Machining track generation method and system of side edge grinding equipment

Abstract

The invention discloses a method and a system for generating a machining track of side edge grinding equipment. The method comprises the following steps: intelligently analyzing a DXF file, performing rule-driven reading and analysis on the obtained DXF file, extracting a geometric entity related to side edge grinding, and associating predefined grinding process parameters according to layer information; performing data processing and trajectory optimization, performing automatic sorting and redundancy elimination processing on the extracted geometric entities, and uniformly converting all types of lines into a continuous trajectory point set; intelligent track planning is carried out, a track point set is automatically converted into equipment coordinates through coordinate system mapping, a machining track is generated, and track connection and smooth optimization are automatically carried out; and track consistency debugging and closed-loop correction are conducted, the theoretical planning track and the actual machining track are automatically compared, and if the precision does not reach the standard, parameters are automatically adjusted, and the track is regenerated till the precision requirement is met.

Description

Technical Field

[0001] This invention belongs to the field of side grinding technology, and particularly relates to a method and system for generating processing trajectories for side grinding equipment. Background Technology

[0002] Side grinding is a critical process in high-end manufacturing fields such as semiconductors, optical glass, and precision molds, and its processing accuracy directly affects the performance and yield of the workpiece. For example, edge chipping on the side of a silicon wafer or contour deviations in the chamfer of optical glass can lead to product failure. Currently, the generation of processing trajectories in this field heavily relies on manual operation. The specific process is as follows: technicians first need to manually read DXF format drawing files, then, based on the drawing information and their personal experience, manually calculate the transformation relationship between the workpiece coordinate system and the equipment coordinate system, then set the grinding process parameters (such as feed rate, grinding wheel compensation, etc.), and finally write the code to drive the movement of the equipment line by line.

[0003] This traditional manual mode has many inherent drawbacks: First, manual editing is extremely inefficient. The process from reading the drawing to generating usable code is time-consuming and cannot meet the flexible production demands of "small batches, multiple varieties." Furthermore, manual calculations and programming are prone to errors, leading to longer production cycles and increased costs. Second, for drawings containing complex geometries (such as nested contours, spline curves, and chamfered corners), manual analysis is exceptionally difficult. The smoothness and optimization of the trajectory highly depend on the experience level of the technicians, making it difficult to guarantee the consistency and stability of the processing results. Third, when the same product needs to be processed on different types of grinding equipment, the different mechanical parameters of each piece of equipment (such as origin position and axis direction) require technicians to recalculate and adjust coordinates, resulting in serious compatibility barriers between multiple equipment and standardized production. Finally, during the debugging phase, manually verifying the correctness of the trajectory is extremely difficult. It usually requires trial cutting of the workpiece and repeated measurements for verification. This process is not only time-consuming and labor-intensive, but also makes it difficult to detect and correct micron-level trajectory deviations, becoming a key bottleneck restricting the improvement of side grinding accuracy and efficiency.

[0004] Therefore, the existing technology lacks a solution that can automatically parse design drawings, intelligently generate high-precision machining trajectories, and automatically close the loop to correct errors, which has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a method and system for generating the processing trajectory of a side grinding device.

[0006] The technical solution of the present invention is as follows:

[0007] A method for generating the machining trajectory of a side grinding device, executed after obtaining a DXF file, includes the following steps:

[0008] Intelligent DXF file parsing: The acquired DXF file is read and parsed in a rule-driven manner to extract geometric entities related to side grinding and associate predefined grinding process parameters based on layer information.

[0009] Data processing and trajectory optimization automatically sort and remove redundancy from the extracted geometric entities, and convert all types of lines into a continuous set of trajectory points.

[0010] Intelligent trajectory planning automatically converts the trajectory point set into equipment coordinates through coordinate system mapping, generates a processing trajectory, and automatically performs trajectory connection and smoothing optimization;

[0011] The trajectory consistency debugging and closed-loop correction automatically compare the theoretically planned trajectory with the actual machining trajectory. If the accuracy does not meet the standard, the parameters are automatically adjusted and the trajectory is regenerated until the accuracy requirements are met.

[0012] Furthermore, the specific method for performing rule-driven reading and parsing of the acquired DXF file, extracting geometric entities related to side grinding, and associating predefined grinding process parameters based on layer information includes:

[0013] By parsing the $ACADVER code block in the DXF file header, the file version identifier is identified, and the corresponding file reading method is matched according to the predefined version mapping table to correctly read different versions of DXF files;

[0014] By identifying the entity types in the DXF file, all axis, text, and dimension entities are filtered out and removed, leaving only the geometric entities related to side grinding, and all geometric data are uniformly converted to millimeters;

[0015] By parsing the layer information of a specified layer associated with the grinding process in the DXF file, the specified layer is associated with predefined grinding process parameters;

[0016] Automatically traverse all geometric entities within the specified layer, and classify them into line segments, polylines, arcs, or splines based on their geometric type identifiers, and generate corresponding initial trajectory groups.

