Intelligent early warning and real-time compensation system for wear of numerical control machining tool

Abstract

The application discloses a numerical control machining tool wear intelligent early warning and real-time compensation system and relates to the technical field of numerical control machining.The system comprises a data acquisition module, a tool wear modeling module, a wear state division module, a real-time working condition monitoring module, a strategy decision module, an execution control module and a storage module.The tool wear modeling module is used for fitting and constructing a correlation relationship between tool wear and single workpiece machining time, and outputs preset wear and preset total life value.The wear state division module is used for dividing wear intervals and outputting machining frequency thresholds.The strategy decision module is used for determining the current wear interval of the tool, generating a treatment strategy and converting the treatment strategy into a control instruction.The execution control module is connected with a numerical control machining machine tool and is used for receiving the control instruction and performing corresponding operations.The application matches corresponding treatment strategies according to different wear intervals, adopts corresponding coping modes in different stages of tool wear, and guarantees machining continuity.

Description

Technical Field

[0001] This invention relates to the field of CNC machining technology, specifically to an intelligent early warning and real-time compensation system for CNC machining tool wear. Background Technology

[0002] In the field of modern mechanical manufacturing, CNC machining is one of the machining methods. As a key component that directly participates in the cutting operation, the cutting tool will inevitably wear during the continuous cutting process. Tool wear will directly cause the workpiece to exceed the dimensional tolerance and the surface quality to decline. In severe cases, it will also cause failures such as tool breakage and tool chipping, resulting in workpiece scrap, machine tool damage, and even production safety accidents, thus restricting processing efficiency and product yield.

[0003] Currently, CNC workshops mostly manage tools by timed tool replacement or manual spot checks. However, timed tool replacement often results in premature replacement of tools that are still capable of machining, leading to a waste of tool resources. Manual spot checks require stopping the machine to disassemble the tools, interrupting production continuity. They rely on the operator's experience to judge the degree of wear, which is highly subjective, has low detection efficiency, and is difficult to adapt to large-scale continuous automated production scenarios.

[0004] Existing tool wear monitoring technologies are mainly divided into two categories: offline detection and online monitoring. Offline detection requires disassembling the tool and measuring the wear amount with instruments, which cannot achieve real-time intervention in the machining process. Existing online monitoring solutions mostly rely on a single sensor to collect vibration and sound signals for simple fault alarms, which can only identify working conditions where the tool has already suffered severe damage. It is difficult to quantitatively model the wear evolution process of the tool throughout its entire life cycle and cannot predict the wear development trend. At the same time, existing technologies generally lack the ability to classify and control the wear in a closed loop, and cannot match differentiated compensation strategies according to different wear stages of the tool, thus failing to achieve classified and automated handling.

[0005] Based on this, we now offer an intelligent early warning and real-time compensation system for CNC machining tool wear, which can eliminate the drawbacks of existing technical solutions. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent early warning and real-time compensation system for CNC machining tool wear, so as to solve the problems in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] The intelligent early warning and real-time compensation system for CNC machining tool wear includes a data acquisition module, a tool wear modeling module, a wear state classification module, a real-time working condition monitoring module, a strategy decision-making module, an execution control module, and a storage module.

[0009] The data acquisition module is used to collect the full life cycle status data and wear measurement data of multiple tools. The collected data includes at least the tool type identifier, workpiece type identifier, single workpiece processing time, single workpiece processing times, actual tool wear, and number of qualified workpieces processed by the tool. The collected data is then uploaded to the tool wear modeling module and the storage module.

[0010] The tool wear modeling module is used to receive the full life cycle status data and wear measurement data of multiple tools. For different tool types and different workpiece types, it fits and constructs the correlation formula between tool wear amount and single workpiece processing time. Based on the correlation formula, it outputs the preset wear amount and preset total life value under the corresponding working conditions and stores them in the storage module.

[0011] The wear state division module is used to receive the correlation formula, the preset wear amount and the preset total life value, and divide the wear into four wear intervals using the cumulative number of processing times of a single workpiece as the scale. It outputs the processing number threshold corresponding to each wear interval and stores it in the storage module.

[0012] The real-time working condition monitoring module is connected to the CNC machine tool and is used to collect real-time working condition data of on-site processing in real time, and transmit the real-time working condition data to the strategy decision module. The real-time working condition data includes the cumulative number of processing times for a single workpiece, the real-time processing time for a single workpiece, the current type of processing tool, and the current type of processing workpiece.

[0013] The strategy decision module is used to receive input data from the real-time working condition monitoring module, and simultaneously retrieve the processing number threshold, preset wear amount and preset total life value that match the current processing tool and the current processing workpiece from the storage module. It compares the real-time cumulative number of processing times of a single workpiece with the processing number threshold corresponding to each wear interval to determine the current wear interval of the tool. Based on the wear interval, it generates a corresponding handling strategy and converts the handling strategy into a control command to be sent to the execution control module.

[0014] The execution control module is connected to the CNC machine tool and is used to receive control commands and execute corresponding operations;

[0015] The storage module is used to store the original acquired data, the correlation formula between tool wear and single workpiece processing time, the processing number threshold for each wear range, the preset wear amount, the preset total life value, and historical processing records.

[0016] Furthermore, the tool wear modeling module constructs a correlation between tool wear and single workpiece machining time, specifically including:

[0017] Regression fitting was performed on multiple sets of sample data under the same tool-workpiece pairing to obtain the tool wear amount-single workpiece machining time function corresponding to different tool types and different workpiece types.

