Nuclear power pressure vessel closure head inner wall self-adapting polishing planning method and related device

Abstract

The present application belongs to the field of intelligent processing of nuclear power equipment and trajectory planning of mechanical arm, and discloses a kind of nuclear power pressure vessel plugging head inner wall self-adapting polishing planning method and related device, based on plugging head inner wall three-dimensional surface model, utilize the feature of rotary surface to extract a plurality of sampling profile lines to construct ideal polishing curve, through arc length segmentation and local normal vector calculation each position polishing allowance, form two-dimensional allowance distribution matrix;Along the row direction of matrix, establish sliding window, extract the maximum allowance of each angle column in each window as the planning polishing amount, determine the polishing times according to this, and generate the trajectory of mechanical arm and rotary table cooperation combined with curve position and normal vector.This method realizes the quantitative control and quality consistency of polishing process;Adapt to the process goal of retaining overall appearance and only modifying local part;Significantly improve the grinding accuracy and efficiency of nuclear power key components.

Description

Technical Field

[0001] This invention belongs to the field of intelligent processing and robotic arm trajectory planning technology for nuclear power equipment, and particularly relates to an adaptive grinding planning method and related device for the inner wall of the sealing head of a nuclear power pressure vessel. Background Technology

[0002] The sealing head of a nuclear power pressure vessel is a key pressure-bearing component in nuclear power equipment. It is typically a large-sized, thick-walled, curved structure with a highly curved, concave, gyroscopic inner wall. To improve the corrosion resistance and service reliability of the sealing head's inner wall, a corrosion-resistant layer is deposited onto the substrate surface during the manufacturing process. After the welding process is completed, the welded area is ground to ensure that the surface morphology of the welded layer meets subsequent quality acceptance standards. Currently, on-site grinding of weld overlays is still mainly done manually. With the development of robotic arm processing, 3D vision inspection, and offline programming technologies, automated grinding technology for large curved workpieces has gradually formed a robotic arm grinding planning method based on 3D measurement data. The mainstream technical routes are divided into two categories: one is a path planning method that does not perform explicit allowance calculation, which generates a covering grinding trajectory based on the actual 3D measurement data of the workpiece and completes surface finishing with fixed processing parameters; the other is an adaptive path planning method that considers the grinding allowance, which calculates the local processing allowance by registering the actual measurement data with the target model, thereby adjusting the grinding trajectory and processing parameters, and has a certain degree of adaptive processing capability.

[0003] Traditional techniques for grinding the weld overlay on the inner wall of nuclear power pressure vessel sealing heads still have several shortcomings. First, manual grinding relies heavily on the operator's experience and subjective judgment, making it difficult to quantify and control the grinding area, removal volume, and number of grinding passes, thus compromising the repeatability and consistency of grinding quality. Second, path planning methods that do not perform explicit allowance calculations, relying solely on the overall surface morphology to generate a covering trajectory and using uniform processing parameters for finishing, cannot accurately match the differentiated material removal requirements at different locations of the weld overlay. Third, planning methods that consider grinding allowances often rely on theoretical CAD. Using existing models or general-purpose finished models as registration benchmarks is incompatible with the grinding target of preserving the overall curved surface morphology of the weld overlay and only adjusting local excess height and surface undulations, resulting in insufficient scenario adaptability. In addition, existing technologies mostly obtain excess information at discrete points, which lacks a clear conversion relationship with the continuous grinding stroke with a certain contact width actually executed by the robotic arm, making it difficult to directly convert it into executable grinding stroke planning and corresponding process parameters. Finally, the trajectory parameter settings of fixed step distance, fixed row spacing and fixed grinding number cannot adapt to the uneven distribution of excess material caused by weld overlap and excess height fluctuations in the weld overlay, making it difficult to achieve differentiated adaptive allocation of grinding intensity while ensuring trajectory continuity.

[0004] It is evident that existing technologies have significant limitations when applied to the grinding of the weld overlay layer on the inner wall of the sealing head of nuclear power pressure vessels: manual grinding relies on operator experience, the process is difficult to quantify and control, and the stability and consistency of processing quality are insufficient; existing automated grinding planning methods have failed to fully adapt to the process objectives of local grinding of the weld overlay layer, and have technical shortcomings in key aspects such as grinding benchmark setting, the corresponding conversion between allowance information and actual grinding stroke, and adaptive allocation of grinding intensity, which cannot fully meet the high-quality grinding requirements of such key pressure-bearing components in nuclear power plants. Summary of the Invention

[0005] This invention provides an adaptive grinding planning method and related apparatus for the inner wall of the sealing head of a nuclear power pressure vessel. This method can fully adapt to the process objective of local grinding of the weld overlay layer and can fully meet the high-quality grinding requirements of this type of key pressure-bearing component in nuclear power.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: An adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head includes: Based on the three-dimensional curved surface model of the inner wall of the sealing head, and combined with the rotational curved surface features of the inner wall of the sealing head, multiple sampling profile lines are uniformly extracted in the rotational direction according to a preset angle interval, and ideal grinding curves are constructed to characterize the grinding morphology of the target, forming a set of ideal grinding curves. Based on the set of ideal grinding curves, each ideal grinding curve is divided into segments of equal length along the arc length direction and the local normal vector of each segment position is calculated. The grinding allowance between the original sampling profile line and the ideal grinding curve is calculated along the direction of the local normal vector. The allowance of each segment is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. Based on a two-dimensional margin distribution matrix, a sliding window is established along the matrix row direction. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The position of the ideal grinding curve and the local normal vector are combined to generate the trajectory points and posture of the robotic arm. The grinding planning result of the robotic arm and the turntable is output.

[0007] Furthermore, the process of constructing the ideal polishing curve set specifically includes: Based on the three-dimensional curved surface model of the inner wall of the sealing head, multiple sampling angles are uniformly selected along the rotation direction according to a preset angle interval to obtain an angle sampling set; the cutting plane passing through the rotation axis of the sealing head and corresponding to each sampling angle intersects with the three-dimensional curved surface model of the inner wall of the sealing head to extract the corresponding sampling profile lines; invalid point removal, outlier processing and interpolation completion preprocessing are performed on each sampling profile line to obtain a sampling profile line set; For each preprocessed sampling profile line, a set of nodes is selected to describe the overall morphological trend of the profile. Based on the node set, a corresponding ideal grinding curve is constructed using cubic spline curves. During the construction process, smoothness constraints are applied to the continuity of function values, first derivatives, and second derivatives. At the same time, second derivative sign consistency constraints and target grinding side position constraints are applied to ensure that the ideal grinding curve maintains a single curvature trend and is located on the target grinding side of the original sampling profile line, thus forming the set of ideal grinding curves.

[0008] Furthermore, the calculation process for the polishing allowance specifically includes: Each ideal grinding curve in the set of ideal grinding curves is parameterized by arc length and divided into equal length segments according to a preset segment length to obtain several continuous ring segments; the ideal grinding curves under different sampling angles are aligned according to the same arc length segmentation rule, and the number of common effective segments is taken as the unified number of ring segments, so that the same number of ring segments under each sampling angle corresponds to the same ring position; For each ring segment, the midpoint of the segment arc length is selected as the representative position, and the local normal vector of the ideal grinding curve at the representative position is calculated. In the neighborhood of the segment corresponding to the original sampling profile line, an effective point set is selected, and the normal projection deviation of each effective point relative to the representative point of the ideal grinding curve is calculated. The maximum positive normal projection deviation is taken as the grinding allowance of the ring segment, and the segment allowance sequence corresponding to each sampling profile is obtained.