[0017] Furthermore, the specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points includes:

[0018] By comparing the endpoint coordinates of each line in the initial trajectory group, they are automatically sorted so that the endpoints of adjacent lines coincide with their starting points, forming a continuous processing path. All coordinate points on the continuous processing path are traversed, and overlapping coordinate points and points that are too close (with a distance less than or equal to a set threshold) are identified and deleted. Furthermore, based on the geometric type of the geometric entity, these points are uniformly converted into trajectory points.

[0019] For geometric entities whose geometric type is identified as line segments, extract their start and end coordinates as trajectory points;

[0020] For geometric entities whose geometric type is identified as multi-segment, they are first decomposed into multiple segments, and then the starting and ending coordinates of each segment are extracted as trajectory points.

[0021] For geometric entities whose geometric type is an arc, extract their start and end coordinates and arc radius as trajectory data;

[0022] For geometric entities whose geometric type is identified as spline curves, extract their control point set, node vector, order, and weight as trajectory data;

[0023] The Deboor algorithm is used to uniformly sample and discretize the spline curve within the effective parameter range, thereby converting it into a series of continuous straight line segment trajectory points.

[0024] Furthermore, the specific method for uniformly sampling and discretizing the spline curve within the effective parameter range using the Deboor algorithm, thereby converting it into a series of continuous straight line segment trajectory points, includes:

[0025] Based on spline curve parameters extracted from DXF files, including control point set P, node vector U, curve order k, and weight vector W;

[0026] Determine the effective parameter range of the spline curve [u] p , u m-p ], u p As the lower limit of the effective parameter, u m-p The upper limit of the effective parameters is defined, and this interval is uniformly divided into N sampling points to obtain the parameter sequence, where the sampling parameter t0 = u in the parameter sequence. p Sampling parameter t1 = t0 + Δt, sampling parameter t N =u m-p The interval between sampling parameters Δt=(u m-p - u p ) / N, where N is set according to the accuracy requirements;

[0027] For each sampling parameter, the Deboor algorithm is used to calculate its corresponding curve coordinates, thereby converting the spline curve into a series of continuous straight line segment trajectory points.

[0028] Furthermore, the specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points includes:

[0029] Using a predefined coordinate system mapping formula, the trajectory point set is mapped from the drawing coordinates (X... d ,Y d Batch conversion to device coordinates (X) m ,Y m The coordinate system mapping formula is:

[0030] X m =(1-K)×(X0+X d ×S x )+K×(Y0+Y d ×S y );

[0031] Y m =(1-K)×(Y0+Y d ×S y )+K×(X0+X d ×S x );

[0032] Where X0 and Y0 are the origin offsets of the corresponding X-axis and Y-axis, respectively; S x S y These are the direction coefficients for the X and Y axes, respectively, with K being the XY axis swap indicator; X d Y d These are the X-axis and Y-axis coordinates in the drawing coordinate system; X m ,Y m These are the X-axis and Y-axis coordinates in the device coordinate system, respectively.

[0033] Based on the device coordinates, connect the trajectories of different layers end to end to reduce redundant trajectories. When there is hardware interference at the connection between different layers, after the processing of one layer trajectory is completed, the Z-axis safety position is automatically raised first, and then the processing start position of the next layer is reached.

[0034] At the connection point of the line trajectory, a transition arc command with a preset radius is automatically inserted to achieve a smooth transition of the motion path.

[0035] Furthermore, the specific method for automatically comparing the theoretically planned trajectory with the actual machining trajectory, and automatically adjusting parameters and regenerating the trajectory if the accuracy does not meet the requirements, until the accuracy requirements are met, includes:

[0036] Based on the device coordinates, trajectory processing simulation is performed in the upper control system, and a theoretically planned trajectory diagram is drawn.

[0037] During the equipment trial operation, the data of each axis point is captured in real time, and the actual machining trajectory diagram is automatically fitted and drawn.

[0038] The actual processing trajectory map is compared with the theoretical planned trajectory map at the pixel level, and the average deviation, maximum deviation, root mean square error and overlap percentage are automatically calculated.

[0039] Determine whether the overlap percentage has reached a preset threshold of 99.9%; if not, automatically adjust the process parameters for generating the actual processing trajectory, and re-execute the intelligent trajectory planning step and trajectory consistency debugging step until the overlap percentage reaches the preset threshold.

