A method and system for splitting allowances for double wear of a milling and polishing consumable

Abstract

The present application belongs to the technical field of numerical control machining process monitoring and multi-machine collaborative scheduling, and provides a milling and polishing consumable double wear allowance apportioning method and system, which converts the consumable wear overrun event into an online task redistribution event. S1, double consumable synchronous acquisition and modeling; S2, overrun determination; S3, non-stop double allowance synchronous apportioning. The consumable wear overrun of the present application no longer causes the whole line to stop, the transfer mechanism continues to run, only the overrun unit itself exits the machining queue, and the capacity gap is taken over by the remaining candidate units according to the allowance apportioning, which significantly reduces the downtime caused by consumable replacement.

Description

Technical Field

[0001] This invention belongs to the field of CNC machining process monitoring and multi-machine collaborative scheduling technology, specifically involving a method and system for online modeling of dual wear of two types of consumables, milling cutters and polishing discs, in a multi-station five-axis flexible parallel CNC island, life determination, and synchronous allocation of machining allowance across units after wear exceeds the limit. Background Technology

[0002] A multi-station, five-axis flexible parallel CNC island (hereinafter referred to as "CNC island") consists of multiple parallel five-axis machining centers. Each center is equipped with a milling spindle and a polishing spindle, enabling the parallel completion of milling and polishing processes. Workpieces are transferred between centers via a transfer mechanism (such as a cam turntable or collaborative robot) to achieve process flow. Typical machining objects of the CNC island include titanium alloy aerospace structural parts, GPU heat dissipation modules, sapphire spherical parts, and diamond tool blanks, with a single material being loaded at a time.

[0003] Milling cutters and polishing discs are two core consumables in CNC islands. Existing process monitoring solutions for production lines have the following shortcomings.

[0004] First, existing compensation programs only perform online identification for a single consumable (usually a milling cutter). They only monitor the wear condition of the milling cutter. The wear of polishing discs relies on manual experience for periodic replacement. However, the material removal rate of the polishing process gradually decreases with the wear of the polishing disc. The wear mechanisms of the two types of consumables differ significantly—milling cutters are mainly subject to abrasive and adhesive wear, with their wear rate increasing non-linearly with load; polishing discs are mainly subject to abrasive passivation and detachment, with their removal capacity decreasing exponentially with cumulative work. A single model cannot accurately characterize the wear behavior of both types of consumables simultaneously.

[0005] Second, the existing production line lacks a dual-process surplus allocation logic across units. Exceeding the wear limit of any consumable will cause the entire machine to stop, or even the entire line to stop while waiting for consumable replacement, resulting in low equipment utilization.

[0006] Third, existing multi-machine task allocation schemes (such as machine tool networking dynamic load balancing practices) take production capacity cycle balance as the trigger source, do not take the wear status of consumables as the trigger condition, and do not provide an algorithm to quantify and allocate the remaining amount of milling and polishing according to the material removal rate.

[0007] Fourth, different workpiece materials (e.g., copper, titanium alloy, sapphire) significantly affect the wear rate of the milling cutter and polishing disc. Existing solutions do not set compensation coefficients for different materials, resulting in large errors in lifespan estimation, leading to premature scrapping and waste, or exceeding the service life and generating scrap.

[0008] Therefore, there is an urgent need in this field for a method and system that can simultaneously identify the wear of milling cutters and polishing discs online, establish two different wear models based on material, and simultaneously distribute the remaining allowance of the milling and polishing processes without stopping the machine when consumables exceed the limit. Summary of the Invention

[0009] This invention aims to solve the problems of low equipment utilization caused by existing CNC islands that only identify single consumables, stop the machine when consumable wear exceeds the limit, and lack cross-unit dual-reserve allocation logic. It provides a method and system for allocating the reserve of milling and polishing consumables under dual wear, so that the consumable wear exceeding the limit event is transformed into an online task reassignment event, thereby significantly improving the overall equipment efficiency of the CNC island.

