Injection molding machine multi-axis servo synchronous drive control method, system and injection molding machine

Abstract

The present invention relates to the field of drive control technology, and discloses a multi-axis servo synchronous drive control method, system and injection molding machine for an injection molding machine. The method comprises: obtaining the screw axis of an injection mechanism and the installation parameters of left and right servo motors, and establishing a load monitoring reference coordinate system; obtaining the motor position coordinates, constructing a virtual load center point and calculating its dynamic position, and implementing differentiated mirror load compensation; calculating the comprehensive stability index of the left and right axes based on the compensation result; accordingly, adopting a progressive switching strategy to determine the reference axis and the follower axis, constructing a virtual load shunt channel and implementing dynamic shunt control; the present invention can realize dynamic balanced distribution of dual-axis loads, eliminate the problem of torque fluctuation transmission, and improve the injection dual-axis synchronization stability and service life.

Description

Technical Field

[0001] The present invention relates to the field of drive control technology, and more specifically, to a multi-axis servo synchronous drive control method and system for an injection molding machine, and an injection molding machine. Background Art

[0002] The injection mechanism of the injection molding machine is the core unit of plastic molding. The synchronization performance of its dual-axis servo drive system directly determines the uniformity of melt filling, product dimensional accuracy and the stability of the production process.

[0003] In the prior art, Chinese patent publication number CN109531572B discloses a method for controlling a dual-drive robot for a PET injection molding machine based on a soft PLC. This method implements dual-axis linkage control by constructing a soft PLC program framework, and uses electronic gear coupling technology to synchronize dual-axis motion, thereby optimizing the system response speed and debugging efficiency. It is mainly aimed at improving the dynamic response and production efficiency of the robot. Chinese patent publication number CN101620421B provides a drive control device and a drive control method for a servo motor. This device adds a learning control unit to the position loop control system, and performs learning control only at a predetermined time before and after the action is reversed to reduce position deviation, focusing on solving the position error problem of the servo system during commutation.

[0004] However, the above technology still has significant limitations. The electronic gear synchronization mechanism in Chinese patent CN109531572B only focuses on the coordination of motion trajectories and does not involve the balance of load distribution between the two axes. When the torque of the two axes differs due to uneven distribution of melt resistance, different mold cavity pressures, or different degrees of screw wear during the injection process, the torque fluctuation will be transmitted to the other axis through mechanical connection components such as gears and belts, destroying the original synchronization state. Although the learning control in Chinese patent CN101620421B optimizes position tracking accuracy, it does not consider the transmission characteristics of torque fluctuations and cannot suppress dynamic disturbances caused by load imbalance. Specifically, when the load on one axis suddenly increases, its torque fluctuation will trigger torque oscillations on the other axis through the mechanical transmission chain, causing the dual-axis position synchronization error to significantly expand from a small range. In severe cases, it will also cause system resonance, resulting in abnormal fluctuations in injection pressure, causing defects such as sink marks or flash on the product, while accelerating the wear of transmission components and shortening the service life of the equipment. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a multi-axis servo synchronous drive control method, system and injection molding machine for an injection molding machine. By establishing a load monitoring reference coordinate system, dynamically calculating the virtual load center point and implementing differentiated mirror compensation, the reference axis selection and dynamic shunt control are optimized in combination with the comprehensive stability index to achieve dynamic balanced distribution of dual-axis loads; effectively suppress torque fluctuation transmission, significantly improve the injection dual-axis synchronization accuracy and system stability, and extend the service life of transmission components.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A multi-axis servo synchronous drive control method for an injection molding machine, comprising:

[0008] Obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors to establish a load monitoring reference coordinate system;

[0009] Obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, calculate the dynamic position of the virtual load center point, and implement differentiated mirror load compensation control;

[0010] Based on differentiated mirror load compensation control, the comprehensive stability index of the left and right axes is calculated;

[0011] Based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis. Based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

[0012] Furthermore, the load monitoring reference coordinate system uses the intersection of the screw axis and the line connecting the centers of the two servo motors as the coordinate origin O, the direction of the screw axis is set as the positive direction of the Z axis, pointing to the direction of melt advancement; the line connecting the center point of the left servo motor and the center point of the right servo motor is set as the X axis.

[0013] Furthermore, the position coordinates of the left and right servo motors in the load monitoring reference coordinate system are determined by the actual installation distance d from the left and right servo motors to the coordinate origin O. The position coordinates of the left servo motor are (-d, 0, 0) and the position coordinates of the right servo motor are (d, 0, 0).

[0014] Furthermore, the method of constructing a virtual load center point and calculating a dynamic position of the virtual load center point includes:

[0015] Get the current value I of the left servo motor L and the current value of the right servo motor I R ;

[0016] According to IL and I R , calculate the real-time torque value T of the left shaft L And the real-time torque value T of the right axis R ;

[0017] Construct a virtual load center point P in the load monitoring reference coordinate system virtual , set P virtual The initial position is at the coordinate origin O;

[0018] According to T L 、T R And the actual installation distance d from the left and right servo motors to the coordinate origin O, dynamically calculate the virtual load center point P virtual X coordinate position X virtual .

[0019] Furthermore, the method for implementing differentiated mirror load compensation control includes:

[0020] Based on the dynamic position of the virtual load center point, a load mirror mapping relationship matrix is ​​established to identify the load imbalance state;

[0021] When it is determined that the load is unbalanced, based on the load mirror mapping relationship matrix M mirror , implement differentiated mirror load compensation control.

[0022] Furthermore, the method for identifying the load imbalance state is: setting an offset threshold, when X virtual When the absolute value of is greater than the offset threshold, it is determined that the system is in a load imbalance state.

[0023] Furthermore, the implementation of differentiated mirror load compensation control further includes:

[0024] According to the real-time torque value T of the left axis L And the real-time torque value T of the right axis R , identify lightly loaded and heavily loaded axles;

[0025] For the identified lightly loaded axis, the load mirror mapping relationship matrix M is used mirror , calculate the mirror load value T that needs to be compensated for the lightly loaded axis mirror ;

[0026] According to the calculated mirror load value T that needs to be compensated for the lightly loaded axis mirror , implement differentiated mirror load compensation control.

[0027] Furthermore, the method of determining the reference axis and the following axis by adopting the progressive switching strategy includes:

[0028] Update the comprehensive stability index of the left and right axes every s control cycles to determine the initial reference axis and initial following axis;

[0029] Based on the determined initial reference axis and initial following axis, a gradual switching is performed to obtain the final reference axis and following axis.