[0040] A processing trajectory generation system for a side grinding equipment includes a DXF file intelligent parsing module, a data processing and trajectory optimization module, an intelligent trajectory planning module, and a trajectory consistency debugging and closed-loop correction module;

[0041] The DXF file intelligent parsing module is used to read and parse the acquired DXF file in a rule-driven manner, extract the geometric entities related to side grinding, and associate predefined grinding process parameters according to the layer information.

[0042] The data processing and trajectory optimization module is used to automatically sort and remove redundancy from the extracted geometric entities, and to uniformly convert all types of lines into a continuous set of trajectory points.

[0043] The intelligent trajectory planning module is used to automatically convert the trajectory point set into equipment coordinates through coordinate system mapping, generate a processing trajectory, and automatically perform trajectory connection and smoothing optimization.

[0044] The trajectory consistency debugging and closed-loop correction module is used to automatically compare the theoretically planned trajectory with the actual processing trajectory. If the accuracy does not meet the standard, the parameters are automatically adjusted and the trajectory is regenerated until the accuracy requirements are met.

[0045] An electronic device includes a memory and a processor, the memory storing a computer program, the processor being configured to invoke and run the computer program stored in the memory to perform the method as described in any of the preceding methods.

[0046] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in any of the preceding claims.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] This invention proposes a method and system for generating the processing trajectory of a side grinding equipment. This method solves the problems of low efficiency in manual drawing reading, version incompatibility, and reliance on manual setting of process parameters by adopting version compatibility processing based on rule-driven (parsing the "$ACADVER" code block in the DXF file header), automatic filtering and unit unification based on entity type, and automatic association of process parameters based on layer information. It realizes fully automatic, high-precision, and standardized parsing from drawings to data, laying a reliable data foundation for subsequent processes.

[0049] This invention solves the problems of error-prone manual calculation, redundant trajectory data, and difficulty in processing complex curves by employing techniques such as automatic trajectory sorting and connection based on endpoint coordinates, intelligent redundancy removal of coordinate points based on a set threshold (e.g., 0.001 mm), and discretization processing of spline curves using the Deboor algorithm. It achieves automated optimization and simplification of trajectory data, significantly improving the efficiency of data processing and trajectory quality.

[0050] This invention solves the problems of difficult adaptation of heterogeneous equipment, easy interference at trajectory connections, and workpiece damage caused by motion impact by using predefined coordinate system mapping formulas for batch coordinate transformation, automatically inserting Z-axis lifting commands at the connection points of different layer trajectories, and automatically inserting transition arcs at path connections. It realizes intelligent planning and smooth optimization of processing trajectories, ensuring the safety of equipment operation and the surface quality of workpiece processing.

[0051] This invention solves the problems of difficulty in manual verification and inability to guarantee accuracy by employing pixel-level automatic comparison between theoretical and actual trajectories, automatic calculation of multiple deviation indicators (average deviation, maximum deviation, root mean square error, and overlap percentage), and automatic adjustment and closed-loop iteration of code parameters based on threshold judgment. It realizes automatic verification and accuracy correction of processing trajectories. Attached Figure Description

[0052] Figure 1 This is a schematic flowchart of the processing trajectory generation method for the side grinding equipment of the present invention. Detailed Implementation

[0053] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0054] Example 1:

[0055] The present invention provides a method for generating the machining trajectory of a side grinding device, which is executed after obtaining the DXF file, such as... Figure 1 As shown, it includes the following steps:

[0056] S1, DXF file intelligent parsing, reads and parses the acquired DXF file in a rule-driven manner, extracts the geometric entities related to side grinding, and associates predefined grinding process parameters according to the layer information;

[0057] S2. Data processing and trajectory optimization: Automatically sort and remove redundancy from the extracted geometric entities, and convert all types of lines into a continuous set of trajectory points.

[0058] S3, Intelligent trajectory planning, automatically converts the trajectory point set into equipment coordinates through coordinate system mapping, generates the processing trajectory, and automatically performs trajectory connection and smoothing optimization;

[0059] S4. Trajectory consistency debugging and closed-loop correction: Automatically compare the theoretically planned trajectory with the actual machining trajectory. If the accuracy does not meet the standard, automatically adjust the parameters and regenerate the trajectory until the accuracy requirements are met.