[0010] The technical solution of the present invention is as follows.

[0011] A method for allocating allowance for the dual wear of milling and polishing consumables is applied to a multi-station five-axis flexible parallel CNC island. The CNC island includes multiple five-axis machining centers arranged in parallel and a transfer mechanism for transferring workpieces between the centers. Each center is equipped with at least a milling spindle and a polishing spindle. The milling spindle is equipped with a milling cutter, and the polishing spindle is equipped with a polishing disc. The method includes the following steps.

[0012] S1. Simultaneous Data Acquisition and Modeling of Dual Consumables: Under the same hardware real-time clock base, the load rate and machining time of the milling spindle and the polishing spindle of each machine unit are simultaneously acquired. A load-time integral wear model is established for the milling cutter, and the milling cutter wear equivalent is accumulated according to the sampling period.

[0013]

[0014] In the formula, Let be the milling spindle load rate in the i-th sampling period. For milling reference load rate, For load index, For milling material coefficient, The sampling period.

[0015] An exponential decay wear model for the removal rate of the polishing disc was established, with the polishing material removal rate as follows:

[0016]

[0017] In the formula, The initial removal rate of the polishing disc. The attenuation constant is For the wear equivalent of the polishing disc, Let be the polishing spindle load rate in the i-th sampling period. For polishing reference load rate, This represents the polishing material coefficient.

[0018] S2, Exceeding Limit Judgment: [This will...] With preset milling cutter over-limit threshold Comparison, will Exceeding the preset polishing disc threshold Comparison; when any wear equivalent reaches its corresponding over-limit threshold, the wear of the milling cutter or polishing disc of the corresponding unit is determined to be over-limit.

[0019] S3. Non-stop dual-reserve synchronous allocation: After determining that the wear exceeds the limit, perform the following allocation operation without stopping the machine.

[0020] Determine the remaining milling allowance for the oversized unit With remaining polishing allowance .

[0021] Select the remaining parallel units that are not under maintenance and whose milling cutters and polishing discs have not reached the corresponding over-limit thresholds as candidate units.

[0022] Calculate the current milling material removal rate for each candidate unit. With polishing material removal rate ,in Output from the polishing disc wear model of step S1 of this unit.

[0023] The remaining milling and polishing allowances of the over-limit unit are simultaneously allocated to each candidate unit according to the removal rate weight, that is:

[0024]

[0025] The allocated milling and polishing tasks are simultaneously distributed to each candidate machine group, and the transfer mechanism is controlled to operate continuously to transfer the corresponding workpiece to the corresponding candidate machine group.

[0026] Preferably, the hardware real-time clock mentioned in step S1 is an FPGA clock with a time base accuracy of [missing information]. The sampling period The milling reference load rate is 1ms. With polishing reference load rate The average load rate was taken in the first stable machining section after the new milling cutter and the new polishing disc were put into operation.

[0027] Preferably, the load-duration integral wear model described in step S1 can be implemented using a piecewise linear model: the ratio of the milling spindle load rate to the reference load rate. The system is divided into multiple intervals, each corresponding to a fixed cumulative slope of wear equivalent. The slope of each interval increases with the increase of the load factor ratio. For example, it can be divided into [0, 0.6), [0.6, 1.0), [1.0, 1.3), etc. There are four intervals, and the slopes of each interval increase sequentially, with a ratio of 0.3:1:2:4.

[0028] Preferably, step S2 further includes a warning threshold. When the wear equivalent reaches the warning threshold but not the over-limit threshold, the allocation scheme of step S3 is pre-generated and candidate units are locked. Wear is considered over-limit when any wear equivalent reaches its over-limit threshold for N consecutive sampling periods, where N is an integer greater than 1. The milling cutter over-limit threshold... The polishing disc exceeds the threshold value, characterized by the equivalent cumulative processing time under a reference load. This refers to the wear equivalent that occurs when the polishing material removal rate decays to a predetermined proportion of its initial removal rate. The aforementioned confirmation mechanism across N consecutive sampling periods eliminates interference from instantaneous signal spikes and avoids misjudgments.