[0009] Furthermore, the process of establishing a sliding window along the matrix row direction is specifically as follows: Sliding windows are set along the row direction of the two-dimensional residual distribution matrix. Each sliding window covers a continuous multi-row annular segment and the sampling angle of all columns, corresponding to the continuous annular area of ​​the inner wall of the sealing head. The center position of the window is taken as the representative annular position of the sliding window. A sliding step size smaller than the window length is set between adjacent sliding windows so that adjacent sliding windows retain overlapping areas. The sliding windows slide sequentially along the row direction until they cover the entire annular area to be polished, resulting in several sliding windows.

[0010] Furthermore, the step of extracting the maximum margin within the window range for each angle column of each sliding window as the planned grinding amount for the corresponding angle sector region includes: For each sliding window, extract the submatrix it covers in the two-dimensional margin distribution matrix; calculate the maximum margin value in each column of the submatrix, and use the maximum margin value as the planned grinding amount of the corresponding sampling angle sector area under the current sliding window; arrange the planned grinding amounts of all sliding windows in rows to form a planned grinding amount matrix.

[0011] Furthermore, the step of determining the corresponding number of grinding operations based on the planned grinding volume, and generating the robotic arm trajectory points and posture by combining the position of the ideal grinding curve and the local normal vector, includes: Based on the preset single equivalent removal amount, each element in the planned grinding amount matrix is ​​rounded up and mapped to the number of grinding times under the corresponding sliding window and the corresponding sampling angle; Select the reference ideal grinding curve corresponding to the fixed execution profile, determine the target grinding point in the modeling coordinate system on the reference ideal grinding curve according to the representative ring position of each sliding window, and determine the constant force pneumatic grinding head attitude in the modeling coordinate system according to the local normal vector at the point. The target grinding point and the constant force pneumatic grinding head posture in the modeling coordinate system are transformed to the robot arm base coordinate system through coordinate transformation to obtain the execution point and execution posture that the robot arm can execute.

[0012] Furthermore, the grinding plan results executed collaboratively by the output robotic arm and the turntable include: The target angle of the turntable corresponding to each sampling angle, the execution point and execution posture in the base coordinate system of the robotic arm, and the planned grinding amount and grinding times under the corresponding turntable angle are combined to form the collaborative execution planning result of the robotic arm and the turntable. During the grinding process, the robotic arm moves along the direction of the ring within the fixed execution profile, and the turntable drives the sealing head to rotate around the rotation axis, sequentially rotating the fan-shaped areas at different angles to the position of the fixed execution profile. When the turntable rotates to the corresponding target angle, the planned grinding amount and grinding number of the sliding window corresponding to that target angle are called to execute the grinding operation.

[0013] An adaptive grinding planning system for the inner wall of a nuclear power pressure vessel sealing head includes: The module is used to extract multiple sampling profile lines uniformly at preset angle intervals in the rotation direction based on the three-dimensional curved surface model of the inner wall of the sealing head and the rotational curved surface features of the inner wall of the sealing head. These lines are then used to construct ideal grinding curves that characterize the grinding morphology of the target, thus forming a set of ideal grinding curves. The calculation module is used to divide each ideal grinding curve into equal-length segments along the arc length direction based on the set of ideal grinding curves and calculate the local normal vector at each segment position. It calculates the grinding allowance between the original sampling profile line and the ideal grinding curve along the direction of the local normal vector and arranges the allowance of each segment according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. The generation module is used to establish sliding windows along the row direction of a two-dimensional margin distribution matrix. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The robot arm trajectory points and posture are generated by combining the position of the ideal grinding curve and the local normal vector. The grinding planning result of the robot arm and the turntable is output.

[0014] An adaptive grinding planning device for the inner wall of a nuclear power pressure vessel sealing head, comprising: Memory, used to store computer programs; A processor is used to implement the above-mentioned adaptive grinding planning method for the inner wall of the sealing head of the nuclear power pressure vessel when executing the computer program.

[0015] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement the above-described adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention provides an adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head. First, based on a three-dimensional curved surface model of the sealing head's inner wall, multiple sampling profile lines are extracted using the features of a surface of revolution to construct an ideal grinding curve. Then, the grinding allowance at each position is calculated through arc length segmentation and local normal vectors, forming a two-dimensional allowance distribution matrix. Subsequently, a sliding window is established along the matrix rows, and the maximum allowance of each angle column within each window is extracted as the planned grinding amount. Based on this, the number of grinding operations is determined, and the trajectory of the robotic arm and turntable is generated by combining the curve position and normal vector. This method, through profile line segmentation and a sliding window mechanism, transforms discrete allowance information into continuous and adaptive planning quantities, accurately matching the local grinding requirements of weld overlap and allowance height fluctuations in the weld overlay layer. This method overcomes the shortcomings of relying on experience in manual processes, and achieves quantitative control and quality consistency in the grinding process. It uses measured surfaces rather than theoretical models as a benchmark, and adapts to the process objective of preserving the overall morphology and only repairing local parts. By taking the maximum value of the window, the difference in the distribution of the allowance is mapped to the differentiated grinding intensity, which completes adaptive allocation while ensuring the continuity of the trajectory, and significantly improves the grinding accuracy and efficiency of key nuclear power components. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the implementation of an adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head, as provided in an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the generation of an ideal polishing curve provided in an embodiment of the present invention; Figure 3 A schematic diagram of the pseudocode algorithm for the adaptive route allocation method provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the robotic arm grinding execution system provided in an embodiment of the present invention; Figure 5 This is a core flowchart of an adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head, provided in an embodiment of the present invention. Figure 6 This is a schematic diagram of an adaptive grinding planning system for the inner wall of a nuclear power pressure vessel sealing head, provided in an embodiment of the present invention. Detailed Implementation

[0018] To further understand the content of this invention, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are merely illustrative and not limiting of the invention.

