A grinding disc CNC machining path control method and system

Abstract

The application relates to a kind of grinding disc CNC machining path control method and system.The method is applied to machining center, machining center has a finishing tool and at least one high-precision sensor, the method comprises the following steps: obtaining the topographic data of grinding disc surface, local temperature data and cutting force data;According to the comprehensive analysis of topographic data, local temperature data and cutting force data, the possibility of protrusion on the surface of grinding disc is evaluated;According to the result of evaluation, the preset machining strategy is selected, wherein the machining strategy comprises: when the possibility of protrusion is higher than the first preset threshold, adjust the finishing tool to remove the finishing protrusion;When the possibility of protrusion is lower than the second preset threshold, cooling intervention / non-cutting scanning cooling is applied to promote the subsidence of protrusion.The application can accurately identify the protrusion properties of the surface of grinding disc, accurately cut for the real finishing protrusion, and promote the natural subsidence of heat-induced protrusion through cooling intervention.

Description

Technical Field

[0001] This application relates to the field of grinding disc processing, and in particular to a method and system for controlling the CNC machining path of grinding discs. Background Technology

[0002] In semiconductor wafer chemical mechanical polishing (CMP), the surface flatness of the polishing pad directly determines the wafer quality, and regular dressing of the polishing pad is crucial to maintaining its performance. Traditional dressing relies on preset paths, while more advanced methods introduce online measurement and closed-loop control. Specifically, a high-precision sensor next to the dressing tool scans the surface topography in real time. After the data is fed back to the CNC controller, it is compared with the ideal plane to calculate the deviation, dynamically adjusting the cutting depth and trajectory to prioritize cutting off those "bumps" that are higher than the target plane. This dynamically generates an optimal processing path, ultimately achieving an extremely high flatness on the entire polishing pad surface.

[0003] However, in actual production environments, the grinding disc itself is not a completely homogeneous material. It is made by mixing and pressing abrasive particles with a matrix material such as resin, resulting in uneven microscopic hardness and density. Cutting parameters set based on the assumption of material homogeneity will cause a sharp increase in cutting resistance in hard spots, leading to minute elastic deformation of the tool and tool holder, also known as "tool deflection" or "tool runout." Although the deformation is small, it is fatal under micrometer-level precision requirements, causing the actual position of the tool tip to deviate from the position required by the controller. At the same time, the matrix material of the grinding disc is very sensitive to temperature changes. This high temperature point will cause local thermal expansion, forming a temporary, heat-induced microbulge on the surface of the grinding disc. Intense friction generates local high temperatures, causing a temporary thermally induced bump on the surface of the resin-based grinding disc. However, the sensor cannot distinguish whether this bump is a geometric defect inherent in the material itself or merely a temporary deformation caused by heat during processing, and will mistakenly identify it as a geometric defect.

[0004] When the controller cannot distinguish the type of protrusion, it will only deepen the cut according to the logic instruction of "remove if it is high". When machining the hard spot again, the greater resistance aggravates the tool wobble and heat generation, forming a larger thermal protrusion; after the sensor detects it, it will cause a deeper cut, falling into a vicious cycle. In the end, the grinding disc will have chatter marks, burn spots and periodic ripples, and be completely scrapped.

[0005] To address this, we propose a method and system for controlling the CNC machining path of a grinding disc. Summary of the Invention

[0006] This application provides a CNC machining path control method and system for grinding discs, which at least solves the problem of misjudgment and vicious cycle caused by material inhomogeneity and local thermal effects during the existing grinding disc dressing process, resulting in a decline in the surface quality of the grinding disc or even scrapping it.

[0007] In a first aspect, this application provides a CNC machining path control method for a grinding disc, which is applied to a machining center, the machining center having a dressing tool and at least one high-precision sensor, the method comprising:

[0008] Acquire surface morphology data, local temperature data, and cutting force data of the grinding disc;

[0009] Based on a comprehensive analysis of the morphology data, the local temperature data, and the cutting force data, the possibility that the protrusions on the surface of the grinding disc are repair protrusions is assessed.

[0010] Based on the evaluation results, a preset processing strategy is selected, wherein the processing strategy includes:

[0011] When the probability that the protrusion is a trimming protrusion is higher than a first preset threshold, the cutting depth and movement trajectory of the trimming tool are adjusted to remove the trimming protrusion;

[0012] When the probability that the protrusion is a repair protrusion is lower than a second preset threshold, cooling intervention / non-cutting scanning cooling is applied to promote the disappearance of the protrusion.

[0013] Optionally, the step of comprehensively analyzing the morphology data, the local temperature data, and the cutting force data, and evaluating the possibility that the protrusion on the surface of the grinding disc is a repair protrusion according to a preset judgment rule, includes:

[0014] When the topography data, the local temperature data, and the cutting force data do not completely match the preset judgment rules, the fuzzy protrusions on the surface of the grinding disc are identified and the dressing tool is adjusted to the micro probe mode. The adjustment of the dressing tool to the micro probe mode includes: raising the dressing tool and finely adjusting the dressing tool downward at a preset speed until the dressing tool makes slight contact with the surface of the grinding disc and reaches a preset low contact force threshold. The dressing tool applies a micro-vibration with a preset frequency and a small amplitude to the grinding disc.

[0015] Collect local temperature response data of the contact area of ​​the trimming tool and force sensor response data of the trimming tool under the micro-vibration and slight contact.

[0016] The force sensor response data is subjected to spectrum analysis and adaptive bandpass filtering to purify the force sensor response signal. Based on the purified force sensor response signal, the damping characteristics of the local area of ​​the grinding disc are analyzed and inferred.

[0017] The thermophysical properties of the local area of ​​the grinding disc are analyzed and inferred based on the local temperature response data.

[0018] Based on the damping characteristics and the thermophysical characteristics, the nature of the ambiguity protrusion is determined, and the possibility that the protrusion on the surface of the grinding disc is a trimming protrusion is evaluated based on the properties.

[0019] Optionally, identifying the fuzzy protrusions on the grinding disc surface when the topography data, the local temperature data, and the cutting force data do not completely match the preset judgment rules includes:

[0020] The inconsistency pattern among the height data, temperature data, and force data of the protrusions on the surface of the grinding disc was identified.

[0021] Calculate the first similarity between the height data pattern, the temperature data pattern, and the force data pattern and a preset trimming protrusion pattern, and the second similarity between the height data pattern, the temperature data pattern, and the force data pattern and a preset thermal protrusion pattern.

[0022] When both the first similarity and the second similarity are within a preset similarity range, the blurry protrusions on the surface of the grinding disc are identified.