[0029] Preferably, the allocation in step S3 also satisfies the load margin constraint: the predicted spindle load after allocation for any candidate unit does not exceed a set proportion of its rated load; if it does, the allocation amount for that unit is truncated according to the load margin, and the truncated remaining amount is redistributed among the remaining candidate units according to the removal rate weight, until all allocations are completed or the number of iterations reaches the set upper limit.

[0030] Preferably, the method further includes a material-specific compensation step: based on the material of the workpiece being machined, selecting corresponding materials from a preset milling material coefficient table and a polishing material coefficient table respectively. and The two sets of coefficients are independent of each other, and respectively correct the cumulative wear equivalent of the milling cutter and the wear equivalent of the polishing disc in step S1. The milling material coefficient table and the polishing material coefficient table are stored according to the combination mapping relationship between workpiece material and consumable type. For example, when machining the copper substrate of the GPU heat dissipation module, the milling material coefficient corresponding to the copper material is selected, and when polishing with a diamond polishing disc, the polishing material coefficient corresponding to the diamond polishing disc is selected.

[0031] Preferably, before determining that the wear exceeds the limit in step S2, the method further includes: adjusting the milling feed depth and polishing dwell time of the corresponding machine group based on the wear rate output by the two sets of wear models, so that the difference between the milling completion time and the polishing completion time of the same workpiece in each parallel machine group does not exceed the set cycle window.

[0032] Preferably, after step S3, the process further includes: the unit with excessive wear is removed from the subsequent process of the current workpiece and enters the consumable replacement and maintenance queue; after the consumable replacement is completed, the corresponding wear equivalent is cleared to zero and the benchmark load rate is re-collected, and after trial processing verification, it is re-added to the candidate unit set.

[0033] Preferably, the workpiece clamping of each unit adopts a unified zero-point tooling, and the transfer mechanism is a cam turntable or a collaborative robot. Before and after the workpiece transfer, the remaining allowance is calculated based on the unified zero point as the coordinate reference.

[0034] The present invention also provides a surplus amortization system for dual wear of milling and polishing consumables, comprising:

[0035] The dual-channel acquisition module is used to synchronously acquire the load rate and processing time of the milling spindle and polishing spindle of each unit under the same hardware real-time clock base;

[0036] The dual-model modeling module is used to perform the modeling and wear equivalent accumulation of the milling cutter load-duration integral wear model and the polishing disc removal rate exponential decay wear model in step S1;

[0037] The limit-out determination module is used to perform the limit-out determination in step S2;

[0038] The surplus allocation module is used to perform the non-stop dual surplus synchronous allocation in step S3 and output transfer instructions to the transfer mechanism.

[0039] Preferably, the system further includes a material compensation module, which is used to select k_f and k_p from a preset milling material coefficient table and a polishing material coefficient table respectively according to the material of the workpiece being processed, and to correct the cumulative wear equivalent of the milling cutter and the wear equivalent of the polishing disc respectively.

[0040] The beneficial effects of the present invention include, but are not limited to, the following aspects.

[0041] In this application, excessive wear of consumables no longer leads to a complete production line shutdown. The transfer mechanism continues to operate, and only the unit with the excessive wear is removed from the processing queue. Its capacity gap is shared by the remaining candidate units according to their surplus capacity, which significantly reduces the downtime caused by consumable replacement.

[0042] The milling cutter and polishing disc each adopt a differentiated model that matches their respective wear mechanisms—the milling cutter adopts a load-duration integral wear model (nonlinear damage accumulation), and the polishing disc adopts a removal rate exponential decay wear model (capacity decay process). With the addition of material-specific dual compensation coefficients, the accuracy of life estimation is significantly improved, effectively avoiding waste caused by premature scrapping and the risk of scrap due to over-service.

[0043] The allowances for milling and polishing are allocated synchronously according to their respective material removal rates, and the cycle times of the two processes are kept in sync to avoid work-in-process accumulation or process waiting after allocation.