[0018] The mapping relationship between the number of machining operations and the theoretical tool wear amount is obtained through conversion in combination with the number of machining operations, so as to determine the preset wear amount under the corresponding working condition. Meanwhile, the preset total life value is derived inversely according to the critical wear amount for tool failure. The number of machining operations refers to the total number of workpieces that have been machined cumulatively after a single tool is put into use, and the preset total life value refers to the maximum number of machined products with qualified dimensions that the tool can process.

[0019] Further, the four wear intervals are respectively a possible wear occurrence interval, a repairable wear interval, an non-repairable but non-failed wear interval, and a failure and replacement required interval. The wear state division module performs interval segmentation according to the four wear intervals, and respectively determines a first threshold N1, a second threshold N2, and a third threshold N3 based on the preset total life value and the preset wear amount in combination with the mapping relationship between the number of machining operations and the theoretical tool wear amount, which are specifically expressed as:

[0020] The cumulative number of machined workpieces 0<N≤N1 is the possible wear occurrence interval;

[0021] The cumulative number of machined workpieces N1<N≤N2 is the repairable wear interval;

[0022] The cumulative number of machined workpieces N2<N≤N3 is the non-repairable but non-failed wear interval;

[0023] The cumulative number of machined workpieces N>N3 is the failure and replacement required interval;

[0024] Wherein, the first threshold N1, the second threshold N2, and the third threshold N3 are all critical machining times obtained by solving based on the critical preset wear amount of each wear interval, and satisfy 0<N1<N2<N3.

[0025] Further, when the strategy decision module determines that the tool is in the possible wear occurrence interval, the disposal strategy is: output a primary warning prompt, and simultaneously issue a cutting parameter fine-tuning instruction to the execution regulation module. The execution regulation module reduces the cutting feed rate, and at the same time, the real-time working condition monitoring module continuously collects the real-time working condition data of on-site machining and transmits it back to the storage module, which is used for iteratively updating the correlation relationship between the tool wear amount and the single workpiece machining time.

[0026] Further, when the strategy decision module determines that the tool is in the repairable wear interval, the disposal strategy is: output an intermediate wear warning prompt, and issue a tool local sharpening compensation instruction to the execution regulation module. The execution regulation module calls the tool local sharpening mechanism to perform local sharpening on the cutting edge part of the tool. After the local sharpening is completed, the tool wear modeling module updates the preset wear amount and the preset total life value based on the real-time working condition data collected after the local sharpening.

[0027] Furthermore, when the strategy decision module determines that the tool is in the wear-unrepairable but not yet failed range, the handling strategy is as follows: output an advanced warning prompt, execute the control module to adjust the cutting parameters of the CNC machine tool, and perform dimensional compensation by reducing the cutting speed and the cutting depth, while recording the workpiece dimensional deviation data of each machining operation in this wear range.

[0028] Furthermore, when the strategy decision module determines that the tool is in the failure and needs to be replaced, the handling strategy is as follows: output a tool failure alarm prompt, execute the control module to issue a tool change command, suspend the machining task, output a tool change reminder or directly execute the tool change action, and after the tool change is completed, the system resets the real-time single workpiece machining count and starts the full life cycle monitoring of the new tool.

[0029] Furthermore, the data acquisition module is connected to an external workpiece size detection device to collect the true value of the actual size of the workpiece after processing. The storage module binds and stores the true value of the actual size of the workpiece with the wear data of the corresponding tool. The tool wear modeling module periodically reads the bound data in the storage module, iteratively corrects the correlation between the tool wear amount and the single workpiece processing time, and synchronously updates the preset wear amount and preset total life value under the corresponding working conditions.

[0030] Furthermore, the real-time working condition monitoring module is also used to receive sensor data of vibration and cutting sound of CNC machine tools in real time, and send the sensor data as auxiliary features into the strategy decision module. When the sensor data is abnormal, the strategy decision module corrects the current wear interval judgment result based on the abnormal data. When the cumulative number of real-time single workpiece processing does not reach the wear interval threshold, the warning and handling strategy of the corresponding upper-level wear interval is triggered in advance.

[0031] Furthermore, the storage module pre-stores a parameter library corresponding to different tool-workpiece pairs. The parameter library includes the types of tools that can be regrinded, the maximum number of regrinding cycles allowed, and the upper and lower limits of adjustable cutting parameters.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] 1. This invention establishes the correlation between tool wear and single workpiece machining time for different tool-workpiece pairs, establishes a mapping between machining times and theoretical wear, realizes quantitative modeling of tool wear, calculates the preset total life value by combining the critical wear value of tool failure, quantifies the evolution law of tool wear under different working conditions, fully explores the usable life of tools, reduces the resource waste caused by premature tool replacement, and reduces machining costs;

[0034] 2. This invention divides the entire life cycle of a cutting tool into four stages: a wear-prone range, a repairable wear range, an unrepairable wear range but not yet failed range, and a failed range requiring replacement range. It sets multiple threshold levels based on the cumulative number of machining operations and matches corresponding handling strategies for different wear ranges. It takes corresponding measures at different stages of tool wear to ensure machining continuity.

[0035] 3. This invention binds and stores the true value of the actual workpiece size detection and the tool wear data, periodically iterates and corrects the wear correlation formula, and synchronously updates the preset wear amount and preset total life value to achieve model self-iteration and improve monitoring reliability. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the system structure of the present invention.

[0037] Figure 2 This is a schematic diagram showing the connection relationship between the system structure and external devices of the present invention.

[0038] Figure 3 This is a schematic diagram of the method flow of the present invention.