[0023] Optionally, identifying the blurred protrusions on the surface of the grinding disc when both the first similarity and the second similarity are within a preset similarity interval includes:

[0024] Acquire information on the material properties of the grinding disc, the wear condition of the dressing tool, and the processing stage;

[0025] The preset similarity interval is dynamically adjusted based on the material properties, the wear state of the dressing tool, and the processing stage information.

[0026] When both the first similarity and the second similarity are within the preset similarity range after dynamic adjustment, the blurry protrusions on the surface of the grinding disc are identified.

[0027] Optionally, the step of performing spectral analysis and adaptive bandpass filtering on the force sensor response data to purify the force sensor response signal includes:

[0028] Spectral analysis of the force sensor response data identifies the preset frequency components applied by the dressing tool, as well as the harmonic and subharmonic components generated by the nonlinear response of the grinding disc material.

[0029] A multi-band filter is constructed based on the preset frequency component, the harmonic component, and the subharmonic component, wherein the multi-band filter has multiple independent passbands, and the multiple independent passbands correspond to the preset frequency component, the harmonic component, and the subharmonic component, respectively;

[0030] The force sensor response data is imported into the multi-band filter for multi-band filtering to filter out other frequency noise components and retain only the preset frequency component, the harmonic component, and the subharmonic component.

[0031] Time-domain analysis is performed on the multi-band filtered signal to distinguish the inherent nonlinear response of the grinding disc material from interference signals, wherein the interference signals include environmental noise and mechanical vibration.

[0032] Optionally, the step of performing time-domain analysis on the multi-band filtered signal to distinguish between the inherent nonlinear response of the grinding disc material and interference signals includes:

[0033] While the dressing tool maintains slight contact with the surface of the grinding disc and applies a slight vibration, the dressing tool is controlled to scan layer by layer in the vertical direction with a preset micron-level step size, and force sensor response data corresponding to different depth layers are collected.

[0034] Spectral analysis of the force sensor response data collected at different depths is performed to identify the characteristic frequencies, amplitudes, and phases of the material's inherent nonlinear response at different depths, and a multilayer material response characteristic map is constructed.

[0035] The signal after multi-band filtering is imported into the response feature spectrum of the multilayer material for comparison, and the components in the signal that highly match the response feature spectrum of the multilayer material are identified to distinguish the inherent nonlinear response of the material from the interference signal.

[0036] Optionally, the identification process involves obtaining the characteristic frequencies, amplitudes, and phases of the inherent nonlinear response of the material corresponding to each depth layer, and constructing a multilayer material response feature map, including:

[0037] The inherent nonlinear response characteristics of the material identified at different depth layers are standardized.

[0038] Identify continuous gradient regions where the feature differences between adjacent depth layers are less than a third preset threshold, and perform weighted averaging or local curve fitting on the feature data of the continuous gradient regions to generate transition feature representations.

[0039] Identify discontinuous regions with feature differences greater than the third preset threshold, divide them into regions with different material properties, and extract representative material-inherent nonlinear response feature vectors for each of the material property regions;

[0040] By integrating the transition feature representation and the representative material's inherent nonlinear response feature vector, the response feature map of the multilayer material is constructed.

[0041] Secondly, this application provides a CNC machining path control system for a grinding disc, the system comprising:

[0042] The data acquisition module is used to acquire morphological data, local temperature data, and cutting force data of the grinding disc surface;

[0043] The probability assessment module is used to comprehensively analyze the morphology data, the local temperature data, and the cutting force data to assess the probability that the protrusion on the surface of the grinding disc is a repair protrusion.

[0044] The strategy selection module is used to select a preset processing strategy based on the evaluation results, wherein the processing strategy includes:

[0045] When the probability that the protrusion is a trimming protrusion is higher than a first preset threshold, the cutting depth and movement trajectory of the trimming tool are adjusted to remove the trimming protrusion;

[0046] When the probability that the protrusion is a repair protrusion is lower than a second preset threshold, cooling intervention / non-cutting scanning cooling is applied to promote the disappearance of the protrusion.

[0047] Thirdly, this application provides a computer device including a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the grinding disc CNC machining path control method provided in the first aspect above.

[0048] Fourthly, this application provides a computer-readable storage medium storing computer instructions for causing a computer to execute the grinding disc CNC machining path control method provided in the first aspect above.

[0049] Compared with related technologies, the CNC machining path control method and system for grinding discs provided in this application have at least the following technical advantages:

[0050] By acquiring surface morphology data, local temperature data, and cutting force data of the grinding disc, and comprehensively analyzing the above multi-dimensional data, the likelihood of the protrusion on the grinding disc surface being a repair protrusion is assessed according to preset judgment rules. Based on this, a preset processing strategy is selected according to the assessment results. That is, when the probability of the protrusion being a repair protrusion is high, the cutting depth and motion trajectory of the repair tool are adjusted to remove the repair protrusion; when the probability of the protrusion being a repair protrusion is low, cooling intervention or non-cutting scanning cooling is applied to promote the disappearance of the protrusion.

[0051] The technical solution of this application, by introducing multi-dimensional data comprehensive analysis and intelligent evaluation mechanism, can accurately identify the true nature of the protrusion. For genuine repair protrusions, the system will perform precise cutting; while for heat-induced protrusions, cooling intervention will be used to promote their natural regression, avoiding unnecessary cutting and heat accumulation. This effectively solves the vicious cycle problem caused by the microscopic inhomogeneity of the grinding disc material, the local thermal effect during the repair process, and the inability of the control system to distinguish between repair protrusions and heat-induced protrusions in the existing technology.

[0052] The technical solution of this application can improve the accuracy and efficiency of grinding disc dressing, avoid damage and scrap of grinding discs caused by misjudgment in traditional methods, thereby extending the service life of grinding discs, reducing production costs, and ultimately improving the surface quality of high-precision processed products such as semiconductor wafers.

[0053] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0054] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0055] Figure 1 This is a flowchart illustrating a CNC machining path control method for a grinding disc according to an exemplary embodiment.

[0056] Figure 2 This is a flowchart illustrating step S20 according to an exemplary embodiment.

[0057] Figure 3 This is a flowchart illustrating step S201 according to an exemplary embodiment.

[0058] Figure 4 This is a flowchart illustrating step S2013 according to an exemplary embodiment.

[0059] Figure 5This is a flowchart illustrating step S203 according to an exemplary embodiment.

[0060] Figure 6 This is a flowchart illustrating step S2034 according to an exemplary embodiment.