[0044] The allocation algorithm includes load margin constraints and an iterative redistribution mechanism to ensure that the allocation scheme can be executed in a closed loop in engineering and will not cause candidate units to be overloaded due to task allocation.

[0045] The beneficial effects listed above are not exhaustive of all advantages. Other potential beneficial effects and detailed technical implementation methods will be further disclosed in the embodiments or other descriptive sections of this application. Attached Figure Description

[0046] Figure 1 shows the topology diagram for dual consumable identification.

[0047] Figure 2 is a flowchart of the surplus allocation process. Detailed Implementation

[0048] Example 1:

[0049] GPU cooling module manufacturing (copper substrate, milling cutter wear exceeding limits)

[0050] Production line configuration and workpieces:

[0051] The CNC island contains four H-type five-axis machining centers, numbered G1, G2, G3, and G4, which process GPU heat dissipation modules in parallel. The workpiece substrate is copper (T2), and a single substrate is loaded per cycle. Each center is equipped with one milling spindle and one polishing spindle. The milling spindle is equipped with a carbide end mill (…). The heat dissipation module is milled, and a diamond polishing disc is mounted on the polishing spindle to polish the bottom surface. The transfer mechanism is a collaborative robot, and each unit uses a unified zero-point tooling to clamp the workpiece. The system uses FPGA hardware for real-time closed-loop control with a time base accuracy of [insert accuracy here]. .

[0052] Parameter settings:

[0053] Load rate sampling period ,based on Time-base stamping enables synchronized data collection across the entire system.

[0054] Milling cutter wear model parameters: material is copper, load index Milling material coefficient After the new milling cutter is put into service, the average load rate within 30 minutes of the first stable cutting section is taken as the milling reference load rate. In this example (Spindle rated power percentage).

[0055] End mill over-threshold The equivalent cumulative processing time under the baseline load is used as the characterization, taken as 480 minutes (supplier recommended lifespan 600 minutes × safety factor 0.8). Warning threshold. .

[0056] Polishing disc wear model parameters: material is copper, attenuation constant = 0.2, Polishing material coefficient = 1.0. After the new polishing disc is put into service, the average load rate within the first stable polishing section is taken as the polishing reference load rate. Polishing disc exceeds threshold. Defined as the wear equivalent when the removal rate of the polishing material decays to 70% of its initial removal rate. From the relationship:

[0057]

[0058] Solving for:

[0059]

[0060] The calibration method is as follows: After the new polishing disc is put into operation, the removal rate is measured once using a standard copper test piece every 60 minutes of cumulative processing. Record the corresponding Value, pair about Performing a least-squares linear fit, the absolute value of the resulting slope is... .

[0061] Exceeding limits confirmation: Wear is determined to be excessive only if the wear threshold is reached for N = 3 consecutive sampling periods (3ms).

[0062] Load margin: The predicted spindle load of each candidate unit after allocation shall not exceed 85% of its rated load.

[0063] Maximum number of iterations: 3.

[0064] Process execution:

[0065] Step S1, Simultaneous Data Acquisition and Modeling of Dual Consumables: The load rates of the milling spindles and polishing spindles of the four machine units are acquired by their respective sensors, synchronously stamped via FPGA, and then imported into the dual-model modeling module. The milling cutter wear equivalent of each machine unit is also included. According to the formula: Real-time cumulative wear of polishing discs. According to the formula: Real-time cumulative polishing material removal rate According to the formula: Updated in real time.

[0066] The roughing cutter of unit G2, after a cumulative machining time of approximately 462 minutes, The model output value continues to rise. The mechanism is that copper chips adhere to the milling cutter edge, leading to increased cutting resistance and a slow increase in load rate, as determined by the model. Nonlinear amplification makes Accelerated growth.

