Control method and control system for magnetic steel polishing

Abstract

The invention discloses a control method and a control system for magnetic steel polishing, and relates to the technical field of polishing control, and the control method comprises the following steps: setting a magnetic steel positioning center coordinate and a polishing reference point, and configuring a polishing control path; a preset fixed position and a preset fixed posture are collected, and initial polishing process parameters are set in combination with the polishing control path; setting a magnetic steel polishing virtual environment, collecting pressure distribution simulation data of a polishing contact surface, dynamically correcting the feeding speed and the axial deflection angle of a polishing head, establishing a nonlinear mapping model, and adaptively adjusting the polishing power; and a laser contour scanning plug-in is arranged at the tail end of the polishing path, the simulated curved surface and the actually-measured point cloud data are compared for simulation compensation, and a polishing process parameter set is output to execute batch machining control. The technical problems that in the prior art, the grinding precision is not high, and technological parameters are not flexible to adjust are solved, and the technical effects of improving the grinding precision and achieving intelligent optimization of the technological parameters are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of grinding control, and in particular to a control method and a control system for grinding magnetic steel. Background Art

[0002] With the rapid development of motor manufacturing and precision machining technology, magnetic steel grinding technology has been widely used in many fields, especially in the production of high-performance motors and precision equipment. However, with the continuous improvement of product performance requirements and the increasing complexity of processing technology, the control and management of magnetic steel grinding process are also facing more and more severe challenges. Traditional magnetic steel grinding methods mainly rely on fixed parameter settings and manual experience adjustments, which often have problems such as low grinding accuracy and inflexible process parameter adjustment, which makes it difficult to meet the modern manufacturing industry's high-precision and high-efficiency processing requirements. Summary of the invention

[0003] The present application provides a control method and control system for magnetic steel grinding, which are used to solve the technical problems of low grinding accuracy and inflexible adjustment of process parameters in the prior art.

[0004] In view of the above problems, the present application provides a control method and control system for magnetic steel grinding.

[0005] In a first aspect of the present application, a control method for grinding magnetic steel is provided, the method comprising: Connect the video recognition unit, locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel; collect the preset fixed position and preset fixed posture of the motor rotor magnetic sheet, and set the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint in combination with the grinding control path; set the magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and use the multi-axis force monitoring plug-in to collect the pressure distribution simulation data of the grinding contact surface in real time; based on the pressure distribution simulation data, dynamically correct the grinding head feed speed and axial deflection angle through the preset pressure threshold, and establish a nonlinear mapping model of the magnetic steel material removal rate, spindle speed and grinding depth, and adaptively adjust the grinding power; at the same time, set a laser contour scanning plug-in at the end of the grinding path, compare the simulated surface with the measured point cloud data for simulation compensation, and output the grinding process parameter set to execute batch processing control after meeting the grinding requirements.

[0006] The second aspect of the present application provides a control system for magnetic steel grinding, the system comprising: The grinding path configuration module is used to connect the video recognition unit, locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and the grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel; the process parameter setting module is used to collect the preset fixed position and preset fixed posture of the motor rotor magnetic sheet, and set the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint in combination with the grinding control path; the simulation grinding module is used to set the magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and use multi-axis force The monitoring plug-in collects the pressure distribution simulation data of the grinding contact surface in real time; the adaptive adjustment module is used to dynamically correct the grinding head feed speed and axial deflection angle through a preset pressure threshold based on the pressure distribution simulation data, and establish a nonlinear mapping model between the magnetic steel material removal rate and the spindle speed and grinding depth to adaptively adjust the grinding power; the processing control module is used to simultaneously set a laser contour scanning plug-in at the end of the grinding path, compare the simulated surface with the measured point cloud data for simulation compensation, and output the grinding process parameter set to execute batch processing control after meeting the grinding requirements.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application is connected to a video recognition unit, positions the motor rotor magnetic sheet, sets the magnetic steel positioning center coordinates and the grinding reference point, and configures the grinding control path according to the basic information of the magnetic steel; collects the preset fixed position and the preset fixed posture of the motor rotor magnetic sheet, and sets the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint in combination with the grinding control path; sets the magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and uses a multi-axis force monitoring plug-in to collect the pressure distribution simulation data of the grinding contact surface in real time; based on the pressure distribution simulation data, dynamically corrects the grinding head feed speed and axial deflection angle through a preset pressure threshold, and establishes a nonlinear mapping model of the magnetic steel material removal rate, the spindle speed, and the grinding depth, and adaptively adjusts the grinding power; at the same time, a laser contour scanning plug-in is set at the end of the grinding path to compare the simulated surface with the measured point cloud data for simulation compensation, and after meeting the grinding requirements, outputs the grinding process parameter set to perform batch processing control. The present invention solves the technical problems of low grinding precision and inflexible adjustment of process parameters in the prior art, and achieves the technical effect of improving grinding precision and realizing intelligent optimization of process parameters through video recognition and positioning, virtual grinding environment simulation, multi-axis force monitoring and dynamic parameter adaptive adjustment, and laser profile scanning compensation technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] 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.

[0009] Figure 1 A schematic flow chart of a control method for grinding magnetic steel provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a control system for magnetic steel grinding provided in an embodiment of the present application.

[0010] Explanation of the reference numerals: grinding path configuration module 11 , process parameter setting module 12 , simulation grinding module 13 , adaptive adjustment module 14 , processing control module 15 . DETAILED DESCRIPTION

[0011] The present application provides a control method and control system for magnetic steel grinding, aiming to solve the technical problems of low grinding precision and inflexible adjustment of process parameters in the prior art. Through video recognition and positioning, virtual grinding environment simulation, multi-axis force monitoring and dynamic parameter adaptive adjustment, and laser contour scanning compensation technology, the technical effect of improving grinding precision and realizing intelligent optimization of process parameters is achieved.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a control method for magnetic steel grinding, the method comprising: Step S100: Connect the video recognition unit, locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel.