[0061] Figure 7 This is a block diagram illustrating a CNC machining path control system for a grinding disc according to an exemplary embodiment. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0063] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any creative effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0064] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0065] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0066] In related technologies, the cutting parameters set by the grinding disc based on the assumption of material homogeneity can cause a sudden increase in cutting resistance in hard spot areas, resulting in minute elastic deformation of the tool and tool holder, also known as "tool deflection" or "tool runout." Although the deformation is small, it is fatal under micron-level precision requirements, causing the actual position of the tool tip to deviate from the position required by the controller. At the same time, the matrix material of the grinding disc is very sensitive to temperature changes. This high temperature point will cause local thermal expansion, resulting in a temporary, heat-induced micro-protrusion on the surface of the grinding disc. Intense friction generates local high temperatures, causing a temporary thermally induced protrusion on the surface of the resin-based grinding disc. However, the sensor cannot distinguish whether this protrusion is a geometric defect inherent in the material itself or merely a temporary deformation caused by heat during processing, and will mistakenly identify it as a geometric defect.

[0067] When the controller cannot distinguish the type of protrusion, it will only deepen the cut according to the logic instruction of "remove if it is high". When machining the hard spot again, the greater resistance aggravates the tool wobble and heat generation, forming a larger thermal protrusion; after the sensor detects it, it will cause a deeper cut, falling into a vicious cycle. In the end, the grinding disc will have chatter marks, burn spots and periodic ripples, and be completely scrapped.

[0068] Based on the above, embodiments of the present invention provide a method and system for controlling the CNC machining path of a grinding disc, which will be described in detail below with reference to specific embodiments and accompanying drawings.

[0069] Example 1

[0070] This invention provides a method for controlling the CNC machining path of a grinding disc. Figure 1 This is a flowchart illustrating a CNC machining path control method for a grinding disc according to an exemplary embodiment. Figure 1 As shown, this method is applied to a machining center, which has a dressing tool and at least one high-precision sensor. The method includes:

[0071] Step S10: Obtain the surface morphology data, local temperature data, and cutting force data of the grinding disc;

[0072] Step S20: Based on a comprehensive analysis of morphology data, local temperature data, and cutting force data, assess the possibility that the protrusions on the surface of the grinding disc are repair protrusions;

[0073] Step S30: Based on the evaluation results, select a preset processing strategy, wherein the processing strategy includes:

[0074] When the probability that the protrusion is a trimming protrusion is higher than the first preset threshold, adjust the cutting depth and movement trajectory of the trimming tool to remove the trimming protrusion;

[0075] When the probability of a protrusion being a repair protrusion is lower than the second preset threshold, a cooling intervention / non-cutting scan cooling is applied to promote the protrusion to disappear.

[0076] In the above embodiment, the grinding disc removes material and smooths the surface through friction between the grinding particles and the workpiece surface. The dressing tool is used to dress the surface of the grinding disc, typically a diamond dressing tool, which restores the surface morphology and cutting performance of the grinding disc. In this embodiment, morphology data refers to the geometric feature information such as the surface height and roughness of the grinding disc obtained by a high-precision displacement sensor (such as a laser displacement sensor); local temperature data refers to the temperature distribution information of a local area on the surface of the grinding disc obtained by a high-precision temperature sensor (such as an infrared thermal imager); cutting force data refers to the magnitude and direction information of the force borne by the dressing tool during the cutting process obtained by a high-precision force sensor (such as a piezoelectric force sensor), while the first preset threshold and the second preset threshold are reference values ​​used to determine the nature of the protrusion.

[0077] Meanwhile, the cooling intervention / non-cutting scanning cooling in the above technical solution refers to reducing the local temperature of the grinding disc by means of spraying coolant, blowing cold air or using dressing tools to scan without cutting, thereby removing heat and promoting the disappearance of heat-induced protrusions.

[0078] The embodiments described above in this application, by introducing a comprehensive analysis of the surface morphology data, local temperature data, and cutting force data of the grinding disc, accurately assess the possibility that the protrusion is a trimming protrusion, thereby distinguishing between trimming protrusions and thermally induced protrusions on the surface of the grinding disc. Subsequently, different processing strategies are selected based on the assessment results, namely, trimming protrusions are removed by cutting, and thermally induced protrusions are subjected to cooling intervention or non-cutting scanning cooling. This method improves the trimming accuracy and efficiency of the grinding disc, extends the service life of the grinding disc, and reduces production costs, providing a more reliable grinding disc trimming solution for high-precision processing fields such as semiconductor wafer manufacturing.

[0079] In this embodiment, it is considered that in the actual processing, due to the complexity of the grinding disc material, the dynamic changes in the processing environment, and the noise interference of sensor data, the morphology data, local temperature data, and cutting force data may not completely match the preset judgment rules. Figure 2 This is a flowchart illustrating step S20 according to an exemplary embodiment. (Refer to the attached document.) Figure 2 Step S20 includes:

[0080] S201. When the morphology data, local temperature data, and cutting force data do not completely match the preset judgment rules, identify the fuzzy protrusions on the grinding disc surface and adjust the dressing tool to micro probe mode.

[0081] The adjustment of the trimming tool to micro probe mode includes: raising the trimming tool and finely adjusting the trimming tool downward at a preset speed until the trimming tool makes slight contact with the surface of the grinding disc and reaches a preset low contact force threshold; the trimming tool applies a micro-vibration with a preset frequency and a small amplitude to the grinding disc.

[0082] S202. Collect local temperature response data of the contact area of ​​the trimming tool and force sensor response data of the trimming tool under slight vibration and slight contact.

[0083] S203. Perform spectrum analysis and adaptive bandpass filtering on the force sensor response data to purify the force sensor response signal, and analyze and inversely deduce the damping characteristics of the local area of ​​the grinding disc based on the purified force sensor response signal.

[0084] S204. Analyze and infer the thermophysical properties of the local area of ​​the grinding disc based on the local temperature response data.

[0085] S205. Based on damping characteristics and thermophysical characteristics, determine the nature of the fuzzy protrusion, and assess the possibility that the protrusion on the grinding disc surface is a repair protrusion based on its properties.

[0086] In the technical solutions of the above embodiments, when the morphology data, local temperature data, and cutting force data do not completely match the preset judgment rules, it means that there is uncertainty or contradiction between these initial data patterns and typical trimmed protrusion patterns or thermally induced protrusion patterns, making it difficult to directly classify the nature of the protrusion.

[0087] At this point, it is necessary to identify any ambiguous protrusions on the surface of the grinding disc and initiate a refined probing process. As a preferred implementation, the trimming tool is adjusted to microprobe mode. In this mode, the trimming tool is first raised to a safe height and then finely adjusted downwards at a preset, slow speed until it makes slight contact with the grinding disc surface. This slight contact is precisely controlled by reaching a preset low contact force threshold, aiming to avoid any damage to the grinding disc surface while ensuring the sensitivity of subsequent data acquisition. While maintaining slight contact, the trimming tool applies micro-vibrations of a preset frequency and small amplitude to the grinding disc. These micro-vibrations can non-destructively excite the material response in a localized area of ​​the grinding disc, providing an excitation signal for characterization analysis.

