Milling method based on material removal amount monitoring

Abstract

The invention discloses a milling machining method based on material removal amount monitoring. The milling machining method comprises the steps that the workpiece contour and the workpiece design material remaining amount of a workpiece and the actual three-phase current signal and spindle machining information of a machine tool spindle servo motor are obtained; performing feature vector processing on the actual three-phase current signal to obtain a current feature vector; inputting the current feature vector and main shaft machining information into a material removal quantity prediction model to obtain the actual material removal quantity of the workpiece; under the condition that the actual material removal amount of the workpiece does not meet the preset removal amount of the workpiece, fitting and calculation are carried out based on the outline of the workpiece, the actual material removal amount of the workpiece and the remaining amount of the designed material of the workpiece, and the tool compensation position and the tool machining compensation direction of the tool are determined; and the steps are repeated until the actual material removal amount of the workpiece meets the preset removal amount of the workpiece. Therefore, the workpiece quality is improved.

Description

Technical Field

[0001] This invention belongs to the field of manufacturing and processing technology, and particularly relates to a milling method based on material removal monitoring. Background Technology

[0002] CNC milling is the primary machining process for thin-walled parts. However, under the combined action of cutting force, clamping force, and residual stress, thin-walled parts are prone to elastic deformation and tool deflection. Tool deflection is particularly prominent, manifesting as follows: during cutting, due to insufficient workpiece rigidity, elastic deformation occurs away from the tool under the action of cutting force, resulting in the actual cutting depth being less than the theoretical value set by the program. After the process is completed, as the cutting force disappears, the workpiece elastically recovers, and the geometry and dimensional accuracy of the machined surface deviate from the design requirements. Common quality problems include uneven wall thickness, out-of-tolerance profile, insufficient straightness or flatness, etc., which can lead to workpiece scrap in severe cases, resulting in waste of materials and time.

[0003] To suppress tool deformation, existing processes typically employ a "high-speed milling" strategy with small depth of cut, small cutting width, high spindle speed, and large feed rate, reducing workpiece deformation by decreasing the cutting force per tooth. However, this method results in low material removal rate, severely impacting machining efficiency. Another approach is to utilize in-machine measurement technology. For example, patent CN104407568A discloses a real-time adaptive optimization method for CNC machining parameters. This method monitors the cutting force of CNC machining in real-time and optimizes and adaptively adjusts the cutting parameters based on this force. It uses the spindle current signal as the cutting force representation signal, and the signal acquisition does not affect the machining process. The real-time cutting force is calculated from the acquired spindle current signal, and suitable cutting parameters are obtained through optimization. Then, a fuzzy controller is used to adjust and control the feed rate and spindle speed, thereby improving the machining quality. However, this method cannot obtain the actual material removal amount of the machined features during machining, making it difficult to use directly when machining deformation is significant. Summary of the Invention

[0004] This invention provides a milling method based on material removal monitoring, which improves the machining quality of the workpiece.

[0005] According to a first aspect of the present invention, embodiments of the present invention provide a milling method based on material removal monitoring, which may include: S1, obtain the workpiece outline and the remaining material of the workpiece design, as well as the actual three-phase current signal of the machine tool spindle servo motor and the spindle machining information, and the workpiece is processed by the machine tool. S2, perform feature vector processing on the actual three-phase current signal to obtain the current feature vector; S3, input the current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece; S4. When the actual material removal amount of the workpiece does not meet the preset material removal amount of the workpiece, the tool compensation position and tool machining compensation direction of the tool are determined by fitting and calculation based on the workpiece outline, the actual material removal amount of the workpiece and the design material remaining amount of the workpiece, and the tool is connected to the machine tool. S5. Repeat steps S1-S4 until the actual material removal amount of the workpiece meets the preset material removal amount of the workpiece.

[0006] Optionally, when the actual material removal amount of the workpiece does not meet the preset material removal amount, the tool compensation position and tool machining compensation direction are determined by fitting and calculating based on the workpiece outline, the actual material removal amount, and the designed remaining material amount, including: The workpiece's external contour is fitted and analyzed to obtain the actual machined surface of the workpiece. A set of workpiece surface points are selected from the actual machined surface of the workpiece, and the actual material removal amount of the workpiece is combined with the fitting calculation to obtain the actual machined bottom surface of the workpiece. Based on the actual machined bottom surface of the workpiece and the remaining material in the workpiece design, the workpiece design machining point is determined, and the workpiece design machining point is the tool compensation position; The normal vector of the workpiece is calculated by performing normal vector calculation on the designed machining points to obtain the external normal vector of the designed surface. The external normal vector of the designed surface is the tool compensation direction.

[0007] Optionally, a fitting analysis is performed on the workpiece's outer contour to obtain the actual machined surfaces of the workpiece, including: A set of in-machine measurement points are extracted from the workpiece's outer contour. By fitting the on-machine measurement points, the actual machined surface of the workpiece can be obtained.

[0008] Optionally, a set of workpiece surface points are selected from the actual machined surface of the workpiece, and a fitting calculation is performed based on the actual material removal amount of the workpiece to obtain the actual machined bottom surface of the workpiece, including: Obtain the current machining step of the workpiece being processed by the machine tool; If the current step is not the first step, the actual remaining amount of material in the workpiece is calculated based on the actual amount of material removed from the workpiece. The actual surface outward normal vector of any set of workpiece surface points in the actual machined surface is calculated according to the normal vector calculation formula. Determine the workpiece surface point, subtract the product between the actual surface outward normal and the actual remaining material of the workpiece, and the difference obtained is the workpiece bottom surface point; Fit the points on the bottom surface of the workpiece to obtain the actual machined bottom surface of the workpiece.