[0060] Example 2:

[0061] This embodiment, based on Embodiment 1, further designs the following: In this example, the method for rule-driven reading and parsing of the acquired DXF file, extracting geometric entities related to side grinding, and associating predefined grinding process parameters based on layer information includes:

[0062] By parsing the $ACADVER code block in the DXF file header, the file version identifier is identified, and the corresponding file reading method is matched according to the predefined version mapping table to correctly read different versions of DXF files;

[0063] By identifying the entity types in the DXF file, all axis, text, and dimension entities are filtered out and removed, leaving only the geometric entities related to side grinding, and all geometric data are uniformly converted to millimeters;

[0064] By parsing the layer information of the specified layer associated with the grinding process in the DXF file, the specified layer is associated with the predefined grinding process parameters. Specifically, the layer associated with the rough grinding process is associated with the first set of feed rate and compensation parameters, the layer associated with the fine grinding process is associated with the second set of feed rate and compensation parameters, the layer associated with the chamfering process is associated with the first set of feed rate and compensation parameters, and the layer associated with the fine grinding process is associated with the third set of feed rate and compensation parameters.

[0065] The following example illustrates the specific implementation of layers and associated parameters. This example uses a scene of beveling and grinding optical glass (corresponding to the layer association process in S1 of Figure 1) as the background. The DXF file pre-defines three specified layers, specifically including:

[0066] Layer “Rough_Grinding” (layer number 101): Associated with rough grinding process parameters, feed speed 10mm / s, grinding wheel compensation 0.2mm, used to remove excess material from the glass surface;

[0067] Layer “Fine_Grinding” (layer number 102): Associated with fine grinding process parameters, feed speed 3mm / s, grinding wheel compensation 0.05mm, used to improve surface roughness, target Ra≤0.1μm;

[0068] Layer “Chamfering” (layer number 103): Associated chamfering process parameters: feed speed 5mm / s, grinding wheel compensation 0.1mm, chamfering angle 45°, used to process the 45° bevel of the glass edge, bevel width 2mm.

[0069] The system automatically matches the predefined parameters by parsing the layer name and number, without manual intervention, thus avoiding processing defects caused by confusion of different process parameters, such as surface scratches caused by using rough grinding parameters for fine grinding.

[0070] Automatically traverse all geometric entities within the specified layer and classify them into line segments, polylines, arcs, or splines based on their geometric type identifiers, and generate the corresponding initial trajectory groups.

[0071] Example 3:

[0072] This embodiment, based on Embodiment 2, further designs the following: The specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points, includes:

[0073] By comparing the endpoint coordinates of each line in the initial trajectory group, the lines are automatically sorted so that the endpoints of adjacent lines coincide with their starting points, forming a continuous processing path. All coordinate points on the continuous processing path are traversed, and overlapping coordinate points and points that are too close (with a distance less than or equal to a set threshold, such as 0.001mm) are identified and deleted. Finally, based on the geometric type of the geometric entity, it is uniformly converted into trajectory points.

[0074] For geometric entities whose geometric type is identified as line segments, extract their start and end coordinates as trajectory points;

[0075] For geometric entities whose geometric type is identified as multi-segment, they are first decomposed into multiple segments, and then the starting and ending coordinates of each segment are extracted as trajectory points.

[0076] For geometric entities whose geometric type is an arc, extract their start and end coordinates and arc radius as trajectory data;

[0077] For geometric entities identified as spline curves, their control point set, node vector, order, and weights are extracted as trajectory data. Control points are feature points that determine the shape of the curve; the curve does not necessarily pass through these points. Node vectors are a sequence of parameters controlling the segmentation and continuity of the curve. The order determines the smoothness of the curve; order = degree + 1, e.g., a 3rd-order spline has an order of 4. Weights affect the curve's "attraction" to the control points; for irrational splines, all weights are 1. These parameters together describe a spline curve.

[0078] The Deboor algorithm is used to uniformly sample and discretize the spline curve within the effective parameter range, thereby converting it into a series of continuous straight line segment trajectory points.

[0079] Example 4:

[0080] This embodiment, based on Embodiment 3, further designs the following: In this example, the Deboor algorithm is used to uniformly sample and discretize the spline curve within the effective parameter range, thereby converting it into a series of continuous straight line segment trajectory points. The specific method includes:

[0081] Based on spline curve parameters extracted from the DXF file, including the control point set P, node vector U, curve order k, and weight vector W; where the control point set P = [P0, P1, ..., P... n ],P0, P1, ..., P n The 0th, 1st, ..., nth element in the control point set, in three-dimensional coordinates; node vector U = [u0, u1, ..., u... m ], non-decreasing sequence; curve order k (or degree p = k-1); weight vector W, non-rational splines are all 1 by default;

[0082] Determine the effective parameter range of the spline curve [u] p , u m-p ], u p As the lower limit of the effective parameter, u m-p The upper limit of the effective parameters is defined, and this interval is uniformly divided into N sampling points to obtain the parameter sequence, where the sampling parameter t0 = u in the parameter sequence. pSampling parameter t1 = t0 + Δt, sampling parameter t N = u m-p The interval between sampling parameters Δt = (u m-p - u p ) / N, where N is set according to the accuracy requirements;