[0039] Figure label annotations: Data acquisition module 10, tool wear modeling module 20, wear state classification module 30, real-time working condition monitoring module 40, strategy decision-making module 50, execution control module 60, storage module 70. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0041] To address the problems of existing CNC machining tool life determination relying heavily on manual experience, failing to match differentiated early warning and compensation strategies based on tool wear stages, and easily leading to premature tool replacement resulting in cost waste or excessive tool wear causing batch workpiece scrap, this invention provides an intelligent early warning and real-time compensation system for CNC machining tool wear. By collecting full lifecycle status data and actual wear measurement data of multiple tools, it constructs a correlation formula between tool wear and single workpiece machining time for different tool types and workpiece types, outputting preset wear amounts and preset total life values. During machining, it monitors machine tool operating parameters in real time, compares them with wear interval thresholds to determine the current tool wear state, and combines the maximum allowable number of tool regrinding times parameter for verification constraints to match corresponding wear interval handling strategies. Simultaneously, it periodically iterates and corrects the wear relationship formula to achieve model self-optimization. This system effectively improves the technical problems of traditional solutions such as fixed tool life threshold settings and poor operating condition adaptability. The specific technical solution is shown below.

[0042] Example 1

[0043] In this embodiment, as Figure 1 and Figure 2 As shown, the present invention provides an intelligent early warning and real-time compensation system for CNC machining tool wear, including a data acquisition module 10, a tool wear modeling module 20, a wear state classification module 30, a real-time working condition monitoring module 40, a strategy decision module 50, an execution control module 60, and a storage module 70.

[0044] The data acquisition module 10 is used to collect the full life cycle status data and wear measurement data of multiple tools. The collected data includes at least tool type identification, workpiece type identification, single workpiece processing time, single workpiece processing times, actual tool wear, and number of qualified workpieces processed by the tool. The collected data is uploaded to the tool wear modeling module 20 and the storage module 70. This module interfaces with CNC machine tools and external equipment, including at least a tool wear microscope and an external workpiece size detection device. This module can also preprocess the collected multi-source raw data, complete data cleaning, outlier removal, and data format unification, and send the processed structured data to the tool wear modeling module 20 for model training and fitting, and write it to the storage module 70 for saving.

[0045] The tool wear modeling module 20 is used to receive the full life cycle status data and wear measurement data of multiple tools. For different tool types and different workpiece types, it fits and constructs the correlation formula between tool wear amount and single workpiece machining time. Based on the correlation formula, it outputs the preset wear amount and preset total life value under the corresponding working condition and stores them in the storage module 70. This module can receive wear data bound to the true value of workpiece size periodically returned by the storage module 70, iteratively correct the wear correlation formula, synchronously update the preset wear amount and preset total life value, and write the updated model parameters into the storage module 70. After the tool completes local regrinding, it receives the working condition acquisition data after regrinding and recalculates and updates the preset wear amount and preset total life value of the tool.

[0046] The wear state division module 30 is used to receive the correlation formula, the preset wear amount, and the preset total life value. Using the cumulative number of processing times of a single workpiece as the benchmark (if the processing time of the workpieces varies greatly, the equivalent processing time or the cumulative cutting time can be used as an auxiliary judgment benchmark to improve the judgment accuracy), it divides the wear into four wear intervals. Based on the preset wear amount and the preset total life value, combined with the mapping relationship between the number of processing times and the theoretical tool wear amount, it determines the processing time threshold, outputs the processing time threshold corresponding to each wear interval, and stores it in the storage module 70. The preset wear amount refers to the theoretical tool wear amount calculated by the correlation formula when processing to a certain preset number of times under a specific tool-workpiece pairing. It serves as the reference benchmark for the wear state division module 30 to divide the intervals.

[0047] The real-time working condition monitoring module 40 is connected to the CNC machine tool and is used to collect real-time working condition data of the on-site processing. The real-time working condition data is transmitted to the strategy decision module 50. The real-time working condition data includes the cumulative number of processing times of a single workpiece, the processing time of a single workpiece, the current type of processing tool, and the current type of processing workpiece. The real-time working condition monitoring module 40 only collects the operating condition signal output by the CNC machine tool and does not collect the true value of the actual size detection of the workpiece output by the external workpiece size detection device. The working condition and sensor data are continuously collected in a loop during the processing. When abnormal characteristics appear in the sensor data, an abnormal mark is marked and uploaded together.

[0048] The strategy decision module 50 receives input data from the real-time working condition monitoring module 40 and retrieves the processing number threshold, preset wear amount, and preset total life value that match the current processing tool and the current processing workpiece from the storage module 70. It compares the real-time cumulative number of single workpiece processing with the processing number threshold corresponding to each wear interval to determine the current wear interval of the tool. Based on the wear interval, it generates a corresponding handling strategy and converts the handling strategy into a control command to be sent to the execution control module 60. At the same time, it receives vibration and cutting sound sensor data. If the sensor data is abnormal, even if the cumulative number of processed parts has not reached the interval threshold, it actively jumps to the next higher wear interval to trigger early warning handling.

[0049] The execution control module 60 is connected to the CNC machine tool and is used to receive control commands, parse command content, distinguish different command types such as early warning prompts, fine adjustment of cutting parameters, local tool regrinding, cutting parameter compensation adjustment, pause machining, and tool change. It is used to issue cutting parameter commands to modify the cutting feed rate, cutting speed, and cutting depth. When modifying, it follows the adjustable upper and lower limits of cutting parameters in the parameter library in the storage module 70 and does not exceed the range of machining parameters allowed by the machine tool. It is also used to interface with the local tool regrinding mechanism. After receiving the regrinding command, it controls the machine tool to pause machining, drives the regrinding mechanism to complete the local regrinding of the cutting edge of the tool, and resumes machining after the regrinding is completed.