[0009] Optionally, if the current step is not the first step, the actual remaining material of the workpiece is calculated based on the actual material removed from the workpiece, including: Get the initial remaining material of the workpiece after the machine tool has processed the workpiece in the previous step of the current step; The difference between the initial remaining material amount of the workpiece and the actual material removal amount of the workpiece is the actual remaining material amount of the workpiece.

[0010] Optionally, after obtaining the current step in the machine tool's machining of the workpiece, the method further includes: If the current step is the first step, acquire the initial measurement point data of the workpiece; The initial measurement point data is mapped to the actual machined surface of the workpiece, and the initial UV parameters corresponding to the initial measurement point data are determined by Newton's algorithm. By inputting the initial UV parameters into the B-spline surface function, the initial material surplus of the workpiece is determined, and the initial material surplus of the workpiece is the actual material surplus of the workpiece.

[0011] Optionally, based on the actual machined bottom surface of the workpiece and the remaining material in the workpiece design, the workpiece design machining points are determined, including: Calculate the normal vector of the workpiece bottom surface point on the actual machined bottom surface to obtain the actual bottom surface external normal vector; Determine the bottom surface point of the workpiece, and subtract the product between the actual bottom surface outward normal and the remaining material of the workpiece design. The difference obtained is the workpiece design machining point.

[0012] Optionally, the actual three-phase current signal is subjected to eigenvector processing to obtain a current eigenvector, including: The actual three-phase current signal is filtered by the cutoff frequency to obtain a low-pass filtered current signal. The low-pass filtered current signal is converted by root mean square to obtain the equivalent DC charge. The equivalent DC charge is compiled into a current feature vector.

[0013] Optionally, before inputting the current feature vector and spindle machining information into the material removal prediction model to obtain the actual material removal amount of the workpiece, the method further includes: Obtain the current control interface status of the tool when it is machining the workpiece; The current feature vector corresponding to the current control interface state is determined as the current current feature vector; By inputting the current feature vector and spindle machining information into the material removal prediction model, the actual material removal amount of the workpiece is obtained, including: Input the current current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece.

[0014] Optionally, before inputting the current feature vector and spindle machining information into the material removal prediction model to obtain the actual material removal amount of the workpiece, the method further includes: Obtain a training sample set, which includes multiple training samples, including: training equivalent DC current, machine tool spindle training parameters, training depth of cut, training width of cut, and corresponding training material removal amount; The equivalent DC current, machine tool spindle training parameters, training cutting depth, and training cutting width are used as model inputs, and the material removal amount is used as the training label. These are then input into the neural network model for training to obtain the material removal amount prediction model.

[0015] The technical solutions provided by the embodiments of the present invention bring at least the following beneficial effects: This invention provides a milling method based on material removal monitoring, comprising: acquiring the workpiece outline and designed remaining material of the workpiece, as well as the actual three-phase current signal of the machine tool spindle servo motor and spindle machining information; performing feature vector processing on the actual three-phase current signal to obtain a current feature vector; inputting the current feature vector and spindle machining information into a material removal prediction model to obtain the actual material removal amount of the workpiece; if the actual material removal amount of the workpiece does not meet the preset material removal amount, fitting and calculating based on the workpiece outline, the actual material removal amount of the workpiece, and the designed remaining material of the workpiece to determine the tool compensation position and tool machining compensation direction; repeating the above steps until the actual material removal amount of the workpiece meets the preset material removal amount. Based on this, by extracting the current feature vector from the acquired actual three-phase current signal and inputting the current feature vector and spindle machining information into the material removal prediction model, the actual material removal amount of the workpiece after completing this machining step can be obtained; thus, by using the actual material removal amount of the workpiece, a direct factor determining the final size of the workpiece, the accurate tool compensation position and tool machining compensation direction can be calculated, enabling proactive compensation for "tool deflection" deformation. This method shifts from "avoiding out-of-tolerance" to "correcting out-of-tolerance," achieving closed-loop control of workpiece machining accuracy. It fundamentally guarantees the shape and dimensional accuracy of the workpiece, effectively solves problems such as uneven wall thickness and out-of-tolerance contours, and improves workpiece quality.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention, but do not constitute an undue limitation of the invention.

[0018] Figure 1 This is a flowchart illustrating a milling method based on material removal monitoring according to an exemplary embodiment; Figure 2 This is a schematic diagram of a milling apparatus based on material removal monitoring, according to an exemplary embodiment. Figure 3 This is a structural block diagram of a milling machine based on material removal monitoring, according to an exemplary embodiment. Detailed Implementation

[0019] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0020] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0021] As described in the background section, after the process is completed and the cutting force disappears, the workpiece elastically recovers, and the geometry and dimensional accuracy of the machined surface deviate from the design requirements. Common quality problems include uneven wall thickness, out-of-tolerance profile, insufficient straightness or flatness, etc., which can lead to workpiece scrapping in severe cases, resulting in waste of materials and time.

[0022] Based on this, the present invention provides a milling method based on material removal monitoring. The milling method based on material removal monitoring provided in the embodiments of the present invention will be described below first.