[0083] For each sampling parameter tᵢ, the Deboor algorithm is used to calculate its corresponding curve coordinates, i.e., three-dimensional coordinates (xᵢ, yᵢ, zᵢ), where xᵢ, yᵢ, and zᵢ are the x-axis, y-axis, and z-axis coordinates of the sampling parameter, respectively. This transforms the spline curve into a series of continuous straight line segment trajectory points. Specifically, the steps are as follows:

[0084] Step 1) Locate the sampling parameter tᵢ within the node vector interval [u] in the node vector. j , u j+1 );u j u is the lower bound of the node vector interval; j+1 The upper limit of the node vector interval;

[0085] Step 2) Select p+1 control points to participate in the calculation, P j-p To P j ;P j-p P is the minimum value of the control point; j The maximum value of the control point;

[0086] Step 3) Calculate the intermediate control points using a recursive formula to finally obtain the curve points corresponding to the sampling parameter tᵢ;

[0087] Step 4) If it is a rational spline (with weights), the calculation result needs to be normalized by weights.

[0088] Example 5:

[0089] This embodiment, based on Embodiment 4, further designs the following: The specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points, includes:

[0090] Using a predefined coordinate system mapping formula, the trajectory point set is transferred from the drawing coordinates (X... d ,Y d Batch conversion to device coordinates (X) m ,Y m The coordinate system mapping formula is:

[0091] X m =(1-K)×(X0+X d ×S x )+K×(Y0+Y d ×S y );

[0092] Y m =(1-K)×(Y0+Y d ×S y )+K×(X0+X d ×S x );

[0093] Where X0 and Y0 are the origin offsets of the corresponding X-axis and Y-axis, respectively; Origin offset X0 (offset of the drawing's X-axis origin on the device's X-axis, used when not swapped) / Offset of the drawing's Y-axis origin on the device's X-axis (used when swapped); Origin offset Y0 (offset of the drawing's Y-axis origin on the device's Y-axis, used when not swapped) / Offset of the drawing's X-axis origin on the device's Y-axis (used when swapped); S x S y X represents the direction coefficients of the X and Y axes, and K is the XY axis swap indicator; X d Y d These are the X-axis and Y-axis coordinates in the drawing coordinate system; X m , Y m These are the X-axis and Y-axis coordinates in the device coordinate system, respectively.

[0094] Based on the equipment coordinates, the trajectories of different layers (each grinding type) are connected end to end to reduce redundant trajectories. When there is hardware interference at the connection between different layers, the Z-axis safety position is automatically raised after the processing of one layer trajectory is completed, and then the processing start position of the next layer is reached.

[0095] At the connection point of the line trajectory, a transition arc command with a preset radius is automatically inserted to achieve a smooth transition of the motion path; the preset radius is 0.05 to 0.1 mm.

[0096] Example 6:

[0097] This embodiment, based on embodiment five, further incorporates the following method: Automatically comparing the theoretically planned trajectory with the actual machining trajectory, and if the accuracy is insufficient, automatically adjusting parameters and regenerating the trajectory until the accuracy requirements are met.

[0098] Based on the equipment coordinates, trajectory processing simulation is performed in the upper control system, and the theoretical planned trajectory diagram is drawn.

[0099] During the equipment trial operation, the data of each axis point is captured in real time, and the actual machining trajectory diagram is automatically fitted and drawn.

[0100] The actual processing trajectory map is compared with the theoretical planned trajectory map at the pixel level, and the average deviation, maximum deviation, root mean square error and overlap percentage are automatically calculated.

[0101] The specific calculation method is as follows:

[0102] Let the set of points for the planned trajectory be P = [P1, P2, ..., P]. n ], where Pᵢ=(xᵢ, yᵢ), corresponding to the theoretical coordinates;

[0103] Let the actual trajectory point set Q = [Q1, Q2, ..., Q...]. n ], where Qᵢ=(x'ᵢ, y'ᵢ), corresponding to the measured coordinates;

[0104] 1) Average deviation δ avg Mean Deviation reflects the overall average deviation; the smaller the value, the higher the overlap. The calculation method is as follows:

[0105] δ avg = (1 / n) × Σ1 n √[(xᵢ - x'ᵢ)² + (yᵢ - y'ᵢ)²]

[0106] The calculation steps are as follows: calculate the Euclidean distance (straight-line distance in the plane) between the planned point and the actual point point point by point; and calculate the arithmetic mean of all distances.

[0107] 2) Maximum deviation δ max Maximum Deviation reflects the most severe deviation in the trajectory and directly determines whether it exceeds the tolerance. It is calculated as follows:

[0108] δ max = max{ √[(x1 - x'1)² + (y1 - y'1)²], ..., √[(x n - x' n )² + (y n -y' n )²]}

[0109] Significance: If the maximum deviation exceeds the machining accuracy requirement (e.g., ±0.1mm), the trajectory overlap does not meet the standard.