[0015] In the embodiment of the present application, the motor rotor magnetic sheet is first positioned by connecting a video recognition unit to ensure the efficiency and stability of the subsequent grinding operation. The video recognition unit is composed of a high-resolution industrial camera (such as a CCD camera) and an image processing system, which can capture the surface image of the motor rotor magnetic sheet in real time. After the image acquisition is completed, the Canny edge detection algorithm is applied to process the image to extract the edge contour of the motor rotor magnetic sheet. The Canny algorithm can accurately identify the significant boundaries of the outer contour of the magnetic steel, and obtain the complete contour information of the magnetic steel by detecting the mutation area of ​​the gray value in the image.

[0016] Based on the extracted contour data, the minimum circumscribed circle fitting algorithm is used to calculate the magnetic steel positioning center coordinates. The minimum circumscribed circle fitting algorithm can find the circle with the minimum radius while ensuring that the magnetic steel contour is covered. The center of the circle is the magnetic steel positioning center. The positioning center coordinates serve as a key reference point to ensure that the grinding path planning has good symmetry and stability.

[0017] After determining the positioning center coordinates, set the grinding reference point as the starting reference point of the grinding path. The grinding reference point is a specific spatial point, usually selected in the surface flat area near the positioning center. To quickly determine the grinding reference point, radially scan outward based on the contour center, and select the point with the smallest surface height change along the radial direction as the grinding reference point.

[0018] Next, spatial matching is performed based on the magnetic steel positioning center coordinates and the grinding reference points to determine the starting point and end point of the grinding control path. After that, a smooth and continuous initial grinding path is generated based on the known starting point and end point through the Bezier curve interpolation algorithm to ensure the uniformity and stability of the grinding process. During this process, the defects on the surface of the magnetic steel are monitored synchronously to identify surface defects such as cracks, pores or scratches. For the detected defective areas, the grinding path is dynamically modified to bypass the defective areas to avoid secondary damage to the magnetic steel. Finally, based on the magnetic steel's positioning center coordinates, grinding reference points and initial grinding path, combined with the basic information of the magnetic steel, such as the size, shape and material hardness of the magnetic steel, the configuration of the grinding control path is completed.

[0019] Furthermore, in the method provided in the embodiment of the application, the magnetic steel positioning center coordinates and the grinding reference point are set, and the grinding control path is configured according to the basic information of the magnetic steel, and further includes: The surface image of the motor rotor magnetic sheet is collected, the contour features of the magnetic steel are extracted, and the coordinates of the magnetic steel positioning center under the preset geometric constraints are calculated; spatial matching is performed according to the coordinates of the magnetic steel positioning center and the grinding reference point, and the starting point and the end point of the grinding control path are located; based on the starting point and the end point of the grinding control path, the initial grinding path is configured, the surface defects of the magnetic steel are monitored synchronously, and decisions on avoiding defective areas are made.

[0020] In an embodiment of the present application, a surface image of a motor rotor magnetic sheet is first captured by a high-resolution industrial camera. Next, the captured image is processed to extract the contour features of the magnetic steel. This process uses the Canny edge detection algorithm to extract the outer contour line of the magnetic steel by identifying areas in the image where the grayscale changes significantly. Contour features refer to the geometric boundary information of the magnetic steel shape, including key features such as straight lines and curves. Based on the extracted contour features, the coordinates of the magnetic steel positioning center are calculated under preset geometric constraints. Geometric constraints refer to specific rules for the shape of the magnetic steel, such as symmetry, standard size range, etc., to ensure the accuracy of the positioning results. The calculation method is based on the centroid calculation method, and the geometric center of the pixel points inside the contour is obtained to obtain the positioning center coordinates of the magnetic steel.

[0021] Next, spatial matching is performed based on the magnetic steel positioning center coordinates and the grinding reference point. This process uses a rigid transformation algorithm, including translation and rotation operations. Specifically, the grinding reference point is first translated and aligned with the positioning center coordinates to ensure that the two are in the same reference coordinate system. Subsequently, the posture of the grinding reference point is adjusted through the rotation matrix to keep it consistent with the geometric axis of the magnetic steel. After completing the spatial matching, the starting point and end point of the grinding control path are located.

[0022] Based on the determined starting and ending points of the grinding path, the initial grinding path is configured. This step uses the Bezier curve interpolation algorithm to generate a smooth curve path between the starting and ending points. The Bezier curve has good smoothness and adjustability, and can automatically generate a smooth grinding trajectory based on the starting point, end point and intermediate control points to ensure that the grinding head moves smoothly on the magnetic steel surface and reduce processing errors. After the configuration is completed, an initial grinding path that conforms to the magnetic steel geometry is obtained.

[0023] While planning the grinding path, the surface defects of the magnetic steel are monitored synchronously, and decisions on avoiding defective areas are made. Surface defects include cracks, pores, scratches, etc., which may affect the grinding quality or even damage the magnetic steel. Using a defect detection algorithm based on image difference, the real-time acquired grinding image is analyzed at the pixel level with the standard reference image to quickly identify abnormal areas. Once a defect is detected, the grinding path is automatically adjusted according to the preset avoidance strategy, and a local path is generated to bypass the defective area, ensuring that the grinding head avoids contact with sensitive areas to prevent further damage.

[0024] Furthermore, the method provided in the application embodiment also includes: The surface defects of the magnetic steel include cracks, pores and scratches; the contour of the defect area is extracted, and the geometric center coordinates and boundary range of the defect area are calculated; through the geometric center coordinates and boundary range of the defect area, a local grinding path that bypasses the defect area is generated.

[0025] In the embodiments of the present application, the surface defects of the magnetic steel are the key factors affecting the grinding quality, and common defect types include cracks, pores and scratches.