[0023] like Figure 1 As shown; Example 1; It may include the following steps: S1, obtain the workpiece outline and the remaining material of the workpiece design, as well as the actual three-phase current signal of the machine tool spindle servo motor and the spindle machining information, and the workpiece is processed by the machine tool. S2, perform feature vector processing on the actual three-phase current signal to obtain the current feature vector; S3, input the current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece; S4. When the actual material removal amount of the workpiece does not meet the preset material removal amount of the workpiece, the tool compensation position and tool machining compensation direction of the tool are determined by fitting and calculation based on the workpiece outline, the actual material removal amount of the workpiece and the design material remaining amount of the workpiece, and the tool is connected to the machine tool. S5. Repeat steps S1-S4 until the actual material removal amount of the workpiece meets the preset material removal amount of the workpiece.

[0024] Based on the above embodiments, by extracting the current feature vector from the acquired actual three-phase current signal, and inputting the current feature vector and spindle machining information into the material removal prediction model, the actual material removal amount of the workpiece after completing this machining step can be obtained. This allows for the calculation of the precise tool compensation position and tool machining compensation direction using the actual material removal amount, a direct factor determining the final size of the workpiece, enabling proactive compensation for "tool deflection" deformation. This method shifts from "avoiding deviations" to "correcting deviations," achieving closed-loop control of workpiece machining accuracy, fundamentally ensuring the shape and dimensional accuracy of the workpiece, effectively solving problems such as uneven wall thickness and out-of-tolerance contours, and improving workpiece quality.

[0025] In S1 above, when a workpiece is designed, it will have a design dimension. The size of the design dimension is the remaining amount of design material for the workpiece. Specifically, the height dimension can be the remaining height of the design material for the workpiece, and the width dimension can be the remaining width of the design material for the workpiece, etc.

[0026] The workpiece is fixed on the machine tool, and the surface contour of the workpiece can be scanned by in-machine measurement to obtain the workpiece's outer contour.

[0027] The actual three-phase current signal of the spindle servo motor during the machine tool machining process is collected by the data acquisition module. Specifically, the data acquisition module may include current transformers and Hall effect current sensors, etc.

[0028] Spindle machining information can include spindle speed and feed rate, which can be obtained directly from the machine tool's machining program or from instruments on the machine tool.

[0029] In step S2 above, the acquired three-phase current signal undergoes feature vector processing to output a current feature vector. This is essentially a process of data dimensionality reduction and information extraction. Through a series of mature signal processing techniques, it condenses thousands of raw current data points per second into a feature vector with only a few to tens of elements. This current feature vector retains information related to the amount of material removed to the greatest extent possible, while discarding redundant information and noise.

[0030] Specifically, in one example, S2 may also include: S21, The actual three-phase current signal is filtered by the cutoff frequency to obtain a low-pass filtered current signal; S22, perform root mean square conversion on the low-pass filtered current signal to obtain the equivalent DC charge; S23, compile the equivalent DC charge into a current feature vector.

[0031] In step S21 above, the selection of the cutoff frequency is crucial to the filtering effect, as it determines which frequency components will be retained and which will be filtered out. In practical applications, the cutoff frequency needs to be set reasonably based on the characteristics of the current signal and the noise level to ensure that the filtered signal can effectively remove high-frequency noise while retaining low-frequency components that are closely related to the amount of material removed.

[0032] In S22 above, root mean square (RMS) transformation is a commonly used signal processing method. It converts an AC signal into an equivalent DC charge by calculating the RMS value of the signal within a period or a certain time window. This process helps eliminate instantaneous fluctuations in the signal, making the obtained equivalent DC charge more reflective of the signal's average energy level, thereby establishing a more stable relationship with the amount of material removed.

[0033] In step S23 above, compiling the equivalent DC charge into a current feature vector is a crucial step in the signal processing flow. This step typically involves feature extraction algorithms, such as Principal Component Analysis (PCA) and Wavelet Transform, which can extract the most representative features from the equivalent DC charge and encode them into vector form. These current feature vectors not only facilitate subsequent data analysis and model training but also significantly improve the accuracy and robustness of the material removal prediction model.

[0034] In S3 above, the material removal prediction model is pre-trained and can predict the amount of material removed from the workpiece based on a preset current feature vector and preset spindle machining information. Therefore, the currently acquired current feature vector and spindle machining information are input into the model for prediction, thereby obtaining the actual material removal amount of the workpiece.

[0035] Specifically, in this application, the actual material removal amount of the workpiece refers to the material that needs to be removed by cutting, which directly affects the geometric dimensions of the machined feature. The actual material removal amount of the workpiece is defined by the distance between a point on the machined surface and the previous machined surface along its outer surface normal. The material removed refers to the workpiece itself, not the amount of workpiece cut by the tool itself.

[0036] In S4 above, the workpiece has a preset material removal amount in the design. The machine tool controls the cutting tool to process the workpiece and obtain the actual material removal amount of the workpiece. The actual material removal amount is then compared with the preset material removal amount to determine whether the workpiece meets the design requirements. The preset material removal amount is the amount that needs to be removed by cutting with the cutting tool, which is calculated based on the design requirements of the workpiece and the overall dimensions of the blank workpiece.

[0037] If the conditions are not met, further processing of the workpiece is required. During this further processing, it is necessary to refit and calculate based on the current workpiece outline, the actual material removal amount of the current workpiece, and the material removal amount required when designing the workpiece. This will determine the tool compensation position and tool processing compensation direction when the tool performs the next processing step based on the current workpiece outline.

[0038] The workpiece is not processed all at once, but is processed step by step. After each step is completed, it is determined whether the actual material removal of the workpiece after the current step meets the preset removal amount. If not, processing continues, and the tool compensation position and tool processing compensation direction for the next step are determined. The workpiece outline and the actual material removal amount refer to the current step.