[0030] A multi-axis servo synchronous drive control system for an injection molding machine, which is used to implement the above-mentioned multi-axis servo synchronous drive control method for an injection molding machine, the system comprising:

[0031] Coordinate system establishment module: used to obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors, and establish the load monitoring reference coordinate system;

[0032] Mirror load compensation module: used to obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, calculate the dynamic position of the virtual load center point, and implement differentiated mirror load compensation control;

[0033] Index calculation module: Calculates the comprehensive stability index of the left and right axes based on differentiated mirror load compensation control;

[0034] Diversion control module: Based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis; based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

[0035] An injection molding machine, comprising:

[0036] An injection mechanism, wherein the injection mechanism is provided with a screw, and the screw is used to advance the melt;

[0037] The left and right servo motors are installed on both sides of the screw to drive the screw movement;

[0038] The multi-axis servo synchronous drive control system for the injection molding machine as described above is electrically connected to the left and right servo motors, and is used to control the left and right servo motors to achieve multi-axis servo synchronous drive.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention provides a unified spatial reference for the quantitative analysis of the dual-axis load state by establishing a load monitoring reference coordinate system, and accurately captures the degree and direction of the uneven load distribution by combining the dynamic position calculation of the virtual load center point; the differentiated mirror load compensation control implements directional adjustment for the load imbalance state and preliminarily balances the dual-axis torque; the calculation of the comprehensive stability index realizes a multi-dimensional evaluation of the dual-axis motion accuracy, dynamic response and load change characteristics, and provides a quantitative basis for the dynamic determination of the reference axis and the follower axis; a virtual load shunt channel is constructed based on the reference axis and dynamic shunt control is implemented to further optimize the dual-axis load distribution, forming a closed-loop control from load monitoring, preliminary compensation to precise optimization. Through the above method, the dynamic balance of the dual-axis load is effectively achieved, the torque difference is reduced, the transmission of torque fluctuations through mechanical connections is suppressed, the system oscillation is avoided, the dual-axis synchronization accuracy and stability during the injection process are improved, and at the same time, the mechanical wear caused by uneven load is reduced, and the service life of the equipment is extended. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying any creative work.

[0042] Figure 1 This is a principle flow chart of a multi-axis servo synchronous drive control method for an injection molding machine in the present invention;

[0043] Figure 2 A flow chart of the method for constructing a virtual load center point and calculating the dynamic position of the virtual load center point in the present invention;

[0044] Figure 3 Schematic diagram of the dynamic position of the virtual load center point in the present invention;

[0045] Figure 4 Schematic diagram of the load mirror mapping relationship matrix structure in the present invention;

[0046] Figure 5 This is a functional module diagram of a multi-axis servo synchronous drive control system for an injection molding machine in the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] Example 1

[0049] See also Figure 1 As shown, this embodiment provides a multi-axis servo synchronous drive control method for an injection molding machine, comprising:

[0050] Step S100: Acquire the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors to establish a load monitoring reference coordinate system; obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system; construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system; calculate the dynamic position of the virtual load center point; and implement differentiated mirror load compensation control;

[0051] Furthermore, step S100 includes:

[0052] Step S110, obtaining the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors, and establishing a load monitoring reference coordinate system;

[0053] Furthermore, step S110 includes:

[0054] Step S111: Obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors. Take the intersection of the screw axis and the line connecting the centers of the two servo motors as the coordinate origin O, and set the screw axis direction as the positive direction of the Z axis, pointing in the direction of melt advancement.

[0055] In step S112, based on the coordinate origin O and the Z-axis direction, the line connecting the center points of the left servo motor and the right servo motor is set as the X-axis, the left side is defined as the negative direction of the X-axis, and the right side is defined as the positive direction of the X-axis. The Y-axis direction is determined by the right-hand rule to form a load monitoring reference coordinate system.

[0056] Specifically, step S110 is used to construct a load monitoring reference coordinate system. Its specific implementation process is as follows: First, the position of the injection mechanism's screw axis and the installation position parameters of the left and right servo motors are obtained. This process requires combining the mechanical design drawings of the injection mechanism with on-site calibration data. The screw axis is marked at multiple points using a laser tracker, for example, marking three points at each end and in the middle of the screw. The spatial linear equation of the screw axis is then fitted using the least squares method. Simultaneously, a high-precision three-dimensional coordinate measuring instrument is used to collect the spatial positions of the centers of the left and right servo motors and record their coordinate values ​​on the mechanical installation reference plane. The intersection of the screw axis and the line connecting the centers of the left and right servo motors is set as the coordinate origin. This origin selection is based on the mechanical transmission path of the injection molding machine's dual-axis drive. The screw is the core component of melt propulsion, and the force and torque in its axial direction directly affect the dual-axis load distribution. The line connecting the two axes is the main path for torque transmission. Using the intersection of the two as the origin allows the origin of the coordinate system to coincide with the simplified center of the load transmission force system, solving the problem of load calculation errors caused by the deviation between the origin of the traditional coordinate system and the actual center of the force system. The Z-axis direction is set as the direction of melt advancement pointing from the screw axis. This direction is consistent with the main direction of melt resistance during injection, ensuring that the change in the Z-axis coordinate can directly reflect the axial component of the melt advancement resistance; the X-axis is set as the line connecting the centers of the left and right servo motors, with the left side being the negative direction of the X-axis and the right side being the positive direction of the X-axis. Since the two axes are symmetrically installed on both sides of the screw, this setting makes the two axes mirror-distributed on the X-axis. The X component of their position coordinates can directly represent the distance relationship between the two axes and the load center; the Y-axis direction is determined by the right-hand rule. This direction is perpendicular to the plane formed by the X-axis and Z-axis, and is used to represent the load component perpendicular to the two axes and the screw plane, such as the lateral force generated by mechanical vibration.

[0057] The coincidence of the origin of the load monitoring reference coordinate system and the center of the force system eliminates the need for additional corrections to the force arm parameters in the load monitoring equation (such as the distance between the two axes and the load center), allowing them to be calculated directly from the coordinate values, thus reducing errors caused by simplified force systems. The coincidence of the X-axis and the line connecting the two axes allows the torque difference between the two axes to be directly expressed as an X-direction load offset in the coordinate system, providing a linear mapping relationship for the subsequent calculation of the X-coordinate of the virtual load center point, and resolving the nonlinear relationship between torque and position offset in traditional coordinate systems. The consistency of the Z-axis with the melt advance direction allows Z-axis coordinate changes to be correlated with melt pressure fluctuations. When the melt pressure increases, the change in the screw axial displacement is reflected in the Z-axis coordinate, providing a data interface for optimizing load compensation by integrating the melt pressure signal. Although the introduction of the Y-axis currently focuses on the X-axis load, it also reserves a dimension for monitoring parallelism errors in the dual-axis mechanical installation, such as the vertical offset of the two axes after long-term operation. When significant coordinate changes in the Y-axis occur, it can indicate installation deviation of the mechanical structure, facilitating preventive maintenance. The symmetrical design of the coordinate system enables the control parameters of the left and right servo motors, such as the current-torque conversion coefficient, to be mutually calibrated through coordinate transformation, thereby improving the stability of the system parameters.

[0058] Step S120, obtaining the position coordinates of the left and right servo motors in the load monitoring reference coordinate system; the position coordinates of the left and right servo motors in the load monitoring reference coordinate system are determined by the actual installation distance d from the left and right servo motors to the coordinate origin O;

[0059] Obtaining the distance d requires mechanical installation calibration. A reference point is marked at the center of the servo motor's mounting flange. A micrometer is used to measure the linear distance from the reference point to the coordinate origin at different angles, including 0°, 90°, 180°, and 270°. After removing outliers, the average value is taken as the actual installation distance d. Based on this, the position of the left servo motor is marked as point L, with coordinates (-d, 0, 0), and the position of the right servo motor is marked as point R, with coordinates (d, 0, 0). This setting, based on the symmetry along the X-axis, results in the dual axes being mirrored on the X-axis. Their coordinate values ​​have opposite signs only in the X-axis component, with the Y-axis and Z-axis components both being 0. This indicates that under ideal installation conditions, the dual axes have no positional deviation in the direction perpendicular to the X-axis.