[0026] When extracting the contour of the defect area, the Canny edge detection algorithm is used to extract the contour line of the defect area. After the contour extraction is completed, each defect area is geometrically analyzed, including the calculation of the geometric center coordinates and boundary range. The geometric center coordinates represent the center of mass position of the defect area and are an important reference point in the polishing path planning. The calculation method is based on the centroid calculation method. The center position of the defect area is obtained by weighted averaging the coordinates of all pixels in the contour. At the same time, by analyzing the circumscribed rectangle of the contour, the boundary range of the defect is determined, and its specific distribution in the polishing area is clarified.

[0027] Based on the geometric center coordinates and boundary range of the defect area, the local grinding path generation stage begins. This process uses a path planning algorithm, such as a simple obstacle avoidance path algorithm, to ensure that the grinding head can automatically adjust its trajectory when approaching the defect area to avoid direct contact with the defect area. Specifically, technical experts pre-set a certain safety distance outside the defect boundary range and plan a smooth path to bypass the defect area, ensuring that the grinding head can maintain stable grinding quality while effectively avoiding potential risk areas. This process generates a local grinding path that bypasses the defect area.

[0028] Furthermore, in the method provided in the embodiment of the application, the surface defects of the magnetic steel are monitored synchronously, and the decision of avoiding the defective area is made, which also includes: The surface defects of the magnetic steel are classified into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation; the subset of surface defects that do not affect the grinding operation is marked and skipped in the grinding path planning; for the subset of surface defects that affect the grinding operation, multiple local grinding paths at a preset safety distance are set, and the grinding control path is determined in combination with the initial grinding path.

[0029] In an embodiment of the present application, the surface defects of the magnetic steel are first accurately classified into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation. Specifically, based on the preset grinding accuracy standard, the key grinding areas that are critical to the grinding quality are marked, and the defects that appear in these areas are classified into the first surface defect subset. Subsequently, an area analysis is performed on all detected defects. If the area of ​​a defect accounts for a proportion of the grinding area where it is located that exceeds a preset threshold, it is included in the second surface defect subset. By merging the first and second surface defect subsets, the surface defect subset that affects the grinding operation is finally determined.

[0030] A subset of surface defects that do not affect the grinding operation are marked. These defects are marked in the reference data for grinding path planning to ensure that their existence can be identified, but they are directly skipped during the path planning process without avoidance processing, so as to maintain the simplicity of the grinding path and processing efficiency.

[0031] For the subset of surface defects that affect the grinding operation, a preset safety distance is set to ensure that the grinding head maintains a sufficient interval when approaching these defective areas to prevent excessive pressure on the defective areas and cause secondary damage. On this basis, the A* path planning algorithm is used to generate multiple local grinding paths. The A* algorithm is a heuristic search algorithm that can quickly calculate the optimal path based on the path length and obstacle avoidance cost. The defective area and its safety distance are regarded as obstacles, and a grinding path that can effectively bypass the defects is planned.

[0032] Finally, the generated local grinding path is merged with the original initial grinding path to form the final grinding control path. During the fusion process, the B-spline curve interpolation algorithm is used to smooth the path, eliminate the sharp turns at the path connection, and ensure that the grinding head maintains stability and continuity during operation.

[0033] Furthermore, in the method provided in the embodiment of the application, the surface defects of the magnetic steel are classified into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation, and further includes: Based on the grinding accuracy standard, the key grinding area is marked, and the first surface defect subset is set; if the ratio of the defect area to the area of ​​the grinding area exceeds the preset area ratio threshold, the second surface defect subset is set; the first surface defect subset and the second surface defect subset are merged to determine the surface defect subset that affects the grinding operation; based on the surface defect subset that affects the grinding operation, the avoidance priority corresponding to the local grinding path is configured, and the defect area is bypassed with the minimum path offset.

[0034] In an embodiment of the present application, firstly, based on the preset polishing accuracy standard, the key polishing areas that need to be focused on are marked. These areas are usually the parts that are critical to the function or appearance of the product. By comparing with the pre-stored CAD design model, combined with the process requirements and product tolerance range, the template matching algorithm is used to quickly identify and mark these key areas. Then, an edge detection algorithm (such as the Canny algorithm) is applied in the key polishing area to identify existing surface defects such as cracks, pores, and scratches. All defects detected in the key polishing area are automatically classified into the first surface defect subset, because even if these defects are small in size, they may have a significant impact on the polishing effect.

[0035] All detected defects are then analyzed for their area ratio to assess their potential risk to the grinding operation. Specifically, a binary processing algorithm (such as the Otsu adaptive threshold method) is used to separate the defect area from the background, and then the actual area of ​​the defect is measured by the pixel area calculation method. The ratio of this area to the total area of ​​the grinding area where the defect is located is calculated. If the ratio exceeds the preset area ratio threshold (such as 10% or 20%), it will be classified into the second surface defect subset. The setting of this threshold is adjusted according to the grinding process requirements to adapt to the quality control standards of different products.

[0036] After the first and second subsets are screened, the two subsets are merged through the set merging operation to form the final subset of surface defects that affect the grinding operation. During the merging process, deduplication is performed based on the spatial position coordinates of the defects to ensure that the same defect is not counted repeatedly. This defect subset contains all defects that may have an adverse effect on the grinding quality and serves as a key reference for subsequent path optimization.

[0037] After identifying the subset of defects that affect the grinding operation, local grinding paths are configured for these defect areas. First, avoidance priorities are assigned to different defect areas, and the priority setting is based on the severity and location of the defects. For example, deep cracks are given higher avoidance priorities, while shallow scratches are given lower priorities. The order in which defects are processed is determined by a simple priority sorting algorithm (such as linear sorting based on defect depth or area).