[0019] As described in the background section, Existing manual grinding methods and robotic arm grinding planning technology for large-size curved workpieces can already accomplish the finishing of complex curved surfaces to a certain extent. However, the weld overlay layer on the inner wall of the sealing head of nuclear power pressure vessels has characteristics such as large concave curvature, local fluctuations in weld height, and limited manual operation space. When existing general curved surface grinding methods are applied to this type of object, the following shortcomings still exist: First, manual grinding methods rely heavily on operator experience, and the grinding process is difficult to quantify and control. Grinding the weld overlay layer on the inner wall of the sealing head usually requires operators to repeatedly adjust the weld height, local protrusions, and uneven surfaces within the concave curved surface space. This process mainly relies on manual observation, tactile judgment, and local measurement results to determine the grinding position and degree of removal. Due to the uneven surface morphology of the weld overlay layer, the actual removal requirements at different locations are different. Manual methods cannot uniformly quantify and express the grinding allowance, grinding times, and grinding range at each location, making it difficult to guarantee the repeatability and consistency of the grinding process. Second, path planning methods that do not perform explicit allowance calculations cannot reflect local removal requirements. Existing robotic arm grinding planning methods primarily generate overlay grinding trajectories based on the actual 3D data of the workpiece. Their focus is on enabling the robotic arm to form a continuous, smooth, and executable motion path along the surface to be processed. These methods typically refine the target area using fixed step distances, fixed row spacings, fixed grinding cycles, or empirically set processing parameters, suitable for surface homogenization or general appearance finishing. However, for the weld overlay layer on the inner wall of the sealing head, the weld overlay height, local protrusion degree, and surface undulation state vary at different locations. Generating an overlay path solely based on the surface morphology makes it difficult to accurately describe the actual amount of material to be removed at each location. Thirdly, directly using theoretical models or finished product models as target references lacks adaptability. In planning methods that consider grinding allowance, a common practice is to register actual measurement data with theoretical CAD models, design surfaces, or finished product models, and calculate the deviation of the actual surface relative to the target model. However, the weld overlay layer on the inner wall of the sealing head is not a typical standard part shape. Its grinding target is not simply to restore a complete theoretical surface, but rather to reasonably adjust the local excess height and undulations of the weld overlay surface while preserving the overall surface morphology of the inner wall of the sealing head. Directly using the overall theoretical surface or a general finished product model as the target reference is insufficient to fully adapt to the actual morphology of the weld overlay layer and the needs of local grinding. Fourth, the correspondence between the calculated local allowance and the actual grinding stroke is not direct enough. Existing planning methods that consider allowance can usually obtain local machining allowances through point-to-surface distance, surface normal distance, or tool contact point direction deviation. However, in actual grinding operations, the robotic arm does not perform isolated point machining, but rather a continuous grinding stroke with a certain contact width and working length. For continuous curved surface areas like the weld overlay layer on the inner wall of the sealing head, obtaining only discrete position allowance information still requires further conversion into specific grinding stroke positions, stroke coverage, grinding times, and corresponding processing intensity.Existing general methods are still not clear enough regarding the conversion relationship between the allowance distribution and the actual grinding stroke planning. Fifth, fixed trajectory parameters are difficult to adapt to the uneven distribution of weld overlay allowance. The surface of the weld overlay on the inner wall of the sealing head usually has morphological features such as weld overlap, local excess height, and random fluctuations. The grinding requirements may vary at different cross-sections, locations, and segments. Existing planning methods that use fixed step distances, fixed strokes, fixed grinding times, or uniform processing parameters are difficult to differentiate for different locations based on the actual allowance distribution of the weld overlay, and are also difficult to achieve adaptive allocation of grinding intensity while ensuring trajectory continuity.

[0020] To address the aforementioned issues, this embodiment provides an adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head. This method is applicable to grinding operations of the weld overlay layer on the inner wall of a nuclear power pressure vessel sealing head. Based on the rotational surface features and three-dimensional morphology data of the inner wall of the sealing head, it can uniformly sample multiple curved surface profiles along the angular direction to construct a target grinding benchmark, calculate the grinding allowance for different annular positions and different angular regions, and form a two-dimensional allowance distribution matrix. Furthermore, it generates the grinding stroke, grinding times, and robotic arm execution trajectory based on the two-dimensional allowance distribution matrix, thereby realizing adaptive grinding planning for the weld overlay layer on the inner wall of the sealing head.

[0021] For example, such as Figure 6 As shown, this embodiment provides an adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel, including: Based on the three-dimensional curved surface model of the inner wall of the sealing head, and combined with the rotational curved surface features of the inner wall of the sealing head, multiple sampling profile lines are uniformly extracted in the rotational direction according to a preset angle interval, and ideal grinding curves are constructed to characterize the grinding morphology of the target, forming a set of ideal grinding curves. Based on the set of ideal grinding curves, each ideal grinding curve is divided into segments of equal length along the arc length direction and the local normal vector of each segment position is calculated. The grinding allowance between the original sampling profile line and the ideal grinding curve is calculated along the direction of the local normal vector. The allowance of each segment is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. Based on a two-dimensional margin distribution matrix, a sliding window is established along the matrix row direction. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The position of the ideal grinding curve and the local normal vector are combined to generate the trajectory points and posture of the robotic arm. The grinding planning result of the robotic arm and the turntable is output.

[0022] The adaptive grinding planning method for the inner wall of the sealing head of the nuclear power pressure vessel provided in this embodiment will be further explained below with reference to the accompanying drawings: For example, this embodiment provides an adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head, with the following specific steps: Step S1: Sampling profile extraction and target grinding benchmark construction. Based on the three-dimensional curved surface model of the inner wall of the sealing head. Based on the characteristics of the rotating curved surface of the inner wall of the sealing head, multiple sampling profile lines are extracted evenly at preset angle intervals in the rotation direction. Smooth and continuous ideal grinding curves were constructed to characterize the grinding morphology of the target. This forms an ideal set of polishing curves:

[0023] in, This indicates the number of sampling profile lines participating in the planning within the current surface area to be planned.

[0024] Further, step S1 includes the following sub-steps: Step S101: Extract sampling profile lines. Based on the three-dimensional curved surface model of the inner wall of the sealing head. Based on the geometric characteristics of the inner wall of the sealing head's curved or approximately curved surface of revolution, multiple sampling angles are uniformly selected along the rotation direction at preset angle intervals to obtain an angle sampling set:

[0025] in, Indicates the first Each sampling angle, Indicates the number of angle samples.

[0026] Furthermore, based on various sampling angles, a three-dimensional curved surface model of the inner wall of the sealing head is created. Extract the corresponding sampling profile lines to obtain the set of sampling profile lines participating in the grinding plan:

[0027] in, This indicates the number of sampling profiles participating in the planning within the current surface area to be planned. Indicates the first A sampling profile line. The sampling profile line is used to characterize the local cross-sectional morphology of the inner wall of the sealing head at different rotation angle positions, so that the subsequent grinding plan can simultaneously consider the morphological changes along the profile direction and the morphological differences along the rotation angle direction.

[0028] Specifically, the rotating shaft of the sealing head and the corresponding sampling angle can be used. Section plane and three-dimensional surface model Intersecting, we obtain the sampling profile line in that angular direction. Finally, for each sampling profile line Preprocessing steps, such as invalid point removal, outlier handling, and necessary interpolation completion, are performed to generate a multi-angle profile input for subsequent ideal grinding curve generation, segmented allowance calculation, and two-dimensional allowance matrix construction.

[0029] Step S102: Generation of the ideal polishing curve. This involves generating the preprocessed sampling profiles obtained in step S101. Based on the cross-sectional morphology, a corresponding ideal polishing curve is constructed. For example... Figure 2 As shown, for any sampled profile line, the original profile contour typically contains localized excess height, fluctuations, and unevenness caused by weld overlay. The ideal grinding curve is used to represent the target profile shape that the sampled profile line is expected to achieve after grinding. Figure 2 In the diagram, the solid line represents the weld overlay region outline corresponding to the original sampling profile, while the dashed line represents the ideal grinding curve generated based on the overall profile morphology. This ideal grinding curve is located on the target grinding side of the original sampling profile and maintains a consistent curvature trend with the inner wall surface of the plug head. Simultaneously, through cubic spline smoothness constraints and second derivative sign consistency constraints, local oscillations, reverse curvature, or deviations from the inner wall curvature change trend are avoided in the ideal grinding curve. Therefore, the ideal grinding curve can serve as the target benchmark for subsequent piecewise normal vector calculation and grinding allowance calculation, allowing the deviation between the original profile and the ideal grinding curve to be transformed into a grinding allowance with a clear processing meaning.