[0060] Symmetrical coordinates unify the form of the load calculation equations for the left and right motors. The same set of calculation logic can be reused only by distinguishing them through symbols, which simplifies the complexity of the control algorithm and solves the problem of designing calculation models for both axes separately under traditional asymmetric coordinates. The setting of the Y-axis and Z-axis components to 0 in the coordinates provides a reference state. When the Y-axis or Z-axis component is detected to be non-zero during actual operation, it can be directly determined that there is installation deviation or mechanical deformation in the two axes. For example, a 0.01-meter deviation on the Y-axis indicates that the left motor is misaligned with the right motor, providing a quantitative basis for mechanical fault diagnosis. The coordinate value is determined by the actual installation distance d, making the method adaptable to different models of injection molding machines. The d value of different models is different. The d value of small injection molding machines is 0.3 meters, and the d value of large injection molding machines is 1.0 meters. The coordinate system scale can be automatically adjusted by simply entering the corresponding d value, which improves the versatility of the method. The setting of symmetrical coordinates makes the torque fluctuations of the two axes present symmetrical spectral characteristics in frequency domain analysis. When abnormal spectral components appear on one axis, such as high-frequency vibration caused by bearing wear, the faulty axis can be quickly located by comparing the spectrum with that of the other axis. However, the symmetry of the spectral characteristics under traditional asymmetric coordinates is destroyed, making this type of comparative analysis difficult to achieve.

[0061] Step S130 , constructing a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, and calculating the dynamic position of the virtual load center point;

[0062] See also Figure 2 As shown, further, step S130 includes:

[0063] Step S131, obtaining the current value I of the left servo motor L and the current value of the right servo motor I R ;

[0064] Step S132, according to I L and I R , calculate the real-time torque value T of the left shaft L And the real-time torque value T of the right axis R ;

[0065] Step S133: construct a virtual load center point P in the load monitoring reference coordinate system. virtual , set P virtual The initial position is at the coordinate origin O;

[0066] Step S134, according to T L 、T R And the actual installation distance d from the left and right servo motors to the coordinate origin O, dynamically calculate the virtual load center point P virtual X coordinate position X virtual .

[0067] Specifically, the current value of the left servo motor I L and the current value of the right servo motor I R The current value collected is processed by digital filtering to eliminate high-frequency noise interference. L And the real-time torque value T of the right axis R The calculation of the current-torque conversion constant K motor Implementation, K motor The acquisition of K requires the combination of the motor's technical parameters and actual calibration: First, obtain the rated current and rated torque from the motor's data sheet. motor is the ratio of rated current to rated torque. Based on this, , .

[0068] Constructing a virtual load center point P virtual The initial position is set at the coordinate origin O. This setting is based on the assumption that the dual-axis load is balanced under the ideal state of the system. At this time, T L With T R The load center should coincide with the force system center (origin) to provide a benchmark for subsequent dynamic offset calculations. Virtual load center point P virtual X coordinate position X virtual The calculation formula is ,in, Characterizes the degree of imbalance of the dual-axle torque. A positive difference indicates that the load on the right axle is greater than that on the left axle, while a negative difference indicates the opposite. is a measure of the total load and is used for normalization so that X virtual The range of variation is limited to between -d and d, that is, between the coordinates (-d,0,0) and (d,0,0). The actual installation distance d is directly involved in the calculation of the virtual load center point as a coordinate parameter. The d in the formula strictly corresponds to the actual mechanical installation size to ensure that the calculation result can accurately reflect the physical distance of the load offset. Figure 3 As shown, T R Equal to T L hour =0, P virtual Located at the origin (i.e. initial position); T R Greater than T L hour Greater than 0, P virtual Offset to the right; T L Greater than T R hour Less than 0, P virtualOffset to the left. Because the two axes are mounted symmetrically on the X-axis and there is no additional drive torque on the Y-axis, the load offset only occurs in the X-axis direction. Therefore, there is no need to calculate the Y-axis and Z-axis coordinates. This simplification reduces the amount of calculation and conforms to the physical characteristics of dual-axis load transmission.

[0069] Step S130 forms a synergistic relationship with the load monitoring reference coordinate system constructed in step S110 and the motor position coordinates determined in step S120. The symmetry of the coordinate system enables the X coordinate of the virtual load center point to directly quantify the load offset direction and degree through the torque difference, solving the problem that the load distribution difference cannot be reflected by only position and speed signals in traditional control. The construction of the virtual load center point converts the abstract dual-axis torque difference into a concrete spatial coordinate offset, making it easier to intuitively judge the degree of load imbalance through geometric position. For example, when the X virtual When the absolute value of is close to d, it indicates that the single axis bears most of the load and needs to be compensated immediately. The dynamic trajectory of the virtual load center point can be used to predict potential failures of the mechanical system. virtual The offset in the same injection stage gradually increases and shows regular changes. For example, each injection deflects more in the same direction, which may reflect uneven screw wear or continuous deterioration of mold cavity pressure distribution, providing data support for preventive maintenance.

[0070] Step S140, based on the dynamic position of the virtual load center point, a load mirror mapping relationship matrix is ​​established to identify the load imbalance state;

[0071] Furthermore, step S140 includes:

[0072] Step S141, establishing a load mirror mapping relationship matrix according to the dynamic position of the virtual load center point;

[0073] Furthermore, step S141 includes:

[0074] Step S1411: Load mirror mapping matrix M mirror is a 2×2 matrix;

[0075] Step S1412, matrix M mirror The first row and first column element M[0,0] is set to T R Divide by T L With T R The sum represents the basic weight of the right axle load on the left axle compensation;

[0076] Step S1413: Set the first row and second column element M[0,1] to X virtual The absolute value of divided by d represents the correction coefficient of the load offset to the right axis compensation;

[0077] Step S1414, set the second row and first column element M[1,0] to T L Divide by T L With T R The sum represents the basic weight of the left axle load on the right axle compensation;

[0078] Step S1415, the second row and second column element M[1,1] is set to 1-|X virtual | / d;

[0079] Step S142, set the offset threshold, when X virtual When the absolute value of is greater than the offset threshold, it is determined that the system is in a load imbalance state.

[0080] Specifically, see Figure 4 As shown, the load mirror mapping relationship matrix M mirror A 2×2 structure is used, and the dimension setting is directly matched with the interaction characteristics of the dual-axis system. The first row corresponds to the load mapping relationship of the left axis to the right axis, and the second row corresponds to the load mapping relationship of the right axis to the left axis. The 2×2 dimension is chosen because the load influence in the dual-axis system is mainly reflected in the interaction between the axes. No higher-dimensional matrix is ​​required to fully represent this bidirectional influence, which simplifies the computational complexity and avoids information interference caused by redundant dimensions. The setting of the matrix elements strictly follows the physical meaning and system characteristics: the first row and first column element M[0,0] is T R / (T L +T R ), which represents the proportion of the right-axle torque in the total torque. Its physical meaning is that the greater the right-axle load, the higher the basic weight of the left-axle compensation. For example, when T R 60N・m, T L When the load is 40N·m, M[0,0] is 0.6, indicating that the right axis load accounts for 60% of the total load, and the compensation of the left axis to the right axis must be based on this ratio; the first row and second column element M[0,1] is |X virtual | / d, which reflects the virtual load center offset (i.e. |X virtual |) to the distance between the two axes, quantifies the correction requirement of the load offset on the right axis compensation. The larger the offset, the higher the correction coefficient, ensuring that the compensation can adapt to the dynamic changes of the load center; the second row and first column element M[1,0] is T L / (T L +T R ), symmetrical with M[0,0], represents the proportion of the left-axis torque in the total load and serves as the basic weight for the right-axis compensation of the left-axis. For example, when T L 50N・m, T RWhen the load is 50N·m, M[1,0] is 0.5, indicating that the load on the left and right axes is balanced and the basic weights of mutual compensation are equal; the second row and second column element M[1,1] is 1 minus X virtual The absolute value of divided by d, that is, 1-|X virtual | / d. Based on the principle of load transfer conservation, this value complements M[0,1] to ensure that the sum of the offset correction coefficients for the left axis relative to the right axis and vice versa is 1, avoiding compensation distortion caused by the superposition of correction coefficients. For example, when M[0,1] is 0.2, M[1,1] is 0.8, and the sum of the two is 1, ensuring that the overall impact of offset on dual-axis compensation is balanced. The matrix normalization constraint, that the sum of the elements in each row is 1, is naturally satisfied by the above element settings. This constraint ensures a logically self-consistent distribution of load influence weights. The offset threshold is determined by continuously monitoring the system's synchronization error and torque fluctuation transmission under different load offset conditions. The maximum offset at which the synchronization error does not exceed the set accuracy requirements and the torque fluctuation does not cause system stability degradation is recorded. The ratio of this offset to the dual-axis spacing d is used as the offset threshold.