[0110] 3) Root mean square error δ rms (RMS Error) focuses more on amplifying the impact of larger deviations and is often used for precision machining evaluation. The calculation method is as follows:

[0111] δ rms = √[ (1 / n) × Σ1 n ( (xᵢ - x'ᵢ)² + (yᵢ - y'ᵢ)² ) ]

[0112] Features: More sensitive to points with large deviations, suitable for evaluating the overall stability of the trajectory.

[0113] 4) The overlap percentage η is applicable to discrete trajectories. When the trajectory consists of line segments, the percentage of overlapping parts η can be calculated as follows:

[0114] η=(L 重合 / L 规划 ) × 100%

[0115] Among them, L 规划 L is the total length of the planned trajectory, which is the sum of the lengths of each line segment. 重合 The length of the overlapping line segment between the actual trajectory and the planned trajectory needs to be determined through geometric calculations to identify the overlapping area.

[0116] Determine whether the overlap percentage has reached the preset threshold of 99.9%; if not, automatically adjust the process parameters for generating the actual processing trajectory, and re-execute the intelligent trajectory planning step and trajectory consistency debugging step until the overlap percentage reaches the preset threshold.

[0117] Example 7:

[0118] The present invention provides a processing trajectory generation system for a side grinding equipment, comprising a DXF file intelligent parsing module, a data processing and trajectory optimization module, an intelligent trajectory planning module, and a trajectory consistency debugging and closed-loop correction module;

[0119] The DXF file intelligent parsing module is used to read and parse the acquired DXF file in a rule-driven manner, extract the geometric entities related to side grinding, and associate predefined grinding process parameters based on the layer information.

[0120] The data processing and trajectory optimization module is used to automatically sort and remove redundancy from the extracted geometric entities, and to convert all types of lines into a continuous set of trajectory points.

[0121] The intelligent trajectory planning module is used to automatically convert the trajectory point set into equipment coordinates through coordinate system mapping, generate the processing trajectory, and automatically perform trajectory connection and smoothing optimization.

[0122] The trajectory consistency debugging and closed-loop correction module is used to automatically compare the theoretically planned trajectory with the actual machining trajectory. If the accuracy does not meet the standard, the parameters are automatically adjusted and the trajectory is regenerated until the accuracy requirements are met.

[0123] Example 8:

[0124] An electronic device according to the present invention includes a memory and a processor. The memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to perform the method as described in any of the above embodiments.

[0125] The present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the above embodiments.

[0126] Example 9:

[0127] This example applies the method of the present invention to the generation of the machining trajectory of a side grinding machine. The machining object is a semiconductor silicon wafer with a diameter of 200mm and a thickness of 775μm. The side contour contains three types of geometric entities: line segments, arcs, and spline curves. The X / Y axis positioning accuracy of the side grinding machine is ±1μm, the Z axis travel is 50mm, and the maximum feed rate is 20mm / s. The corresponding DXF file parameters are: AutoCAD 2020 version, containing 3 specified layers (rough grinding, fine grinding, and chamfering), with a total of 86 geometric entities, including 32 line segments, 28 arcs, and 26 spline curves.

[0128] In contrast, this example also uses a traditional manual method to generate the processing trajectory of the same side grinding machine. The traditional manual method is operated by the same technician with 5 years of experience. The method of this invention is referred to as Method A and the traditional manual method is referred to as Method B.

[0129] The values ​​for each step in the method of this invention are as follows:

[0130] S1 (DXF parsing): The version mapping table matches the AutoCAD 2020 reading method, the unit is converted to millimeters, 32 axis / text / dimension entities are filtered, and 86 geometric entities are retained;

[0131] S2 (Data Processing): The distance threshold is set to 0.001 mm, and 158 redundant points are deleted; the number of sampling points for the spline curve is N=1000, the effective parameter range is [0,1], and Δt=0.001;

[0132] S3 (Trajectory Planning): Coordinate system mapping parameters X0=10mm, Y0=5mm, S x =1,S y =1, K=0 (not interchanged); transition radius 0.08mm; Z-axis safety position lifting height 5mm;

[0133] S4 (Closed-loop correction): The preset overlap threshold is 99.9%. The average deviation in the first simulation is 8μm. After adjusting the feed speed (from 5mm / s to 3mm / s) and the grinding wheel compensation amount (from 0.1mm to 0.12mm), the system is redesigned.