[0088] Subsequently, local temperature response data of the contact area of ​​the dressing tool and force sensor response data of the dressing tool under micro-vibration and slight contact were simultaneously acquired. The local temperature response data is used to reflect the thermal behavior of the grinding disc material under micro-vibration excitation, such as heat conduction and thermal expansion; the force sensor response data is used to capture the mechanical behavior of the grinding disc material under mechanical excitation, such as elasticity, plasticity, and viscoelasticity.

[0089] Next, to extract useful information from the complex force sensor response data, spectral analysis and adaptive bandpass filtering are performed to identify the main frequency components and harmonics in the signal. The adaptive bandpass filtering dynamically adjusts the filtering parameters based on these identified frequency components to filter out interference signals such as environmental noise and mechanical vibrations to the greatest extent possible, thus purifying the force sensor response signal. The purified force sensor response signal can more accurately reflect the inherent mechanical properties of the grinding disc material, and the damping characteristics of local areas of the grinding disc can then be analyzed and inferred from this signal. Damping characteristics are an important parameter for measuring a material's ability to dissipate mechanical energy and can be used to distinguish between trimmed protrusions (typically with low damping) and thermally induced protrusions (which may exhibit higher viscoelastic damping). Simultaneously, based on the collected local temperature response data, the thermophysical properties of local areas of the grinding disc, such as the local coefficient of thermal expansion and thermal conductivity, can be analyzed and inferred to determine whether the protrusion is caused by local heat accumulation.

[0090] Ultimately, based on the damping and thermophysical properties obtained through reverse engineering, the nature of the fuzzy protrusions can be determined more accurately, and the likelihood of the protrusions on the grinding disc surface being trimming protrusions can be assessed based on these properties. For example, if a local area exhibits low damping and a low coefficient of thermal expansion, it is more likely to be a trimming protrusion; conversely, if it exhibits high damping and a high coefficient of thermal expansion, it is more likely to be a thermally induced protrusion.

[0091] In the embodiments described above, when there is ambiguity in the initial data evaluation, a micro-probe detection mechanism is introduced. The micro-vibration of the trimming tool excites the material response in a local area of ​​the grinding disc, and mechanical and thermal response data are collected by a highly sensitive sensor. Subsequently, the mechanical and thermal response data are subjected to refined spectral analysis and filtering to purify the signals that reflect the inherent properties of the material, thereby inferring the damping and thermophysical properties of the local area of ​​the grinding disc. The trimming protrusions and thermally induced protrusions are distinguished based on the damping and thermophysical properties, which can effectively make up for the shortcomings of relying solely on surface morphology, macroscopic temperature, and cutting force data for judgment, thereby solving the problem of ambiguity in the judgment of protrusion properties when the data is not completely matched.

[0092] In one example, suppose that during the grinding process, the system detects a local protrusion using a topography sensor. The height data of this protrusion is close to the preset pattern for trimming protrusions, but the local temperature sensor shows that the temperature in this area is slightly higher than normal, and the cutting force sensor data fluctuates significantly, not perfectly matching the typical trimming or thermally induced protrusion patterns, making it impossible for the system to definitively determine its nature. In this case, the system identifies the protrusion as an ambiguous protrusion and activates a micro-probe mode. The trimming tool is precisely adjusted to make slight contact with the protrusion surface and applies a micro-amplitude vibration at a preset frequency. Simultaneously, the high-precision force sensor and local temperature sensor begin acquiring response data. After spectral analysis and adaptive bandpass filtering, the purified force sensor response signal is used to infer the damping characteristics of the protrusion area. The results show that the damping coefficient in this area is relatively high; and the local temperature response data analysis indicates that the coefficient of thermal expansion in this area is also relatively high. Based on the damping and thermophysical characteristics, the system determines that the ambiguous protrusion is more likely a thermally induced protrusion than a trimming protrusion. Ultimately, a cooling intervention strategy was chosen instead of adjusting the cutting depth for removal. After a period of cooling, the protrusion successfully subsided, avoiding unnecessary cutting and protecting the surface of the grinding disc.

[0093] In this embodiment, it is considered that in the actual implementation process, the condition of "not completely matched" alone may not be enough to accurately and consistently identify fuzzy protrusions, thus affecting the accurate selection of subsequent processing strategies. Figure 3 This is a flowchart illustrating step S201 according to an exemplary embodiment. (Refer to the attached document.) Figure 3 Step S201 includes:

[0094] S2011. Identify inconsistencies between height, temperature, and force data of protrusions on the grinding disc surface. In this embodiment, the height, temperature, and force data obtained from the grinding disc surface are analyzed to identify contradictions or inconsistencies in these data when characterizing the properties of the protrusions. For example, a protrusion may exhibit significant geometric features in terms of height, but its local temperature or cutting force response may not fully conform to the characteristics of a typical trimmed protrusion. This inconsistency pattern is a preliminary basis for judging the ambiguity of the protrusion properties.

[0095] S2012, calculate the first similarity between the height data pattern, temperature data pattern, and force data pattern and the preset trimming protrusion pattern, and the second similarity between the height data pattern, temperature data pattern, and force data pattern and the preset thermally induced protrusion pattern. In this embodiment, the matching degree between the current protrusion's data pattern and the preset patterns of two typical protrusions (trimming protrusions and thermally induced protrusions) is quantitatively compared. The preset trimming protrusion pattern is typically characterized by significant height, relatively small temperature changes, and direct cutting force response; while the preset thermally induced protrusion pattern may be characterized by insignificant height changes but abnormal local temperature increases, and cutting force response exhibiting thermal softening or thermal expansion characteristics. Furthermore, the first and second similarities in this embodiment can be calculated using various algorithms, such as correlation coefficients, Euclidean distance, cosine similarity, or probability values ​​output by a machine learning-based classifier.

[0096] S2013. When both the first similarity and the second similarity are within a preset similarity interval, an ambiguous protrusion on the surface of the grinding disc is identified. In this embodiment, the preset similarity interval is a pre-defined numerical range used to define the degree of similarity between the current protrusion's data pattern and both the trimmed protrusion pattern and the thermally induced protrusion pattern, which is in an intermediate, ambiguous state. This means that the protrusion does not fully conform to the characteristics of either the trimmed protrusion or the thermally induced protrusion, and is therefore identified as an ambiguous protrusion. For example, if the first similarity (similarity to the trimmed protrusion pattern) is 0.6, the second similarity (similarity to the thermally induced protrusion pattern) is 0.5, and the preset similarity interval is [0.4, 0.7], then the protrusion is identified as an ambiguous protrusion. Specifically,

[0097] Figure 4 This is a flowchart illustrating step S2013 according to an exemplary embodiment. In this embodiment, refer to the attached... Figure 4 Step S2013 includes:

[0098] S20131. Obtain information on the material properties of the grinding disc, the wear condition of the dressing tool, and the processing stage.