[0081] The load mirror mapping relationship matrix converts the abstract load offset and torque difference into weight parameters that can be directly used for compensation calculation, solving the problem of blind compensation caused by the inability to quantify the load influence in traditional synchronous control. The streamlined structure of the 2×2 matrix retains the core information of the mutual influence of the two axes, has low computational complexity, and can complete the calculation within a single control cycle to meet real-time control requirements. The physical meaning of the matrix elements is clear and follows the normalization constraint, ensuring that the compensation weight distribution reflects both the basic influence of the load ratio and the dynamic correction of the load offset, such as T R Much larger than T L When M[0,0] increases, the basic compensation weight increases, and X virtual The rightward shift causes M[0,1] to increase, further strengthening the correction compensation. The two work together to precisely match the compensation amount to the actual load state. Changes in matrix elements can indirectly reflect the system's mechanical state. For example, a long-term and gradually increasing deviation of the ratio of M[0,0] to M[1,0] from 1 may indicate a decrease in the right-shaft mechanical transmission efficiency (e.g., belt wear leading to reduced torque output at the same current). This provides a quantitative indicator for preventive maintenance of the mechanical system and expands the system's condition monitoring capabilities. Without step S140, the subsequent differentiated mirror load compensation would lack a quantitative basis, potentially resulting in insufficient compensation (failure to suppress torque fluctuations) or overcompensation (inducing new synchronization errors), causing the compensation control in step S150 to lose precision.

[0082] Step S150: When it is determined that the load is in an unbalanced state, based on the load mirror mapping relationship matrix M mirror ,implementing differentiated mirror load compensation control;

[0083] Furthermore, step S150 includes:

[0084] Step S151, according to T L and T R , identify lightly loaded and heavily loaded axles;

[0085] Step S152: for the identified lightly loaded axis, use the load mirror mapping relationship matrix M mirror , calculate the mirror load value T that needs to be compensated for the lightly loaded axis mirror ;

[0086] Furthermore, step S152 includes:

[0087] Step S1521, constructing a torque difference vector V diff =[T heavy ,T light ], where T heavy is the torque value of the heavily loaded shaft, T light is the torque value for a lightly loaded shaft;

[0088] Step S1522: If the lightly loaded shaft is the right shaft, then T mirror Equal to M mirror The first row with V diff The dot product of

[0089] Step S1523: If the lightly loaded shaft is the left shaft, then T mirror Equal to M mirror The second line with V diff The dot product of

[0090] Step S153: calculate the mirror load value T that needs to be compensated for the light-load axis. mirror , implement differentiated mirror load compensation control.

[0091] Specifically, step S151 is based on the left shaft real-time torque value T L And the right axis real-time torque value T R Identify lightly loaded and heavily loaded shafts. The decision logic is based on direct comparison of the torques of the two shafts: When T L Greater than T R When , it indicates that the left shaft bears a heavier load, so the left shaft is the heavy load shaft (denoted as T heavy =T L ), the right axis is the light load axis (denoted as T light =T R ); When T R Greater than T L When the right axis is the heavy load axis (T heavy =T R ), the left axis is the light load axis (T light =T LThis determination is directly related to the actual state of load distribution and clarifies the target for subsequent compensation. By increasing the load capacity of the lightly loaded axle, load balancing is achieved on both axes, preventing increased torque fluctuations on the heavily loaded axle due to continuous overload.

[0092] Step S152 calculates the mirror load value T that needs to be compensated for the light-load axis. mirror The specific process is as follows: First, construct the torque difference vector V diff , the vector is defined as [T heavy ,T light ], where the first element is the torque value of the heavy-loaded axis corresponding to the light-loaded axis, and the second element is the torque value of the light-loaded axis itself. The construction of the vector is to use the torque difference of the two axes as the input of the matrix operation in a structured form to ensure the correspondence between the operation logic and the load state. When the light-loaded axis is the right axis, T mirror Through the load mirror mapping relationship matrix M mirror The first row with V diff The dot product is calculated and the expansion is T mirror =M[0,0]×T heavy +M[0,1]×(T heavy -T light ), where M[0,0]×T heavy is the basic compensation term based on the right axle load weight, reflecting the influence of the proportion of the heavy load axle torque in the total load on the compensation, M[0,1]×(T heavy -T light ) is a correction compensation item based on load offset, which reflects the dynamic adjustment of the compensation amount by the virtual load center offset; when the light load axis is the left axis, T mirror By combining the second row of the matrix with V diff The dot product calculation is T mirror =M[1,0]×T heavy +M[1,1]×(T heavy -T light ), which is logically symmetrical with the right axis compensation, ensuring that the left axis compensation also takes into account the basic load weight and offset correction.

[0093] Step S153 implements differentiated mirror load compensation control based on the calculated T mirror , compensation is achieved by adjusting the current loop proportional gain of the light-load axis servo driver: the current loop proportional gain is positively correlated with the motor output torque. When the gain is increased, the response speed of the torque output is faster under the same current change. By adjusting the gain adjustment amount and T mirror Binding, so that the light load shaft maintains the original speed command while increasing the output torque T mirror The corresponding value dynamically improves its load capacity.

[0094] Step S150 forms a deep synergy with the load mirror mapping relationship matrix constructed in step S140. The basic weight and correction coefficient provided by the load mirror mapping relationship matrix make T mirror The calculation considers both the inherent ratio of the dual-axis load and the dynamic offset of the load center, solving the problem of insufficient adaptability caused by traditional compensation methods based solely on linear compensation based on torque difference. Accurate identification of the lightly loaded axis ensures that compensation resources are concentrated on the axis with weaker load capacity, avoiding excessive intervention on the heavily loaded axis, balancing the dual-axis load distribution while reducing energy loss. The calculation method of the torque difference vector and the matrix dot product transforms complex load interactions into quantifiable compensation values, making the calculated compensation amount have clear physical meaning and ensuring the consistency of the compensation strategy under different load conditions.