[0039] The tool position point is composed of the tool position (tool compensation position) and the tool direction (tool machining compensation direction). It refers to the specific xyz coordinates and ijk direction when the tool moves to that point. If the tool compensation position and the tool machining compensation direction are known, then the tool position point of the tool at this time can be determined.

[0040] In S5 above, steps S1-S4 are executed cyclically until, in a certain step, the actual material removal amount of the workpiece predicted by S3 meets the preset removal amount of the workpiece, then the workpiece is considered to have met the design requirements.

[0041] It should be noted that the literature in the background section can only derive the cutting force, and cannot simply infer the actual amount of material removed from the workpiece. This is because in conventional cutting force calculations, the cutting force is composed of both the plowing action of the tool on the workpiece and the pushing action of the tool on the workpiece. The amount of material removed mainly contributes to the plowing action; simply put, the larger the amount removed, the greater the plowing force. However, in the cutting force model, the amount of material removed is mainly characterized by the thickness of the undeformed chips, which is only a small part of the amount removed. The thickness of the undeformed chips cannot be directly equated with the amount of material removed. Therefore, the cutting force cannot be directly used to calculate the amount of material removed. Furthermore, during machining, the cutting force is also affected by unstable factors such as vibration and tool wear, causing fluctuations. These fluctuations are not equivalent to electrical noise, and these factors directly affect the actual value of the cutting force. This further increases the difficulty of using the cutting force to calculate the amount of material removed, because it is impossible to know whether the increase in cutting force is due to unstable factors or an increase in the amount of material removed.

[0042] Alternatively, in one example, the above S4 may include: S41, perform fitting analysis on the workpiece's outer contour to obtain the actual machined surface of the workpiece; S42, Select a set of workpiece surface points from the actual machined surface of the workpiece, and perform fitting calculations based on the actual material removal amount of the workpiece to obtain the actual machined bottom surface of the workpiece; S43, based on the actual machined bottom surface of the workpiece and the remaining material in the workpiece design, determine the workpiece design machining point, which is the tool compensation position; S44, calculate the normal vector of the workpiece design machining point to obtain the design surface external normal vector, which is the tool compensation direction.

[0043] In S41 above, the workpiece outline is obtained through measurement by the measuring head, which yields hundreds or thousands of discrete points (i.e., the workpiece outline is composed of discrete points). However, CNC programming and toolpath calculation are based on continuous mathematical surfaces. It is impossible to directly calculate the precise normal direction of a point from a discrete point cloud, and the normal direction is crucial for determining the tool's compensation direction. Therefore, it is necessary to refit the workpiece outline (these discrete points) into a smooth, continuous parametric surface (the actual machined surface of the workpiece). On the actual machined surface of the workpiece, for any (u, v) parameter, the precise coordinate point and normal vector can be calculated.

[0044] In S42 above, a workpiece surface point is selected from the actual surface of the workpiece. The UV parameters of the point are fitted and calculated with the actual material removal amount of the workpiece to derive the surface where the bottom surface of the workpiece is located, that is, the actual machined bottom surface of the workpiece. The actual machined bottom surface of the workpiece is also a smooth, continuous parameterized surface, just like the actual machined surface of the workpiece, which is different from the surface obtained by scanning with the measuring head.

[0045] In S43 above, the actual machined bottom surface of the workpiece is the bottom surface of the target workpiece as required by the design. The remaining material of the workpiece is the amount of material that should remain in the design workpiece. Therefore, based on the actual machined bottom surface of the workpiece, and combined with the remaining material of the workpiece, the position of the surface opposite to the actual machined bottom surface of the target workpiece is deduced. Any point on this surface satisfies the design requirements of the workpiece design machining point. In other words, the tool will eventually need to move to this point. Therefore, this point is the tool compensation position.

[0046] In S44 above, the normal vector of the workpiece design machining point is calculated. The result is the normal vector outside the design surface, which indicates the angle at which the tool should contact the workpiece design machining point.

[0047] Alternatively, in one example, the above S41 may include: S411, Extract a set of in-machine measurement points from the workpiece outline; S412, fits the in-machine measurement points to obtain the actual machined surface of the workpiece.

[0048] In S411 above, the outer contour of the workpiece is composed of N in-machine measurement points (discrete points), from which a set of in-machine measurement points is extracted.

[0049] In S412 above, B-spline surface fitting can be performed on the on-machine measurement points to obtain a continuous and smooth actual machined surface of the workpiece.

[0050] Alternatively, in one example, the above S42 may include: S421, Obtain the current step when the machine tool is processing the workpiece; S422, when the current step is not the first step, calculate the actual remaining amount of material in the workpiece based on the actual amount of material removed from the workpiece; S423, calculate the actual surface out-of-surface normal of any set of workpiece surface points in the actual machined surface of the workpiece according to the normal vector calculation formula; S424, determine the workpiece surface point, subtract the product between the actual surface external normal and the actual remaining material of the workpiece, and the difference obtained is the workpiece bottom surface point; S425, fits the bottom surface points of the workpiece to obtain the actual machined bottom surface of the workpiece.

[0051] In S421 above, the current step of the current workpiece processing can be obtained based on the workpiece processing program input into the machine tool.

[0052] In S422 above, if the current step is not the first step, it means that the workpiece has been processed by the tool and is no longer in a blank state. Therefore, it is necessary to calculate the actual remaining amount of material in the workpiece based on the actual amount of material removed after processing.