[0038] In the process of planning the avoidance path, the A* path planning algorithm is used, and the minimum path offset strategy is introduced in the path search process. Specifically, the initial grinding path is defined as the reference datum, and the offset distance between each path node and the initial path is calculated. When planning the avoidance path, the path offset is evaluated in real time, and those paths that can bypass the defect area while maintaining the minimum offset are selected as much as possible. In this way, the grinding head can quickly return to the vicinity of the initial path after avoiding the defect, maintaining the consistency and continuity of the grinding process.

[0039] Step S200: collecting the preset fixed position and the preset fixed posture of the motor rotor magnetic sheet, and setting the initial grinding process parameters of the grinding reference point and the grinding head under the grinding stability constraint in combination with the grinding control path.

[0040] In the embodiment of the present application, the preset fixed position and preset fixed posture of the motor rotor magnetic sheet are first collected. This process relies on the combined use of a laser ranging sensor and an industrial camera. The laser ranging sensor is used to accurately measure the position coordinates of the magnetic sheet in the grinding tooling to ensure its consistency with the equipment reference point; the industrial camera uses image capture analysis to identify the angle and orientation of the magnetic sheet based on a template matching algorithm to obtain its fixed posture in three-dimensional space.

[0041] After completing the acquisition of position and posture data, it is integrated with the grinding control path. Through coordinate system conversion technology, the actual position of the magnetic sheet is spatially aligned with the grinding path to ensure that the path planning accurately matches the actual workpiece.

[0042] On this basis, according to the grinding control path and the fixed position and posture data of the workpiece, the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint are set. At this time, the parameters are set according to the grinding stability constraint to ensure that the grinding head maintains stable contact with the workpiece surface during the grinding process to avoid vibration, offset or uneven grinding. Grinding stability constraints mainly include precise control of parameters such as contact pressure, feed speed, grinding angle, spindle speed and grinding depth.

[0043] When setting the initial grinding process parameters, first set the appropriate contact pressure according to the characteristics of the grinding path and the hardness and geometry of the magnetic sheet surface to ensure that the grinding head will not cause surface damage due to excessive pressure when in contact with the workpiece, nor will it cause insufficient grinding due to too little pressure. The setting of this parameter is usually combined with historical grinding data and feedback from real-time force control sensors to ensure that the pressure fluctuates within a stable range. Secondly, set the feed speed of the grinding head, taking into account the complexity of the grinding path and the material characteristics of the workpiece, to ensure that the grinding head can maintain a uniform grinding effect in different areas, and avoid knife jumping or uneven grinding due to excessive speed. In response to the angle changes in different grinding areas, adjust the grinding angle of the grinding head according to the fixed posture of the magnetic sheet to ensure that the grinding head maintains the best contact state with the workpiece surface and reduce processing deviations. In addition, set the spindle speed according to the process requirements to balance the grinding efficiency and surface quality, and avoid thermal damage or surface scratches caused by improper speed. Finally, the grinding depth is set based on the height difference between the grinding reference point and the workpiece surface, combined with the grinding path planning and the workpiece material removal requirements to ensure uniform material removal during the grinding process and avoid local over-grinding or uneven residue. Through this step, the initial grinding process parameters are set.

[0044] Step S300: Setting a magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and using a multi-axis force monitoring plug-in to collect pressure distribution simulation data of the grinding contact surface in real time.

[0045] In the embodiment of the present application, a highly simulated magnetic steel grinding virtual environment is first constructed based on the planned grinding control path and the set initial grinding process parameters. The construction of this virtual environment relies on the existing digital twin technology and multi-physics field simulation technology to simulate the interaction between the grinding head and the magnetic steel surface.

[0046] In the virtual environment, the grinding path is highly matched with the three-dimensional model of the workpiece, and the dynamic movement of the grinding head along the control path is simulated. At the same time, the initial grinding process parameters are loaded, covering key process conditions such as feed speed, contact pressure, grinding angle, spindle speed and grinding depth. The contact behavior between the grinding head and the workpiece surface is modeled through a physical contact model, taking into account complex physical effects such as friction, normal force, tangential force and material removal, thereby constructing a highly realistic grinding simulation environment.

[0047] In order to monitor the force on the contact surface between the grinding head and the workpiece in real time during the grinding process, a multi-axis force monitoring plug-in is integrated. Based on the finite element analysis (FEA) technology, the plug-in meshes the grinding contact area and divides the workpiece surface into multiple small unit grids to accurately calculate the force borne by each grid node during the grinding process. By using the contact force calculation algorithm, combined with the motion state and process parameters of the grinding head, the normal pressure and tangential stress distribution of the grinding contact surface are calculated in real time to capture the force changes between the grinding head and the workpiece surface at different positions and at different time points. In the grinding virtual environment, the multi-axis force monitoring plug-in continuously collects real-time data of the grinding contact surface, including information such as pressure distribution in the contact area, force direction, and stress concentration area. In order to improve the accuracy and stability of the data, the data filtering and noise reduction algorithms (such as Kalman filtering) are used to process the collected raw data to remove noise and errors in the virtual simulation and ensure high accuracy and availability of the data.

[0048] Finally, through the collaborative work of the grinding control path, initial grinding process parameters and multi-axis force monitoring plug-in, we successfully achieved a high degree of simulation and real-time monitoring of the grinding process in a virtual environment, and obtained simulation data of the pressure distribution of the grinding contact surface.

[0049] Step S400: Based on the pressure distribution simulation data, the grinding head feed speed and axial deflection angle are dynamically corrected by a preset pressure threshold, and a nonlinear mapping model of magnetic steel material removal rate, spindle speed, and grinding depth is established to adaptively adjust the grinding power.

[0050] In the embodiment of the present application, based on the acquired pressure distribution simulation data, the grinding process parameters are dynamically optimized and adjusted through the preset pressure threshold value pre-set by technical experts to improve the grinding accuracy and process stability. The preset pressure threshold value is determined based on the physical properties of the magnetic steel material, the grinding process requirements and the actual grinding experience, and usually includes an upper limit and a lower limit, which are used to determine whether there is excessive or insufficient pressure on the grinding contact surface.