[0030] Specifically, let the first The discrete representation of the sampled profile line is:

[0031] in, Indicates the first The first sampling profile line discrete points, This indicates the number of discrete points on the sampling profile.

[0032] Furthermore, based on the sampling profile line Based on the overall morphological change trend, a set of nodes is selected to generate the ideal polishing curve:

[0033] in, The nodes in the sample profile can be composed of representative points from the point set, smoothed feature points, or sampling points selected at preset intervals. This indicates the number of nodes. The set of nodes is used to describe the overall shape trend of the current sampled profile and to provide constraints for the construction of cubic spline curves.

[0034] Based on the node set The first cubic spline curve is used to construct the second cubic spline curve. The ideal polishing curve corresponding to each sampling profile line. Let the ideal polishing curve be represented as:

[0035] in, This is a piecewise cubic polynomial function. Within the interval of adjacent nodes... Within this context, the ideal polishing curve can be expressed as:

[0036] in, , , , For the first The cubic spline coefficients are defined within each segmented interval. To ensure the smooth and continuous ideal grinding curve, continuity constraints on function values, first derivatives, and second derivatives are applied to adjacent segments, such that:

[0037]

[0038]

[0039] Furthermore, to ensure that the ideal grinding curve conforms to the overall bending trend of the inner wall profile of the plug head and to avoid local oscillations or reverse bending, a sign consistency constraint is applied to the second derivative of the ideal grinding curve. Let the sign of the overall bending direction corresponding to the current sampling profile line be: Then the ideal polishing curve satisfies:

[0040] in, This constraint ensures that the second derivative sign of the ideal grinding curve remains consistent within the effective planning interval, maintaining a single curvature trend overall. If the generated cubic spline curve does not satisfy the second derivative sign consistency constraint in a local interval, the curve is corrected by adjusting node positions, reducing the influence of abnormal nodes, adding smoothing constraints, or reselecting representative nodes until the second derivative constraint is satisfied.

[0041] Meanwhile, to ensure that the subsequent grinding allowance calculation has clear processing significance, a positional constraint can be applied to the ideal grinding curve according to the preset removal direction, so that the ideal grinding curve is located on the target grinding side of the original sampling profile line. The direction indicates the direction of the weld overlay height, which can be constrained.

[0042] in, This represents the profile height of the original sampling profile line at the corresponding position. If the locally generated ideal polishing curve exceeds the constraint range, a downward correction or refit is performed on that local interval to maintain a reasonable removal margin relationship between the generated ideal polishing curve and the original sampling profile line.

[0043] After the above processing, the ideal polishing curves corresponding to each sampling profile line can be obtained, forming a set of ideal polishing curves:

[0044] in, Indicates the first The ideal polishing curves corresponding to each sampling angle. The set of ideal polishing curves. Based on the actual morphological characteristics of each sampling profile line, curves are generated. Through constraints such as cubic spline smoothness, second derivative sign consistency, and target grinding side position, the curves can characterize the target profile morphology after grinding of the weld overlay layer on the inner wall of the plug head, providing a basis for subsequent processing.

[0045] Step S2: Grinding allowance calculation and two-dimensional allowance matrix construction. Based on the ideal grinding curve set obtained in Step S1. Each ideal grinding curve is divided into segments of equal length along the arc length direction, and the local normal vector at each segment position is calculated. Then, the morphological deviation between the original sampling profile line and the ideal grinding curve is calculated along the local normal direction to obtain the grinding allowance at different ring positions and different sampling angles. Finally, the allowance of each segment is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix.

[0046] Further, step S2 includes the following sub-steps: Step S201, ideal polishing curve segmentation. Based on the set of ideal polishing curves obtained in step S102. Each ideal grinding curve is parameterized by arc length. For the first... The ideal polishing curve corresponding to each sampling angle Let its arc length parameter be in the form of:

[0047] in, This represents the arc length parameter along the ideal grinding curve. Indicates the first The total arc length of the ideal polishing curve. Based on the preset segment length. The ideal polishing curve is divided into equal lengths. Preferably, the following is taken:

[0048] Then the first An ideal grinding curve can be divided into several continuous annular segments:

[0049] in, Indicates the first The sampling angle at the th sampling angle Each 10mm ring belt segment Indicates the first The number of segments obtained by dividing the ideal grinding curve along the cross-sectional direction.

[0050] To facilitate the subsequent construction of a unified two-dimensional residual matrix, the ideal grinding curves under different sampling angles can be aligned according to the same arc length segmentation rule, and the number of common effective segments is taken as:

[0051] Thus, the first sampling angle at each sampling angle Each segment corresponds to the same annular zone position along the cross-sectional direction on the inner wall of the plug head.

[0052] Step S202: Calculate the piecewise normal vector and residual. For any piecewise normal vector... We select the midpoint of the arc length of this segment as the representative position. Let the arc length interval corresponding to this segment be: Then the position of the representative arc length of this segment is:

[0053] This gives the representative point of this segment on the ideal polishing curve:

[0054] Furthermore, calculate the unit tangent vector of the ideal polishing curve at this representative location:

[0055] Rotate the unit tangent vector by 90° to obtain the unit normal vector at that segment position:

[0056] in, The unit normal vector The preferred direction is the material removal direction pointing from the ideal grinding curve to the original sampling profile line.

[0057] To reduce the impact of single-point derivative fluctuations on the normal direction, segmentation can also be used. Multiple sampling locations are selected within the segment, and the corresponding unit normal vectors are calculated for each location. These normal vectors are then averaged and normalized to obtain the average unit normal vector for that segment.

[0058] in, Indicates segmentation Inner Unit normal vector at each sampling location This indicates the number of sampling points used for normal averaging within this segment.

[0059] Next, in the Original sampling profile line Selection and segmentation The corresponding set of valid points is denoted as:

[0060] in, This represents the set of valid points in the original sampling profile that fall within the neighborhood of the current segment. This indicates the number of points in the valid point set. For any point... Its representative point relative to the ideal polishing curve The projection deviation in the normal direction is:

[0061] The maximum positive normal deviation of the valid points within the current segment is taken as the grinding allowance for that segment:

[0062] in, Indicates the first The 10mm loop position is at the... Grinding allowance in each sampling angle direction. If there are local noise points or abnormal protrusions, the grinding allowance can be adjusted before calculating the maximum normal deviation. Outlier removal is performed on the projection deviation, or local high quantile values ​​are used as the residual for that segment to improve the stability of the residual calculation. Through the above processing, the [number]th [segment] can be obtained. The segmented residual sequence corresponding to each sampling profile line:

[0063] By repeating the above steps for all sampling profiles corresponding to sampling angles, segmented residual values ​​at different annular positions and sampling angles can be obtained.