[0053] Specifically, S422 may include: S4221, obtain the initial remaining amount of workpiece material after the machine tool has processed the workpiece in the previous step of the current step; S4222, determine the difference between the initial remaining material of the workpiece and the actual material removed from the workpiece, which is the actual remaining material of the workpiece.

[0054] In S4221 above, it is known that the current step is not the first step, so the workpiece has already been processed. After the previous step is completed, the actual amount of material removed from the workpiece in this step will be known. Therefore, the amount of material remaining in the workpiece after the completion of this step can be obtained based on the actual amount of material removed from the workpiece, that is, the initial amount of material remaining in the workpiece.

[0055] In S4222 above, the initial material remaining amount of the workpiece - the actual material removal amount of the workpiece = the actual material remaining amount of the workpiece.

[0056] Specifically, you can refer to the following formula:

[0057] in, Actual material removal amount of the workpiece This represents the initial remaining material amount of the workpiece. This represents the actual remaining material in the workpiece.

[0058] In S423 above, the workpiece surface points in the actual machined surface of the workpiece are calculated using the following formula to obtain the external normal vector of the actual surface.

[0059]

[0060] in, U represents the external normal vector of the actual surface, and U and V represent the UV parameters of points on the workpiece surface.

[0061] In S424 above, the workpiece surface point is calculated according to the following formula. The difference between the actual surface normal vector and the actual remaining material of the workpiece is the workpiece bottom point.

[0062]

[0063] in, Point on the bottom surface of the workpiece. For points on the workpiece surface, For the actual surface outward normal, This represents the actual amount of remaining material.

[0064] In S425 above, the bottom surface points of the workpiece are fitted using the B-spline surface fitting method to obtain a continuous and smooth actual machined bottom surface of the workpiece.

[0065] Optionally, in one embodiment, after S421 described above, S42 may further include: S426, when the current step is the first step, acquire the initial measurement point data of the workpiece; S427, Map the initial measurement point data to the actual machined surface of the workpiece, and determine the initial UV parameters corresponding to the initial measurement point data using Newton's algorithm; S428, by inputting the initial UV parameters into a B-spline surface function, the initial material remaining amount of the workpiece is determined, and the initial material remaining amount of the workpiece is the actual material remaining amount of the workpiece.

[0066] In the above S426, the current step is the first step, which means that the workpiece has not been processed yet and is still in the blank state. Therefore, it is necessary to re-acquire the initial measurement point data of the workpiece when it is in the blank state. The initial measurement point data can be obtained by measuring with the machine probe or by obtaining it from the supplier of the blank workpiece.

[0067] In S427 above, when mapping the initial measurement point data to the actual machined surface of the workpiece, the geometric correspondence between the measurement point and the design surface is first established, and the initial UV parameters of the point in the parameter domain are solved by Newton's iterative algorithm to ensure that the convergence accuracy meets the set threshold, thereby obtaining a high-precision parameter mapping result.

[0068] In step S428 above, by inputting the initial UV parameters into a pre-constructed B-spline surface function, which generates a continuous parametric surface based on the workpiece design model, the initial material surplus of each part of the workpiece in its blank state can be accurately calculated. This calculation process is achieved through a reverse mapping mechanism, that is, using the positioning information of the initial measurement points in the parameter domain, combined with the surface equation, to inversely deduce the material distribution. The final initial material surplus of the workpiece is equivalent to the actual material surplus under the current machining step. This value will serve as the benchmark parameter for subsequent machining compensation calculations.

[0069] Alternatively, in one example, S43 may include: S431, calculate the normal vector of the workpiece bottom surface point on the actual machined bottom surface of the workpiece to obtain the actual bottom surface external normal vector; S432, determine the bottom surface point of the workpiece, subtract the product between the actual bottom surface normal and the remaining material of the workpiece design, and the difference obtained is the workpiece design machining point.

[0070] In S431 above, the workpiece bottom surface points in the actual machined bottom surface of the workpiece are calculated using the following formula to obtain the external normal vector of the actual bottom surface.

[0071]

[0072] in, U represents the actual outward normal vector of the bottom surface, and U and V represent the UV parameters of the bottom surface points of the workpiece.

[0073] In S432 above, the workpiece bottom surface point can be calculated by subtracting the product between the actual bottom surface outward normal and the workpiece design material remaining amount. The difference obtained is determined as the workpiece design machining point.

[0074]

[0075] in, Design machining points for the workpiece. Point on the bottom surface of the workpiece. The actual external normal vector of the base. Design the remaining material amount for the workpiece.

[0076] Example 2: Alternatively, in one example, before S3 above, the method may further include: S6, Obtain the current control interface status of the tool when it is machining the workpiece; S6, determine the current feature vector corresponding to the current control interface state as the current current feature vector; The above-mentioned S3 may also include: S31, input the current current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece.

[0077] In S6 above, when the tool is machining the workpiece, it changes the control interface state (high or low level) of the control interface through control commands; the current control interface state is the current control interface state when the tool is machining. For example, when the tool is machining the workpiece, if the control command can instruct the control interface state to change from low level to high level, then the current control interface state is high level.

[0078] Specifically, the current control interface status corresponding to the tool machining can be obtained through the edge controller.

[0079] Similarly, in S7 above, the three-phase current signal will also have a corresponding control interface state. The current three-phase current signal corresponding to the current control interface state will correspond to the tool position when the tool is machining the workpiece. Therefore, the current current feature vector obtained by further feature processing of the corresponding current three-phase current signal also corresponds to the current tool position.

[0080] The tool position and current characteristic vector during machining are not directly related, but they can be indirectly related in time through the control interface state. Under the same control interface state, the tool position and current characteristic vector correspond to each other.

