Self-adaptive pressure adjusting system and method for double-roller rolling machine

Abstract

The invention relates to the technical field of pressure adjusting devices, in particular to a self-adaptive pressure adjusting system and method for a double-roller rolling machine, and the system comprises a pressure intensity judgment module, a section analysis module, a deformation characteristic module, a hydraulic adjusting module and a coupling correction module. According to the method, the yield risk is recognized in advance by obtaining and analyzing the roller contact pressure difference value in real time, recognizing the asymmetric trend of the left pressure and the right pressure in combination with a support vector mechanism to judge the deviation direction, dynamically segmenting the section diameter and calculating the difference to improve the deformation resolution, and vectorizing and analyzing the included angle relation between the section deformation and the roller displacement; in hydraulic analysis, time delay and valve control flow velocity are coupled, pressure regulation continuity is guaranteed, roller posture change is judged through cooperation of a roller shaft and a section trend, hydraulic output is dynamically corrected, and stable pressure regulation and intelligent adaptation are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of pressure regulating devices, and in particular to an adaptive pressure regulating system and method for a double-roller roller compactor. Background Art

[0002] The technical field of pressure regulating devices encompasses technologies related to regulating pressure in mechanical systems, particularly hydraulic and pneumatic systems and their application in construction machinery. The core objective is to ensure that equipment maintains a reasonable and stable operating pressure under varying operating conditions, thereby guaranteeing operational safety and system reliability. This technical field systematically encompasses pressure sensing, pressure feedback, automatic regulation, and energy conversion. It is widely used in pressure control and response regulation solutions in compaction machinery, forming equipment, transmission mechanisms, and other fields, involving multi-faceted integration of mechanical structure, control logic, and actuators.

[0003] Among them, a dual-roller adaptive pressure regulation system and method refers to a control system and operating procedure for adaptively adjusting the pressure applied between the two rollers in road or foundation compaction equipment. The technical matters involved include load monitoring of the compaction device, identification of travel status, and adjustment of hydraulic drive pressure output. Specifically, by detecting changes in road resistance and roller vibration in the rolling area, combined with a hydraulic control system, dynamic regulation of the roller compaction force is achieved. The methods used include setting a parameter threshold control loop, establishing a roller pressure feedback model, and using an electro-hydraulic proportional controller to adjust the pressure output in a coordinated manner, achieving independent pressure adjustment of the left and right rollers to adapt to different compaction environments.

[0004] While existing technologies for pressure regulation in rolling equipment include hydraulic feedback and vibration recognition mechanisms, they are significantly deficient in accurately identifying uneven roller pressure distribution and deformation trends. Pressure sensing often relies on fixed threshold control and is unable to dynamically capture subtle changes in pressure fluctuations, resulting in lag in the adjustment results. During roller movement, the synergistic relationship between roller structural deformation and road resistance changes is not fully integrated, resulting in a deviation between roller displacement and pressure output, particularly on irregular terrain. Furthermore, hydraulic system responses are often executed based on preset parameters, ignoring the flow rate imbalance caused by differences in signal response time during dynamic operations. This can easily lead to insufficient or overshooting of roller pressure, resulting in risks such as uneven rolling effects and structural stress concentration. For example, when operating in soft-hard interface areas, the existing system is unable to identify local roller deformation and pressure differences in real time, often resulting in insufficient compaction on one side and overpressure and cracking on the other side, affecting the overall compaction quality and the life of the equipment structure. Summary of the Invention

[0005] In order to solve the technical problems existing in the prior art, the embodiment of the present invention provides a system and method for adjusting the pressure of a double-roller compactor. The technical solution is as follows: In one aspect, a double-drum roller compactor adaptive pressure regulation system is provided, the system comprising: The pressure discrimination module obtains the pressure in the contact area of ​​the two rollers and calculates the difference between the pressure on both sides and the average pressure. It constructs the roller pressure asymmetry judgment value through the support vector machine, generates the pressure asymmetry parameter and passes it to the cross-section analysis module; A cross-sectional analysis module, which obtains the cross-sectional diameter of the bicycle frame tube based on the pressure asymmetry parameter, segments the cross-sectional diameter change and calculates the difference, extracts the cross-sectional deformation trend through a segmented regression algorithm, and transmits it to the deformation feature module; The deformation feature module obtains the roller displacement data based on the cross-sectional deformation trend, constructs the roller deformation gradient, calculates whether the vector angle exceeds the yield strain threshold, and generates the pipe deformation feature to transmit to the hydraulic adjustment module; A hydraulic adjustment module collects the time difference between the hydraulic command issuance time and the pressure detection time based on the pipe deformation characteristics, calculates the hydraulic proportional valve flow rate adjustment value when the response delay threshold is exceeded, and generates a hydraulic adjustment flow rate command to transmit to the coupling correction module; The coupling correction module obtains the offset position of the central axis of the roller based on the hydraulic adjustment flow rate instruction, constructs the offset pressure coupling judgment condition based on the cross-sectional deformation trend, adjusts the hydraulic flow rate, and generates a roller pressure adjustment instruction set.

[0006] As a further solution of the present invention, the pressure asymmetry parameters include the average pressure value of the contact area, the unilateral pressure values ​​of the left and right rollers, and the pressure imbalance coefficient; the cross-sectional deformation trend includes the cross-sectional diameter change segment, the segmented regression fitting curvature value, and the deformation slope change amplitude; the pipe deformation characteristics include the deformation gradient vector, the yield threshold angle, and the cross-sectional critical strain value; the hydraulic flow rate regulation instruction includes the response delay time difference, the threshold judgment mark, and the proportional valve adjustment flow rate value; the roller pressure regulation instruction set includes the roller axial offset, the cross-sectional deformation correlation rate, and the coupling condition matching flow rate parameters.