[0050] The storage module 70 is used to store the original acquired data, the correlation formula between tool wear and single workpiece processing time, the processing number threshold for each wear range, the preset wear amount, the preset total life value, and historical processing records.

[0051] Specifically, the tool wear modeling module 20 constructs the correlation between tool wear and single-workpiece machining time, including:

[0052] Receive multi-tool lifecycle status data and wear measurement data from data acquisition module 10, group the samples according to tool-workpiece pairing, collect multiple complete machining samples under the same tool and the same machining workpiece combination, remove abnormal noise samples, and obtain a standardized sample set for fitting.

[0053] Regression fitting was performed on multiple sets of sample data for the same tool-workpiece pairing to obtain a tool wear amount versus single workpiece machining time function corresponding to different tool types and different workpiece types. The function input is the single workpiece machining cutting time, and the output is the theoretical tool wear amount. The regression fitting yielded... , This refers to the machining time for a single workpiece (the cutting time required to process one workpiece). This represents the tool wear, assuming the tool has already been machined. For each workpiece, the total cutting time is... (processing) (Total cutting time of the entire workpiece).

[0054] Based on the single-workpiece machining time and the total number of workpieces machined by the tool, the total cutting time of the tool is calculated. The mapping relationship between the number of machining operations and the theoretical tool wear is then obtained by converting the number of operations to determine the preset wear amount under the corresponding working condition. Simultaneously, the preset total lifespan is derived from the critical tool failure wear amount. Substituting these values ​​into the tool wear amount - single-workpiece machining time function, the theoretical total wear amount is obtained. This establishes a mapping relationship between the number of processing cycles and the theoretical wear amount. Based on this mapping, a certain number of processing cycles is selected. The theoretical wear value corresponding to this value is used as the preset wear value under this working condition, and the critical wear value for tool failure is retrieved. ,make Find the maximum value, The maximum value is the preset total lifespan, which is the maximum number of workpieces that the tool can process. The number of processing times is the total number of workpieces processed after a single tool is put into use (cumulative number of workpiece processing times). That is, the number of processing times is incremented by 1 for each workpiece processed after the tool is put into use. For example, if a new tool has processed 85 workpieces, then the current number of processing times for the tool is 85. The preset total lifespan is the maximum number of qualified products that the tool can process.

[0055] The system receives the true value of the actual size of the workpiece and the data bound to the tool wear in the storage module 70, refits periodically, iteratively corrects the correlation between the tool wear amount and the single workpiece processing time, and synchronously updates the preset wear amount and preset total life value, and writes them back to the storage module 70.

[0056] After the tool completes the partial regrinding operation, the on-site working condition data after the regrinding is completed is read, the fitting calculation is re-executed, and the preset wear amount and preset total life value of the tool are updated.

[0057] Specifically, the four wear intervals are respectively the possible wear occurrence interval, the repairable wear interval, the non-repairable but non-failed wear interval, and the failure requiring replacement interval. The wear state division module 30 performs interval segmentation according to the four wear intervals, and respectively determines a first threshold N1, a second threshold N2, and a third threshold N3 based on the preset total service life and the preset wear amount in combination with the mapping relationship between the number of machining times and the theoretical tool wear amount, which are specifically expressed as:

[0058] The cumulative number of machined workpieces 0<N≤N1 is the possible wear occurrence interval;

[0059] The cumulative number of machined workpieces N1<N≤N2 is the repairable wear interval;

[0060] The cumulative number of machined workpieces N2<N≤N3 is the non-repairable but non-failed wear interval;

[0061] The cumulative number of machined workpieces N>N3 is the failure requiring replacement interval;

[0062] Wherein, the first threshold N1, the second threshold N2, and the third threshold N3 are all critical machining times obtained by solving based on the critical preset wear amount of each wear interval, and satisfy 0<N1<N2<N3.

[0063] Specifically, when the strategy decision module 50 determines that the tool is in the possible wear occurrence interval, the disposal strategy is: output a primary warning prompt, and simultaneously issue a cutting parameter fine-tuning instruction to the execution regulation module 60. The execution regulation module 60 reduces the cutting feed rate, and at the same time, the real-time working condition monitoring module 40 continuously collects the real-time working condition data of on-site machining and transmits it back to the storage module 70, for iteratively updating the correlation relationship between the tool wear amount and the single-workpiece machining time;

[0064] When the strategy decision module 50 determines that the tool is in the repairable wear interval, the disposal strategy is: output an intermediate wear warning prompt, issue a tool partial sharpening compensation instruction to the execution regulation module 60, the execution regulation module 60 calls the tool partial sharpening mechanism to perform partial sharpening on the tool cutting edge part. After the partial sharpening is completed, the tool wear modeling module 20 updates the preset wear amount and the preset total service life based on the real-time working condition data collected after partial sharpening. The partial sharpening includes: the strategy decision module 50 determines the sharpening allowance to be removed from the cutting edge according to the current predicted wear amount, the execution regulation module 60 controls the tool partial sharpening mechanism to complete the cutting edge removal. After the sharpening is completed, the size of the first workpiece is collected. If there is still a deviation in the size of the first workpiece, a tiny coordinate axis offset compensation is superimposed to eliminate the residual dimensional error caused by sharpening;

