0.5 s. Based on this, axial compensation parameters are generated to adjust the axial distribution model of the pressure data to accommodate the effects of the concentrated load.

[0081] The spatial grid segmentation mechanism is applied in this scenario as follows: During large-scale matching, the overall pressure field of the mother plate is observed to shift to the left; during medium-scale matching, the mother plate is divided into a left-side load-bearing zone and a right-side non-load-bearing zone, with the pressure in the left-side load-bearing zone being significantly higher than that in the right-side zone; during small-scale matching, the pressure changes of each grid cell within the left-side load-bearing zone are further analyzed. For example, the pressure in the central grid cell (coordinates 50, 100) reaches a peak of 1.8 MPa at the moment of contact, and the pressure in adjacent grid cells decreases with distance from the center. Through multi-scale matching processing, the system can accurately identify the distribution characteristics and variation patterns of the pressure field under concentrated loads.

[0082] The implementation of the dynamic field strength clustering method is also reflected in the real-time matching of standardized pressure and deformation data. For example, in the application of motherboards in the aerospace field, the motherboards are subjected to alternating temperature loads (-20℃ to 80℃) and periodic vibration loads, resulting in complex changes in the pressure and deformation fields. Real-time monitoring showed that when the temperature rose from 20℃ to 50℃, the pressure in the central region of the motherboard increased from 1.0MPa to 1.3MPa, accompanied by a deformation of 0.15mm. The dynamic field strength clustering method first generates axial compensation parameters based on the temperature and vibration operating parameters to reflect the influence of temperature changes on the material's elastic modulus and the interference of vibration on pressure transmission. Then, multi-scale matching is performed through a spatial grid segmentation mechanism. On a large scale, it was found that the overall pressure field increased due to the temperature rise; on a medium scale, a large pressure gradient was observed at the joints of different materials; and on a small scale, the focus was on the pressure abrupt changes in temperature-sensitive areas (such as bolted connections).

[0083] During the matching process, the system performs correlation analysis on standardized pressure and deformation data. For example, when the pressure in a certain grid cell increases by 0.2 MPa, the corresponding deformation data increases by 0.05 mm. The correlation between these two data points is determined using a dynamic field strength clustering method, providing a basis for subsequent pressure compensation. Simultaneously, for compensation sequences with inconsistent numbers of pressure nodes, the system performs virtual field strength interpolation based on the operating parameters of the last pressure node of the sequence with fewer pressure nodes, ensuring data consistency and comparability.

[0084] The entire implementation process of the dynamic field strength clustering method closely revolves around real-time operating parameters. Through a multi-scale spatial grid segmentation mechanism, it achieves accurate analysis and matching of pressure and deformation data. This method can adapt to different types of loads and operating conditions, providing a reliable benchmark sequence for the pressure standardization module, thus laying the foundation for subsequent pressure field equilibrium identification and pressure compensation command generation. For example, in the stamping process of automobile manufacturing, the mother plate is subjected to instantaneous high-pressure impacts. The dynamic field strength clustering method can quickly identify the abrupt changes in the pressure field, generate accurate axial compensation parameters and matching results, enabling the system to respond promptly and implement pressure compensation, ensuring the zero-point positioning accuracy of the mother plate and the normal operation of the equipment.