[0099] In this embodiment, the material properties of the grinding disc include physical parameters such as its hardness, elastic modulus, thermal conductivity, and coefficient of thermal expansion. These parameters directly affect the deformation response and heat conduction characteristics of the grinding disc when heated or subjected to stress. For example, materials with different hardness may produce different degrees of deformation under the same cutting force, thus affecting the morphology data; materials with different thermal conductivity will also have different local temperature distributions under the same heat source.

[0100] Furthermore, it is considered that worn dressing tools may cause abnormal cutting force data or changes in contact characteristics with the grinding disc surface in micro-probe mode. The wear condition of a dressing tool refers to the degree of wear on its cutting edge or working surface during long-term use.

[0101] Finally, the processing stage information can indicate whether the grinding disc is currently in the roughing, semi-finishing, or finishing stage. Different processing stages have different requirements for the surface quality of the grinding disc, and the processing parameters (such as cutting speed and feed rate) will also differ. These differences may lead to different formation mechanisms and manifestations of protrusions.

[0102] S20132. Based on material properties, wear status of dressing tools, and processing stage information, dynamically adjust the preset similarity interval;

[0103] In this embodiment, based on the acquired material properties of the grinding disc, the wear state of the dressing tool, and the processing stage information, the similarity interval used to identify ambiguous protrusions is adjusted in real-time or near real-time using a pre-established model, lookup table, or machine learning algorithm. For example, when the grinding disc material is softer, its sensitivity to thermal deformation may be higher, and the similarity interval of the thermal protrusion pattern may be appropriately widened; when the dressing tool is severely worn, the cutting force signal it generates may be more unstable, and the similarity interval of the force data pattern may need to be adjusted to adapt to this uncertainty; in the finishing stage, the surface quality requirements are higher, and the similarity interval may be tightened to more accurately identify potential ambiguous protrusions.

[0104] S20133. When both the first similarity and the second similarity are within the preset similarity interval after dynamic adjustment, identify the fuzzy protrusions on the surface of the grinding disc.

[0105] In the technical solution of the above embodiments, by introducing the identification of inconsistency patterns between height data, temperature data, and force data, and by performing similarity calculations with preset trimming protrusion patterns and thermally induced protrusion patterns, the complex characteristics of protrusions on the surface of the grinding disc can be captured more precisely. When the initial data does not completely match the preset judgment rules, the similarity between the current protrusion and two typical protrusion patterns is quantified, and it is determined whether these similarities fall within the preset similarity interval. Furthermore, based on the actual material characteristics of the grinding disc, the wear degree of the trimming tool, and the current processing stage, the judgment criteria for identifying ambiguous protrusions are intelligently adjusted, which can effectively identify ambiguous protrusions whose properties are between trimming protrusions and thermally induced protrusions, thereby avoiding misjudgments that may be caused by simple binary judgment.

[0106] The above embodiments of this application can improve the recognition accuracy and robustness of ambiguous protrusions on the surface of the grinding disc. Compared with the rough judgment that relies solely on "imperfect matching", by introducing inconsistent pattern recognition and dual similarity calculation, the ambiguity of the protrusions can be defined more accurately, thereby avoiding the risk of misjudging thermal protrusions as trimming protrusions for cutting, or misjudging trimming protrusions as thermal protrusions for cooling.

[0107] In one example, suppose that during the grinding process of a grinding disc, the data acquisition module acquires data on a protrusion at a certain point on the grinding disc surface. The height data of this protrusion shows that it is slightly higher than the surrounding surface, but not to the significant extent of a typical trimmed protrusion; the local temperature data shows that its temperature is slightly increased, but not reaching the critical value of a typical thermally induced protrusion; the cutting force data shows that its response is in between, not completely conforming to any typical pattern. That is, an inconsistency pattern is identified among the height data, temperature data, and force data.

[0108] Subsequently, the system calculates the first similarity between the data pattern of the protrusion and a preset trimming protrusion pattern, for example, 0.6; simultaneously, it calculates the second similarity between the data pattern of the protrusion and a preset thermally induced protrusion pattern, for example, 0.55. If the preset similarity interval is set to [0.4, 0.7], since both the first similarity of 0.6 and the second similarity of 0.55 fall within this interval, the system will accurately identify the protrusion as a blurred protrusion on the surface of the grinding disc. The machining center then adjusts the trimming tool to micro-probe mode for more in-depth physical property analysis to ultimately determine its properties and select the most suitable processing strategy. This method ensures accurate protrusion classification even when data is unclear, thereby optimizing the grinding disc processing.

[0109] Figure 5 This is a flowchart illustrating step S203 according to an exemplary embodiment. In this embodiment, refer to the attached... Figure 5 Step S203 includes:

[0110] S2031. Spectral analysis of force sensor response data to identify the preset frequency components applied by the dressing tool, as well as the harmonic and subharmonic components generated by the nonlinear response of the grinding disc material.

[0111] In this embodiment, when performing spectral analysis on the force sensor response data, methods such as Fast Fourier Transform (FFT) can be used to convert the time-domain signal into a frequency-domain signal. Through spectral analysis, the preset frequency component applied by the dressing tool can be accurately identified. This preset frequency component is the fundamental frequency of the dressing tool's micro-amplitude vibration. Simultaneously, due to the nonlinear characteristics of the grinding disc material, when subjected to micro-amplitude vibration excitation, harmonic components with frequencies that are integer multiples of the fundamental frequency and subharmonic components with frequencies that are fractional multiples of the fundamental frequency will be generated.

[0112] S2032. Construct a multi-band filter based on preset frequency components, harmonic components, and subharmonic components. The multi-band filter has multiple independent passbands, and the multiple independent passbands correspond to the preset frequency components, harmonic components, and subharmonic components, respectively.

[0113] In this embodiment, the multi-band filter is a digital or analog filter capable of simultaneously processing multiple signals within a specific frequency range. It is designed to have multiple independent passbands, each precisely configured to allow the corresponding frequency component to pass while suppressing other frequency components. For example, a passband can be set for a preset frequency component, and a separate passband can be set for each identified harmonic and subharmonic component.

[0114] S2033. Import the force sensor response data into the multi-band filter for multi-band filtering to filter out other frequency noise components and retain only the preset frequency components, harmonic components and subharmonic components.