[0081] In S31 above, the current current feature vector that reflects the current step and the spindle machining information are input into the material removal amount prediction model to obtain the actual material removal amount of the workpiece, thereby improving the accuracy of the material removal amount prediction model.

[0082] In the above embodiment 2, the various methods of embodiment 1 are included, and the various processes in embodiment 1 can be implemented to achieve the same technical effect. To avoid repetition, they will not be described again here.

[0083] Example 3: Optionally, in one example, before S3 above, the following is also included: S8. Obtain the training sample set, which includes multiple training samples, including: training equivalent DC current, machine tool spindle training parameters, training cutting depth, training cutting width, and corresponding training material removal amount. S9 uses the equivalent DC current, machine tool spindle training parameters, training cutting depth, and training cutting width as model inputs, and the material removal amount as training labels, and inputs them into the neural network model for training to obtain the material removal amount prediction model.

[0084] In S9 above, the construction of the training sample set needs to cover typical working conditions in actual machining. High-precision sensors are used to collect data on equivalent DC current, spindle speed, spindle feed rate, depth of cut, and width of cut under different combinations of cutting parameters, while simultaneously recording the corresponding actual material removal amount as a label value. During the data collection process, it is necessary to ensure the uniformity of parameter distribution to avoid bias in the training data towards specific machining ranges.

[0085] The machine tool spindle training parameters include: spindle speed training parameters and spindle feed speed training parameters.

[0086] In S10 above, the neural network model adopts a multilayer perceptron structure. The number of nodes in the input layer corresponds to four cutting parameters, and the output layer has a single node predicting the material removal amount. During training, the Adam optimizer is used, and the mean squared error loss function is selected. Cross-validation is used to determine the optimal hidden layer dimension. After model training, parameter sensitivity analysis is performed to verify the contribution of each input feature to the prediction result, ensuring that the model possesses the generalization ability required for engineering applications. This prediction model can replace traditional empirical formulas, achieving real-time and accurate estimation of material removal amount under different working conditions (equivalent DC current, machine tool spindle training parameters, training cutting depth, and training cutting width).

[0087] In the above embodiment 3, the methods of embodiment 1 or 2 are included, and the processes of embodiment 1 or 2 can be implemented to achieve the same technical effect. To avoid repetition, they will not be described again here.

[0088] Example 4: like Figure 2 As shown, optionally, in one example, the present invention also provides a milling apparatus 200 based on material removal monitoring, which includes: The first acquisition module 210 is used to acquire the workpiece outline and the remaining material of the workpiece design, as well as the actual three-phase current signal of the machine tool spindle servo motor and the spindle machining information, and the workpiece is processed by the machine tool. The feature vector processing module 220 is used to perform feature vector processing on the actual three-phase current signal to obtain the current feature vector; Prediction module 230 is used to input the current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece; The calculation module 240 is used to fit and calculate based on the workpiece outline, the actual material removal amount of the workpiece, and the designed remaining material amount of the workpiece when the actual material removal amount of the workpiece does not meet the preset material removal amount of the workpiece, and to determine the tool compensation position and tool machining compensation direction of the tool. The tool is connected to the machine tool. The repeat module 250 is used to repeat the above steps until the actual material removal amount of the workpiece meets the preset removal amount of the workpiece.

[0089] Based on the above embodiments, the first acquisition module 210 acquires the workpiece outline and the remaining material quantity designed for the workpiece, as well as the actual three-phase current signal of the machine tool spindle servo motor and spindle machining information; the feature vector processing module 220 performs feature vector processing on the actual three-phase current signal to obtain a current feature vector; the prediction module 230 inputs the current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece; the calculation module 240, when the actual material removal amount of the workpiece does not meet the preset material removal amount of the workpiece, performs fitting and calculation based on the workpiece outline, the actual material removal amount of the workpiece, and the remaining material quantity designed for the workpiece to determine the tool compensation position and the tool machining compensation direction; the repeat module 250 repeats the above steps until the actual material removal amount of the workpiece meets the preset material removal amount of the workpiece. Based on this, by extracting the current feature vector from the acquired actual three-phase current signal and inputting the current feature vector and spindle machining information into the material removal prediction model, the actual material removal amount of the workpiece after completing this machining step can be obtained. Thus, by using the actual material removal amount, a direct factor determining the final size of the workpiece, the precise tool compensation position and tool machining compensation direction can be calculated, enabling proactive compensation for "tool deflection" deformation. This method shifts from "avoiding deviations" to "correcting deviations," achieving closed-loop control of workpiece machining accuracy, fundamentally ensuring the shape and dimensional accuracy of the workpiece, effectively solving problems such as uneven wall thickness and out-of-tolerance contours, and improving workpiece quality.

[0090] Optionally, the computing module 240 may include: The first fitting analysis submodule is used to fit and analyze the workpiece's outer contour to obtain the actual machined surface of the workpiece. The first fitting calculation submodule is used to select a set of workpiece surface points from the actual machined surface of the workpiece, and perform fitting calculations in combination with the actual material removal amount of the workpiece to obtain the actual machined bottom surface of the workpiece. The first determination submodule is used to determine the workpiece design machining point based on the actual machined bottom surface of the workpiece and the remaining material of the workpiece design. The workpiece design machining point is the tool compensation position. The first normal vector calculation submodule is used to calculate the normal vector of the workpiece design machining point to obtain the design surface external normal vector, which is the tool compensation direction.