[0051] When it is detected that the pressure in a certain grinding area exceeds the preset threshold range, the dynamic correction mechanism is automatically triggered to adjust the feed speed and axial deflection angle of the grinding head in real time. The feed speed adjustment is based on a closed-loop feedback control algorithm. By real-time monitoring of pressure distribution data, the deviation between the current pressure value and the preset threshold is calculated, and the feed speed of the grinding head is adjusted. If the pressure in a local area exceeds the upper threshold, the feed speed is reduced and the processing time of the grinding head in this area is extended to reduce the instantaneous pressure; conversely, when the pressure is lower than the lower threshold, the feed speed is appropriately increased to enhance the grinding effect and ensure uniform material removal.

[0052] At the same time, the axial deflection angle of the grinding head is adjusted to optimize the pressure distribution on the grinding contact surface. The angle adjustment of the grinding head relies on the geometric compensation algorithm. Based on the area of ​​uneven pressure distribution, the angle between the grinding head and the workpiece surface is fine-tuned to reduce the local pressure concentration area, ensure that the grinding head maintains uniform contact with the workpiece surface, and thus improve the consistency and stability of grinding.

[0053] On the basis of dynamically correcting the grinding parameters, a nonlinear mapping model between the magnetic steel material removal rate and the spindle speed and grinding depth is established. The model uses a polynomial fitting algorithm to fit the complex nonlinear relationship between the spindle speed, grinding depth and material removal rate by analyzing historical grinding experimental data. Specifically, the actual material removal rate data under different spindle speeds and grinding depths are collected, and the polynomial equation is used for fitting to determine the coefficients, thereby obtaining a mathematical model that can accurately describe the trend of material removal rate changes.

[0054] Based on the output results of the nonlinear mapping model, the grinding power is further adaptively adjusted. The adaptive adjustment relies on the adaptive control algorithm to monitor the changes in the grinding power in real time. Combined with the prediction results of the material removal rate model, the spindle speed and grinding depth are dynamically adjusted to achieve intelligent power distribution and optimization. When abnormal power consumption is detected or the material removal rate deviates from the ideal value, the spindle speed or grinding depth is automatically adjusted to optimize the power output and ensure the efficiency of the grinding process and the rationality of energy utilization. For example, when the removal rate is too high and causes an abnormal increase in power consumption, the spindle speed is reduced or the grinding depth is reduced to avoid excessive grinding and energy waste; conversely, when the removal rate is low and affects the grinding efficiency, the power output is appropriately increased to improve the processing efficiency.

[0055] Finally, through the dynamic correction mechanism and adaptive adjustment strategy based on the pressure distribution simulation data, the real-time optimization of the grinding head feed speed and axial deflection angle is achieved, an accurate nonlinear mapping model of material removal rate is established, and the intelligent control of the grinding power is realized.

[0056] Step S500: At the same time, a laser profile scanning plug-in is set at the end of the polishing path to compare the simulated surface with the measured point cloud data for simulation compensation. After the polishing requirements are met, the polishing process parameter set is output to perform batch processing control.

[0057] In the embodiment of the present application, in order to ensure the processing accuracy and consistency of mass production, a laser profile scanning plug-in is integrated at the end of the grinding path to perform high-precision detection of the workpiece surface after grinding. The core of this process is to compare and analyze the simulated surface with the measured point cloud data to achieve refined process compensation and optimization.

[0058] Among them, the simulated surface is the ideal processing surface obtained by simulating the grinding process in the magnetic steel grinding virtual environment based on the grinding control path and the initial grinding process parameters. It reflects the grinding results predicted under theoretical conditions and represents the standard target of the grinding process. Correspondingly, the measured point cloud data is the three-dimensional data obtained by scanning the surface of the workpiece after the actual grinding through the laser contour scanning plug-in, reflecting the real morphology of the workpiece surface in actual processing.

[0059] After obtaining the measured point cloud data, it is registered with the simulated surface generated in the virtual environment of magnetic steel grinding. The registration process uses the iterative closest point (ICP) algorithm to minimize the geometric deviation between them by continuously adjusting the spatial position and posture of the two to ensure the accuracy of data comparison. After completing the registration, the surface profile error is calculated, that is, the geometric difference between the actual machined surface and the ideal simulated surface. These errors reflect the deviations in the grinding process, such as insufficient or excessive local grinding.

[0060] Based on the surface profile error data, the error distribution characteristics are further extracted to analyze the spatial distribution of the error on the workpiece surface. Areas where the errors are significantly concentrated are identified and marked as high compensation areas. These high compensation areas are usually key areas that affect the surface quality of the workpiece, and they are optimized first to improve the consistency of grinding and surface quality. The identification of high compensation areas is based on the error threshold judgment method. When the local error exceeds the preset tolerance range, it is marked as an area that needs compensation.

[0061] For high compensation areas, simulated compensation is performed in a virtual environment. The core purpose of simulated compensation is to adjust the grinding process parameters and grinding paths so that the compensated simulated surface meets the grinding requirements and reduces the deviation from the actual processing results. According to the error distribution, the feed speed, grinding depth and grinding angle of the grinding head are dynamically adjusted to optimize the local grinding effect. During the compensation process, multiple iterative simulations are performed, and the grinding process is re-simulated after each adjustment to verify the compensation effect until the error range of the simulated surface meets the grinding requirements and ensures the processing accuracy.