[0067] Material-specific independent compensation step: The system determines that the material of the currently being machined workpiece is copper, and selects from the milling material coefficient table. = 0.6, selected from the polishing material coefficient table = 1.0, with two sets of coefficients independently correcting the cumulative calculation of wear equivalent. The milling material coefficient table and polishing material coefficient table are stored according to the combination mapping relationship between workpiece material and consumable type. For example, the coefficient for copper material corresponding to the milling cutter is 0.6, and the coefficient for copper material corresponding to the diamond polishing disc is 1.0; the coefficient for titanium alloy corresponding to the milling cutter is 1.0, and the coefficient for titanium alloy corresponding to the polishing disc is separately calibrated.

[0068] Step S2, Exceeding Limit Determination: G2 When the warning threshold of 408 minutes is reached, the system pre-generates the allocation scheme for step S3 and locks G1, G3, and G4 as candidate units (these three units are not under maintenance, and their milling cutters and polishing discs have not reached the corresponding over-limit thresholds). When G2's... If the wear of the G2 rough milling cutter exceeds the limit after 480 minutes and remains above the limit for three consecutive sampling cycles (3ms), the system determines that the wear of the G2 rough milling cutter has exceeded the limit.

[0069] Step S3, simultaneous allocation of dual redundancy without stopping the machine:

[0070] Determine the remaining milling allowance of the current workpiece in over-limit unit G2. Remaining polishing allowance = 0 (Polishing of this workpiece has not yet started. The polishing allowance will be transferred to the candidate unit along with the workpiece and then executed according to the normal process flow).

[0071] Read the current milling material removal rate of the candidate unit (Based on the cutting parameters of the current toolpath segment) calculate): , , .

[0072] Calculate the weighting of milling allowance allocation:

[0073]

[0074] The weights of each unit are as follows: , , .

[0075] Milling allowance allocation:

[0076]

[0077]

[0078]

[0079] Load margin verification: After allocation, the predicted spindle load rates of G1, G3 and G4 are 78%, 72% and 68% respectively, all of which do not exceed 85% of the rated load. They converge after one iteration and do not need to be truncated and redistributed, thus satisfying the load margin constraint described in step S3.

[0080] Simultaneously distribute the allocated milling tasks and polishing task list (G2's) to G1, G3, and G4. = 0, so the polishing list is an empty set, but the synchronous distribution process is still consistent. The collaborative robot continues to operate, transferring the workpiece in G2 to G1 for further processing.

[0081] Maintenance and Regression: G2 exits the subsequent processes of the current workpiece and enters the consumables replacement and maintenance queue. After replacing with a new end mill, Reset to zero and re-acquire milling reference load rate. After a trial run of one unit was completed and verified to be qualified, G2 was re-added to the candidate unit set.

[0082] Result: The production line did not stop during the entire process. The capacity gap during the G2 cutter replacement was shared by G1, G3, and G4 according to the remaining capacity, and the transfer mechanism continued to operate.

[0083] Example 2:

[0084] Mixed-line machining of titanium alloy aerospace structural components and sapphire spherical components (scenario where polishing disc wear exceeds limits).

[0085] Production line configuration and workpieces:

[0086] The CNC island comprises three five-axis machining centers (G1, G2, and G3) operating in parallel. This batch of workpieces includes titanium alloy aerospace structural components (primarily milled) and sapphire spherical components (primarily polished using diamond polishing discs). Each loading consists of a single substrate, processed in two batches sequentially. The transfer mechanism is a cam-driven rotary table. Zero-point tooling is used uniformly for clamping.

[0087] Parameter settings:

[0088] When milling titanium alloys with a milling cutter: Load index = 2.0, Milling Material Coefficient = 1.0, =120min equivalent processing time.

[0089] Attenuation constant when polishing sapphire with a polishing disc = 0.35, Polishing material coefficient = 1.4 (Sapphire is a hard and brittle material, which accelerates the shedding of abrasive grains, so the material coefficient is adjusted upwards compared to copper). The wear equivalent corresponding to a removal rate decaying to 70% of its initial value can be obtained from the following formula:

[0090]

[0091] The remaining parameters are the same as in Example 1. = 1ms, N = 3, load margin 85%, iteration limit 3 times).