[0065] When the strategy decision module 50 determines that the tool is in the unrepairable but not yet failed wear range, the handling strategy is as follows: An advanced warning is output; the execution control module 60 adjusts the cutting parameters of the CNC machine tool, compensating for dimensions by reducing the cutting speed and depth of cut; and simultaneously records the workpiece dimensional deviation data for each machining operation within this wear range. Parameter compensation includes: real-time dimensional compensation by the execution control module 60, which includes machine tool coordinate offset compensation and cutting process parameter compensation; querying the theoretical workpiece dimensional deviation corresponding to the current predicted wear amount; and issuing a coordinate axis offset compensation command to the CNC machine tool based on the theoretical workpiece dimensional deviation to compensate for the workpiece dimensional error. Simultaneously, dynamic adjustment of cutting parameters is performed, using preset benchmark cutting parameters as a reference, and adjusting the parameters according to the proportion of the predicted tool wear to the critical failure wear amount. , To predict tool wear, attenuation calculations are performed on cutting speed and depth of cut within the adjustable upper and lower limits of the parameter library:

[0066] , ;

[0067] in, As the reference cutting speed, As the reference cutting depth, To reduce the cutting speed, This is the depth of cut after attenuation. , The attenuation coefficient has a value of 0 to 0.5. The adjustment result shall not exceed the upper and lower limits of the preset cutting parameters in the parameter library of the storage module 70.

[0068] When the strategy decision module 50 determines that the tool is in the failure and needs to be replaced, the handling strategy is as follows: output a tool failure alarm, execute the control module 60 to issue a tool change command, suspend the machining task, output a tool change reminder or directly execute the tool change action. After the tool change is completed, the system resets the real-time single workpiece machining count and starts the full life cycle monitoring of the new tool.

[0069] like Figure 1 and Figure 2 As shown, the data acquisition module 10 is connected to an external workpiece size detection device to collect the actual size detection value of the workpiece after processing. The storage module 70 binds and stores the actual size detection value of the workpiece with the wear data of the corresponding tool. The tool wear modeling module 20 periodically reads the data bound in the storage module 70, iteratively corrects the correlation between the tool wear amount and the single workpiece processing time, and synchronously updates the preset wear amount and preset total life value under the corresponding working conditions.

[0070] Specifically, in this invention, the data acquisition module 10 and the real-time working condition monitoring module 40 are functionally independent. The data acquisition module 10 is mainly used to collect the operating data of the CNC machine tool and the actual tool wear data and the true value of the actual workpiece size detection output by the external workpiece size detection device, for model construction and iterative optimization. The real-time working condition monitoring module 40 only collects the real-time working condition data output by the CNC machine tool body, for real-time status judgment during the processing, and does not acquire the detection true value of the external detection equipment.

[0071] Specifically, such as Figure 1 and Figure 2 As shown, the real-time working condition monitoring module 40 is also used to receive sensor data of vibration and cutting sound of CNC machine tool in real time, and send the sensor data as auxiliary features to the strategy decision module 50. When the sensor data is abnormal, the strategy decision module 50 corrects the current wear interval judgment result based on the abnormal data. When the cumulative number of real-time single workpiece processing does not reach the wear interval threshold, the warning and handling strategy of the corresponding upper level wear interval is triggered in advance.

[0072] Specifically, such as Figure 1 and Figure 2 As shown, the storage module 70 pre-stores a parameter library corresponding to different tool-workpiece pairs. The parameter library also needs to record the historical actual number of times each tool has been regrinded. This data is also stored in the storage module 70. When the strategy decision module 50 is verifying, it reads the historical actual number of times the tool has been regrinded and compares it with the maximum number of times allowed to be regrinded in the parameter library. The parameter library includes the types of tools that are allowed to be regrinded, the maximum number of times allowed to be regrinded, and the upper and lower limits of adjustable cutting parameters. When the strategy decision module 50 generates control instructions, it calls the parameter library for verification. When the tool has reached the maximum number of times allowed to be regrinded, even if it is in the wear repairable range, the wear unrepairable and non-failed range handling strategy is executed, and the local regrinding operation is not executed.

[0073] Specifically, although some cutting tools are still within the wear-repairable range and have the physical conditions for regrinding, the number of times the tool can be regrinded has a physical limit. There is a maximum number of regrindings allowed. Regrinding will consume the tool body. After repeated regrinding, the remaining material of the tool body will be exhausted, and the tool will be scrapped if regrinding is performed again. Therefore, this invention abandons regrinding at this time and directly adopts the cutting parameter optimization compensation strategy corresponding to the wear-unrepairable and non-failure range. It uses cutting parameter adjustment and size compensation instead. The tool is replaced when the failure threshold is reached. This operation can avoid ineffective regrinding and prevent the tool from being damaged due to insufficient tool strength after regrinding. The machining accuracy of the workpiece can be continued without regrinding.

[0074] Example 2

[0075] The difference from Example 1 is that, as in Example 1, Figure 3As shown, the present invention also provides a method for intelligent early warning and real-time compensation of CNC machining tool wear, which is applied to the intelligent early warning and real-time compensation system for CNC machining tool wear shown in Embodiment 1 above. The method specifically includes the following steps:

[0076] Model building phase: Data acquisition module 10 collects full life cycle status data and wear measurement data of multiple tools, and the true value of actual workpiece size detection, and sends them to tool wear modeling module 20. Tool wear modeling module 20 fits different tool-workpiece pairs to obtain the correlation between tool wear amount and single workpiece processing time, and outputs preset wear amount and preset total life value. The preset wear amount and preset total life value are sent to wear state division module 30, which divides the wear state into possible wear intervals, wear repairable intervals, wear unrepairable and non-failed intervals, and failure intervals that need to be replaced based on the cumulative number of single workpiece processing. The threshold number of processing times for each wear interval is output. All model parameters, thresholds, and tool-workpiece pairing parameters are stored in storage module 70.