[0007] As a further solution of the present invention, the pressure determination module includes: The pressure acquisition submodule captures the contact area of ​​the dual rollers during the rolling process of bicycle frame tubes. It collects the spatial coordinates of multiple contact points on the roller surface, records the corresponding pressures through pressure sensors, calculates the point pressures within the contact area, and linearly arranges and integrates them to generate a linear pressure distribution value. The left-right difference calculation submodule divides the left and right contact areas based on the linear pressure distribution value, calculates the average pressure value of the left and right areas and the difference between the average pressures of the two sides by counting the total pressure values ​​and the number of contact points in multiple areas, and corrects the pressure difference based on the left and right contact areas to obtain the left-right pressure difference; The asymmetry discrimination submodule performs discrimination and classification based on the left and right pressure difference using the model's internal support vectors and interval boundaries, calculates the drum pressure asymmetry judgment value using the support vector machine model, and converts it into a continuous asymmetry expression to obtain the pressure asymmetry parameter; The support vector machine model is composed of input features, left and right pressure difference, support vectors, interval boundaries and discriminant hyperplanes.

[0008] As a further solution of the present invention, the cross-section analysis module includes: A diameter extraction submodule detects the boundary points of the bicycle frame tube cross-section profile at the roller connection based on the pressure asymmetry parameter and matching the corresponding roller numbers. The diameter values ​​at the corresponding cross-section positions are calculated based on the transverse projection distance between the boundary points and are arranged in order of distance to obtain a bicycle tube cross-section diameter sequence. A segmented difference submodule is configured to divide the bicycle tube cross-sectional diameter sequence into a number of equidistant segments according to continuous distance intervals, calculate first-order difference values ​​for adjacent diameters within each segment, and simultaneously calculate the direction and magnitude of the difference values ​​within each segment to generate segmented difference variation; The deformation trend submodule performs linear fitting on each segment through a segmented regression algorithm based on the segmented difference change, calculates the fitting slope and intercept, compares the slope change directions of adjacent segments and classifies them to form a trend structure sequence, extracts the overall trend change interval and integrates it to generate the cross-sectional deformation trend.

[0009] As a further solution of the present invention, the deformation feature module includes: The displacement acquisition submodule uses the roller position encoder signal to record the continuous displacement data of the roller along the axial direction, maps the roller displacement data with the cross-sectional deformation trend position based on the timestamp synchronization, and calculates the change amplitude at the same position node to generate the roller dynamic displacement sequence; The gradient construction submodule extracts the incremental vector of the displacement direction and the cross-sectional trend change vector based on the dynamic displacement sequence value of the roller, combines them into a roller deformation gradient vector group, calculates the cosine value of the angle between adjacent vectors, and filters the nodes and positions according to the yield strain angle threshold to obtain the number of positions exceeding the yield angle; The morphology generation submodule calls the vector information of the number of positions exceeding the yield angle and the corresponding positions in the roller deformation gradient vector group, reconstructs the deformation curve structure according to the node position, calculates the corresponding structural change index based on the inflection point density and angle change characteristics of the bicycle tube deformation curve, and establishes the tube deformation characteristics.

[0010] As a further solution of the present invention, the yield strain threshold is set by subjecting typical pipe samples to uniaxial compression and bending loading in an experimental platform, recording stress changes and corresponding surface displacement fields in real time, calculating angle changes between consecutive nodes, selecting angle values ​​corresponding to nonlinear mutation points, and combining statistical results from multiple batches of material samples.

[0011] As a further solution of the present invention, the hydraulic adjustment module includes: The response detection submodule collects the hydraulic command issuance time and the pressure sensor feedback detection time during the hydraulic control cycle of the bicycle frame tube rolling based on the tube deformation characteristics, synchronizes and matches them by number, calculates the difference between two time points with the same number, arranges them in time sequence, and generates a hydraulic response time difference sequence; The delay determination submodule calls the hydraulic response time difference sequence, compares the multiple time difference values ​​with the response delay threshold, marks the time difference exceeding the response delay threshold and records the corresponding time sequence number, calculates the proportion of the exceeding threshold number to the total number and extracts the position range to obtain the response delay identification interval; The flow rate calculation submodule selects the hydraulic proportional valve control parameters and pressure change curve in the corresponding time period according to the response delay identification interval, calculates the current pressure change rate and combines it with the target pressure increase value, establishes the control amount adjustment ratio required for time and converts the valve opening amplitude to obtain the hydraulic regulation flow rate instruction.

[0012] As a further solution of the present invention, the hydraulic response delay threshold is set by collecting time difference data between the control instructions issued by the hydraulic proportional valve and the feedback of the pressure sensor under multiple strokes in the test phase, statistically analyzing the response time distribution of multiple groups of data, and selecting the corresponding time difference with a high cumulative frequency ratio as the upper limit of the response time.

[0013] As a further solution of the present invention, the coupling correction module includes: The center offset submodule detects the coordinate values ​​output by the position sensors at both ends of the drum based on the hydraulic flow rate instruction and aligns them in the same time sequence. It calculates and classifies the differences between the coordinate values ​​at both ends at adjacent time nodes, averages the consecutive differences, and generates the drum center axis offset. The trend extraction submodule calls the offset of the central axis of the drum, obtains the position of the cross-sectional contour point cloud of the drum in the corresponding time period and extracts the cross-sectional curvature, calculates the change amplitude of the cross-sectional curvature at consecutive time nodes and sorts them, selects the nodes with the most dramatic change amplitude and numbers them, and obtains the change amplitude of the cross-sectional curvature; The coupling adjustment submodule performs data pairing and alignment based on the cross-sectional curvature variation and the roller center axis offset, selects a coupling section where the offset and curvature variation are greater than a curvature offset threshold, calculates the corresponding flow rate increase or decrease ratio, converts it into a control instruction format, and generates a roller pressure adjustment instruction set; The curvature offset threshold is set by collecting response data of the roller center axis offset and the cross-sectional curvature variation under typical working conditions, combined with comparative analysis of the performance stable area and the abnormal area.