[0062] After completing the simulation compensation and confirming that the compensation effect meets the process requirements, all adjusted data are integrated into the final grinding process parameter set. This parameter set contains the optimized grinding path, process parameter adjustment plan and compensation strategy to ensure that each workpiece can maintain consistent high quality standards during batch processing. Finally, the grinding process parameter set is output and enters the batch processing control stage. During the batch production process, the parameter set is automatically called to ensure that each workpiece is processed according to the optimized process flow, effectively improving the processing accuracy and consistency, and reducing the quality risks caused by process fluctuations.

[0063] Furthermore, in the method provided in the embodiment of the application, a laser profile scanning plug-in is provided at the end of the grinding path to compare the simulated surface with the measured point cloud data for simulation compensation, and further includes: Based on the simulated surface, the measured point cloud data is used for registration to calculate the surface contour error; through the surface contour error, the error distribution characteristics are extracted, the error concentration area is identified and marked as a high compensation area, and the high compensation area is optimized first.

[0064] In the embodiment of the present application, the measured point cloud data is first aligned with the simulated surface. By using the iterative closest point algorithm, the rotation and translation parameters of the measured point cloud data are repeatedly adjusted to minimize the Euclidean distance between it and the simulated surface until the error converges to ensure the precise alignment of the two. After the alignment is completed, the simulated surface is compared with the measured point cloud data to calculate the surface contour error. The error calculation is based on the point-to-surface distance calculation method, that is, each data point in the measured point cloud is mapped to the simulated surface, and the shortest vertical distance from the point to the simulated surface is measured to form error distribution data. These data reveal the geometric differences between the actual polished surface and the ideal simulated surface, such as local depressions, protrusions, or uneven polishing.

[0065] After obtaining the surface profile error data, the error distribution feature extraction is performed. This process is based on error statistical analysis and heat map visualization technology. The error value is mapped to the three-dimensional model surface of the workpiece, and the error size and distribution trend are intuitively displayed through different color gradients. The error value is statistically processed, its mean, standard deviation and range are calculated, the distribution density of the error is analyzed, and the error concentration area is quickly located. These areas usually have large deviations, which may have adverse effects on product performance and appearance. Subsequently, the areas with significant errors are identified and marked as high compensation areas. The identification of high compensation areas is based on the threshold judgment algorithm. The preset error threshold is obtained, and the areas exceeding this threshold are automatically marked as key areas that need priority compensation. The marking of these areas helps to concentrate resources for fine-tuning during the compensation process to improve compensation efficiency and grinding quality.

[0066] Compensation strategies are implemented for identified high compensation areas. Compensation strategies mainly include fine-tuning parameters such as grinding head feed speed, grinding depth, contact pressure and grinding angle to achieve effective correction of local errors. For example, for detected recessed areas, the grinding depth is appropriately increased and the feed speed is reduced to ensure sufficient removal of excess material; while in raised or over-grinded areas, the grinding depth is reduced or the contact pressure is adjusted to prevent further material loss. In addition, for areas with poor surface uniformity, the grinding angle is fine-tuned to optimize the contact state between the grinding head and the workpiece surface to ensure more uniform material removal.

[0067] During the implementation of the compensation strategy, the iterative simulation verification mechanism is used to continuously adjust the process parameters and simulate the grinding effect in a virtual environment. After each parameter adjustment, the error change in the compensation area is recalculated to evaluate the compensation effect. If the error fails to meet the preset process standards, the parameters are further fine-tuned until the compensation effect reaches the ideal state.

[0068] Finally, the compensated optimized process parameters are integrated into a complete grinding process parameter set, including optimized key parameters such as feed speed, grinding depth, contact pressure and grinding angle.

[0069] Furthermore, the method provided in the application embodiment also includes: Based on the simulated surface, a rigid body transformation matrix is ​​set for spatial alignment to generate a surface contour error distribution map; based on the surface contour error distribution map, a first compensation area corresponding to the error mean, a second compensation area corresponding to the variance, and a third compensation area corresponding to the extreme value are determined; an overlapping area between the first compensation area and the third compensation area, and an overlapping area between the second compensation area and the third compensation area are determined; a random selection operation is performed, with the boundary of the high compensation area as the starting point, and the operation is expanded along the normal direction to the overlapping area between the first compensation area and the third compensation area, or, the operation is expanded along the normal direction to the overlapping area between the second compensation area and the third compensation area.

[0070] In the embodiment of the present application, the measured point cloud data is first used to align the simulated surface in space. To achieve high-precision alignment, a rigid body transformation matrix is ​​used, which includes rotation and translation operations. It can accurately align the measured point cloud data with the simulated surface without changing the geometric dimensions of the workpiece, ensuring that the two are compared in the same three-dimensional coordinate system.

[0071] After completing the spatial alignment, the surface profile error distribution map is calculated based on the registered data. The error distribution map intuitively displays the error size and distribution characteristics of different areas on the workpiece surface. It is usually visualized using a color gradient, with red indicating a large error and blue indicating a small error. The error distribution map is used to extract key statistical features and further determine the areas that need compensation.

[0072] Based on the error distribution diagram, the first compensation area, the second compensation area and the third compensation area are identified. The first compensation area is determined based on the mean value of the error, mainly covering the area where the error is concentrated and the deviation is close to the average level, reflecting the overall grinding deviation trend; the second compensation area is determined based on the variance of the error, and the positioning error fluctuates greatly and changes dramatically, usually corresponding to the part with uneven surface quality; the third compensation area is determined based on the extreme value of the error, directly identifying the key areas with the largest or smallest local errors, which are usually the high-risk points that need compensation the most.