[0077] Online machining monitoring and handling stage: The real-time working condition monitoring module 40 reads the real-time working condition data of the CNC machine tool output on-site machining, retrieves the matching model parameters and wear interval machining number threshold from the storage module 70, and sends them to the strategy decision module 50. The strategy decision module 50 compares the cumulative machining number to determine the wear interval of the tool, and reads the parameter library to verify the tool's historical regrinding number.

[0078] If the tool is in a wear-prone zone, a primary warning and a fine-tuning command for cutting parameters are output and handed over to the execution control module 60 for execution. If the tool is in a repairable wear zone, the tool's historical regrinding count is checked. If the maximum allowable regrinding count has not been reached, a local regrinding command is issued. If the maximum allowable regrinding count has been reached, regrinding is skipped, and the handling strategy for the wear-unrepairable but not yet failed zone is executed. If the tool is in a wear-unrepairable but not yet failed zone, a high-level warning is output, and cutting parameters are adjusted and dimensions are compensated. If the tool is in a failed zone requiring replacement, a tool failure alarm is output, triggering a tool change command. The execution control module 60 interfaces with the CNC machine tool and the tool local regrinding mechanism to execute the corresponding actions, and the execution results are written back to the storage module 70.

[0079] Model Iteration and Update Phase: After the workpiece is processed, the data acquisition module 10 acquires the true value of the actual size of the workpiece output by the external workpiece size detection device, binds it with the tool wear data and stores it in the storage module 70. After the set iteration cycle is reached, the tool wear modeling module 20 reads the bound data to iteratively correct the correlation between the tool wear amount and the single workpiece processing time, updates the preset wear amount and preset total life value and writes it back to the storage module 70. The updated parameters are used for real-time judgment in the next round of processing and run in a loop.

[0080] The present invention also provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor is used to execute the computer program to implement the above-described intelligent early warning and real-time compensation method for CNC machining tool wear. The present invention also provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by the processor to implement the above-described intelligent early warning and real-time compensation method for CNC machining tool wear.

[0081] To further demonstrate the application value of the intelligent early warning and real-time compensation system for CNC machining tool wear in this invention, the system structure, workflow and method steps of this invention are described in detail below, taking the machining of 45# steel workpieces with carbide end mills as an example.

[0082] The data acquisition module 10 connects to the tool wear microscope detector and the external workpiece size detection device (such as a coordinate measuring machine) to collect tool type identification, workpiece type identification, single workpiece processing time, single workpiece processing times, actual tool wear, and number of qualified workpieces processed by the tool. At the same time, it collects the true value of the actual size detection of the workpiece after processing and uploads the collected data to the tool wear modeling module 20 and the storage module 70.

[0083] The real-time working condition monitoring module 40 is connected to the CNC machine tool and is equipped with vibration sensors and cutting sound sensors. It collects the cumulative number of times a single workpiece is processed, the processing time of a single workpiece, the current type of cutting tool, and the current type of workpiece in real time. At the same time, it collects machine tool vibration and cutting sound sensor data and transmits them to the strategy decision module 50. The execution control module 60 is electrically connected to the CNC machine tool, the tool local regrinding mechanism, and the tool changing mechanism. It can receive control commands to complete cutting parameter modification, tool local regrinding, pause processing, and automatic tool changing.

[0084] The storage module 70 has a pre-stored parameter library corresponding to different tool-workpiece pairs. The parameter library includes the types of tools that can be regrinded, the maximum number of regrindings allowed, and the upper and lower limits of adjustable cutting parameters. In this example, the maximum number of regrindings allowed for the carbide end mill is 2, the feed rate is adjustable in the range of 0.05~0.25mm / r, and the cutting speed is adjustable in the range of 60~120m / min.

[0085] The tool wear modeling module 20 receives multiple sets of full life cycle sample data under the same tool-workpiece pairing, uses least squares regression fitting to construct the correlation formula between tool wear and single workpiece processing time under this pairing, and combines single workpiece processing time conversion to establish the mapping relationship between processing times and theoretical tool wear. Using the tool failure critical back face wear of 0.3mm as the failure criterion, the preset total life value of this tool-workpiece pairing is calculated to be 320 pieces, that is, the tool can process a maximum of 320 qualified workpieces.

[0086] The wear state classification module 30 calculates the critical processing number threshold for each wear interval based on the preset wear amount and preset total life value, and obtains: the first threshold N1 is 80 pieces, the second threshold N2 is 180 pieces, and the third threshold N3 is 280 pieces, satisfying 0 < N1 < N2 < N3, then -0 < N ≤ 80 is the wear interval that may occur, -80 < N ≤ 180 is the wear interval that can be repaired, -180 < N ≤ 280 is the wear interval that cannot be repaired and has not failed, and -N > 280 is the interval that has failed and needs to be replaced. The storage module 70 saves the correlation formula, the N1 / N2 / N3 thresholds, the preset wear amount, and the preset total life value.

[0087] The processing and operation process specifically includes:

[0088] The machine tool starts processing, and a brand-new carbide end mill begins machining a 45# steel workpiece. The real-time working condition monitoring module 40 continuously collects data on the cumulative number of processed parts, processing time, vibration, and cutting sound, and sends this data to the strategy decision module 50. When the cumulative number of processed parts N is 45, and 0 < 45 ≤ 80, it is determined that the machine is in a wear range that may occur. The strategy decision module 50 outputs a primary warning prompt (human-machine interface pop-up window + yellow indicator light prompt) and issues a fine-tuning command for cutting parameters. For example, the control module 60 reduces the cutting feed rate from 0.15 mm / r to 0.12 mm / r. The real-time working condition monitoring module 40 continuously collects working condition data and sends it back to the storage module 70. The data acquisition module 10 simultaneously collects the true value of the workpiece dimensions, binds the tool wear data, and stores it in the storage module 70 for subsequent iteration updates.