[0095] Differentiated mirror load compensation control optimizes the dynamic response characteristics of the two axes during implementation: when the current loop proportional gain of the light-load axis increases, its torque output bandwidth increases, and its response speed to load changes accelerates. This improvement in dynamic characteristics will inversely suppress the torque fluctuation amplitude of the heavy-load axis, forming a coordinated optimization of the dynamic performance of the two axes, rather than a simple single-axis compensation. In addition, T mirror During the calculation process, the product of the matrix elements and the torque difference essentially implements feedforward compensation for load disturbances. By predicting the impact of load differences on the system in advance and applying reverse compensation, the dual axes begin adjusting at the initial stage of load changes, reducing the accumulation of synchronization errors. This goes beyond the scope of simple feedback compensation and significantly improves the system's anti-interference capability. If step S150 is missing, the load imbalance identified in step S140 will not be converted into specific control actions, and the problem of uneven load distribution will persist, leading to increased torque fluctuations, reduced synchronization accuracy, and even system oscillations. Step S150 ensures the closed-loop integrity of the entire solution, from monitoring to execution.

[0096] In step S200, based on differentiated mirror load compensation control, the comprehensive stability index of the left and right axes is calculated; based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis; based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

[0097] Step S200 builds on the differentiated mirror load compensation control of step S100. Through quantitative assessment and dynamic optimization of the compensated system state, it achieves a profound improvement in dual-axis load balancing and synchronization accuracy. In step S100, the current loop proportional gain of the lightly loaded axis is adjusted to achieve balanced torque distribution between the two axes. This process alters the actual position, speed, and torque dynamics of the two axes. After compensation, the position deviation between the left and right axes is reduced, the speed fluctuation amplitude is reduced, and the torque difference is minimized. This compensated real-time data (position, speed, and torque) becomes the core input for step S200. Step S100 achieves a preliminary balance of loads on both axes through mirror load compensation, resolving the issue of significantly uneven load distribution. However, due to the limitations of single-axis compensation (which only adjusts the lightly loaded axis), there may still be differences in stability between the two axes after compensation. For example, due to the greater mechanical inertia of one axis, the speed fluctuation of the other axis after compensation may still be greater. Step S200 is precisely aimed at this residual stability difference after compensation. By calculating the comprehensive stability index, the more stable axis is dynamically selected as the reference axis, so that the following axis can adjust the motion trajectory with a better reference, further reducing the synchronization error; at the same time, dynamic diversion control is implemented based on the torque state of the reference axis and the following axis. On the basis of the single-axis compensation in step S100, dual-axis coordinated adjustment is achieved, so that the load distribution moves from preliminary balance to dynamic and precise balance.

[0098] Furthermore, step S200 includes:

[0099] Step S210, calculating the comprehensive stability index of the left and right axes;

[0100] Furthermore, step S210 includes:

[0101] Step S211, calculate the position stability index S of the left and right axes according to the position coordinates of the left and right servo motors position,L and S position,R And the speed stability index S of the left and right axes velocity,L and S velocity,R ;

[0102] Step S212: Obtain the historical torque value sequence of the left axis and the historical torque value sequence of the right axis for m consecutive control cycles, and calculate the load change slope k of the left axis. L and the load change slope k of the right axis R ;

[0103] Step S213, according to k L and k R , calculate the torque stability index S of the left and right axes torque,L and S torque,R ;

[0104] Step S214: Based on the position stability index, speed stability index and torque stability index of the left and right axes, the comprehensive stability index S of the left and right axes is obtained. total,L and S total,R .

[0105] Specifically, step S210 is used to calculate the comprehensive stability index of the left and right axes, and the specific implementation process is as follows: Step S211 calculates the position stability index S position,L 、S position,R and speed stability index S velocity,L 、S velocity,R , where S position,L is the position stability index of the left axis, S position,R is the position stability index of the right axis, S velocity,L is the speed stability index of the left axis, S velocity,R is the speed stability index of the right axis. The calculation of the position stability index is based on the position deviation data in the last r control cycles. The control cycle is the minimum repetition time interval of the whole process of "acquisition-calculation-control" including the differential mirror load compensation control in step S100. The value of r needs to balance the timeliness of data and statistical reliability. It is usually determined to be 5 to 20 cycles based on the system response speed. In the specific calculation, first record the deviation between the actual position and the target position of the left axis in each control cycle (denoted as e L1 ,e L2 ,...,e Lr ) and the corresponding deviation on the right axis (e R1 ,e R2 ,...,e Rr ), where e Lr is the deviation between the actual position and the target position of the left axis in the rth control cycle, e Rr is the deviation between the actual position of the right axis and the target position in the rth control cycle, and then the standard deviation of the left axis deviation is calculated to obtain S position,L ; The calculation logic of the right axis is the same, and S is obtained position,R The smaller the standard deviation, the more concentrated the position deviation is and the higher the position stability of the axis is.

[0106] The calculation of the speed stability index is based on the derivation of the speed signal based on the position data: the position difference between two consecutive control cycles is divided by the control cycle time interval Δt (i.e., the control cycle duration) to obtain the instantaneous speed of each cycle; then the speed signal is subjected to spectrum analysis using fast Fourier transform to calculate the energy proportion of different frequency components and define f low It is the critical value below the mechanical resonance frequency of the system, for example 10Hz, determined by the natural frequency test of the transmission mechanism. When the frequency in the speed signal is lower than f lowWhen the energy proportion of the left axis exceeds 70%, a higher speed stability index value is assigned. The higher the energy proportion, the greater the index value. A high proportion of low-frequency energy indicates a stable speed change. A high proportion of high-frequency components may indicate speed jitter caused by mechanical vibration or torque fluctuation transmission. For example, if the energy below 10Hz in the left axis speed spectrum accounts for 85% and that of the right axis is 60%, then S velocity,L Higher than S velocity,R .

[0107] Step S212 calculates the left axle load change slope k L and the right axis load change slope k R : Get the left axis historical torque value sequence of m consecutive control cycles (T L1 ,T L2 ,...,T Lm ) and the right axis historical torque value series (T R1 ,T R2 ,...,T Rm ), where T Lm is the historical torque value of the left axis in the mth control cycle, T Rm is the historical torque value of the right axis in the mth control cycle. The value of m needs to reflect the torque change trend. For example, in 5 to 10 cycles, the slope of the left axis load change k L The difference between the torque values ​​of two adjacent cycles is divided by the control cycle time interval Δt, k R Similarly, the smaller the absolute value of the slope, the smoother the torque change. Step S213 is based on k L and k R Calculate the left shaft torque stability index S torque,L and right shaft torque stability index S torque,R : The torque stability index value is inversely proportional to the absolute value of the slope, which can be expressed as S torque,L =1 / (1+|k L |) is normalized, and the same is true for the right axis. The smaller the slope, the closer the index value is to 1, indicating higher torque stability. A slope threshold can also be set, such as 20N·m / s. When the absolute value of the load change slope is less than the slope threshold, a higher score is given, otherwise the score is reduced. For example, k L =8N・m / s is less than the slope threshold, S torque,L =0.8; right axis k R =25N・m / s is greater than the slope threshold, S torque,R =0.4.

[0108] Step S214 obtains the comprehensive stability index S of the left axis by weighted summation. total,L And the comprehensive stability index S of the right axis total,R: The weight distribution is based on the degree of influence of each indicator on the system synchronization performance and is determined by fitting experimental data. For example, position stability has the greatest impact on synchronization accuracy, with a weight of 0.4, speed stability has the second greatest impact, with a weight of 0.3, and torque stability has a weight of 0.3. In precision injection molding scenarios, the position weight can be increased to 0.5.