[0073] Then, the spatial relationship between different compensation areas is analyzed to determine the overlapping areas of the first compensation area and the third compensation area, as well as the overlapping areas of the second compensation area and the third compensation area. These overlapping areas are the key areas of concern because they contain both the extreme values ​​of the error and the characteristics of overall deviation or fluctuation anomaly, representing the areas that need the most accurate compensation. After determining these key areas, a random selection operation is performed to optimize the compensation path. Specifically, starting from the boundary of the high compensation area, the compensation range is extended to the overlapping area along the normal direction of the surface. This extension operation can be selectively performed along two paths. One is to expand to the overlapping area of ​​the first compensation area and the third compensation area to correct the superposition effect of the average deviation and the extreme deviation; the other is to expand to the overlapping area of ​​the second compensation area and the third compensation area, mainly to solve the problem of violent local error fluctuations. The random selection strategy aims to introduce a certain degree of flexibility to avoid the local optimal trap of the compensation path, thereby improving the comprehensiveness and robustness of the compensation effect.

[0074] Finally, the error compensation optimization is completed efficiently through spatial alignment based on rigid body transformation, generation of error distribution map, identification of key compensation areas and random expansion compensation operation.

[0075] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application is connected to a video recognition unit, positions the motor rotor magnetic sheet, sets the magnetic steel positioning center coordinates and the grinding reference point, and configures the grinding control path according to the basic information of the magnetic steel; collects the preset fixed position and the preset fixed posture of the motor rotor magnetic sheet, and sets the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint in combination with the grinding control path; sets the magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and uses a multi-axis force monitoring plug-in to collect the pressure distribution simulation data of the grinding contact surface in real time; based on the pressure distribution simulation data, dynamically corrects the grinding head feed speed and axial deflection angle through a preset pressure threshold, and establishes a nonlinear mapping model of the magnetic steel material removal rate, the spindle speed, and the grinding depth, and adaptively adjusts the grinding power; at the same time, a laser contour scanning plug-in is set at the end of the grinding path to compare the simulated surface with the measured point cloud data for simulation compensation, and after meeting the grinding requirements, outputs the grinding process parameter set to perform batch processing control. The present invention solves the technical problems of low grinding precision and inflexible adjustment of process parameters in the prior art, and achieves the technical effect of improving grinding precision and realizing intelligent optimization of process parameters through video recognition and positioning, virtual grinding environment simulation, multi-axis force monitoring and dynamic parameter adaptive adjustment, and laser profile scanning compensation technology.

[0076] Embodiment 2, based on the same inventive concept as the control method for grinding magnetic steel in the above embodiment, Figure 2 As shown, the present application provides a control system for magnetic steel grinding, and the system and method embodiments in the present application are based on the same inventive concept. The system includes: The grinding path configuration module 11 is used to connect the video recognition unit, locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and the grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel; the process parameter setting module 12 is used to collect the preset fixed position and preset fixed posture of the motor rotor magnetic sheet, and set the grinding reference point and the initial grinding process parameters of the grinding head under the grinding stability constraint in combination with the grinding control path; the simulation grinding module 13 is used to set the magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and use the multi-axis The force monitoring plug-in collects the pressure distribution simulation data of the grinding contact surface in real time; the adaptive adjustment module 14 is used to dynamically correct the grinding head feed speed and axial deflection angle through a preset pressure threshold based on the pressure distribution simulation data, and establish a nonlinear mapping model between the magnetic steel material removal rate and the spindle speed and the grinding depth to adaptively adjust the grinding power; the processing control module 15 is used to simultaneously set a laser contour scanning plug-in at the end of the grinding path, compare the simulated surface with the measured point cloud data for simulation compensation, and output the grinding process parameter set to execute batch processing control after meeting the grinding requirements.

[0077] Furthermore, the system is also used to implement the following functions: The surface image of the motor rotor magnetic sheet is collected, the contour features of the magnetic steel are extracted, and the coordinates of the magnetic steel positioning center under the preset geometric constraints are calculated; spatial matching is performed according to the coordinates of the magnetic steel positioning center and the grinding reference point, and the starting point and the end point of the grinding control path are located; based on the starting point and the end point of the grinding control path, the initial grinding path is configured, the surface defects of the magnetic steel are monitored synchronously, and decisions on avoiding defective areas are made.

[0078] Furthermore, the system is also used to implement the following functions: The surface defects of the magnetic steel include cracks, pores and scratches; the contour of the defect area is extracted, and the geometric center coordinates and boundary range of the defect area are calculated; through the geometric center coordinates and boundary range of the defect area, a local grinding path that bypasses the defect area is generated.

[0079] Furthermore, the system is also used to implement the following functions: The surface defects of the magnetic steel are classified into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation; the subset of surface defects that do not affect the grinding operation is marked and skipped in the grinding path planning; for the subset of surface defects that affect the grinding operation, multiple local grinding paths at a preset safety distance are set, and the grinding control path is determined in combination with the initial grinding path.

[0080] Furthermore, the system is also used to implement the following functions: Based on the grinding accuracy standard, the key grinding area is marked, and the first surface defect subset is set; if the ratio of the defect area to the area of ​​the grinding area exceeds the preset area ratio threshold, the second surface defect subset is set; the first surface defect subset and the second surface defect subset are merged to determine the surface defect subset that affects the grinding operation; based on the surface defect subset that affects the grinding operation, the avoidance priority corresponding to the local grinding path is configured, and the defect area is bypassed with the minimum path offset.

[0081] Furthermore, the system is also used to implement the following functions: Based on the simulated surface, the measured point cloud data is used for registration to calculate the surface contour error; through the surface contour error, the error distribution characteristics are extracted, the error concentration area is identified and marked as a high compensation area, and the high compensation area is optimized first.

[0082] Furthermore, the system is also used to implement the following functions: Based on the simulated surface, a rigid body transformation matrix is ​​set for spatial alignment to generate a surface contour error distribution map; based on the surface contour error distribution map, a first compensation area corresponding to the error mean, a second compensation area corresponding to the variance, and a third compensation area corresponding to the extreme value are determined; an overlapping area between the first compensation area and the third compensation area, and an overlapping area between the second compensation area and the third compensation area are determined; a random selection operation is performed, with the boundary of the high compensation area as the starting point, and the operation is expanded along the normal direction to the overlapping area between the first compensation area and the third compensation area, or, the operation is expanded along the normal direction to the overlapping area between the second compensation area and the third compensation area.