[0091] Optionally, the first fitting analysis submodule may include: The extraction unit is used to extract a set of in-machine measurement points from the workpiece's outline. The fitting unit is used to fit the on-machine measurement points to obtain the actual machined surface of the workpiece.

[0092] Optionally, the first fitting calculation submodule may include: The acquisition unit is used to acquire the current step of the machine tool when processing a workpiece; The first calculation unit is used to calculate the actual remaining amount of material in the workpiece based on the actual amount of material removed in the current step when the current step is not the first step. The second calculation unit is used to calculate the actual surface out-of-surface normal vector of any set of workpiece surface points in the actual machined surface of the workpiece according to the normal vector calculation formula. The first determining unit is used to determine the workpiece surface point by subtracting the product between the actual surface external normal and the actual remaining material of the workpiece, and the difference obtained is the workpiece bottom surface point. The fitting unit is used to fit the bottom surface points of the workpiece to obtain the actual machined bottom surface of the workpiece.

[0093] Optionally, the first computing unit may include: The acquisition sub-unit is used to acquire the initial remaining amount of workpiece material after the machine tool has processed the workpiece in the previous step of the current step. The first determining subunit is used to determine the difference between the initial material remaining amount of the workpiece and the actual material removal amount of the workpiece, which is the actual material remaining amount of the workpiece.

[0094] Optionally, the milling apparatus 200 based on material removal monitoring may further include: The second acquisition module is used to acquire the initial measurement point data of the workpiece when the current step is the first step. The first determining module is used to map the initial measurement point data to the actual machined surface of the workpiece and determine the initial UV parameters corresponding to the initial measurement point data through Newton's algorithm. The second determining module is used to determine the initial material remaining amount of the workpiece by inputting the initial UV parameters into the B-spline surface function. The initial material remaining amount of the workpiece is the actual material remaining amount of the workpiece.

[0095] Optionally, the first determining submodule may include: The normal vector calculation unit is used to calculate the normal vector of the workpiece bottom surface point on the actual machined bottom surface of the workpiece, and obtain the normal vector outside the actual bottom surface; The second determining unit is used to determine the bottom surface point of the workpiece. The difference between the actual bottom surface outward normal vector and the workpiece design material remaining amount is the workpiece design machining point.

[0096] Optionally, the feature vector processing module 220 may include: The filtering submodule is used to filter the actual three-phase current signal by the cutoff frequency to obtain a low-pass filtered current signal. The conversion module is used to perform root mean square conversion on the low-pass filtered current signal to obtain the equivalent DC charge. The compilation module is used to compile the equivalent DC charge into a current feature vector.

[0097] Optionally, the milling apparatus 200 based on material removal monitoring may further include: The third acquisition module is used to acquire the current control interface status of the machine tool when the tool is machining the workpiece; The third determining module is used to determine the current feature vector corresponding to the current control interface state that is consistent with the current control interface state from multiple preset control interface states. Prediction module 230 may include: The prediction submodule is used to input the current current feature vector and spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece.

[0098] Optionally, the milling apparatus 200 based on material removal monitoring may further include: The fourth acquisition module is used to acquire the training sample set, which includes multiple training samples, including: training equivalent DC current, machine tool spindle training parameters, training cutting depth, training cutting width, and corresponding training material removal amount. The training module is used to input the equivalent DC current, machine tool spindle training parameters, training cutting depth, and training cutting width as model inputs, and the material removal amount as training labels into the neural network model for training, so as to obtain the material removal amount prediction model.

[0099] In the above embodiment 4, each method described in any one of embodiments 1-3 is included, and each process in any one of embodiments 1-3 can be implemented, achieving the same technical effect. To avoid repetition, it will not be described again here.

[0100] Example 5: Figure 3 The diagram shows a hardware structure schematic of a milling equipment based on material removal monitoring according to an embodiment of the present invention.

[0101] A milling machine based on material removal monitoring may include a processor 301 and a memory 302 storing computer program instructions.

[0102] Specifically, the processor 301 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0103] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 32 is non-volatile solid-state memory.

[0104] In certain embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, typically, memory 302 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors 301), it is operable to perform the operations described with reference to one aspect of the invention.

[0105] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any of the milling methods based on material removal monitoring in the above embodiments.

[0106] In one example, the milling equipment based on material removal monitoring may also include a communication interface 303 and a bus 304. As shown in the figure, the processor 301, memory 302, and communication interface 303 are connected via the bus 304 and communicate with each other.

[0107] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0108] Bus 304 may include hardware, software, or both. For example, and not as a limitation, bus 304 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, a Wireless Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Control Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 304 may include one or more buses 304. Although specific bus 304s are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus 304 or interconnect.

[0109] This milling equipment based on material removal monitoring can combine current milling methods based on material removal monitoring. Figure 1 , 2 The milling method and milling apparatus 200 based on material removal monitoring are described.

[0110] In addition, this embodiment of the invention also provides a computer program product, including computer program instructions, which, when executed by processor 301, can implement the steps and corresponding content of the aforementioned method embodiments.