[0089] Processing continues until the cumulative number of processed parts N reaches 120. If 80 < 120 ≤ 180, it is determined to be in the wear repairable range. The strategy decision module 50 retrieves the parameter library of the storage module 70 for verification. The current number of tool regrinding times is 0, which has not reached the maximum allowed number of regrinding times of 2. A medium wear warning is output (orange indicator light), and a tool local regrinding compensation command is issued. The execution control module 60 suspends processing and calls the tool local regrinding mechanism to perform local regrinding on the cutting edge of the tool. After regrinding is completed, the system resumes processing. The tool wear modeling module 20 reads the working condition and dimensional detection data collected after regrinding and updates the preset wear amount and preset total life value. After the update, the preset total life value is corrected to 245 pieces.

[0090] Processing continues until the cumulative number of processed parts N reaches 210. If 180 < 210 ≤ 280, it is determined that the parts are in the range of unrepairable wear but not yet failed. The strategy decision module 50 outputs an advanced warning prompt (red flashing indicator light), and the execution control module 60 adjusts the cutting parameters. The cutting speed is reduced from 100m / min to 75m / min, and the cutting depth is reduced from 1.2mm to 0.8mm to achieve processing size compensation. At the same time, the workpiece size deviation data corresponding to each processing is recorded and stored in the storage module 70.

[0091] Continue processing until the cumulative number of processed parts N reaches 285. Since N > 280, the system is determined to have entered the failure and replacement range. The strategy decision module 50 outputs a tool failure alarm, the execution control module 60 issues a tool change command, the processing task is paused, a tool change reminder pops up on the human-machine interface, the automatic tool changer is triggered to complete the tool change, and after the tool change is completed, the system resets the real-time cumulative number of single workpiece processing times and starts full life cycle monitoring of the new tool.

[0092] If there is an abnormal jump condition: the cumulative number of parts N is 62, which has not yet reached N1=80, but the machine tool vibration amplitude collected by the real-time condition monitoring module 40 exceeds the set threshold, and the cutting sound spectrum shows abnormal characteristics, the strategy decision module 50 receives the abnormal sensor auxiliary characteristics, jumps to the next higher wear range in advance, directly triggers the primary warning and feed rate reduction handling strategy, and intervenes in tool wear in advance.

[0093] If there are special working conditions for parameter library verification: In the tool-workpiece pairing example above, the tool has already undergone two partial re-sharpening operations, and the cumulative number of parts N to be machined is 130, falling into the wear-repairable range. The strategy decision module 50 calls the parameter library for verification and finds that the tool has reached the maximum allowed number of re-sharpening operations. Therefore, it no longer performs partial re-sharpening operations and directly executes the wear-unrepairable but not yet failed range handling strategy, reducing the cutting speed and depth of cut for dimensional compensation.

[0094] The data acquisition module 10 periodically reads the true value of the actual workpiece size output by the coordinate measuring machine, binds and stores the true value of the workpiece size with the corresponding tool wear data, and the tool wear modeling module 20 reads the dataset bound by the storage module 70 every 50 workpieces processed, iteratively corrects the correlation formula between tool wear amount and single workpiece processing time, and synchronously updates the preset wear amount, preset total life value and processing number threshold of each interval, so as to realize the model continuously self-optimizes with the on-site processing data.

[0095] As can be seen from the above examples, this system divides wear into four wear ranges for graded treatment. At the same time, it introduces vibration and cutting sound sensors as auxiliary features to achieve early intervention for anomalies. Combined with the periodic iterative wear model based on the true value of the workpiece's measured dimensions, it adapts to fluctuations in on-site working conditions, effectively reduces the workpiece scrap rate, and fully utilizes the tool's full life cycle processing capabilities.

[0096] In summary, this invention discloses an intelligent early warning and real-time compensation system for CNC machining tool wear. The system achieves graded early warning and real-time compensation control of tool wear based on the number of machining operations. It can also combine sensor information such as vibration and cutting sound to predict the wear state in advance. At the same time, it considers constraints such as the maximum number of tool regrinding operations to avoid ineffective regrinding. After regrinding or tool replacement, the system can update model parameters or reset the count to start monitoring new tools. Relying on the true value of workpiece size detection, the wear model is continuously iterated and optimized to achieve full life cycle monitoring and early warning of tools. This effectively reduces the risk of workpiece scrap due to tool wear, improves the stability of the CNC machining process, and has good application value.

[0097] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A CNC machining tool wear intelligent early warning and real-time compensation system, characterized in that, It includes a data acquisition module, a tool wear modeling module, a wear state classification module, a real-time working condition monitoring module, a strategy decision-making module, an execution control module, and a storage module; The data acquisition module is used to collect the full life cycle status data and wear measurement data of multiple tools. The collected data includes at least the tool type identifier, workpiece type identifier, single workpiece processing time, single workpiece processing times, actual tool wear, and number of qualified workpieces processed by the tool. The collected data is then uploaded to the tool wear modeling module and the storage module. The tool wear modeling module is used to receive the full life cycle status data and wear measurement data of multiple tools. For different tool types and different workpiece types, it fits and constructs the correlation formula between tool wear amount and single workpiece processing time. Based on the correlation formula, it outputs the preset wear amount and preset total life value under the corresponding working conditions and stores them in the storage module. The wear state division module is used to receive the correlation formula, the preset wear amount and the preset total life value, and divide the wear into four wear intervals using the cumulative number of processing times of a single workpiece as the scale. It outputs the processing number threshold corresponding to each wear interval and stores it in the storage module. The real-time working condition monitoring module is connected to the CNC machine tool and is used to collect real-time working condition data of on-site processing in real time, and transmit the real-time working condition data to the strategy decision module. The real-time working condition data includes the cumulative number of processing times for a single workpiece, the real-time processing time for a single workpiece, the current type of processing tool, and the current type of processing workpiece. The strategy decision module is used to receive input data from the real-time working condition monitoring module, and simultaneously retrieve the processing number threshold, preset wear amount and preset total life value that match the current processing tool and the current processing workpiece from the storage module. It compares the real-time cumulative number of processing times of a single workpiece with the processing number threshold corresponding to each wear interval to determine the current wear interval of the tool. Based on the wear interval, it generates a corresponding handling strategy and converts the handling strategy into a control command to be sent to the execution control module. The execution control module is connected to the CNC machine tool and is used to receive control commands and execute corresponding operations; The storage module is used to store the original acquired data, the correlation formula between tool wear and single workpiece processing time, the processing number threshold for each wear range, the preset wear amount, the preset total life value, and historical processing records.

2. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 1, characterized in that, The tool wear modeling module constructs a correlation between tool wear and single workpiece machining time, specifically including: Regression fitting was performed on multiple sets of sample data under the same tool-workpiece pairing to obtain the tool wear amount-single workpiece machining time function corresponding to different tool types and different workpiece types. By combining the number of processing operations to obtain the mapping relationship between the number of processing operations and the theoretical tool wear, the preset wear amount under the corresponding working conditions is determined. At the same time, the preset total life value is derived from the critical wear amount of tool failure. The number of processing operations is the total number of workpieces processed after a single tool is put into use. The preset total life value is the maximum number of products that the tool can process with qualified dimensions.

3. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 1, characterized in that, The four wear intervals are respectively a possible wear occurrence interval, a repairable wear interval, a non-repairable but non-failed wear interval, and a failure requiring replacement interval. The wear state division module performs interval segmentation according to the four wear intervals, and respectively determines a first threshold N1, a second threshold N2, and a third threshold N3 based on the preset total life value and the preset wear amount in combination with the mapping relationship between the number of machining times and the theoretical tool wear amount, which are specifically expressed as: The cumulative number of machined workpieces 0<N≤N1 is the possible wear occurrence interval; The cumulative number of machined workpieces N1<N≤N2 is the repairable wear interval; The cumulative number of machined workpieces N2<N≤N3 is the non-repairable but non-failed wear interval; The cumulative number of machined workpieces N>N3 is the failure requiring replacement interval; Wherein, the first threshold N1, the second threshold N2, and the third threshold N3 are all critical machining times solved based on the preset critical wear amount of each wear interval, and satisfy 0<N1<N2<N3.

4. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 3, characterized in that, When the strategy decision module determines that the tool is in the possible wear occurrence interval, the disposal strategy is: output a primary warning prompt, and simultaneously issue a cutting parameter fine-tuning instruction to the execution regulation module. The execution regulation module reduces the cutting feed rate, and at the same time the real-time working condition monitoring module continuously collects real-time working condition data of on-site machining and transmits it back to the storage module, which is used for iteratively updating the correlation relationship between the tool wear amount and the single-workpiece machining time.

5. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 3, characterized in that, When the strategy decision module determines that the tool is in the repairable wear interval, the disposal strategy is: output an intermediate wear warning prompt, issue a tool local sharpening compensation instruction to the execution regulation module. The execution regulation module invokes the tool local sharpening mechanism to perform local sharpening on the tool cutting edge part. After the local sharpening is completed, the tool wear modeling module updates the preset wear amount and the preset total life value based on the real-time working condition data collected after the local sharpening.

6. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 3, characterized in that, When the strategy decision module determines that the tool is in the non-repairable but non-failed wear interval, the disposal strategy is: output an advanced warning prompt. The execution regulation module adjusts the cutting parameters of the CNC machining machine tool to perform dimension compensation by reducing the cutting speed and decreasing the cutting depth, and simultaneously records the workpiece dimension deviation data of each machining in this wear interval.

7. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 3, characterized in that, When the strategy decision module determines that the tool is in the failure requiring replacement interval, the disposal strategy is: output a tool failure alarm prompt. The execution regulation module issues a tool change instruction, suspends the machining task, outputs a tool change reminder or directly executes the tool change action. After the tool change is completed, the system resets the real-time number of single-workpiece machining times and starts full life cycle monitoring of the new tool.

8. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 1, characterized in that, The data acquisition module is connected with an external workpiece dimension detection device to collect the true detection value of the actual workpiece dimension after the workpiece is machined. The storage module binds and stores the true detection value of the actual workpiece dimension with the wear data of the corresponding tool. The tool wear modeling module periodically reads the bound data in the storage module to iteratively correct the correlation relationship between the tool wear amount and the single-workpiece machining time, and synchronously updates the preset wear amount and the preset total life value under the corresponding working condition.

9. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 7, characterized in that, The real-time working condition monitoring module is also used to receive sensor data of vibration and cutting sound of CNC machine tools in real time, and send the sensor data as auxiliary features into the strategy decision module. When the sensor data is abnormal, the strategy decision module corrects the current wear interval judgment result based on the abnormal data. When the cumulative number of real-time single workpiece processing does not reach the wear interval threshold, the warning and handling strategy of the corresponding upper-level wear interval is triggered in advance.

10. The intelligent early warning and real-time compensation system for CNC machining tool wear according to claim 7, characterized in that, The storage module contains a parameter library corresponding to different tool-workpiece pairs. The parameter library includes the types of tools that can be regrinded, the maximum number of regrinding cycles allowed, and the upper and lower limits of adjustable cutting parameters.

Images