[0083] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0085] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A control method for grinding magnetic steel, characterized in that: The method comprises: Connect the video recognition unit to locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel; Collecting the preset fixed position and the preset fixed posture of the motor rotor magnetic sheet, and setting the initial grinding process parameters of the grinding reference point and the grinding head under the grinding stability constraint in combination with the grinding control path; The magnetic steel grinding virtual environment is set by using the grinding control path and the initial grinding process parameters, and the multi-axis force monitoring plug-in is used to collect the pressure distribution simulation data of the grinding contact surface in real time; Based on the pressure distribution simulation data, the grinding head feed speed and axial deflection angle are dynamically corrected by presetting the pressure threshold, and a nonlinear mapping model of the magnetic steel material removal rate, the spindle speed, and the grinding depth is established to adaptively adjust the grinding power; At the same time, a laser contour scanning plug-in is set at the end of the polishing path to compare the simulated surface with the measured point cloud data for simulation compensation. After meeting the polishing requirements, the polishing process parameter set is output to execute batch processing control.

2. A control method for grinding magnetic steel as claimed in claim 1, characterized in that: The magnetic steel positioning center coordinates and the grinding reference point are set, and the initial grinding path is configured according to the basic information of the magnetic steel. The method includes: Collect the surface image of the motor rotor magnetic sheet, extract the contour features of the magnetic steel, and calculate the magnetic steel positioning center coordinates under the preset geometric constraints; Perform spatial matching based on the magnetic steel positioning center coordinates and the grinding reference point to locate the starting point and end point of the grinding control path; Based on the starting point and the end point of the grinding control path, the initial grinding path is configured, the surface defects of the magnetic steel are monitored synchronously, and defect area avoidance decisions are made.

3. A control method for grinding magnetic steel as claimed in claim 2, characterized in that: The magnetic steel surface defects include cracks, pores and scratches; Extract the contour of the defect area and calculate the geometric center coordinates and boundary range of the defect area; A local grinding path that bypasses the defect area is generated through the geometric center coordinates and boundary range of the defect area.

4. A control method for grinding magnetic steel as claimed in claim 3, characterized in that: The method of synchronously monitoring the surface defects of magnetic steel and making decision on avoiding defective areas comprises: Classifying the surface defects of the magnetic steel into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation; A subset of surface defects that do not affect the grinding operation are marked and skipped in the grinding path planning; For a subset of surface defects that affect the grinding operation, a plurality of local grinding paths at a preset safety distance are set, and the grinding control path is determined in combination with the initial grinding path.

5. A control method for grinding magnetic steel as claimed in claim 4, characterized in that: The surface defects of the magnetic steel are classified into a subset of surface defects that affect the grinding operation and a subset of surface defects that do not affect the grinding operation. The method further includes: Based on the grinding accuracy standard, the key grinding area is marked and the first surface defect subset is set; If the ratio of the defect area to the area of ​​the polishing region exceeds a preset area ratio threshold, a second surface defect subset is set; The first surface defect subset and the second surface defect subset are combined to determine a surface defect subset that affects a grinding operation; Based on the subset of surface defects that affect the grinding operation, the avoidance priority corresponding to the local grinding path is configured, and the defect area is bypassed with a minimum path offset.

6. A control method for grinding magnetic steel as claimed in claim 1, characterized in that: A laser profile scanning plug-in is provided at the end of the grinding path to compare the simulated surface with the measured point cloud data for simulation compensation. The method further includes: Based on the simulated surface, the measured point cloud data is used for registration to calculate the surface contour error; Through the surface profile error, the error distribution characteristics are extracted, the error concentration area is identified and marked as a high compensation area, and the high compensation area is optimized first.

7. A control method for grinding magnetic steel as claimed in claim 6, characterized in that: The method comprises: Based on the simulated surface, a rigid body transformation matrix is ​​set to perform spatial alignment, and a surface profile error distribution map is generated; Based on the surface profile error distribution diagram, determining a first compensation area corresponding to the error mean, a second compensation area corresponding to the variance, and a third compensation area corresponding to the extreme value; Determine an overlapping area between the first compensation area and the third compensation area, and an overlapping area between the second compensation area and the third compensation area; A random selection operation is performed, with the boundary of the high compensation area as the starting point, and the area is expanded along the normal direction to the overlapping area of ​​the first compensation area and the third compensation area, or the area is expanded along the normal direction to the overlapping area of ​​the second compensation area and the third compensation area.

8. A control system for magnetic steel grinding, characterized in that: The system comprises: The grinding path configuration module is used to connect the video recognition unit, locate the motor rotor magnetic sheet, set the magnetic steel positioning center coordinates and grinding reference point, and configure the grinding control path according to the basic information of the magnetic steel; A process parameter setting module, used to collect the preset fixed position and preset fixed posture of the motor rotor magnetic sheet, and set the initial grinding process parameters of the grinding reference point and the grinding head under the grinding stability constraint in combination with the grinding control path; A simulation grinding module, used to set a magnetic steel grinding virtual environment through the grinding control path and the initial grinding process parameters, and use a multi-axis force monitoring plug-in to collect pressure distribution simulation data of the grinding contact surface in real time; An adaptive adjustment module is used to dynamically correct the grinding head feed speed and axial deflection angle through a preset pressure threshold based on the pressure distribution simulation data, and to establish a nonlinear mapping model between the magnetic steel material removal rate and the spindle speed and grinding depth to adaptively adjust the grinding power; The processing control module is used to set a laser contour scanning plug-in at the end of the grinding path, compare the simulated surface with the measured point cloud data for simulation compensation, and output the grinding process parameter set to execute batch processing control after meeting the grinding requirements.

Images