[0115] In this embodiment, the force sensor response data is imported into a multi-band filter for multi-band filtering processing. The purpose is to effectively filter out all frequency noise components other than the target frequency component (i.e., the preset frequency component, harmonic components, and subharmonic components). These other frequency noise components include mechanical vibrations originating from environmental vibrations, electromagnetic interference, or other non-grinding disc material responses. Multi-band filtering significantly improves the signal-to-noise ratio, thereby obtaining a pure signal that reflects the true response of the grinding disc material.

[0116] S2034. Perform time-domain analysis on the signal after multi-band filtering to distinguish the inherent nonlinear response of the grinding disc material from interference signals, including environmental noise and mechanical vibration. In this embodiment, although multi-band filtering has filtered out most of the noise, some interference signals may still exist that overlap with the target frequency components or are difficult to completely separate by frequency domain methods. For example, some environmental noise or mechanical vibration may be close to harmonic or subharmonic frequencies. Time-domain analysis can further identify and separate these residual interferences by observing the signal envelope, phase changes, or specific patterns, thereby more accurately extracting the inherent nonlinear response characteristics of the grinding disc material. Specifically,

[0117] In this embodiment, Figure 6 This is a flowchart illustrating step S2034 according to an exemplary embodiment. (Refer to the attached document.) Figure 6 Step S2034 further includes:

[0118] S20341. When the dressing tool is in slight contact with the surface of the grinding disc and a small vibration is applied, the dressing tool is controlled to scan layer by layer in the vertical direction with a preset micron-level step size, and force sensor response data corresponding to different depth layers are collected.

[0119] In this embodiment, the micrometer-level step size refers to an extremely small scanning depth step size, ranging from several micrometers to tens of micrometers, to ensure precise detection of the response characteristics of the grinding disc material at different depths. Through this precise layer-by-layer scanning, force sensor response data corresponding to different depth layers can be collected, thereby reflecting the mechanical response of the grinding disc material at different depths.

[0120] S20342. Spectral analysis of force sensor response data collected at different depths is performed to identify the characteristic frequencies, amplitudes, and phases of the inherent nonlinear responses of materials at different depths, and a multilayer material response characteristic map is constructed.

[0121] In this embodiment, the inherent nonlinear response of the material is a unique physical property exhibited by the grinding disc material under micro-amplitude vibration and slight contact. Its characteristic frequency, amplitude, and phase are key parameters characterizing this unique physical property. By identifying this unique physical property, a multi-layer material response feature map is constructed, which includes the inherent nonlinear response characteristics of the grinding disc material at different depths.

[0122] S20343. The signal after multi-band filtering is imported into the multilayer material response feature map for comparison to identify components in the signal that highly match the multilayer material response feature map, thereby distinguishing between the material's inherent nonlinear response and interference signals. In this embodiment, the comparison can distinguish between the grinding disc material's inherent nonlinear response and interference signals. Highly matched components are considered the material's inherent response, while mismatched or poorly matched components are classified as interference signals, such as environmental noise and mechanical vibration. Specifically, step S20343 includes:

[0123] Standardize the inherent nonlinear response characteristics of materials identified at different depth layers;

[0124] In this embodiment, the characteristic frequency, amplitude, and phase data obtained from different depth layers are subjected to a unified scaling transformation or normalization process to eliminate inconsistencies in dimensions or numerical ranges between different depth layers caused by factors such as measurement conditions, material density, or differences in sensor response. This ensures that the characteristic data from different depth layers are comparable in subsequent analysis and comparison, thereby improving the accuracy of feature recognition.

[0125] Identify continuous gradient regions where the feature differences between adjacent depth layers are less than a third preset threshold, and perform weighted averaging or local curve fitting on the feature data of the continuous gradient regions to generate transition feature representations.

[0126] In this embodiment, the difference in the standardized inherent nonlinear response characteristics (e.g., characteristic frequency, amplitude, or phase) of the material between adjacent depth layers is compared. When this difference is less than a preset third threshold, these depth layers are considered to belong to a continuously changing region. Subsequently, the feature data of the continuous gradient region is processed by weighted averaging or local curve fitting. The purpose is to smooth the data, remove measurement noise, and generate a transition feature representation that accurately reflects the gradual change trend of the material properties in that region. For example, moving average, Gaussian filtering, or polynomial fitting methods can be used to process these data to better characterize the continuous changes of the material in these regions.

[0127] Identify discontinuous regions with feature differences greater than a third preset threshold, divide them into regions with different material properties, and extract representative material-inherent nonlinear response feature vectors for each material property region;

[0128] In this embodiment, when the difference in the inherent nonlinear response characteristics of materials between adjacent depth layers exceeds a preset third threshold, it indicates that there are significant changes in material properties between these depth layers, which may represent different material layers or structural interfaces. Therefore, these regions are divided into different material property regions, and a representative material inherent nonlinear response feature vector is extracted from each material property region to capture the typical nonlinear response characteristics of the material within that region. For example, the average, median, or center point obtained through cluster analysis of all feature data within that region can be taken as the representative feature vector; the purpose is to accurately distinguish the different material layers or structures existing inside the grinding disc.

[0129] By integrating transition feature representations and representative material intrinsic nonlinear response feature vectors, a multilayer material response feature map is constructed.

[0130] In this embodiment, the transition feature representation obtained by weighted averaging or local curve fitting is combined with the representative material intrinsic nonlinear response feature vector extracted from discontinuous regions to generate a multilayer material response feature map. This multilayer material response feature map can not only reflect the continuous gradual change characteristics of the material, but also clearly identify the discontinuous interfaces between different material layers.

[0131] In the technical solution of the above embodiment, step S2032 of this embodiment eliminates the dimensional differences between data by standardizing the inherent nonlinear response characteristics of materials identified at different depth layers, thus ensuring the comparability of features. Simultaneously, by identifying discontinuous regions with significant feature differences and extracting representative feature vectors, the interfaces of different material layers or structures within the grinding disc can be defined. Finally, by integrating the transition feature representation and representative feature vectors, a multi-layer material response feature map is constructed, thereby overcoming the potential accuracy limitations of traditional methods when processing complex material structures.

[0132] In one example, suppose the grinding disc is composed of two different materials, A and B, where material A occupies the surface layer and material B occupies the core layer, with a thin transition layer between A and B. After the dressing tool performs a vertical scan in micrometer-level steps and acquires force sensor response data, the inherent nonlinear response characteristics of the materials identified at different depth layers are first standardized to eliminate measurement errors and dimensional differences. Subsequently, the system identifies continuous gradient regions within the surface material A region where the characteristic differences between adjacent depth layers are less than a third preset threshold, and performs local curve fitting on the feature data of these regions to generate a transition feature representation reflecting the microscopic changes within material A.