[0064] Step S203: Construct the two-dimensional residual distribution matrix. There are a total of [number missing] along the profile direction. Each 10mm ring belt segment, along the rotation direction, has a total of [number missing] segments. If we sample the profile lines at each angle, then the segmented residuals obtained in step S202 can form a two-dimensional residual distribution matrix:

[0065] in, , This indicates the number of 10mm ring segments obtained along the cross-sectional direction. This represents the number of angles obtained by uniformly sampling along the direction of rotation. Matrix The Behavior:

[0066] Used to indicate the first The variation of the margin at the position of each 10mm ring within a sector region at different sampling angles. Matrix The Listed as:

[0067] Used to indicate the first Each sampling angle corresponds to the residual change of the sector region along the cross-sectional direction. Therefore, the two-dimensional residual distribution matrix... The margin distribution corresponding to each sampling angle is preserved, enabling it to simultaneously characterize the margin variation characteristics of the inner wall of the plug head along the cross-sectional direction and along the rotation angle direction.

[0068] Furthermore, when a certain position When local data is missing and cannot be directly calculated, interpolation can be performed based on the remaining values ​​at adjacent ring positions or adjacent angle positions. When the remaining value at a certain position is less than the preset effective grinding threshold, the position can be marked as an area that does not need grinding or requires low-intensity grinding.

[0069] After the above processing, a two-dimensional margin distribution matrix is ​​obtained for subsequent adaptive grinding planning. This provides quantitative input for extracting grinding volume in sliding windows, mapping grinding times, and generating robotic arm trajectories.

[0070] Step S3: Adaptive grinding planning based on the two-dimensional margin matrix. Based on the two-dimensional margin distribution matrix obtained in step S2... A sliding window is established along the matrix rows to plan the grinding of the inner wall of the sealing head in terms of rings and angles. Specifically, each sliding window covers several continuous 10mm rings and simultaneously covers all sampling angle columns. For each angle column within the window, the maximum margin within the window's coverage area is taken as the planned grinding amount for that angle's sector area at the current window position. Subsequently, the corresponding grinding number is determined based on the planned grinding amount, and the robot arm trajectory points and posture are generated by combining the position and normal information of the ideal grinding curve, thus forming an adaptive grinding plan result that the robot arm can execute.

[0071] like Figure 3 As shown, this embodiment is based on a two-dimensional residual distribution matrix. Adaptive polishing planning is performed. The algorithm establishes a sliding window along the matrix row direction and determines the annular zone range corresponding to each window. and representative position and the reference ideal grinding curve corresponding to the fixed execution profile. The robot arm trajectory points and posture are generated; simultaneously, the maximum margin is extracted for each sampled angle column within each window. And based on the single equivalent removal amount and effective polishing threshold Determine the number of polishing cycles Finally, the target angle of the turntable will be determined. Robotic arm execution point Execution posture Planning and polishing volume and number of polishing times The combined result of the robotic arm and turntable collaborative execution plan .

[0072] Furthermore, step S3 includes the following sub-steps: Step S301, row-oriented sliding window partitioning. Based on the two-dimensional residual distribution matrix obtained in step S203:

[0073] in, This indicates the number of 10mm ring segments obtained along the cross-sectional direction. This represents the number of angles obtained by uniformly sampling along the direction of rotation. Indicates the first The position of the annulus at the first Grinding allowance in each sampling angle direction. Along the two-dimensional allowance matrix. Establish a sliding window along the row direction. Let the first row be... Initial behavior of a sliding window Window length is Then the range of rows covered by the sliding window is:

[0074] in, Indicates the first The continuous ring band range covered by each sliding window. This indicates the starting position of the sliding window loop. This indicates the number of consecutive rings covered in a single grinding stroke. Correspondingly, the first... The submatrix covered by a sliding window in the two-dimensional residual matrix can be represented as:

[0075] That is, the window coverage matrix From the middle Arrive at the The entire range of angle columns. This submatrix is ​​used to represent the local allowance distribution of different ring positions and different angle sector regions within the current grinding window's effective range.

[0076] Furthermore, let the first The position of the representative annulus corresponding to each sliding window is: The center position of the window is preferably selected as the representative position of the window:

[0077] in, Used to determine the trajectory output position corresponding to the current grinding window. For grinding tools of different lengths or different process requirements, the center position of the window, the starting position of the window, or a specified position within the window can also be used as the representative ring position of the window.

[0078] Step S302: Extracting the window polishing amount and generating the planned polishing amount matrix. For the first... For each sliding window, the maximum margin within the window's coverage area is calculated for each sampling angle column, and this margin is used as the planned grinding amount for the fan-shaped region at the current window position. Specifically, for the The number of sampling angles is as follows: The planned grinding amount corresponding to the current window is:

[0079] in, Indicates the first The sliding window in the first Each sampling angle corresponds to the planned grinding amount within a sector area. The above calculation process is equivalent to: within several continuous rings covered by the current sliding window, finding the highest margin within the window range for each angle sector area, and taking this highest margin as the grinding requirement for that angle area in the current grinding window. This reflects both the effective range of the constant force pneumatic grinding head within its continuous stroke and preserves the margin differences between different angle sector areas.

[0080] Performing the above calculation on all sampled angle columns yields the result of the first... The planning and polishing vector corresponding to each sliding window:

[0081] in, Indicates the first The planned grinding amount distribution of each sliding window across the entire angular sector area. After calculating the grinding amount for the current window, the sliding window is moved along the two-dimensional residual matrix. The window continues to move in the row direction to process subsequent annular regions. Let the sliding step size between adjacent windows be... Then the next window will begin to behave as follows:

[0082] Wherein, the sliding step size The sliding step size can be determined based on the working length of the grinding tool, the desired overlap range, and the continuity requirements of adjacent strokes. Preferably, the sliding step size is smaller than the window length. This ensures that there is some overlap between adjacent sliding windows, thereby guaranteeing a continuous transition between adjacent grinding strokes.

[0083] Repeat the above process of extracting and sliding the polishing amount within the window until the sliding window covers the entire area to be polished, thus obtaining the planned polishing amount matrix:

[0084] in, This represents the number of sliding windows. (Matrix) Each row represents the planned grinding amount of a sector area at different angles at a sliding window position, and each column represents the change in the planned grinding amount along the ring direction at the corresponding sampling angle.

[0085] Step S303: Mapping the number of grinding operations to the robotic arm trajectory. Based on the planned grinding quantity matrix obtained in step S302. The planned grinding amount at each window position and sampling angle is mapped to the corresponding number of grinding passes. Specifically, based on the preset equivalent removal amount per pass, process experience parameters, or grinding test calibration relationships, a mapping relationship from the planned grinding amount to the number of grinding passes is established:

[0086] in, This represents the mapping function from the planned amount of polishing to the number of polishing passes. Indicates the first The first sliding window, the... The number of grinding cycles corresponding to each sampling angle.

[0087] Preferably, the equivalent removal amount corresponding to a single grinding stroke is set as follows: At that time, the amount of grinding can be determined according to the plan. Determine the required number of polishing passes. The number of polishing passes can be determined using the following method:

[0088] in, This indicates rounding up. For different pressure levels, different abrasive belt conditions, or different process stages, a segmented mapping method can also be used to determine the number of grinding passes and processing intensity.

[0089] Furthermore, during the actual grinding process, the robotic arm does not need to move within the entire rotation angle range of the inner wall of the sealing head. Instead, it drives the constant-force pneumatic grinding head to move along the belt direction within the preset execution profile. The sealing head is supported by a turntable with an encoder and rotates around a preset axis, causing different angle sector areas to rotate sequentially to the execution profile positions corresponding to the robotic arm. Therefore, the first... The margin column corresponding to each sampling angle does not indicate that the robotic arm needs to move to that angle position, but rather indicates the planned grinding amount and number of grinding operations required under the current execution profile when the turntable rotates to the corresponding angle.