[0014] On the other hand, a method for adaptive pressure regulation of a double-roller roller is provided, which is applied to an adaptive pressure regulation system of a double-roller roller roller, and the method comprises: S1: Obtain the pressure in the contact area of ​​the two rollers and calculate the difference between the pressure on both sides and the average pressure. Use the support vector machine to construct the roller pressure asymmetry judgment value and generate the pressure asymmetry parameter. S2: Obtaining the cross-sectional diameter of the bicycle tube based on the pressure asymmetry parameter, segmenting the cross-sectional diameter change and calculating the difference, and extracting the cross-sectional deformation trend through a segmented regression algorithm; S3: Based on the cross-sectional deformation trend, roller displacement data is obtained, a roller deformation gradient is constructed, and a vector angle is calculated to determine whether it exceeds a yield strain threshold, thereby generating a deformation feature of the pipe. S4: Based on the deformation characteristics of the pipe, the time difference between the hydraulic command issuance time and the pressure detection time is collected, and when the time difference exceeds the response delay threshold, the flow rate adjustment value of the hydraulic proportional valve is calculated to generate a hydraulic adjustment flow rate instruction; S5: Based on the hydraulic flow rate adjustment instruction, the offset position of the drum center axis is obtained, and the offset pressure coupling judgment condition is constructed in combination with the cross-sectional deformation trend, and the hydraulic flow rate is adjusted to generate a drum pressure adjustment instruction set.

[0015] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: By acquiring and differentially analyzing the contact pressure during the dual-roller rolling process, the degree of pressure asymmetry between the left and right rollers can be accurately identified. Combined with a support vector recognition mechanism, the asymmetry trend can be identified, effectively avoiding pressure regulation errors caused by misjudgment. Dynamic segmentation and differential calculation of cross-sectional diameters accurately extract subtle trends in the pipe deformation process, further improving the spatial resolution of deformation perception. Vectorized analysis of cross-sectional deformation trends and roller displacement behavior, and the relationship between their angle and yield strain threshold, enables early identification of potential structural yield risks during rolling, enabling proactive early warning before pressure regulation. In hydraulic response analysis, coupling time delay with proportional valve flow rate enables dynamic adjustment of hydraulic flow output, ensuring continuous and real-time pressure control despite response lags. By collaboratively determining roller center axis offset and cross-sectional deformation trends, roller posture changes caused by uneven pressure are identified and finely adjusted accordingly, ultimately achieving synchronized correction and stable regulation of roller pressure. The overall processing logic is based on multi-dimensional perception fusion, dynamic trend prediction and feedback control coupling. It not only improves the response accuracy to rolling pressure, but also enhances the system's perception of structural deformation under complex working conditions, effectively improving the intelligence level and on-site adaptability of the rolling equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 is a system flow chart of the present invention; Figure 2 is a system block diagram of the present invention; Figure 3 Schematic diagram of the steps of the method of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0019] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0020] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0021] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0022] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0023] The embodiment of the present invention provides a dual roller compactor adaptive pressure regulating system, please refer to Figures 1 to 2 The present invention provides a technical solution, a double-roller roller compactor adaptive pressure regulation system comprising: The pressure discrimination module obtains the pressure in the contact area of ​​the two rollers and calculates the difference between the pressure on both sides and the average pressure. It constructs the roller pressure asymmetry judgment value through the support vector machine, generates the pressure asymmetry parameter and passes it to the cross-section analysis module; The cross-sectional analysis module obtains the cross-sectional diameter of bicycle frame tubes based on the pressure asymmetry parameter, divides the cross-sectional diameter change into segments and calculates the difference, extracts the cross-sectional deformation trend through the segmented regression algorithm and transmits it to the deformation feature module; The deformation feature module obtains the roller displacement data based on the cross-sectional deformation trend, constructs the roller deformation gradient, and calculates whether the vector angle exceeds the yield strain threshold. The generated pipe deformation feature is then transmitted to the hydraulic adjustment module. The hydraulic adjustment module collects the time difference between the hydraulic command issuance time and the pressure detection time based on the pipe deformation characteristics. When the response delay threshold is exceeded, it calculates the hydraulic proportional valve flow rate adjustment value and generates a hydraulic adjustment flow rate command, which is then transmitted to the coupling correction module. The coupling correction module obtains the offset position of the roller center axis based on the hydraulic adjustment flow rate instruction, constructs the offset pressure coupling judgment condition based on the cross-sectional deformation trend, adjusts the hydraulic flow rate, and generates a roller pressure adjustment instruction set.

[0024] The pressure asymmetry parameters include the average pressure value of the contact area, the unilateral pressure value of the left and right rollers, and the pressure imbalance coefficient. The cross-sectional deformation trend includes the cross-sectional diameter change segment, the segmented regression fitting curvature value, and the deformation slope change amplitude. The pipe deformation characteristics include the deformation gradient vector, the yield threshold angle, and the cross-sectional critical strain value. The hydraulic regulation flow rate instructions include the response delay time difference, the threshold judgment mark, and the proportional valve regulation flow rate value. The roller pressure regulation instruction set includes the roller axial offset, the cross-sectional deformation correlation rate, and the coupling condition matching flow rate parameters.