[0109] The differentiated mirror load compensation control in step S210 forms a deep synergy with step S150. The compensated position, speed, and torque data serve as the basis for calculating the stability index, ensuring that the index truly reflects the system state after compensation. After compensation, the load on both axes is more balanced, and position deviation, speed fluctuation, and torque variation are all reduced. The calculated stability index better reflects the actual stability of the system, avoiding misjudgments caused by data distortion in the uncompensated state. The construction of multi-dimensional indicators breaks through the limitations of traditional single-position synchronization error assessment. Position stability reflects motion accuracy, speed stability reflects the smoothness of dynamic response, and torque stability reflects the gentleness of load changes. The combination of these three provides a comprehensive portrayal of system stability. For example, if an axis has a small position synchronization error but severe torque fluctuations, a single indicator may misjudge its stability, while a comprehensive index can identify potential risks. The introduction of load change slope expands torque stability assessment from static deviation to dynamic trend, capturing the precursors of load mutations in advance. For example, a sudden increase in slope may indicate a sudden increase in melt resistance, providing a predictive basis for subsequent reference axis switching. The comprehensive stability index provides a quantitative standard for the selection of the reference axis, ensuring that the reference axis is always the axis with better performance in multiple dimensions in the current system, rather than the best in a single dimension. For example, if one axis has more stable position but large torque fluctuations, while another axis has the opposite, the comprehensive index can balance the selection of the more suitable reference axis.

[0110] The changing trends of various stability indicators are correlated and can be used to diagnose potential mechanical failures in the system: a sudden increase in the high-frequency component of the speed stability indicator, coupled with a simultaneous increase in the torque variation slope, may indicate drive belt slippage or bearing wear. Without this step, the reference axis switching in step S220 lacks a scientific and quantitative basis and may be determined solely based on subjective experience or a single indicator, leading to irrational reference axis selection and increased synchronization error. Furthermore, the system's stable state differences in position, speed, and torque cannot be fully identified, making it difficult to achieve dynamic and balanced load distribution. This can lead to a loss of precision in the diversion control in step S230, ultimately impacting the continuous optimization of dual-axis synchronization performance.

[0111] Step S220 , based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis;

[0112] Furthermore, step S220 includes:

[0113] Step S221, updating the comprehensive stability index of the left and right axes every s control cycles to determine the initial reference axis and the initial following axis;

[0114] Set the stability difference threshold ΔS threshold , when S total,L t consecutive control cycles are greater than S total,R , and S total,L With S total,R The difference is greater than ΔS threshold When , the left axis is set as the initial reference axis and the right axis is set as the initial following axis;

[0115] When S total,R t consecutive control cycles are greater than S total,L , and S total,R With S total,L The difference is greater than ΔS threshold When the right axis is set as the initial reference axis, the left axis is set as the initial following axis.

[0116] Step S222: Based on the determined initial reference axis and initial following axis, a gradual switching is performed to obtain the final reference axis and following axis;

[0117] Step S223: After the switching is completed, the synchronization errors before and after the switching are compared. If the synchronization error increases, the state before the switching is automatically restored.

[0118] Specifically, step S220 uses a progressive switching strategy to determine the reference axis and the following axis based on the comprehensive stability index of the left and right axes. The specific implementation process is as follows: Step S221 updates the comprehensive stability index of the left and right axes every s control cycles and determines the initial reference axis and the initial following axis. The value of s needs to be combined with the dynamic response characteristics of the system and the computing resource limitations. It is usually set to 5 to 10 control cycles. This range ensures the real-time nature of the index update, avoids the reference axis lagging behind the actual stability changes due to slow updates, and does not increase the computing burden due to too frequent updates. Stability difference threshold ΔS threshold The determination is based on the quantitative relationship between the allowable range of synchronization error and the stability index. The synchronization error changes under different difference values ​​are tested experimentally. For example, let ΔS threshold is 0.1, the comprehensive stability index of the left axis is 0.8, and that of the right axis is 0.65, and the difference of 0.15 is greater than ΔS threshold , the switching condition is met. The t value of t consecutive control cycles is set to 3 to 5. Through multiple continuous judgments to eliminate instantaneous fluctuation interference, the left axis is determined to be the initial reference axis and the right axis is determined to be the initial following axis.

[0119] Step S222 performs a gradual switching of the initial reference axis and initial following axis to obtain the final reference axis and following axis. Gradual switching involves gradually increasing the control gain of the initial reference axis by v%, while gradually decreasing the control gain of the initial following axis by v%. The total adjustment, v%, is dynamically determined based on the current synchronization error. When the synchronization error is large, v is set to a larger value, such as 5%. When the synchronization error is small, v is set to a smaller value, such as 2%, to avoid excessive adjustment amplitudes that may cause system oscillation. The total adjustment, v%, is evenly distributed over p control cycles, with p typically ranging from 3 to 5. This value matches the inertia of the mechanical system, with p taking a larger value when the inertia is large. Within each control cycle, the control gain of the initial reference axis increases by v% / p, while the control gain of the initial following axis decreases by v% / p. For example, when v = 5% and p = 5, the control gain of the reference axis increases by 1% and the control gain of the following axis decreases by 1% within each control cycle. The total adjustment is completed after five cycles, ensuring smooth and seamless gain changes. The adjustment direction of the control gain is consistent with the trend of the comprehensive stability index. The reference axis has higher stability, and increasing the gain can enhance its anti-interference ability; the following axis has lower stability, and reducing the gain can reduce its disturbance transmission to the reference axis.

[0120] After the switch is completed, step S223 compares the synchronization errors before and after the switch. If the synchronization error increases, it will automatically restore to the state before the switch. The synchronization error is the absolute value of the difference between the actual position of the left axis and the actual position of the right axis. The error comparison before and after the switch needs to be performed in the same injection stage, such as the melt filling stage, to ensure that the comparison conditions are consistent. The recovery operation includes restoring the setting relationship between the reference axis and the follower axis, as well as the corresponding control gain parameters. For example, if the left axis is the reference axis before the switch, it is reset to the reference axis and the gain value stored before the switch is called to avoid continuous instability of the system due to improper switching. If the synchronization error increases after the switch, the reference axis setting and gain value before the switch are immediately restored to bring the error back to the original level.

[0121] The progressive switching strategy adjusts control gains in stages, avoiding system oscillations caused by sudden gain changes. In large injection molding machines, the mechanical transmission system has significant inertia. A single, large gain adjustment (e.g., 5%) can increase torque fluctuations. However, adjusting over multiple cycles (e.g., five) can reduce fluctuations, balancing switching efficiency and system stability. A post-switching synchronization error verification mechanism enhances the system's fault tolerance. When external disturbances, such as sudden changes in melt pressure, cause poor switching performance, an automatic recovery function quickly curbs synchronization error expansion, ensuring continuous injection. Compared to existing technologies, the gain adjustments of the reference and following axes create complementary damping characteristics. Increasing the reference axis's gain enhances its ability to suppress load disturbances, while decreasing the following axis's gain reduces its reaction force transmission to the reference axis. These two factors synergistically reduce the risk of resonance in the dual-axis mechanical transmission system. The gradual gain changes in progressive switching produce a low-frequency modulation effect that filters out high-frequency components in torque fluctuations, smoothing dual-axis torque transmission and indirectly extending the service life of transmission components such as belts and gears. Without this step, the relationship between the reference axis and the following axis remains fixed. If the original reference axis loses stability due to mechanical wear or load changes, the following axis will continue to use the unstable axis as its reference, leading to cumulative synchronization errors. Furthermore, the lack of progressive gain adjustment can cause sudden changes that can cause the system to frequently enter an oscillatory state, exacerbating torque fluctuations and destabilizing the dynamic shunt control in step S230, ultimately leading to a re-imbalance in the dual-axis load distribution. Therefore, step S220 ensures that the system always maintains dual-axis synchronization and load balancing in the most stable state by dynamically optimizing reference axis selection and smoothly switching.