[0090] Specifically, let the fixed execution profile corresponding to the robotic arm be the reference profile, and its corresponding ideal grinding curve be denoted as: ; According to the Each sliding window corresponds to a position of the ring. In the corresponding ideal grinding curve The target grinding point is determined in the modeling coordinate system. Let this point be:

[0091] in, Indicates the first The sliding window is located at a point in the modeling coordinate system on the fixed execution profile. Simultaneously, the attitude of the constant-force pneumatic grinding head at the end of the robotic arm in the modeling coordinate system is determined based on the local normal vector at that point. This attitude is represented as a rotation matrix of the end-tool coordinate system relative to the modeling coordinate system, denoted as: ; in, Used to represent the constant force pneumatic grinding head relative to the target grinding point in the modeling coordinate system. The contact posture allows the grinding tool to contact the surface of the weld overlay layer on the inner wall of the sealing head along the local normal direction of the ideal grinding curve.

[0092] To enable the aforementioned trajectory points and postures to be used in actual robotic arm execution, the trajectory information in the modeling coordinate system needs to be transformed to the robotic arm's base coordinate system or the actual machining coordinate system. Let the coordinate transformation relationship from the modeling coordinate system to the robotic arm's base coordinate system be:

[0093] in, This represents the coordinates of a point in the modeling coordinate system. This represents the coordinates of a point in the robot arm's base coordinate system. This represents the rotation matrix from the modeling coordinate system to the robot arm's base coordinate system. This represents the corresponding translation vector. Therefore, the target grinding point in the modeling coordinate system can be represented... Convert to execution point in the robot arm's base coordinate system:

[0094] in, This represents the trajectory points required for actual execution within a fixed execution profile.

[0095] Correspondingly, the end-effector attitude rotation matrix in the modeling coordinate system Also according to the rotation matrix The transformation is performed to obtain the rotation matrix in the robot arm's base coordinate system:

[0096] Subsequently, based on the attitude expression format adopted by the robotic arm controller, the end-effector attitude rotation matrix is... Converted into attitude parameters that can be executed by the robotic arm, denoted as:

[0097] in, This represents the transformation relationship from the attitude rotation matrix to the robot arm attitude parameters, which can be Euler angles, RPY angles, or attitude angle parameters defined by the robot arm controller. This indicates the execution posture of the constant force pneumatic grinding head at the end of the robotic arm at this trajectory point.

[0098] Furthermore, let the first Each sampling angle The corresponding target angle of the turntable is Then when the turntable encoder feedback angle reaches When, it indicates the first The first angular sector area has been rotated to the fixed execution profile position of the robotic arm. At this time, for the first... A sliding window, using the corresponding elements in the two-dimensional planning grinding quantity matrix. and the corresponding number of polishing times As the current execution parameter.

[0099] This allows for the formation of the grinding plan required for the coordinated execution of the robotic arm and the turntable:

[0100] in, Indicates the first The target angle of the turntable corresponds to each angular sector area. This represents the trajectory points within a fixed execution profile in the robot arm's base coordinate system. Indicates the corresponding execution posture. Indicates the current turntable angle at the first... The amount of planning and polishing corresponding to each window. This indicates the corresponding number of grinding cycles. Based on the grinding plan, the robotic arm moves along the belt direction within a fixed execution profile; the turntable rotates different angled sector areas to the execution profile position sequentially according to the encoder angle. By coordinating the robotic arm's fixed profile movement with the turntable's angle switching, adaptive grinding operations are completed for different angled areas of the sealing head's inner wall.

[0101] In the above execution method, the trajectory points and end effector posture of the robotic arm are mainly determined by the position of the ring belt within the fixed execution profile. The turntable angle is used to select the angular sector area to be processed and the corresponding planned grinding amount and number of grinding cycles. Therefore, this embodiment does not achieve full-area grinding by directly moving the robotic arm within the full rotation range of the inner wall of the sealing head, but rather achieves sequential processing of different angular areas by coordinating the fixed profile movement of the robotic arm with the rotational positioning of the turntable.

[0102] For example, in order to implement the above-mentioned adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel, this embodiment also provides an adaptive grinding planning system for the inner wall of the sealing head of a nuclear power pressure vessel, specifically including: like Figure 4 As shown, the implementation system includes a robotic arm, a constant force pneumatic grinding head, a turntable with an encoder, and a control processing unit.

[0103] The robotic arm carries a constant-force pneumatic grinding head and, based on the adaptive grinding plan, drives the grinding head to move along the belt direction within a preset execution profile, bringing it to the corresponding grinding position on the inner wall of the sealing head. Switching between different angle zones is achieved by a rotary table with an encoder rotating the sealing head workpiece.

[0104] The constant force pneumatic grinding head is installed at the end of the robotic arm and is used to contact the surface of the weld overlay layer on the inner wall of the sealing head during the grinding process, and to provide stable contact pressure, so as to complete the grinding operation of the corresponding area according to the planned trajectory and grinding number of times.

[0105] The encoder-equipped turntable is used to carry the workpiece to be ground and to drive the workpiece to rotate around a preset axis. The encoder is used to output the turntable angle information in real time, so that the control processing unit can determine the angle sector area to be processed and call the corresponding planned grinding amount and grinding times.

[0106] The control processing unit is connected to the robotic arm, the constant-force pneumatic grinding head, and the encoder-equipped turntable, and can be composed of an industrial computer, a PLC controller, a robotic arm control cabinet, or a combination thereof. It receives the three-dimensional morphological data of the inner wall of the sealing head, executes the adaptive grinding planning process described in Example 1, and coordinates the various execution components to complete the grinding operation. Specifically, the control processing unit generates a planned grinding amount matrix and a grinding number matrix based on the two-dimensional margin distribution matrix, generates the robotic arm trajectory points and end-effector posture by combining them with the reference ideal grinding curve corresponding to the fixed execution profile, and selects the grinding parameters corresponding to the current angle region based on the turntable angle information fed back by the encoder.

[0107] In this embodiment, the robotic arm moves along the annular direction within a fixed execution profile. The turntable sequentially rotates different angled fan-shaped areas to the execution profile position according to the target angle. The constant-force pneumatic grinding head completes the grinding according to the set pressure and planned number of grinding cycles. Through the coordinated operation of the robotic arm's fixed profile movement, the turntable's rotational positioning, and the constant-force pneumatic grinding head's stable contact pressure, adaptive grinding operations can be achieved for different annular positions and angled areas of the weld overlay layer on the inner wall of the sealing head.

[0108] For example, such as Figure 6 As shown, this embodiment also provides another adaptive grinding planning system for the inner wall of the sealing head of a nuclear power pressure vessel, including: The module is used to extract multiple sampling profile lines uniformly at preset angle intervals in the rotation direction based on the three-dimensional curved surface model of the inner wall of the sealing head and the rotational curved surface features of the inner wall of the sealing head. These lines are then used to construct ideal grinding curves that characterize the grinding morphology of the target, thus forming a set of ideal grinding curves. The calculation module is used to divide each ideal grinding curve into equal-length segments along the arc length direction based on the set of ideal grinding curves and calculate the local normal vector at each segment position. It calculates the grinding allowance between the original sampling profile line and the ideal grinding curve along the direction of the local normal vector and arranges the allowance of each segment according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. The generation module is used to establish sliding windows along the row direction of a two-dimensional margin distribution matrix. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The robot arm trajectory points and posture are generated by combining the position of the ideal grinding curve and the local normal vector. The grinding planning result of the robot arm and the turntable is output.