[0133] Next, at the interface between material A and the transition layer, and at the interface between the transition layer and material B, due to significant changes in material properties, the feature differences exceed the third preset threshold. These regions are identified as discontinuous regions and divided into different material property regions. Representative material intrinsic nonlinear response feature vectors are extracted from the material A region, the transition layer region, and the material B region, respectively.

[0134] Finally, these transition feature representations and representative material intrinsic nonlinear response feature vectors are integrated to construct a multilayer material response feature map that can accurately reflect the material properties of the surface layer, transition layer and core layer of the grinding disc.

[0135] In summary, the embodiments of this application identify and retain the preset frequency components, harmonic components, and subharmonic components applied by the trimming tool, and construct a multi-band filter for targeted filtering, significantly improving the signal-to-noise ratio and purity of the signal. Furthermore, through subsequent time-domain analysis, interference signals similar to the inherent response of the material are further distinguished and eliminated, ensuring that the obtained signal can truly and accurately reflect the nonlinear characteristics of the grinding disc material. Through this step-by-step and refined signal processing mechanism, a pure signal reflecting the characteristics of the grinding disc material can be accurately extracted from complex force sensor response data, achieving efficient and precise purification of the force sensor response signal.

[0136] In summary, the CNC machining path control method for grinding discs provided in this invention acquires surface morphology data, local temperature data, and cutting force data of the grinding disc, and comprehensively analyzes these multi-dimensional data. Based on preset judgment rules, it assesses the likelihood that a protrusion on the grinding disc surface is a trimming protrusion. Furthermore, based on the assessment results, it selects a preset machining strategy: when the probability of a protrusion being a trimming protrusion is high, it adjusts the cutting depth and motion trajectory of the trimming tool to remove the trimming protrusion; when the probability is low, it applies cooling intervention or non-cutting scanning cooling to promote protrusion regression. The technical solution of this application can improve the accuracy and efficiency of grinding disc trimming, avoid damage and scrapping of the grinding disc caused by misjudgment in traditional methods, thereby extending the service life of the grinding disc, reducing production costs, and ultimately improving the surface quality of high-precision processed products such as semiconductor wafers.

[0137] Example 2

[0138] Embodiment 2 of the present invention provides a CNC machining path control system for a grinding disc. Figure 7 This is a block diagram illustrating a CNC machining path control system for a grinding disc according to an exemplary embodiment. Figure 7 As shown, the system includes:

[0139] Data acquisition module 1 is used to acquire morphology data, local temperature data, and cutting force data of the grinding disc surface;

[0140] The probability assessment module 2 is used to comprehensively analyze the morphology data, local temperature data and cutting force data to assess the probability that the protrusion on the surface of the grinding disc is a repair protrusion.

[0141] Strategy selection module 3 is used to select a preset processing strategy based on the evaluation results, wherein the processing strategies include:

[0142] When the probability that the protrusion is a trimming protrusion is higher than the first preset threshold, adjust the cutting depth and movement trajectory of the trimming tool to remove the trimming protrusion;

[0143] When the probability of a protrusion being a repair protrusion is lower than the second preset threshold, a cooling intervention / non-cutting scan cooling is applied to promote the protrusion to disappear.

[0144] The CNC machining path control system for the grinding disc provided in Embodiment 2 of this invention acquires surface morphology data, local temperature data, and cutting force data of the grinding disc. Based on the comprehensive analysis of these multi-dimensional data, it assesses the likelihood that a protrusion on the grinding disc surface is a trimming protrusion according to preset judgment rules. Furthermore, based on the assessment results, it selects a preset machining strategy: when the probability of a protrusion being a trimming protrusion is high, it adjusts the cutting depth and motion trajectory of the trimming tool to remove the trimming protrusion; when the probability is low, it applies cooling intervention or non-cutting scanning cooling to promote the protrusion's disappearance. The technical solution of this application can improve the accuracy and efficiency of grinding disc trimming, avoid damage and scrapping of the grinding disc caused by misjudgment in traditional methods, thereby extending the service life of the grinding disc, reducing production costs, and ultimately improving the surface quality of high-precision processed products such as semiconductor wafers.

[0145] Example 3

[0146] Embodiment 3 of the present invention provides a computer device, including a memory and a processor, which are interconnected and communicate with each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the grinding disc CNC machining path control method provided in Embodiment 1.

[0147] Example 4

[0148] Embodiment 4 of the present invention provides a computer-readable storage medium storing computer instructions, which are used to cause a computer to execute the grinding disc CNC machining path control method provided in Embodiment 1.

[0149] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0150] In a possible implementation, the present invention can also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, is used to cause the terminal device to execute the grinding disc CNC machining path control method of Embodiment 1.

[0151] The program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for controlling a CNC machining path of a grinding disc, which is applied to a machining center, wherein a dressing tool is provided in the machining center, characterized in that, The method comprises: acquiring topography data, local temperature data and cutting force data of the surface of the grinding disc; comprehensively analyzing the topography data, the local temperature data and the cutting force data to evaluate the possibility of the protrusion on the surface of the grinding disc being a dressing protrusion; selecting a preset machining strategy according to the evaluation result, wherein the machining strategy comprises: when the possibility of the protrusion being a dressing protrusion is higher than a first preset threshold, adjusting the cutting depth and the motion trajectory of the dressing tool to remove the dressing protrusion; when the possibility of the protrusion being a dressing protrusion is lower than a second preset threshold, applying cooling intervention / non-cutting scanning cooling to promote the subsidence of the protrusion; wherein the comprehensively analyzing the topography data, the local temperature data and the cutting force data to evaluate the possibility of the protrusion on the surface of the grinding disc being a dressing protrusion comprises: when the topography data, the local temperature data and the cutting force data do not completely match a preset judgment rule, identifying a fuzzy protrusion on the surface of the grinding disc and adjusting the dressing tool to a micro-probe mode, wherein the adjusting the dressing tool to the micro-probe mode comprises: lifting the dressing tool and finely adjusting the dressing tool downward at a preset speed until the dressing tool slightly contacts the surface of the grinding disc and reaches a preset low contact force threshold, and the dressing tool applies a micro-amplitude vibration of a preset frequency and a micro-amplitude to the grinding disc; collecting local temperature response data of the contact area of the dressing tool and force sensor response data of the dressing tool under the micro-amplitude vibration and slight contact; spectrum analyzing and adaptive band filtering the force sensor response data to purify the force sensor response signal, and analyzing and backstepping the thermal physical properties of the local area of the grinding disc according to the purified force sensor response signal; analyzing and backstepping the thermal physical properties of the local area of the grinding disc according to the local temperature response data; judging the properties of the fuzzy protrusion based on the damping properties and the thermal physical properties, and evaluating the possibility of the protrusion on the surface of the grinding disc being a dressing protrusion according to the properties.