[0092] Process (Polishing disc over-limit branch):

[0093] Step S1, Simultaneous Data Acquisition and Modeling of Dual Consumables: Data is acquired and modeled synchronously according to the method in Example 1. When polishing sapphire spherical parts in batches using Unit G3, the equivalent wear of the polishing disc is... according to Continuous accumulation. Polishing disc removal rate. From initial value according to Gradually decrease.

[0094] Material-specific independent compensation steps: The system identifies the current workpiece material as sapphire and selects from the polishing material coefficient table. = 1.4 Cumulative correction of polishing disc wear equivalent.

[0095] Step S2, Exceeding Limit Determination: When Decay to approximately When (approximately 71% of the initial value), the corresponding Reaching the warning threshold The system pre-generates an allocation scheme and identifies G1 and G2 as candidate units. When decay to (70% of the initial value), and confirmed for 3 consecutive sampling cycles, it is determined that the wear of the G3 polishing disc exceeds the limit.

[0096] Step S3, simultaneous allocation of dual redundancy without stopping the machine:

[0097] Determine the remaining polishing allowance for G3. Convert the remaining spherical area to the equivalent volume by multiplying the remaining thickness. Remaining milling allowance = 0 (The milling process of this workpiece has been completed).

[0098] The polishing material removal rate of candidate units G1 and G2 is output in real time by their respective polishing disc wear models: , (Equivalent volumetric rate, from) ×Polishing contact area conversion).

[0099] Polishing allowance allocation:

[0100]

[0101] The weights of each unit are as follows: and .

[0102]

[0103]

[0104] Load margin verification: G1's predicted load rate after allocation is 81%, passing; G2's predicted load rate after allocation reaches 87%, exceeding the rated load margin limit of 85%. The allocation amount for G2 is truncated according to the load margin constraint, limiting G2's allocation amount to [amount missing]. Truncation The polishing allowance is redistributed on G1 according to the removal rate weight (G1 still has a margin, no further iteration is needed). After the second distribution, the total polishing allowance of G1 is approximately 275 + 25 = The iterations converged in two iterations, meeting the upper limit requirement for the number of iterations.

[0105] Two types of allowance lists were issued simultaneously: Milling Allowance List = 0 Issue with an empty set. The polishing allowance list is issued according to the above allocation results to ensure process consistency.

[0106] Maintenance and Reset: The cam turntable continues to rotate. G3 exits the current workpiece's subsequent process and enters the polishing disc replacement maintenance queue. After replacement... Clear to zero and re-collect. and After trial processing and verification, it was re-added to the candidate unit set.

[0107] Quantitative calculation of downtime:

[0108] Assuming that the consumable overrun event for a single unit is a Poisson occurrence: each unit experiences approximately 4 overruns of milling cutters and approximately 3 overruns of polishing discs per month, totaling 7 overruns.

[0109] Traditional solution: Each time the limit is exceeded, the entire line is shut down to replace consumables, taking approximately 25 minutes per instance (including waiting and replacement time). The total downtime for the four units per month is approximately 4 × 7 × 25 = 700 minutes.

[0110] The proposed solution is as follows: Over-limit events do not halt operation; only the over-limit unit itself is removed from the processing queue, and its capacity shortfall is absorbed by the allocated units. Only the increased cycle time loss due to allocation coordination is considered, approximately 3 minutes per instance (additional time spent on task rescheduling and transfer). The monthly loss is approximately 4 × 7 × 3 = 84 minutes.

[0111] Downtime caused by consumables was reduced from 700 minutes to about 84 minutes, a reduction of about 88%.

[0112] Example 3:

[0113] Piecewise linear milling cutter model implementation and multi-configuration adaptation

[0114] Piecewise linear model implementation:

[0115] To adapt to FPGA fixed-point computing resources, the milling cutter load-time integral wear model is implemented as a piecewise linear model. The load ratio is... Divided into four intervals: [0, 0.6), [0.6, 1.0), [1.0, 1.3), ... The cumulative slopes of the wear equivalent for each interval are as follows: , , , The slope ratio is 0.3:1:2:4, which increases as the load ratio increases.