[0109] The present invention also provides an adaptive grinding planning device for the inner wall of the sealing head of a nuclear power pressure vessel, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the adaptive grinding planning method for the inner wall of the sealing head of the nuclear power pressure vessel.

[0110] The present invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the adaptive grinding planning method for the inner wall of the sealing head of the nuclear power pressure vessel.

[0111] When the processor executes the computer program, it implements the above-mentioned steps of adaptive grinding planning for the inner wall of the sealing head of the nuclear power pressure vessel. For example, based on the three-dimensional curved surface model of the inner wall of the sealing head, combined with the rotational curved surface features of the inner wall of the sealing head, multiple sampling profile lines are uniformly extracted in the rotational direction according to a preset angle interval, and ideal grinding curves for characterizing the target grinding morphology are constructed respectively, forming a set of ideal grinding curves. Based on the set of ideal grinding curves, each ideal grinding curve is divided into segments of equal length along the arc length direction and the local normal vector of each segment position is calculated. The grinding allowance between the original sampling profile line and the ideal grinding curve is calculated along the direction of the local normal vector. The allowance of each segment is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. Based on a two-dimensional margin distribution matrix, a sliding window is established along the matrix row direction. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The position of the ideal grinding curve and the local normal vector are combined to generate the trajectory points and posture of the robotic arm. The grinding planning result of the robotic arm and the turntable is output.

[0112] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing preset functions. These instruction segments describe the execution process of the computer program in the adaptive grinding planning device for the inner wall of the nuclear power pressure vessel sealing head. For example, the computer program can be divided into a construction module, a calculation module, and a generation module; the specific functions are as follows: The construction module is used to extract multiple sampling profile lines uniformly at preset angle intervals in the rotation direction based on the three-dimensional curved surface model of the inner wall of the sealing head, combined with the rotational curved surface features of the inner wall of the sealing head, to construct ideal grinding curves for characterizing the target grinding morphology, forming a set of ideal grinding curves; The calculation module is used to divide each ideal grinding curve into equal-length segments along the arc length direction based on the set of ideal grinding curves and calculate the local normal vector at each segment position, and calculate the relationship between the original sampling profile line and the ideal grinding curve along the direction of the local normal vector. The grinding allowance between grinding curves is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. The generation module is used to establish sliding windows along the matrix row direction based on the two-dimensional allowance distribution matrix. Each sliding window covers several continuous ring segments and all sampling angle columns. For each sliding window, the maximum allowance of each angle column within the window range is extracted as the planned grinding amount of the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The robot arm trajectory points and posture are generated by combining the position of the ideal grinding curve and the local normal vector. The grinding planning result of the robot arm and the turntable is output.

[0113] The adaptive grinding and planning device for the inner wall of the nuclear power pressure vessel plug head can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above are examples of adaptive grinding and planning devices for the inner wall of nuclear power pressure vessel plug heads and do not constitute a limitation on such devices. The device may include more components than described above, or combine certain components, or use different components. For example, the adaptive grinding and planning device for the inner wall of nuclear power pressure vessel plug heads may also include input / output devices, network access devices, buses, etc.

[0114] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or any conventional processor. The processor is the control center for the adaptive grinding design of the inner wall of the nuclear power pressure vessel sealing head, connecting various parts of the entire adaptive grinding design equipment for the inner wall of the nuclear power pressure vessel sealing head through various interfaces and lines.

[0115] The memory can be used to store the computer program and / or modules. The processor realizes various functions of the adaptive grinding planning device for the inner wall of the nuclear power pressure vessel sealing head by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory.

[0116] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.). The data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0117] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the adaptive grinding planning method for the inner wall of a nuclear power pressure vessel sealing head.

[0118] If the modules / units integrated by the adaptive grinding planning system for the inner wall of the nuclear power pressure vessel sealing head are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0119] Based on this understanding, the present invention can implement all or part of the processes in the above-mentioned adaptive grinding planning method for the inner wall of the nuclear power pressure vessel sealing head, or it can be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-mentioned adaptive grinding planning method for the inner wall of the nuclear power pressure vessel sealing head. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form, etc.

[0120] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0121] It should be noted that the content contained in the computer-readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0122] Compared with existing grinding planning methods, this invention provides an adaptive grinding planning method and related apparatus for the inner wall of the sealing head of a nuclear power pressure vessel, which has the following advantages: First, this invention, based on the rotational curved surface characteristics of the inner wall of the sealing head of a nuclear power pressure vessel, uniformly extracts multiple sampling profile lines along the angular direction, and constructs corresponding ideal grinding curves according to the actual morphology of each profile line, forming a target grinding reference. This method does not directly rely on manual experience, fixed templates, or overall theoretical surfaces to determine the grinding target, but rather combines the actual three-dimensional morphology of the inner wall of the sealing head to generate a target reference suitable for grinding the weld overlay layer. This improves the adaptability between the target grinding reference and the actual surface to be ground, providing a more reasonable geometric basis for subsequent allowance calculation and trajectory planning.

[0123] Secondly, this invention calculates the morphological deviation between the original profile line and the ideal grinding curve along the local normal direction of the ideal grinding curve, and constructs a two-dimensional allowance distribution matrix for the grinding allowance at different ring positions and different sampling angles. This method can transform the local allowance height and morphological undulations of the weld overlay surface into a quantifiable allowance expression, avoiding grinding based solely on a covering trajectory or fixed processing parameters; at the same time, the two-dimensional allowance matrix can simultaneously characterize the allowance changes of the inner wall of the sealing head along the profile direction and the rotation angle direction, providing a more complete quantitative basis for the material removal requirements at different locations.

[0124] Third, this invention establishes a row-oriented sliding window planning method based on a two-dimensional margin distribution matrix. Each sliding window covers several continuous rings, and the maximum margin within the window range is extracted in each sampling angle column as the planned grinding amount for the corresponding position. This method can transform discrete margin information into planned grinding amounts corresponding to continuous grinding strokes, making a more direct correspondence between margin distribution and actual grinding windows, stroke coverage, and grinding times, which is beneficial to improving the matching degree between grinding planning results and actual processing.

[0125] Fourth, based on the planning results obtained from the sliding window, this invention further generates the corresponding number of grinding passes, robotic arm trajectory points, and posture information, and executes them collaboratively through the robotic arm, constant force pneumatic grinding head, encoder-equipped turntable, and control processing unit. By moving the robotic arm within a fixed execution profile and cooperating with the encoder-equipped turntable for angle switching, the processing intensity can be adaptively allocated according to the actual allowance differences in different ring positions and angle regions. This reduces the risk of local under-grinding or over-grinding caused by fixed step distances, fixed strokes, and fixed grinding passes, and improves the stability, consistency, and engineering feasibility of the grinding process for the weld overlay layer on the inner wall of the sealing head.