[0134] Results Analysis: The application results of each method were compared and analyzed from four dimensions: trajectory generation efficiency, processing accuracy, product quality, and changeover adaptability. Regarding the total time spent on trajectory generation, the method of this invention (Method A) only required 2.5 minutes, while the traditional manual method (Method B) required 4.2 hours. The efficiency of the method of this invention is significantly higher than the comparative method, greatly shortening the preparation time before processing. Regarding the trajectory overlap percentage, Method A ultimately reached 99.96%, exceeding the preset threshold, while Method B only achieved 94.8%. The trajectory accuracy of the method of this invention is superior, better meeting design requirements. Regarding the maximum deviation, the maximum deviation of Method A was only 1.2 μm, while that of Method B was 12.5 μm. The method of this invention effectively controlled trajectory deviation and reduced the risk of processing errors. Regarding the edge chipping rate of silicon wafers, the chipping rate after processing by Method A was only 0.3%, while that of Method B was 9.2%. The method of this invention significantly improved the product qualification rate. Regarding the time spent on trajectory adjustment during changeover, Method A only required 8 minutes, while Method B required 2 minutes. The method of this invention exhibits stronger adaptability to production changes, meeting the flexible production needs of "small batches, multiple varieties." Furthermore, through 10 repeated tests, the trajectory overlap fluctuation of Method A was only ±0.02%, while that of Method B reached ±3.5%, indicating that the method of this invention has extremely strong processing stability and can achieve stable high-precision processing without relying on human experience. The trajectory generation time of Method A is only 1 / 100.8 of that of Method B, and the production changeover adjustment time is only 1 / 360 of that of Method B, fully meeting the needs of flexible production, demonstrating the efficiency advantage of this method. The trajectory overlap of Method A exceeds the preset threshold, with the maximum deviation only 1 / 10.4 of that of Method B, and the edge breakage rate is far lower than the comparative methods, meeting the high-precision processing requirements of semiconductor silicon wafers, demonstrating the precision advantage of this method. Method A achieves automatic parameter adjustment through closed-loop correction, requiring no human intervention; the trajectory overlap fluctuation in 10 repeated tests was only ±0.02%, while that of Method B reached ±3.5%, demonstrating extremely strong processing stability, demonstrating the stability advantage of this method.

[0135] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for generating the machining trajectory of a side grinding device, executed after acquiring a DXF file, characterized in that, Includes the following steps: Intelligent DXF file parsing: The acquired DXF file is read and parsed in a rule-driven manner to extract geometric entities related to side grinding and associate predefined grinding process parameters based on layer information. Data processing and trajectory optimization automatically sort and remove redundancy from the extracted geometric entities, and convert all types of lines into a continuous set of trajectory points. Intelligent trajectory planning automatically converts the trajectory point set into equipment coordinates through coordinate system mapping, generates a processing trajectory, and automatically performs trajectory connection and smoothing optimization; The trajectory consistency debugging and closed-loop correction automatically compare the theoretically planned trajectory with the actual machining trajectory. If the accuracy does not meet the standard, the parameters are automatically adjusted and the trajectory is regenerated until the accuracy requirements are met.

2. The method for generating the processing trajectory of the side grinding equipment according to claim 1, characterized in that, The specific method for performing rule-driven reading and parsing of the acquired DXF file, extracting geometric entities related to side grinding, and associating predefined grinding process parameters based on layer information includes: By parsing the $ACADVER code block in the DXF file header, the file version identifier is identified, and the corresponding file reading method is matched according to the predefined version mapping table to correctly read different versions of DXF files; By identifying the entity types in the DXF file, all axis, text, and dimension entities are filtered out and removed, leaving only the geometric entities related to side grinding, and all geometric data are uniformly converted to millimeters; By parsing the layer information of a specified layer associated with the grinding process in the DXF file, the specified layer is associated with predefined grinding process parameters; Automatically traverse all geometric entities within the specified layer, and classify them into line segments, polylines, arcs, or splines based on their geometric type identifiers, and generate corresponding initial trajectory groups.

3. The method for generating the processing trajectory of the side grinding equipment according to claim 2, characterized in that, The specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points includes: By comparing the endpoint coordinates of each line in the initial trajectory group, they are automatically sorted so that the endpoints of adjacent lines coincide with their starting points, forming a continuous processing path. All coordinate points on the continuous processing path are traversed, and overlapping coordinate points and points that are too close (with a distance less than or equal to a set threshold) are identified and deleted. Furthermore, based on the geometric type of the geometric entity, these points are uniformly converted into trajectory points. For geometric entities whose geometric type is identified as line segments, extract their start and end coordinates as trajectory points; For geometric entities whose geometric type is identified as multi-segment, they are first decomposed into multiple segments, and then the starting and ending coordinates of each segment are extracted as trajectory points. For geometric entities whose geometric type is an arc, extract their start and end coordinates and arc radius as trajectory data; For geometric entities whose geometric type is identified as spline curves, extract their control point set, node vector, order, and weight as trajectory data; The Deboor algorithm is used to uniformly sample and discretize the spline curve within the effective parameter range, thereby converting it into a series of continuous straight line segment trajectory points.