[0122] Step S230: Based on the determined reference axis and follower axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

[0123] Furthermore, step S230 includes:

[0124] Step S231, constructing a virtual load diversion channel and calculating the load transfer efficiency;

[0125] Step S232: Obtaining the torque adjustment required for the reference axis and the following axis based on the load transfer efficiency;

[0126] Furthermore, step S232 includes:

[0127] Step S2321, obtain the real-time torque value T of the reference shaft base and the real-time torque value T of the following axis follow ;

[0128] Step S2322: Calculate the total torque value T of the reference axis and the following axis. total , set the target torque values ​​of the reference axis and the following axis to Ttotal Divide by 2;

[0129] Step S2323, based on the load transfer efficiency, correct the target torque values ​​of the reference axis and the following axis to obtain a corrected target torque value;

[0130] Step S2324, according to T base 、T follow The torque adjustment required for the reference axis and the following axis is calculated based on the corrected target torque value.

[0131] Step S233 : Implement dynamic flow splitting control according to the torque adjustment amounts required by the reference axis and the following axis respectively.

[0132] Specifically, step S230 constructs a virtual load diversion channel based on the determined reference axis and follower axis and implements dynamic diversion control. The specific implementation process is as follows: Step S231 constructs a virtual load diversion channel. With the reference axis as the reference point, a virtual load diversion channel is established in the load monitoring reference coordinate system, connecting the reference axis position coordinates and the follower axis position coordinates. The spatial path of the channel is strictly consistent with the dual-axis mechanical transmission link, covering the actual force transmission path such as the gear meshing path and the belt drive direction, ensuring that the load diversion direction is completely matched with the actual force transmission direction in the mechanical system, making the virtual channel not only a geometric connection line, but also a control carrier that reflects the actual load transfer characteristics. Load transfer efficiency α transfer The calculation of is integrated into this channel construction process, and its value is based on the virtual load center point P virtual The X coordinate position Xvirtual is determined by This calculation directly relates the degree of load offset to the mechanical transmission efficiency. When the load is symmetrically distributed, X virtual is 0, α transfer When the load is offset to a certain axis, X virtual The absolute value of α increases, transfer Reduce accordingly. For example, X virtual When α is 0.15d, transfer It is 0.85, reflecting a 15% loss in transfer efficiency due to load offset. This quantitative relationship enables the virtual channel to perceive the efficiency degradation in load transfer.

[0133] Step S232 calculates the torque adjustment required for the reference axis and the following axis. First, the real-time torque value T of the reference axis is obtained. base and the real-time torque value T of the following axis follow These torque values ​​are converted by the servo motor's current-torque conversion constant K motor The same method as in step S132 is used to obtain the total torque value T total Tbase With T follow The ideal target torque values ​​of the reference axis and the following axis are set to T total One half of the load transfer efficiency α transfer Correct the target torque value: The target torque value after the reference axis correction is This is because the reference axis is the main path for load transmission, and its output torque needs to be transmitted through the virtual channel. transfer The higher the value, the closer the actual torque transmitted to the load is to the ideal value. Otherwise, the target torque needs to be reduced to avoid actual load overload due to transmission loss. The corrected target torque value of the following axis is Since the following axis needs to compensate for the insufficient transmission efficiency of the reference axis, when α transfer When lowered, Increase to ensure that the total torque actually transmitted to the load by the dual shafts is still T total And the distribution is balanced, and the sum of the corrected two-axis target torque values ​​is equal to T total , satisfying load conservation. The calculation of torque adjustment is the difference between the corrected target torque and the real-time torque: the adjustment amount of the reference axis for , the adjustment amount of the following axis for .

[0134] Step S233 implements dynamic shunt control according to the torque adjustment amount, and adjusts the current gain parameters of the reference axis and the following axis servo driver based on the path characteristics of the virtual load shunt channel: when adjusting the reference axis current gain, the gain correction coefficient is proportional to α transfer Positive correlation, α transfer The closer it is to 1 (high channel efficiency), the smaller the gain adjustment range is, and there is no need for excessive adjustment due to stable transmission; α transfer The smaller the value (low channel efficiency), the greater the gain adjustment range, and the need to strengthen the stability of the reference axis output to offset the transmission loss. When adjusting the follower axis current gain, the gain correction coefficient is ( ) is positively correlated, ( ) is larger, the gain adjustment range is larger, ensuring that its torque output can accurately match the transfer characteristics of the reference shaft. Real-time monitoring of the actual transfer effect of the virtual load diversion channel is carried out by comparing the theoretical load value (T base ×α transfer ) and the actual load value of the following axis output torque (T follow × (1-α transfer )), set the threshold value T of the difference between the two th , for example, let T total 5% of T, when the difference is lower than T for three consecutive control cycles th, determine whether the shunt control matches the virtual channel characteristics, and gradually reduce the adjustment amplitude until the system is stable.

[0135] Compared with the differentiated mirror load compensation control of step S150, step S230 achieves a deep optimization from single-axis compensation to dual-axis collaborative diversion. Step S150 only adjusts the torque of the light-load axis, while step S230 adjusts the reference axis and the following axis at the same time. Through bidirectional torque adjustment, the load distribution reaches a balanced state faster, and the response speed is significantly improved. The gradual reduction mechanism of the dynamic diversion control intensity avoids the interference of continuous control on the dynamic characteristics of the system. When the torque difference stabilizes, the high-intensity control is exited, allowing the system to return to a natural stable state and reduce energy loss. If this step is missing, the reference axis and the following axis determined in step S220 will lose their specific load adjustment strategy, and the two axes may still maintain the residual torque difference after compensation, resulting in the continuous transmission of torque fluctuations. At the same time, the control intensity cannot be dynamically adjusted according to the load transfer efficiency, and new synchronization errors may be caused by excessive adjustment, making it difficult to fully achieve the load balancing goal of the entire solution. Therefore, step S230 guides the diversion direction and dynamically adjusts the torque distribution through the virtual channel to ensure that the dual-axis system transitions from the stability optimization stage to the coordinated optimization stage of load balancing and stability, and ultimately achieves the improvement of the injection dual-axis synchronization performance and service life.

[0136] Example 2

[0137] This embodiment provides a multi-axis servo synchronous drive control system for an injection molding machine based on the embodiment 1. Figure 5 Shown, including:

[0138] Coordinate system establishment module: used to obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors, and establish the load monitoring reference coordinate system;

[0139] Mirror load compensation module: used to obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, calculate the dynamic position of the virtual load center point, and implement differentiated mirror load compensation control;

[0140] Index calculation module: Calculates the comprehensive stability index of the left and right axes based on differentiated mirror load compensation control;

[0141] Diversion control module: Based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis; based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

[0142] Example 3

[0143] This embodiment provides an injection molding machine, which applies the above-mentioned multi-axis servo synchronous drive control system to achieve high-precision synchronous drive of the two injection axes.

[0144] The structure of the injection molding machine:

[0145] Injection mechanism: includes a screw, barrel, and melt channel. The screw is arranged along a preset axis (Z axis), which is the core path for melt propulsion and is used to propel the melt. The melt channel is coaxial with the screw axis and is used to guide the melt from the barrel to the mold cavity.