[0126] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

[0127] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for adaptive grinding planning of the inner wall of a sealing head for nuclear power pressure vessels, characterized in that, include: Based on the three-dimensional curved surface model of the inner wall of the sealing head, and combined with the rotational curved surface features of the inner wall of the sealing head, multiple sampling profile lines are uniformly extracted in the rotational direction according to a preset angle interval, and ideal grinding curves are constructed to characterize the grinding morphology of the target, forming a set of ideal grinding curves. Based on the set of ideal grinding curves, each ideal grinding curve is divided into segments of equal length along the arc length direction and the local normal vector of each segment position is calculated. The grinding allowance between the original sampling profile line and the ideal grinding curve is calculated along the direction of the local normal vector. The allowance of each segment is arranged according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. Based on a two-dimensional margin distribution matrix, a sliding window is established along the matrix row direction. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The position of the ideal grinding curve and the local normal vector are combined to generate the trajectory points and posture of the robotic arm. The grinding planning result of the robotic arm and the turntable is output.

2. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 1, characterized in that, The process of constructing the ideal polishing curve set specifically includes: Based on the three-dimensional curved surface model of the inner wall of the sealing head, multiple sampling angles are uniformly selected along the rotation direction according to a preset angle interval to obtain an angle sampling set; the cutting plane passing through the rotation axis of the sealing head and corresponding to each sampling angle intersects with the three-dimensional curved surface model of the inner wall of the sealing head to extract the corresponding sampling profile lines; invalid point removal, outlier processing and interpolation completion preprocessing are performed on each sampling profile line to obtain a sampling profile line set; For each preprocessed sampling profile line, a set of nodes is selected to describe the overall morphological trend of the profile. Based on the node set, a corresponding ideal grinding curve is constructed using cubic spline curves. During the construction process, smoothness constraints are applied to the continuity of function values, first derivatives, and second derivatives. At the same time, second derivative sign consistency constraints and target grinding side position constraints are applied to ensure that the ideal grinding curve maintains a single curvature trend and is located on the target grinding side of the original sampling profile line, thus forming the set of ideal grinding curves.

3. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 1, characterized in that, The calculation process for the polishing allowance specifically includes: Each ideal grinding curve in the set of ideal grinding curves is parameterized by arc length and divided into equal length segments according to a preset segment length to obtain several continuous ring segments; the ideal grinding curves under different sampling angles are aligned according to the same arc length segmentation rule, and the number of common effective segments is taken as the unified number of ring segments, so that the same number of ring segments under each sampling angle corresponds to the same ring position; For each ring segment, the midpoint of the segment arc length is selected as the representative position, and the local normal vector of the ideal grinding curve at the representative position is calculated. In the neighborhood of the segment corresponding to the original sampling profile line, an effective point set is selected, and the normal projection deviation of each effective point relative to the representative point of the ideal grinding curve is calculated. The maximum positive normal projection deviation is taken as the grinding allowance of the ring segment, and the segment allowance sequence corresponding to each sampling profile is obtained.

4. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 1, characterized in that, The process of establishing a sliding window along the row direction of the matrix is ​​as follows: Sliding windows are set along the row direction of the two-dimensional residual distribution matrix. Each sliding window covers a continuous multi-row annular segment and the sampling angle of all columns, corresponding to the continuous annular area of ​​the inner wall of the sealing head. The center position of the window is taken as the representative annular position of the sliding window. A sliding step size smaller than the window length is set between adjacent sliding windows so that adjacent sliding windows retain overlapping areas. The sliding windows slide sequentially along the row direction until they cover the entire annular area to be polished, resulting in several sliding windows.

5. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 4, characterized in that, The step of extracting the maximum margin within the window range for each angle column of each sliding window as the planned grinding amount for the corresponding angle sector area includes: For each sliding window, extract the submatrix it covers in the two-dimensional margin distribution matrix; calculate the maximum margin value in each column of the submatrix, and use the maximum margin value as the planned grinding amount of the corresponding sampling angle sector area under the current sliding window; arrange the planned grinding amounts of all sliding windows in rows to form a planned grinding amount matrix.

6. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 5, characterized in that, The process of determining the corresponding number of grinding operations based on the planned grinding volume, and generating the robotic arm trajectory points and posture by combining the position of the ideal grinding curve and the local normal vector, includes: Based on the preset single equivalent removal amount, each element in the planned grinding amount matrix is ​​rounded up and mapped to the number of grinding times under the corresponding sliding window and the corresponding sampling angle; Select the reference ideal grinding curve corresponding to the fixed execution profile, determine the target grinding point in the modeling coordinate system on the reference ideal grinding curve according to the representative ring position of each sliding window, and determine the constant force pneumatic grinding head attitude in the modeling coordinate system according to the local normal vector at the point. The target grinding point and the constant force pneumatic grinding head posture in the modeling coordinate system are transformed to the robot arm base coordinate system through coordinate transformation to obtain the execution point and execution posture that the robot arm can execute.

7. The adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel according to claim 6, characterized in that, The output of the grinding plan, executed collaboratively by the robotic arm and the turntable, includes: The target angle of the turntable corresponding to each sampling angle, the execution point and execution posture in the base coordinate system of the robotic arm, and the planned grinding amount and grinding times under the corresponding turntable angle are combined to form the collaborative execution planning result of the robotic arm and the turntable. During the grinding process, the robotic arm moves along the direction of the ring within the fixed execution profile, and the turntable drives the sealing head to rotate around the rotation axis, sequentially rotating the fan-shaped areas at different angles to the position of the fixed execution profile. When the turntable rotates to the corresponding target angle, the planned grinding amount and grinding number of the sliding window corresponding to that target angle are called to execute the grinding operation.

8. An adaptive grinding planning system for the inner wall of a nuclear power pressure vessel sealing head, characterized in that, include: The module is used to extract multiple sampling profile lines uniformly at preset angle intervals in the rotation direction based on the three-dimensional curved surface model of the inner wall of the sealing head and the rotational curved surface features of the inner wall of the sealing head. These lines are then used to construct ideal grinding curves that characterize the grinding morphology of the target, thus forming a set of ideal grinding curves. The calculation module is used to divide each ideal grinding curve into equal-length segments along the arc length direction based on the set of ideal grinding curves and calculate the local normal vector at each segment position. It calculates the grinding allowance between the original sampling profile line and the ideal grinding curve along the direction of the local normal vector and arranges the allowance of each segment according to the ring position and sampling angle to construct a two-dimensional allowance distribution matrix. The generation module is used to establish sliding windows along the row direction of a two-dimensional margin distribution matrix. Each sliding window covers several continuous ring segments and all sampled angle columns. For each sliding window, the maximum margin of each angle column within the window range is extracted as the planned grinding amount for the corresponding angle sector area. The corresponding grinding number is determined according to the planned grinding amount. The robot arm trajectory points and posture are generated by combining the position of the ideal grinding curve and the local normal vector. The grinding planning result of the robot arm and the turntable is output.

9. An adaptive grinding and planning device for the inner wall of a nuclear power pressure vessel sealing head, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the adaptive grinding planning method for the inner wall of the sealing head of a nuclear power pressure vessel as described in any one of claims 1-7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the adaptive grinding planning method for the inner wall of the sealing head of the nuclear power pressure vessel as described in any one of claims 1-7.

Images