[0025] See also Figure 2 , the pressure discrimination module includes: The pressure acquisition submodule captures the contact area of ​​the dual rollers during the rolling process of bicycle frame tubes. It collects the spatial coordinates of multiple contact points on the roller surface, records the corresponding pressures through pressure sensors, calculates the point pressures within the contact area, and linearly arranges and integrates them to generate a linear pressure distribution value. During the rolling process of bicycle frame tubes, array pressure sensors are placed on the roller surface at 5mm intervals to collect real-time pressure data in the contact area. The data is converted using a 24-bit AD sampling chip model HX711 with a sampling frequency of 1kHz. The physical position of the sensor is mapped to a three-dimensional coordinate system using a spatial coordinate conversion algorithm. Coordinates, including arranging 36 sensors on the surface of the rolling drum with a diameter of 300mm, forming a 6×6 format, setting the pressure threshold when identifying the contact area When a sensor exceeds the threshold value within three consecutive sampling periods, it is determined to be a valid contact point. The contact area is calculated by Delaunay triangulation of the valid point, and the pressure of the three adjacent points is smoothed by the moving average method. In this example, the coordinates of sensor No. 3 are measured to be (152.3, 86.4, 0), with a pressure value of 2.7 MPa, and the pressure value of sensor No. 5 is (153.1, 85.9, 0) is 2.9 MPa. The pressure at the center of the contact area is 2.8 MPa calculated by linear interpolation. A linear distribution sequence of 72 pressure values ​​is generated and sorted along the rolling direction according to spatial position.

[0026] Table 1 Contact point pressure data

[0027] As shown in Table 1, the contact area model is constructed by selecting data from three adjacent sensors, and the contact area is calculated. , pressure gradient .

[0028] The left-right difference calculation submodule divides the left and right contact areas based on the linear pressure distribution value. By counting the total pressure values ​​and the number of contact points in multiple areas, it calculates the average pressure value of the left and right areas and the difference between the average pressures on both sides. It then corrects the pressure difference based on the left and right contact areas to obtain the left-right pressure difference. The linear distribution sequence is divided into left and right areas with the rolling center line as the boundary. The total pressure of the 18 points on the left is , the sum of the 20 points on the right , the average pressure in the left area is calculated , right district , contact area ratio , the area of ​​the right area was measured by a 3D scanner , the area of ​​the left zone , the difference correction formula is , represents the corrected pressure difference, and substituting the data into , set the difference threshold , when the calculated value exceeds the threshold, an early warning is triggered.

[0029] The asymmetry discrimination submodule performs discrimination and classification based on the left and right pressure difference using the model's internal support vectors and interval boundaries. It calculates the drum pressure asymmetry judgment value using the support vector machine model and converts it into a continuous asymmetry expression to obtain the pressure asymmetry parameter. The support vector machine model consists of input features, left and right pressure difference, support vector, interval boundary and discriminant hyperplane; Select n=5 support vectors to calculate the asymmetry, Represents the number of support vectors involved in the calculation. 32 abnormal samples were screened using the K-means clustering algorithm, and the cluster center spacing threshold was set to 2 mm. : The pressure on the left side of point 2 is collected by the PT124B-150 pressure sensor with a sampling period of 10ms and is obtained after 3 sliding average filtering. : The pressure on the right side of the same point, the collection method is the same as above, the difference , : Take the maximum left and right pressure difference in the last 200 rolling cycles and record the abnormal working conditions at 10-15 and 14:22:35. : The pressure weight on the left is determined by feature criticality analysis. The analysis data set includes 1500 groups of normal samples and 200 groups of abnormal samples. : The pressure weight on the right, and Together they constitute the eigenvector modulus , : Normalized reference value, taking the maximum value of all support vector weight modulo values, : The offset of point 2 is measured by LK-G5000 laser displacement sensor with a sampling frequency of 1kHz. : Maximum offset, derived from the 09-20 roller bearing wear failure record, : Bias constant, set according to GBT28700-2018 standard, corresponding to tolerance grade IT7, : Maximum offset reference, take the maximum adjustment allowed by the process specification, and substitute it into the formula: , Compared with the preset threshold value of 0.2, it shows that the symmetry of the pressure distribution in the current rolling process meets the requirements.

[0030] See also Figure 2 , the cross-section analysis module includes: The diameter extraction submodule detects the boundary points of the bicycle frame tube cross-section at the roller connection based on the pressure asymmetry parameter and the corresponding roller number. It calculates the diameter values ​​at the corresponding cross-section positions based on the transverse projection distance between the boundary points and arranges them in order of distance to obtain a bicycle tube cross-section diameter sequence. Based on the pressure asymmetry parameter D = 0.164, the profile data of roller No. 3 (diameter Φ300±0.05mm) was matched, and a laser displacement sensor (LK-H020) was used to scan the pipe cross section at a resolution of 0.01mm. The boundary judgment threshold was set to a diameter change rate ≥ 0.05mm / mm. One point was collected every 5° in the circumferential direction, for a total of 72 measurement points. In this example, the coordinates at 12 o'clock (150.02, 0.00) and 3 o'clock (0.00, 150.15) were measured. The lateral projection distance was calculated to obtain a diameter of 300.17mm. A diameter sequence was generated in the measurement order, with the diameter value of the 15th point being 299.83mm and the diameter of the 32nd point being 300.25mm. The spacing error between adjacent points was controlled within ±0.03mm. The sequence of 72 diameter values ​​was stored in the array Dia-Array72 in the order of 0°-360°.

[0031] Table 2 Cross-section diameter measurement table

[0032] As shown in Table 2, key angle diameter data were collected using synchronized clocks, with a timestamp error of <1ms. Diameter calculations used an ellipse fitting algorithm, with the major axis a = 150.08mm, the minor axis b = 149.97mm, and a roundness error of 0.11mm.