4. The method for generating the processing trajectory of the side grinding equipment according to claim 3, characterized in that, The specific method for uniformly sampling and discretizing the spline curve within the effective parameter range using the Deboor algorithm, thereby converting it into a series of continuous straight line segment trajectory points, includes: Based on spline curve parameters extracted from DXF files, including control point set P, node vector U, curve order k, and weight vector W; Determine the effective parameter range of the spline curve [u] p , u m-p ], u p As the lower limit of the effective parameter, u m-p The upper limit of the effective parameters is defined, and this interval is uniformly divided into N sampling points to obtain the parameter sequence, where the sampling parameter t0 = u in the parameter sequence. p Sampling parameter t1 = t0 + Δt, sampling parameter t N =u m-p The interval between sampling parameters Δt=(u m-p - u p ) / N, where N is set according to the accuracy requirements; For each sampling parameter, the Deboor algorithm is used to calculate its corresponding curve coordinates, thereby converting the spline curve into a series of continuous straight line segment trajectory points.

5. The method for generating the processing trajectory of the side grinding equipment according to claim 4, characterized in that, The specific method for automatically sorting and deduplicating the extracted geometric entities, and uniformly converting all types of lines into a continuous set of trajectory points includes: Using a predefined coordinate system mapping formula, the trajectory point set is mapped from the drawing coordinates (X... d ,Y d Batch conversion to device coordinates (X) m ,Y m The coordinate system mapping formula is: X m =(1-K)×(X0+X d ×S x )+K×(Y0+Y d ×S y ); Y m =(1-K)×(Y0+Y d ×S y )+K×(X0+X d ×S x ); Where X0 and Y0 are the origin offsets of the corresponding X-axis and Y-axis, respectively; S x S y These are the direction coefficients for the X and Y axes, respectively, with K being the XY axis swap indicator; X d Y d These are the X-axis and Y-axis coordinates in the drawing coordinate system; X m ,Y m These are the X-axis and Y-axis coordinates in the device coordinate system, respectively. Based on the device coordinates, connect the trajectories of different layers end to end to reduce redundant trajectories. When there is hardware interference at the connection between different layers, after the processing of one layer trajectory is completed, the Z-axis safety position is automatically raised first, and then the processing start position of the next layer is reached. At the connection point of the line trajectory, a transition arc command with a preset radius is automatically inserted to achieve a smooth transition of the motion path.

6. The method for generating the processing trajectory of the side grinding equipment according to claim 5, characterized in that, The specific method for automatically comparing the theoretically planned trajectory with the actual machining trajectory, and automatically adjusting parameters and regenerating the trajectory if the accuracy is not up to standard, until the accuracy requirements are met, includes: Based on the device coordinates, trajectory processing simulation is performed in the upper control system, and a theoretically planned trajectory diagram is drawn. During the equipment trial operation, the data of each axis point is captured in real time, and the actual machining trajectory diagram is automatically fitted and drawn. The actual processing trajectory map is compared with the theoretical planned trajectory map at the pixel level, and the average deviation, maximum deviation, root mean square error and overlap percentage are automatically calculated. Determine whether the overlap percentage has reached a preset threshold of 99.9%; if not, automatically adjust the process parameters for generating the actual processing trajectory, and re-execute the intelligent trajectory planning step and trajectory consistency debugging step until the overlap percentage reaches the preset threshold.

7. A machining trajectory generation system for a side grinding equipment, characterized in that, It includes a DXF file intelligent parsing module, a data processing and trajectory optimization module, an intelligent trajectory planning module, and a trajectory consistency debugging and closed-loop correction module; The DXF file intelligent parsing module is used to read and parse the acquired DXF file in a rule-driven manner, extract the geometric entities related to side grinding, and associate predefined grinding process parameters according to the layer information. The data processing and trajectory optimization module is used to automatically sort and remove redundancy from the extracted geometric entities, and to uniformly convert all types of lines into a continuous set of trajectory points. The intelligent trajectory planning module is used to automatically convert the trajectory point set into device coordinates through coordinate system mapping, generate a processing trajectory, and automatically perform trajectory connection and smoothing optimization. The trajectory consistency debugging and closed-loop correction module is used to automatically compare the theoretically planned trajectory with the actual processing trajectory. If the accuracy does not meet the standard, the parameters are automatically adjusted and the trajectory is regenerated until the accuracy requirements are met.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor being configured to invoke and run the computer program stored in the memory to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6 above.

Images