[0116] The cumulative wear equivalent is simplified as follows:

[0117]

[0118] in This represents the cumulative processing time within each interval. This implementation only requires a comparator and an accumulator, eliminating the need for floating-point multiplication and exponentiation, thus adapting to FPGA fixed-point computing resources.

[0119] Multi-line configuration adaptation:

[0120] The method of this invention is applicable to various flexible production line configurations, including but not limited to circular cam turntable type, triangular collaborative robot type, and H-type robot type. Different configurations only require modification to the determination logic of the candidate unit set and the transfer path table; the core logic of the allocation algorithm (steps S1 to S3) remains unchanged. A unified zero-point tooling ensures consistent coordinate reference conversion accuracy for the remaining workpiece balance before and after transfer across different configurations.

[0121] This invention is implemented using FPGA synchronous acquisition and fixed-point arithmetic, without adding extra sensor hardware. It can be directly deployed in the existing multi-station five-axis flexible parallel CNC island CNC system and cycle equalization module. It is suitable for milling and polishing parallel production lines of single substrates such as titanium alloy, copper and sapphire, and has good industrial applicability.

[0122] While exemplary embodiments of this disclosure have been described, those skilled in the art will understand that various changes and modifications can be made to the exemplary embodiments of this disclosure without departing from the spirit and scope thereof. Therefore, all changes and modifications are included within the scope of protection of this disclosure as defined by the claims. This disclosure is defined by the appended claims, and equivalents of those claims are also included.

Claims

1. A method for allocating allowance for the dual wear of milling and polishing consumables, applied to a multi-station five-axis flexible parallel CNC island, wherein the CNC island includes multiple five-axis machining centers arranged in parallel and a transfer mechanism for transferring workpieces between the centers, each center being equipped with at least a milling spindle and a polishing spindle, the milling spindle being equipped with a milling cutter and the polishing spindle being equipped with a polishing disc, characterized in that... Includes the following steps: S1. Simultaneous Data Acquisition and Modeling of Dual Consumables: Under the same hardware real-time clock base, the load rate and machining time of the milling spindle and the polishing spindle of each unit are synchronously acquired; a load-duration integral wear model is established for the milling cutter, and the milling cutter wear equivalent is accumulated according to the sampling period. In the formula, Let be the milling spindle load rate in the i-th sampling period. For milling reference load rate, For load index, For milling material coefficient, The sampling period is defined; an exponential decay wear model for the removal rate is established for the polishing disc, with the polishing material removal rate as follows: In the formula, The initial removal rate of the polishing disc. It is the attenuation constant. For the wear equivalent of the polishing disc, Let be the polishing spindle load rate in the i-th sampling period. For polishing reference load rate, For polishing material coefficient; S2, Exceeding Limit Judgment: [This will...] With preset milling cutter over-limit threshold Comparison, will Exceeding the preset polishing disc threshold Comparison; when any wear equivalent reaches its corresponding over-limit threshold, the wear of the milling cutter or polishing disc of the corresponding unit is determined to be over-limit; S3. Non-stop dual-reserve synchronous allocation: After determining that the wear exceeds the limit, perform the following allocation operation without stopping the machine: Determine the remaining milling allowance for the oversized unit With remaining polishing allowance ; Select the remaining parallel units that are not under maintenance and whose milling cutters and polishing discs have not reached the corresponding over-limit thresholds as candidate units; Calculate the current milling material removal rate for each candidate unit. With polishing material removal rate ,in Output from the polishing disc wear model in step S1 of this unit; The remaining milling and polishing allowances of the over-limit unit are simultaneously allocated to each candidate unit according to the removal rate weight, that is: The allocated milling and polishing tasks are simultaneously distributed to each candidate machine group, and the transfer mechanism is controlled to operate continuously to transfer the corresponding workpiece to the corresponding candidate machine group.

2. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: In step S1, the hardware real-time clock is an FPGA clock with a time base accuracy of [missing information]. The sampling period The milling reference load rate is 1ms; With polishing reference load rate The average load rate was taken in the first stable machining section after the new milling cutter and the new polishing disc were put into operation.

3. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: In step S1, the load-duration integral wear model is implemented as a piecewise linear model: the ratio of the milling spindle load rate to the reference load rate is divided into multiple intervals, each interval corresponds to a fixed cumulative slope of wear equivalent, and the slope of each interval increases as the load rate ratio increases.

4. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: In step S2, a warning threshold is also set. When the wear equivalent reaches the warning threshold but does not reach the over-limit threshold, the allocation scheme of step S3 is pre-generated and candidate units are locked; when any wear equivalent reaches its over-limit threshold for N consecutive sampling cycles, the wear is determined to be over-limit, where N is an integer greater than 1; the milling cutter over-limit threshold The polishing disc exceeds the threshold value, characterized by the equivalent cumulative processing time under a reference load. The wear equivalent corresponding to the decay of the polishing material removal rate to a set percentage of its initial removal rate.

5. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: In step S3, the allocation also satisfies the load margin constraint: the predicted spindle load after allocation for any candidate unit does not exceed a set proportion of its rated load; If the amount exceeds the limit, the allocation for that unit will be truncated according to the load margin, and the remaining amount will be redistributed among the other candidate units according to the removal rate weight, until all allocations are completed or the number of iterations reaches the set limit.

6. The method for allocating allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that, It also includes a material-specific compensation step: based on the material of the workpiece being machined, the corresponding material coefficients are selected from the preset milling material coefficient table and polishing material coefficient table respectively. and The two sets of coefficients are independent of each other and respectively correct the cumulative wear equivalent of the milling cutter and the wear equivalent of the polishing disc in step S1; the milling material coefficient table and the polishing material coefficient table are stored according to the combination mapping relationship between workpiece material and consumable type.

7. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: Before step S2 determines that the wear exceeds the limit, it also includes: based on the wear rate output by the two sets of wear models, the linkage cycle equalization module adjusts the milling feed depth and polishing dwell time of the corresponding machine group so that the difference between the milling completion time and the polishing completion time of the same workpiece in each parallel machine group does not exceed the set cycle window.

8. The method for allocating the allowance for dual wear of milling and polishing consumables according to claim 1, characterized in that: Step S3 and beyond also includes: the unit with excessive wear is removed from the subsequent process of the current workpiece and enters the consumable replacement and maintenance queue; after the consumable is replaced, the corresponding wear equivalent is cleared to zero and the benchmark load rate is re-collected, and after trial processing verification, it is added back to the candidate unit set; the workpiece clamping of each unit adopts a unified zero-point tooling, and the transfer mechanism is a cam turntable or a collaborative robot. Before and after the workpiece is transferred, the remaining margin is calculated with the unified zero point as the coordinate benchmark.

9. A system for allocating allowances for dual wear of milling and polishing consumables, characterized in that, include: The dual-channel acquisition module is used to synchronously acquire the load rate and processing time of the milling spindle and polishing spindle of each unit under the same hardware real-time clock base; The dual-model modeling module is used to perform the modeling and wear equivalent accumulation of the milling cutter load-duration integral wear model and the polishing disc removal rate exponential decay wear model in step S1 of claim 1; An over-limit determination module is used to perform the over-limit determination in step S2 of claim 1; The surplus allocation module is used to perform the non-stop dual surplus synchronous allocation in step S3 of claim 1, and output transfer instructions to the transfer mechanism.

10. The system according to claim 9, characterized in that, It also includes a material compensation module, used to select from preset milling material coefficient tables and polishing material coefficient tables according to the material of the workpiece being processed. and The cumulative wear equivalents of the milling cutter and the polishing disc are corrected separately.

Images