[0033] The segmented difference submodule is based on the bicycle tube cross-section diameter sequence, which is divided into several equidistant segments according to the continuous distance interval. The first-order difference value is calculated for the adjacent diameters within each segment, and the change direction and amplitude of the difference value within each segment are counted to generate the segmented difference change value. The diameter sequence was divided into 24 segments of 15° each. The fifth segment (60°-75°) included five data points: 300.05, 300.12, 300.08, 300.15, and 300.10. The difference scores of adjacent points were calculated as +0.07, -0.04, +0.07, and -0.05. The change amplitude threshold Δ=0.06mm was set. The first and third points exceeding the threshold were marked as significant change points. Statistically, there were two positive changes and two negative changes, with a total amplitude of 0.23mm. The difference vectors were generated as +0.07, -0.04, +0.07, and -0.05, and the change after normalization was 0.15mm.

[0034] The deformation trend submodule performs linear fitting on each segment using a segmented regression algorithm based on the segmented difference variation, calculates the fitting slope and intercept, compares the slope change directions of adjacent segments and classifies them into a trend structure sequence, extracts the overall trend change interval and integrates it to generate the cross-sectional deformation trend; Perform linear fitting on the 5th segment, taking the horizontal coordinate x as the angle value (60, 63, 66, 69, 72) and the vertical coordinate y as the diameter value. The number of data points representing the 5th segment is determined by the angle division algorithm, with 5 points collected every 15° interval. Represents the starting angle, recorded by encoder ZSP-3206 with a resolution of 0.001°. Represents the starting point diameter, measured by LK-H020 sensor, sampling time 1425ms, Representative average calculation formula , Representative range calculation formula , Representative range calculation formula , represents the diameter of the end point, and the measurement time is 1429ms; Substituting the above parameters into the formula, , slope value It forms an upward trend with the slopes of 0.42 and 0.35 in the adjacent segments. The slope difference threshold is set at 0.05, and it is determined to be a continuous positive trend area. After integration, the trend interval 60°-90° has a continuous diameter increase of 0.15mm.

[0035] See also Figure 2 , the deformation feature module includes: The displacement acquisition submodule uses the roller position encoder signal to record the continuous displacement data of the roller along the axial direction. It maps the roller displacement data with the cross-sectional deformation trend position based on the timestamp synchronization, calculates the change amplitude at the same position node, and generates the roller dynamic displacement sequence. Call the encoder model HEIDENHAINER N480 to collect the axial displacement of the roller at a sampling rate of 500Hz. The time stamp synchronization error is controlled within ±0.1ms. The displacement data and the cross-sectional deformation trend are aligned according to the position coordinates. In this example, the displacement value of measurement point 32 corresponds to 2.15mm and the deformation trend value is 0.12mm. Calculate the change amplitude , a displacement sequence including 120 nodes is generated, of which the displacement of the 55th node is 2.33mm, corresponding to a trend value of 0.15mm and an amplitude difference of 0.18mm. The tolerance threshold is set to 0.2mm, and the nodes exceeding the threshold are marked as abnormal points. A total of 3 abnormal points are detected.

[0036] Table 3 Displacement synchronization data table

[0037] As shown in Table 3, the displacement data and deformation trend values ​​maintain a synchronization accuracy of 0.1ms. The moving window method is used for outlier detection, with a window size of 5 nodes and a standard deviation threshold of 0.15mm.

[0038] The gradient construction submodule extracts the incremental vector of the displacement direction and the cross-sectional trend change vector based on the dynamic displacement sequence value of the roller, and combines them into a roller deformation gradient vector group. It calculates the cosine value of the angle between adjacent vectors and filters the nodes and positions based on the yield strain angle threshold to obtain the number of positions exceeding the yield angle. Extract displacement increment vector , trend change vector , construct the deformation gradient vector group , calculate the angle between adjacent vectors: , set the yield strain angle threshold (correspond ), two nodes were detected exceeding the threshold, with position numbers 48 and 79.

[0039] The morphology generation submodule calls the vector information of the number of positions exceeding the yield angle and the corresponding positions in the roller deformation gradient vector group, reconstructs the deformation curve structure according to the node position, calculates the corresponding structural change index based on the inflection point density and angle change characteristics of the bicycle tube deformation curve, and establishes the tube deformation characteristics; When reconstructing the deformation curve, cubic spline interpolation was performed on node 48 (displacement 2.28 mm, angle 82°) and node 79 (2.41 mm, 87°). The inflection point density was calculated to be 1.2 inflection points per 10 mm length. The normal range was set to 0.8-1.0 / 10 mm. Abnormal deformation was determined, and the structural change index was calculated as: , the index exceeds the preset threshold by 1.5%, triggering the roller adjustment instruction, and the adjustment amount .

[0040] See also Figure 2 , the hydraulic adjustment module includes: The response detection submodule collects the hydraulic command issuance time and the pressure sensor feedback detection time during the hydraulic control cycle of the bicycle frame tube rolling based on the tube deformation characteristics, synchronizes them by number, calculates the difference between two time points with the same number, and arranges them in time sequence to generate a hydraulic response time difference sequence; A hard real-time communication link is established between the BOSCHRexroth VT-HNC100 hydraulic controller and the IFMPN2594 pressure sensor. The controller sends a command packet including a 32-bit CRC checksum every 1ms, and the sensor returns a response data with a timestamp within 0.05ms. In this example, the sending time of command No. 502 is (Timed by the controller's internal crystal oscillator, frequency error ±5ppm), sensor feedback time (Based on FPGA timing unit, resolution 10ns), calculate time difference , 500 sets of data are collected continuously to form a time difference sequence, and a moving average filter (window width 5 points) is used for smoothing. The original value of data No. 38 is 0.42ms and is filtered to 0.38ms. The abnormality detection threshold is set to (in ), It represents the arithmetic mean of the hydraulic response time difference series, and 12 abnormal points are detected.