[0146] Two servo motors, symmetrically mounted on either side of the screw, drive the screw. The motor centerline (X-axis) intersects the screw axis (Z-axis) at right angles. The intersection serves as the origin of the load monitoring reference coordinate system. The motors are equidistant from the origin, ensuring a mirror-image distribution on the X-axis.

[0147] Injection molding machine multi-axis servo synchronous drive control system: integrated in the injection molding machine control cabinet, including a coordinate system establishment module, a mirror load compensation module, an index calculation module and a shunt control module; the system is electrically connected to the left and right servo motors, and is used to control the left and right servo motors to achieve multi-axis servo synchronous drive.

[0148] Working process of injection molding machine:

[0149] When the injection molding machine performs the injection action, the multi-axis servo synchronous drive control system of the injection molding machine works according to the following process:

[0150] Coordinate system initialization: The coordinate system establishment module automatically collects the installation position parameters of the screw axis and the left and right servo motors, and establishes a load monitoring reference coordinate system with the intersection of the screw axis and the center line of the two motors as the origin and the Z axis pointing to the direction of melt advancement.

[0151] Load monitoring and compensation: The mirror load compensation module collects the current values ​​of the left and right servo motors in real time and calculates the real-time torque of the two axes based on the current values. Based on the torque values ​​and the distances from the two motors to the origin, it constructs a virtual load center point and dynamically calculates its X-coordinate position to determine the direction of load offset. By establishing a load mirror mapping relationship matrix, it identifies the lightly loaded and heavily loaded axes, calculates the mirror load value to be compensated for the lightly loaded axis, and implements differentiated mirror load compensation control to balance the loads on the two axes.

[0152] Stability assessment and reference axis switching: The index calculation module calculates the comprehensive stability index of the left and right axes from multiple dimensions, including position stability, velocity stability, and torque stability, based on the compensated state. The index is updated at regular control intervals. Combined with the stability difference threshold, a progressive switching strategy is used to determine the reference and following axes. Smooth switching is achieved by gradually adjusting the control gains of the two axes. The synchronization effect after switching is verified to ensure system stability.

[0153] Dynamic shunt control: The shunt control module constructs a virtual load shunt channel connecting the determined reference axis and follower axis, and calculates the load transfer efficiency. It sets the target torque of the two axes based on the total torque value and corrects the target torque based on the load transfer efficiency. By adjusting the current gain of the two-axis servo drive, the torque is dynamically adjusted to ensure continuous and balanced load distribution on the two axes, thereby ensuring the synchronization accuracy of the injection process.

[0154] By integrating the above-mentioned multi-axis servo synchronous drive control system, this injection molding machine solves the problems of large synchronization errors and rapid mechanical wear caused by uneven load distribution in traditional dual-axis drive injection molding machines, improves the stability and precision of the injection molding process, and extends the service life of transmission components.

[0155] The methods and systems of the present application may be implemented in many ways. For example, the methods and systems of the present application may be implemented using software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps used in the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified.

[0156] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0157] The above-described specific embodiments further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above description is merely a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A multi-axis servo synchronous drive control method for an injection molding machine, characterized in that: The method comprises: Obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors to establish a load monitoring reference coordinate system; Obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, calculate the dynamic position of the virtual load center point, and implement differentiated mirror load compensation control; The method for implementing differentiated mirror load compensation control includes: establishing a load mirror mapping relationship matrix based on the dynamic position of the virtual load center point to identify the load imbalance state; when it is determined to be in the load imbalance state, based on the load mirror mapping relationship matrix M mirror ,implementing differentiated mirror load compensation control; The method for identifying the load imbalance state is as follows: setting an offset threshold value, when the X coordinate position of the virtual load center point is X virtual When the absolute value of is greater than the offset threshold, the system is determined to be in a load imbalance state; The implementation of the differential mirror load compensation control further includes: according to the real-time torque value T of the left shaft L And the real-time torque value T of the right axis R , identify the light-load axis and the heavy-load axis; for the identified light-load axis, use the load mirror mapping relationship matrix M mirror , calculate the mirror load value T that needs to be compensated for the lightly loaded axis mirror ; Based on the calculated mirror load value T that needs to be compensated for the lightly loaded axis mirror , implement differential mirror load compensation control; the real-time torque value T of the left shaft L And the real-time torque value T of the right axis R By obtaining the current value I of the left servo motor L and the current value of the right servo motor I R , according to I L and I R Calculated; Based on differentiated mirror load compensation control, the comprehensive stability index of the left and right axes is calculated; Based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis. Based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented. The method for determining the reference axis and the following axis using a progressive switching strategy includes: updating the comprehensive stability index of the left and right axes every s control cycles to determine the initial reference axis and the initial following axis; and performing progressive switching based on the determined initial reference axis and initial following axis to obtain the final reference axis and following axis.

2. The multi-axis servo synchronous drive control method for an injection molding machine according to claim 1, characterized in that: The load monitoring reference coordinate system uses the intersection of the screw axis and the line connecting the centers of the two servo motors as the coordinate origin O, the direction of the screw axis is set as the positive direction of the Z axis, pointing to the direction of melt advancement; the line connecting the center points of the left servo motor and the right servo motor is set as the X axis.

3. The multi-axis servo synchronous drive control method for an injection molding machine according to claim 2, characterized in that: The position coordinates of the left and right servo motors in the load monitoring reference coordinate system are determined by the actual installation distance d from the left and right servo motors to the coordinate origin O. The position coordinates of the left servo motor are (-d, 0, 0) and the position coordinates of the right servo motor are (d, 0, 0).

4. The multi-axis servo synchronous drive control method for an injection molding machine according to claim 3, characterized in that: The method of constructing a virtual load center point and calculating the dynamic position of the virtual load center point includes: Construct a virtual load center point P in the load monitoring reference coordinate system virtual , set P virtual The initial position is at the coordinate origin O; According to T L 、T R And the actual installation distance d from the left and right servo motors to the coordinate origin O, dynamically calculate the virtual load center point P virtual X coordinate position X virtual .

5. A multi-axis servo synchronous drive control system for an injection molding machine, which is used to implement the multi-axis servo synchronous drive control method for an injection molding machine according to any one of claims 1 to 4, characterized in that: The system comprises: Coordinate system establishment module: used to obtain the screw axis position of the injection mechanism and the installation position parameters of the left and right servo motors, and establish the load monitoring reference coordinate system; Mirror load compensation module: used to obtain the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, construct a virtual load center point based on the load monitoring reference coordinate system and the position coordinates of the left and right servo motors in the load monitoring reference coordinate system, calculate the dynamic position of the virtual load center point, and implement differentiated mirror load compensation control; Index calculation module: Calculates the comprehensive stability index of the left and right axes based on differentiated mirror load compensation control; Diversion control module: Based on the comprehensive stability index of the left and right axes, a progressive switching strategy is adopted to determine the reference axis and the following axis; based on the determined reference axis and the following axis, a virtual load diversion channel is constructed and dynamic diversion control is implemented.

6. An injection molding machine, characterized in that: include: An ejection mechanism, wherein the ejection mechanism is provided with a screw; The left and right servo motors are installed on both sides of the screw to drive the screw movement; The multi-axis servo synchronous drive control system for an injection molding machine according to claim 5, wherein the system is electrically connected to the left and right servo motors and is used to control the left and right servo motors to achieve multi-axis servo synchronous drive.

Images