2. The grinding disc CNC machining path control method of claim 1, wherein, when the topography data, the local temperature data and the cutting force data do not completely match a preset judgment rule, identifying a fuzzy protrusion on the surface of the grinding disc, comprises: identifying that there is an inconsistent mode among the height data, the temperature data and the force data of the protrusion on the surface of the grinding disc; calculating a first similarity between the height data mode, the temperature data mode and the force data mode and a preset dressing protrusion mode, and a second similarity between the height data mode, the temperature data mode and the force data mode and a preset thermal protrusion mode; when the first similarity and the second similarity are both within a preset similarity interval, identifying the fuzzy protrusion on the surface of the grinding disc.

3. The grinding disc CNC machining path control method of claim 2, wherein, when the first similarity and the second similarity are both within a preset similarity interval, identifying the fuzzy protrusion on the surface of the grinding disc, comprises: acquiring the material properties of the grinding disc, the dressing tool wear state and the machining stage information; According to the material characteristics, the dressing tool wear state and the processing stage information, the preset similarity interval is dynamically adjusted; When the first similarity and the second similarity are both in the preset similarity interval after the dynamic adjustment, the surface blur protrusion of the grinding disc is identified.

4. The grinding disc CNC machining path control method of claim 1, wherein, The force sensor response data is subjected to frequency spectrum analysis and adaptive band-pass filtering processing, and the force sensor response signal is purified, including: The force sensor response data is subjected to frequency spectrum analysis, and a preset frequency component applied by the dressing tool and a harmonic component and a sub-harmonic component generated by the nonlinear response of the grinding disc material are identified and obtained; According to the preset frequency component, the harmonic component and the sub-harmonic component, a multi-band filter is constructed, wherein the multi-band filter has a plurality of independent passbands, and the plurality of independent passbands correspond to the preset frequency component, the harmonic component and the sub-harmonic component respectively; The force sensor response data is introduced into the multi-band filter for multi-band filtering processing, and other frequency component noises are filtered out, and only the preset frequency component, the harmonic component and the sub-harmonic component are retained; The signal after the multi-band filtering processing is subjected to time domain analysis, and the inherent nonlinear response of the grinding disc material and the interference signal are distinguished, wherein the interference signal includes environmental noise and mechanical vibration.

5. The grinding disc CNC machining path control method of claim 4, wherein, The signal after the multi-band filtering processing is subjected to time domain analysis, and the inherent nonlinear response of the grinding disc material and the interference signal are distinguished, including: When the dressing tool and the grinding disc surface are in slight contact and apply micro-vibration, the dressing tool is controlled to scan layer by layer in the vertical direction with a preset micron-level step size, and force sensor response data corresponding to different depth layers is collected; The force sensor response data collected at different depth layers is subjected to frequency spectrum analysis, and the characteristic frequency, amplitude and phase of the inherent nonlinear response of the material corresponding to different depth layers are identified and obtained, and a multi-layer material response characteristic map is constructed; The signal after the multi-band filtering processing is introduced into the multi-layer material response characteristic map for comparison, and components in the signal that are highly matched with the multi-layer material response characteristic map are identified, so as to distinguish the inherent nonlinear response of the material and the interference signal.

6. The grinding disc CNC machining path control method of claim 5, wherein, The characteristic frequency, amplitude and phase of the inherent nonlinear response of the material corresponding to each depth layer are identified and obtained, and a multi-layer material response characteristic map is constructed, including: The inherent nonlinear response characteristic of the material identified at different depth layers is subjected to standardization processing; A continuous gradient region with a feature difference between adjacent depth layers being less than a third preset threshold is identified, and weighted average or local curve fitting processing is performed on the feature data of the continuous gradient region to generate a transition feature representation; A non-continuous region with a feature difference greater than the third preset threshold is identified, and is divided into different material characteristic regions, and a representative inherent nonlinear response feature vector of each material characteristic region is extracted; The transition feature representation and the representative inherent nonlinear response feature vector of the material are integrated to construct the multi-layer material response characteristic map.

7. A grinding disc CNC machining path control system characterized by, The system comprises: a data acquisition module configured to acquire topography data, local temperature data, and cutting force data of a surface of the polishing disc; a possibility evaluation module configured to comprehensively analyze the topography data, the local temperature data, and the cutting force data, and evaluate a possibility that a protrusion on the surface of the polishing disc is a dressing protrusion according to a preset judgment rule; a strategy selection module configured to select a preset machining strategy according to a result of the evaluation, wherein the machining strategy comprises: when the possibility that the protrusion is the dressing protrusion is higher than a first preset threshold, adjusting a cutting depth and a motion trajectory of a dressing tool to remove the dressing protrusion; when the possibility that the protrusion is the dressing protrusion is lower than a second preset threshold, applying cooling intervention / non-cutting scanning cooling to promote the protrusion to subside; and wherein the evaluating the possibility that the protrusion on the surface of the polishing disc is the dressing protrusion according to the comprehensive analysis of the topography data, the local temperature data, and the cutting force data comprises: when the topography data, the local temperature data, and the cutting force data do not completely match the preset judgment rule, identifying a fuzzy protrusion on the surface of the polishing disc and adjusting the dressing tool to a micro-probe mode, wherein the adjusting the dressing tool to the micro-probe mode comprises: lifting the dressing tool, and finely adjusting the dressing tool downward at a preset speed until the dressing tool slightly contacts the surface of the polishing disc and reaches a preset low contact force threshold, the dressing tool applying micro-amplitude vibration of a preset frequency and a micro-amplitude to the polishing disc; acquiring local temperature response data of a contact area of the dressing tool and force sensor response data of the dressing tool under the micro-amplitude vibration and the slight contact; performing frequency spectrum analysis and adaptive band-pass filtering processing on the force sensor response data, purifying the force sensor response signal, and analyzing and backstepping damping characteristics of a local area of the polishing disc according to the purified force sensor response signal; analyzing and backstepping thermal physical characteristics of the local area of the polishing disc according to the local temperature response data; judging a property of the fuzzy protrusion based on the damping characteristics and the thermal physical characteristics, and evaluating the possibility that the protrusion on the surface of the polishing disc is the dressing protrusion according to the property.

8. A computer device, comprising: a memory and a processor, which are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the polishing disc CNC machining path control method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, a computer readable storage medium stores computer instructions, and the computer instructions are used to make a computer execute the polishing disc CNC machining path control method in any one of claims 1 to 6.

Images