[0041] Table 4 Hydraulic response time difference data table

[0042] As shown in Table 4, the data acquisition interval is strictly controlled within 1ms ± 0.01ms. Time difference calculation is implemented using a hardware subtractor module, with processing latency less than 0.01ms. Outlier verification is performed by repeating the test command three times to eliminate accidental errors.

[0043] The delay determination submodule calls the hydraulic response time difference sequence, compares the multiple time difference values ​​with the response delay threshold, marks the time difference that exceeds the response delay threshold and records the corresponding time sequence number, calculates the proportion of the exceeding threshold number to the total number, extracts the position range, and obtains the response delay identification interval; In the instruction interval 300-400, the time difference sequence is extracted for distribution analysis and the statistics are calculated: the maximum value is 0.41ms, the minimum value is 0.17ms, and the median value is 0.21ms. The dynamic delay threshold is set. (According to ISO4413 hydraulic system response standard), 28 limit-exceeding points were detected, such as the time difference of instruction No. 327 was 0.32ms, and that of instruction No. 385 was 0.35ms, and the limit-exceeding ratio was calculated. ,The density clustering algorithm (neighborhood radius 10 points, minimum number of points 5) was used to locate the abnormal ,aggregation area as instructions 335-345. The density of the ,exceeding-limit points in the interval reached 80%, and the pressure fluctuation value ,during the synchronous detection period was ±0.12MPa (normal range ±0.05MPa).

[0044] The flow rate calculation submodule selects the hydraulic proportional valve control parameters and pressure change curve within the corresponding period according to the response delay identification interval, calculates the current pressure change rate and combines it with the target pressure increase value to establish the control amount adjustment ratio required for the time and convert the valve opening amplitude to obtain the hydraulic regulation flow rate instruction; For the abnormal interval No. 335-345, the proportional valve control current value (12.5-13.2mA) and the pressure change curve (slope 0.008MPa / ms) are extracted, and the target pressure slope is calculated to be 0.012MPa / ms (according to the process specification GBT3766). The hydraulic adjustment ratio Calculated as: , check the valve core displacement-flow curve (data point: 4mA corresponds to 5L / min, 20mA corresponds to 25L / min), the current opening of 13.2mA corresponds to a flow of 14.2L / min, and the conversion valve opening is calculated as: , generating an adjustment instruction to increase the current to 15.6mA, the measured flow rate increased to 16.1L / min, and the pressure fluctuation decreased to ±0.04MPa.

[0045] See also Figure 2 , the coupling correction module includes: The center offset submodule, based on the hydraulic flow rate control command, detects the coordinate values ​​output by the position sensors at both ends of the drum and aligns them in the same time series. It calculates and classifies the differences between the coordinate values ​​at both ends at adjacent time nodes, averages the consecutive differences, and generates the drum center axis offset. The SICKOD5000 position sensor is used to synchronously collect the coordinates of both ends of the roller at a sampling rate of 500Hz, and the left end sensor records the position sequence. , right end , the timestamp alignment error is controlled within ±0.1ms, and the difference between adjacent time nodes is calculated , , classified into axial offset group and radial offset group, taking the average of 10 consecutive cycle data to get the center offset , set the drift threshold to 0.2mm. In the example, the offset of node 25 is 0.22mm, which triggers the alarm.

[0046] Table 5 Position sensor data sheet

[0047] As shown in Table 5, the data collection interval is 2 ms, and the coordinate difference calculation is performed using a hardware subtractor, with a processing delay of < 0.01 ms. The offset classification threshold is set according to the ISO9283 industrial robot accuracy standard.

[0048] The trend extraction submodule calls the roller center axis offset, obtains the cross-sectional contour point cloud position of the roller within the corresponding time period, extracts the cross-sectional curvature, calculates the change amplitude of the cross-sectional curvature at consecutive time nodes and sorts them, selects nodes with drastic change amplitudes, numbers them, and obtains the cross-sectional curvature change amplitude; The cross-sectional point cloud data obtained by a 3D scanner (accuracy 0.02mm) was used to select 36 points within a 120° azimuth angle range to calculate the curvature radius. , point 15 in the example , point 16 , the range of change , set the threshold to 0.3mm, and detected 8 mutation points, such as the change amplitude of point 22 , marked as P22, generates a sequence of amplitude changes of 0.2, 0.4, and 0.35 in chronological order.

[0049] The coupling adjustment submodule pairs and aligns data based on the cross-sectional curvature variation and the offset of the drum center axis. It selects coupling sections where the offset and curvature variation are greater than the curvature offset threshold, calculates the corresponding flow rate increase or decrease ratio, converts it into a control instruction format, and generates a drum pressure adjustment instruction set. The curvature deviation threshold is set by collecting response data of the roller center axis deviation and the cross-sectional curvature variation under typical working conditions, combined with comparative analysis of the performance stable area and abnormal area; Pair the data of the offset of 0.15mm and the curvature change of 0.35mm to set the coupling threshold (Based on the GBT1800.1-2020 tolerance standard), three over-limit sections were detected: 1425-1427ms, 1430-1432ms, and 1435-1437ms. The hydraulic flow demand increment ratio for the corresponding time periods was calculated: , check the hydraulic valve flow-pressure curve (10L / min corresponds to 2MPa, 15L / min corresponds to 3MPa), increase the current flow rate of 12L / min to , generating a pressure regulation instruction set of +7.02L / min@1425ms, +5.6L / min@1430ms, +6.8L / min@1435ms.