[0111] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0112] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the required tasks. The programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0113] It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or apparatus. However, this invention is not limited to the order of the steps described above; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0114] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, special-purpose computer, or other programmable processor of a milling method, apparatus, and device based on material removal monitoring, to produce a machine such that these instructions, executed via the processor of the computer or other programmable processor of the milling method, apparatus, and device based on material removal monitoring, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0115] The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A milling method based on material removal amount monitoring, characterized by, The steps of the method are as follows: S1, obtaining a workpiece contour profile and a workpiece design material remaining amount of a workpiece, and actual three-phase current signals of a machine tool main shaft servo motor and main shaft machining information, the workpiece being machined by the machine tool; S2, performing feature vector processing on the actual three-phase current signals to obtain a current feature vector; S3, inputting the current feature vector and the main shaft machining information into a material removal amount prediction model to obtain a workpiece actual material removal amount; S4, in the case that the workpiece actual material removal amount does not meet a workpiece preset removal amount, fitting and calculating based on the workpiece contour profile, the workpiece actual material removal amount and the workpiece design material remaining amount to determine a tool compensation position and a tool machining compensation direction of a tool, the tool being connected with the machine tool; S5, repeating the above steps S1-S4 until the workpiece actual material removal amount meets the workpiece preset removal amount.

2. The material-removal-amount-based milling process method according to claim 1, wherein In the case that the workpiece actual material removal amount does not meet the workpiece preset removal amount, fitting and calculating based on the workpiece contour profile, the workpiece actual material removal amount and the workpiece design material remaining amount to determine the tool compensation position and the tool machining compensation direction of the tool, comprising: performing fitting analysis on the workpiece contour profile to obtain a workpiece actual machining surface; selecting a group of workpiece surface points from the workpiece actual machining surface, and fitting and calculating based on the workpiece actual material removal amount to obtain a workpiece actual machining bottom surface; determining a workpiece design machining point according to the workpiece actual machining bottom surface and the workpiece design material remaining amount, the workpiece design machining point being the tool compensation position; performing normal vector calculation on the workpiece design machining point to obtain a design surface outer normal vector, the design surface outer normal vector being the tool compensation direction.

3. The material-removal-amount-based milling process method according to Claim 2, wherein The fitting analysis on the workpiece contour profile to obtain the workpiece actual machining surface comprises: extracting a group of in-machine measurement points from the workpiece contour profile; fitting the in-machine measurement points to obtain the workpiece actual machining surface.

4. The material-removal-amount-based milling process method according to Claim 2, wherein The selecting a group of workpiece surface points from the workpiece actual machining surface, and fitting and calculating based on the workpiece actual material removal amount to obtain the workpiece actual machining bottom surface comprises: obtaining a current working step of the machine tool machining the workpiece; in the case that the current working step is not the first working step, calculating a workpiece actual material remaining amount based on the workpiece actual material removal amount; calculating actual surface outer normal vectors of any group of workpiece surface points in the workpiece actual machining surface according to a normal vector calculation formula; determining the workpiece surface points, subtracting the product of the actual surface outer normal vectors and the workpiece actual material remaining amount to obtain a workpiece bottom surface point; fitting the workpiece bottom surface point to obtain the workpiece actual machining bottom surface.

5. The material-removal-amount-based milling process method according to Claim 4, wherein The calculating the workpiece actual material remaining amount based on the workpiece actual material removal amount in the case that the current working step is not the first working step comprises: obtaining a workpiece initial material remaining amount of the workpiece after the machine tool machined the workpiece in a previous working step of the current working step; Determine the difference between the initial material remaining amount of the workpiece and the actual material removal amount of the workpiece as the actual material remaining amount of the workpiece.

6. The material-removal-amount-based milling process method according to Claim 4, wherein After obtaining the current working step of the machine tool machining the workpiece, the method further comprises: In the case that the current working step is the first working step, obtaining the initial measurement point data of the workpiece; Mapping the initial measurement point data to the actual machining surface of the workpiece, and determining the initial UV parameters corresponding to the initial measurement point data by Newton algorithm; Inputting the initial UV parameters into the B-spline surface function to determine the initial material remaining amount of the workpiece, which is the actual material remaining amount of the workpiece.

7. The material-removal-amount-based milling process of claim 2, wherein, The determination of the workpiece design machining point according to the actual machining bottom surface of the workpiece and the workpiece design material remaining amount comprises: Calculating the normal vector of the workpiece bottom surface point of the actual machining bottom surface of the workpiece to obtain the actual bottom surface outer normal vector; Determining the workpiece bottom surface point, subtracting the product of the actual bottom surface outer normal vector and the workpiece design material remaining amount, and obtaining the difference as the workpiece design machining point.

8. The material-removal-amount-based milling method according to any one of claims 1 to 7, wherein The feature vector processing of the actual three-phase current signal to obtain the current feature vector comprises: Filtering the actual three-phase current signal by cutoff frequency to obtain a low-pass filtered current signal; Converting the low-pass filtered current signal into a root mean square to obtain an equivalent direct current amount; Compiling the equivalent direct current amount into a current feature vector.

9. The material-removal-amount-based milling process according to any one of claims 1 to 7, wherein Before inputting the current feature vector and the spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece, the method further comprises: Obtaining the current control interface state corresponding to the tool when machining the workpiece; Determining the current feature vector corresponding to the current control interface state as the current feature vector; Inputting the current feature vector and the spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece comprises: Inputting the current feature vector and the spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece.

10. The material-removal-amount-based milling method according to any one of claims 1 to 7, wherein Before inputting the current feature vector and the spindle machining information into the material removal amount prediction model to obtain the actual material removal amount of the workpiece, the method further comprises: Obtaining a training sample set, the training sample set comprising a plurality of training samples, the training samples comprising: training equivalent direct current, machine tool spindle training parameters, training cutting depth, training cutting width, and corresponding training material removal amount; Inputting the training equivalent direct current, the machine tool spindle training parameters, the training cutting depth, and the training cutting width as model inputs, and the training material removal amount as a training label into a neural network model for training to obtain the material removal amount prediction model.

Images