[0085] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0086] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A pressure self-adaptive compensation system for a zero-point positioning motherboard of a pneumatic type, characterized in that, The application relates to a zero-point positioning module for multi-source pressure field fusion processing of initial pressure field data, wherein the initial pressure field data comprises a compensation sequence corresponding to static reference characteristic data, a feedback sequence corresponding to dynamic deformation characteristic data and a calibration sequence corresponding to axial offset characteristic data; the static reference characteristic data comprises a pressure label generated by a motherboard bearing distribution parameter and a zero-point label generated by a positioning reference parameter; a pressure compensation module for dynamic fluctuation suppression processing of a real-time deformation data stream and inputting of the real-time deformation data stream into a multi-field coupling layer for pressure field equalization identification; a pressure compensation instruction is generated according to an output result of the multi-field coupling layer; the multi-field coupling layer comprises a pressure standardization module and a field strength reconstruction module; the pressure standardization module is used for spatial grid division and abnormal field strength filtering of an original pressure data stream; the field strength reconstruction module is obtained through joint training based on multi-working-condition historical pressure data and a real-time deformation track; the field strength reconstruction module comprises a pressure field analysis layer, a zero-point adaptive layer and a compensation trigger layer which are connected in sequence; the pressure field analysis layer is used for spatial coupling processing of different compensation sequences in the original pressure data stream to generate pressure field fusion characteristic data; the zero-point adaptive layer is used for modeling of a dynamic correlation relationship between pressure field fusion characteristic data corresponding to each compensation sequence to generate zero-point compensation field data; and the compensation trigger layer is used for multi-dimensional field strength fusion based on the zero-point compensation field data and the pressure field fusion characteristic data to generate a pressure compensation instruction. The modeling of the dynamic correlation relationship between the pressure field fusion characteristic data corresponding to each compensation sequence to generate the zero-point compensation field data comprises the following steps: pressure key nodes in the pressure field fusion characteristic data are identified by using a deformation segmentation algorithm, and zero-point compensation sequences corresponding to each compensation sequence are determined based on a working condition type corresponding to each pressure key node; offset coefficients between pressure nodes of the same working condition in zero-point compensation sequences corresponding to any two compensation sequences are calculated, and zero-point compensation field data between the any two compensation sequences is generated based on the offset coefficients.

2. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 1, wherein, The calculation of the offset coefficients between the pressure nodes of the same working condition in the zero-point compensation sequences corresponding to the any two compensation sequences comprises the following steps:

3. The system of claim 2, wherein, when the number of pressure nodes in the any two compensation sequences is inconsistent, virtual field strength points are interpolated based on working condition parameters corresponding to terminal pressure nodes in the compensation sequence with less pressure nodes, and offset coefficients between pressure nodes of the same working condition are calculated based on the interpolated data. The pressure standardization module is specifically used for the following steps: equal field strength division is performed on the compensation sequence, the feedback sequence and the calibration sequence according to a preset spatial grid to generate standardized pressure data, standardized deformation data and standardized calibration data; 4. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 3, wherein, real-time matching is performed on the standardized pressure data and the standardized deformation data by using a dynamic field strength clustering method, and a fixed space sliding mechanism is used for steady-state optimization of the standardized calibration data to output a first reference sequence, a second reference sequence and a third reference sequence; wherein the first reference sequence comprises a calibrated pressure label and a calibrated zero-point label. The pressure standardization module is further used for the following steps:

5. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 1, wherein, ​ ​ ​ 6. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 5, wherein, ​ Calculate a deformation offset coefficient of the calibrated pressure label and the calibrated zero point label in a historical monitoring period; According to the deformation offset coefficient and the calibrated pressure label, predict an expected distribution value of the calibrated zero point label in a real-time monitoring period under a working condition parameter of the real-time monitoring period; Generate target zero point compensation data based on the calibrated zero point label and the expected distribution value, and take a compensation sequence corresponding to the target zero point compensation data as a first reference sequence.

7. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 2, wherein, The zero point adaptive layer specifically includes: A field strength tracing unit configured to perform pressure field tracing on each compensation sequence in the pressure field fusion feature data respectively, so as to extract a corresponding pressure conduction chain from each compensation sequence; A working condition matching unit configured to perform spatial superposition on the pressure conduction chain extracted from each compensation sequence and the corresponding pressure field fusion feature data, so as to generate zero point compensation field data.

8. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 7, wherein, The zero point adaptive layer further includes: A fluctuation suppression unit configured to perform spatial lag effect elimination processing on the zero point compensation field data.

9. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 2, wherein, The compensation trigger layer specifically includes: A multi-dimensional field strength coordination unit including a plurality of compensation decision nodes, each compensation decision node being connected to the zero point compensation field data and each compensation sequence in the pressure field fusion feature data through parameter configuration; A dynamic field strength optimization unit configured to perform iterative optimization on the parameter configuration through a dynamic field strength adjustment algorithm, so as to minimize the error between the pressure compensation instruction and the actual pressure distribution; A pressure equalization unit configured to perform high-pressure working condition positioning based on the zero point compensation field data and the pressure field fusion feature data, so as to generate a pressure compensation instruction.

10. The pneumatic zero positioning mother board pressure adaptive compensation system of claim 5, wherein, The dynamic field strength clustering method specifically includes: Generating an axial compensation parameter based on the pressure field distribution characteristics of the real-time working condition parameter; Performing multi-scale matching processing on the standardized pressure data by using a spatial grid segmentation mechanism.

Images