[0050] See also Figure 3 A method for adaptively regulating pressure of a double-roller roller compactor is provided. The method is used to implement the above-mentioned adaptive pressure regulating system for the double-roller roller compactor. The method comprises: S1: Obtain the pressure in the contact area of ​​the two rollers and calculate the difference between the pressure on both sides and the average pressure. Use the support vector machine to construct the roller pressure asymmetry judgment value and generate the pressure asymmetry parameter. S2: Obtain the cross-sectional diameter of the bicycle tube based on the pressure asymmetry parameter, segment the cross-sectional diameter change and calculate the difference, and extract the cross-sectional deformation trend through the segmented regression algorithm; S3: Based on the cross-sectional deformation trend, the roller displacement data is obtained, the roller deformation gradient is constructed, and the vector angle is calculated to see whether it exceeds the yield strain threshold, thereby generating the pipe deformation characteristics. S4: Based on the deformation characteristics of the pipe, the time difference between the hydraulic command issuance time and the pressure detection time is collected. When the response delay threshold is exceeded, the flow rate adjustment value of the hydraulic proportional valve is calculated to generate a hydraulic adjustment flow rate instruction; S5: Based on the hydraulic adjustment flow rate instruction, the offset position of the drum center axis is obtained, and the offset pressure coupling judgment condition is constructed in combination with the cross-sectional deformation trend and the hydraulic flow rate is adjusted to generate a drum pressure adjustment instruction set.

[0051] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0052] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0053] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0054] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0055] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0056] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, and can be electrical, mechanical, or other forms.

[0057] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0058] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0059] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0060] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A double roller compactor adaptive pressure regulation system, characterized in that: The system comprises: The pressure discrimination module obtains the pressure in the contact area of ​​the two rollers and calculates the difference between the pressure on both sides and the average pressure. It constructs the roller pressure asymmetry judgment value through the support vector machine, generates the pressure asymmetry parameter and passes it to the cross-section analysis module; A cross-sectional analysis module, which obtains the cross-sectional diameter of the bicycle frame tube based on the pressure asymmetry parameter, segments the cross-sectional diameter change and calculates the difference, extracts the cross-sectional deformation trend through a segmented regression algorithm, and transmits it to the deformation feature module; The deformation feature module obtains the roller displacement data based on the cross-sectional deformation trend, constructs the roller deformation gradient, calculates whether the vector angle exceeds the yield strain threshold, and generates the pipe deformation feature to transmit to the hydraulic adjustment module; A hydraulic adjustment module collects the time difference between the hydraulic command issuance time and the pressure detection time based on the pipe deformation characteristics, calculates the hydraulic proportional valve flow rate adjustment value when the response delay threshold is exceeded, and generates a hydraulic adjustment flow rate command to transmit to the coupling correction module; The coupling correction module obtains the offset position of the central axis of the roller based on the hydraulic adjustment flow rate instruction, constructs the offset pressure coupling judgment condition based on the cross-sectional deformation trend, adjusts the hydraulic flow rate, and generates a roller pressure adjustment instruction set.

2. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The pressure asymmetry parameters include the average pressure value of the contact area, the unilateral pressure values ​​of the left and right rollers, and the pressure imbalance coefficient; the cross-sectional deformation trend includes the cross-sectional diameter change segment, the segmented regression fitting curvature value, and the deformation slope change amplitude; the pipe deformation characteristics include the deformation gradient vector, the yield threshold angle, and the cross-sectional critical strain value; the hydraulic flow rate regulation instruction includes the response delay time difference, the threshold judgment mark, and the proportional valve adjustment flow rate value; the roller pressure regulation instruction set includes the roller axial offset, the cross-sectional deformation correlation rate, and the coupling condition matching flow rate parameters.

3. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The pressure determination module includes: The pressure acquisition submodule captures the contact area of ​​the dual rollers during the rolling process of bicycle frame tubes. It collects the spatial coordinates of multiple contact points on the roller surface, records the corresponding pressures through pressure sensors, calculates the point pressures within the contact area, and linearly arranges and integrates them to generate a linear pressure distribution value. The left-right difference calculation submodule divides the left and right contact areas based on the linear pressure distribution value, calculates the average pressure value of the left and right areas and the difference between the average pressures of the two sides by counting the total pressure values ​​and the number of contact points in multiple areas, and corrects the pressure difference based on the left and right contact areas to obtain the left-right pressure difference; The asymmetry discrimination submodule performs discrimination and classification based on the left and right pressure difference using the model's internal support vectors and interval boundaries, calculates the drum pressure asymmetry judgment value using the support vector machine model, and converts it into a continuous asymmetry expression to obtain the pressure asymmetry parameter; The support vector machine model is composed of input features, left and right pressure difference, support vectors, interval boundaries and discriminant hyperplanes.

4. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The cross-section analysis module includes: A diameter extraction submodule detects the boundary points of the bicycle frame tube cross-section profile at the roller connection based on the pressure asymmetry parameter and matching the corresponding roller numbers. The diameter values ​​at the corresponding cross-section positions are calculated based on the transverse projection distance between the boundary points and are arranged in order of distance to obtain a bicycle tube cross-section diameter sequence. A segmented difference submodule is configured to divide the bicycle tube cross-sectional diameter sequence into a number of equidistant segments according to continuous distance intervals, calculate first-order difference values ​​for adjacent diameters within each segment, and simultaneously calculate the direction and magnitude of the difference values ​​within each segment to generate segmented difference variation; The deformation trend submodule performs linear fitting on each segment through a segmented regression algorithm based on the segmented difference change, calculates the fitting slope and intercept, compares the slope change directions of adjacent segments and classifies them to form a trend structure sequence, extracts the overall trend change interval and integrates it to generate the cross-sectional deformation trend.

5. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The deformation feature module includes: The displacement acquisition submodule uses the roller position encoder signal to record the continuous displacement data of the roller along the axial direction, maps the roller displacement data with the cross-sectional deformation trend position based on the timestamp synchronization, and calculates the change amplitude at the same position node to generate the roller dynamic displacement sequence; The gradient construction submodule extracts the incremental vector of the displacement direction and the cross-sectional trend change vector based on the dynamic displacement sequence value of the roller, combines them into a roller deformation gradient vector group, calculates the cosine value of the angle between adjacent vectors, and filters the nodes and positions according to the yield strain angle threshold to obtain the number of positions exceeding the yield angle; The morphology generation submodule calls the vector information of the number of positions exceeding the yield angle and the corresponding positions in the roller deformation gradient vector group, reconstructs the deformation curve structure according to the node position, calculates the corresponding structural change index based on the inflection point density and angle change characteristics of the bicycle tube deformation curve, and establishes the tube deformation characteristics.

6. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 5, characterized in that: The yield strain threshold is determined by subjecting typical pipe samples to uniaxial compression and bending loading in an experimental platform, recording stress changes and corresponding surface displacement fields in real time, calculating angle changes between consecutive nodes, selecting angle values ​​corresponding to nonlinear mutation points, and combining the statistical results of multiple batches of material samples to set the yield strain threshold.

7. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The hydraulic adjustment module includes: The response detection submodule collects the hydraulic command issuance time and the pressure sensor feedback detection time during the hydraulic control cycle of the bicycle frame tube rolling based on the tube deformation characteristics, synchronizes and matches them by number, calculates the difference between two time points with the same number, arranges them in time sequence, and generates a hydraulic response time difference sequence; The delay determination submodule calls the hydraulic response time difference sequence, compares the multiple time difference values ​​with the response delay threshold, marks the time difference exceeding the response delay threshold and records the corresponding time sequence number, calculates the proportion of the exceeding threshold number to the total number and extracts the position range to obtain the response delay identification interval; The flow rate calculation submodule selects the hydraulic proportional valve control parameters and pressure change curve in the corresponding time period according to the response delay identification interval, calculates the current pressure change rate and combines it with the target pressure increase value, establishes the control amount adjustment ratio required for time and converts the valve opening amplitude to obtain the hydraulic regulation flow rate instruction.

8. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 7, characterized in that: The hydraulic response delay threshold is set by collecting time difference data between the control command issued by the hydraulic proportional valve and the feedback of the pressure sensor under multiple strokes in the test phase, statistically analyzing the response time distribution of multiple groups of data, and selecting the corresponding time difference with a high cumulative frequency as the upper limit of the response time.

9. The self-adaptive pressure regulating system for a double-roller roller compactor according to claim 1, characterized in that: The coupling correction module includes: The center offset submodule detects the coordinate values ​​output by the position sensors at both ends of the drum based on the hydraulic flow rate instruction and aligns them in the same time sequence. It calculates and classifies the differences between the coordinate values ​​at both ends at adjacent time nodes, averages the consecutive differences, and generates the drum center axis offset. The trend extraction submodule calls the offset of the central axis of the drum, obtains the position of the cross-sectional contour point cloud of the drum in the corresponding time period and extracts the cross-sectional curvature, calculates the change amplitude of the cross-sectional curvature at consecutive time nodes and sorts them, selects the nodes with the most dramatic change amplitude and numbers them, and obtains the change amplitude of the cross-sectional curvature; The coupling adjustment submodule performs data pairing and alignment based on the cross-sectional curvature variation and the roller center axis offset, selects a coupling section where the offset and curvature variation are greater than a curvature offset threshold, calculates the corresponding flow rate increase or decrease ratio, converts it into a control instruction format, and generates a roller pressure adjustment instruction set; The curvature offset threshold is set by collecting response data of the roller center axis offset and the cross-sectional curvature variation under typical working conditions, combined with comparative analysis of the performance stable area and the abnormal area.

10. A method for adaptive pressure regulation of a double-roller roller compactor, characterized in that: The method is used to implement the double-roller roller compactor adaptive pressure regulation system according to any one of claims 1 to 9, and the method comprises: S1: Obtain the pressure in the contact area of ​​the two rollers and calculate the difference between the pressure on both sides and the average pressure. Use the support vector machine to construct the roller pressure asymmetry judgment value and generate the pressure asymmetry parameter. S2: Obtaining the cross-sectional diameter of the bicycle tube based on the pressure asymmetry parameter, segmenting the cross-sectional diameter change and calculating the difference, and extracting the cross-sectional deformation trend through a segmented regression algorithm; S3: Based on the cross-sectional deformation trend, roller displacement data is obtained, a roller deformation gradient is constructed, and a vector angle is calculated to determine whether it exceeds a yield strain threshold, thereby generating a deformation feature of the pipe. S4: Based on the deformation characteristics of the pipe, the time difference between the hydraulic command issuance time and the pressure detection time is collected, and when the time difference exceeds the response delay threshold, the flow rate adjustment value of the hydraulic proportional valve is calculated to generate a hydraulic adjustment flow rate instruction; S5: Based on the hydraulic flow rate adjustment instruction, the offset position of the drum center axis is obtained, and the offset pressure coupling judgment condition is constructed in combination with the cross-sectional deformation trend, and the hydraulic flow rate is adjusted to generate a drum pressure adjustment instruction set.

Images