A method and system for optimizing cutting parameters of CNC lathes

Abstract

This application relates to the field of CNC lathe machining technology, specifically to a method and system for optimizing cutting parameters of a CNC lathe. The method includes: receiving a machining task for a batch of workpieces, dividing the machining task into machining batches, and determining detection machining steps; during the detection machining steps, acquiring cutting process signals and machining parameters of the detection machine tool in real time; dynamically establishing a expected stable signal range based on the cutting process signals and their corresponding machine tool machining parameters, identifying and quantifying transient impact events; statistically and normally processing the transient impact events to generate risk assessment information; sharing the risk assessment information with other machine tools within the machining batch; pre-adjusting the cutting parameters of subsequent machining steps based on the risk assessment information; and executing the machining of the batch of workpieces after pre-adjustment. This solves the problem of decreased production efficiency and stability caused by information sharing issues in multi-machine collaborative machining scenarios for existing CNC lathes.

Description

Technical Field

[0001] This application relates to the field of CNC lathe machining technology, and more specifically, to a method and system for optimizing CNC lathe cutting parameters. Background Technology

[0002] In modern industrial production, CNC lathes, as core machining equipment, directly affect workpiece machining quality, production efficiency, and tool life through the setting of their cutting parameters. However, traditional evaluation methods based on local single-machine data and empirical models have significant limitations when dealing with the inhomogeneity of the workpiece material's microstructure. Especially when the workpiece material has randomly distributed hard particles due to new processing techniques, traditional methods struggle to identify and predict the resulting instantaneous impact wear on the tool, leading to decreased efficiency and increased costs.

[0003] In typical workshop settings, each CNC lathe is usually equipped with an independent tool wear condition assessment system, relying on sensor data such as spindle power, average cutting force, and vibration, combined with empirical models for judgment. This type of method is suitable for macroscopically smooth signals but struggles to capture transient anomalies caused by hard particles. When the tool encounters a hard particle, the cutting force spikes dramatically within a very short time, accompanied by high-frequency vibrations, leading to micro-chipping of the cutting edge, rapid expansion of the wear band, and even localized fracture. Because traditional models cannot analyze such transient signals, the output results continuously deviate from the actual tool condition, increasing the difficulty of tool management.

[0004] The problems are exacerbated in multi-machine collaborative machining. Each machine tool's evaluation model operates independently, relying on its own local data and lacking information sharing. For example, when a machine tool experiences rapid tool wear due to frequent encounters with hard points, other machines are unaware and unable to adjust their machining strategies. This information silo effect amplifies the impact of localized problems: if a machine tool is forced to stop for tool replacement due to unexpected wear, the downtime disrupts the overall production rhythm. Because other machine tools' scheduling relies on independent forecasts, they cannot respond promptly to emergencies, often leading to material accumulation or waiting in subsequent processes, damaging production line coordination and efficiency, and ultimately extending delivery cycles. Traditional CNC lathe cutting parameter optimization methods, when facing new workpiece materials and parallel machining modes, fail to accurately identify transient impact events and lack cross-machine risk information sharing, resulting in significantly limited tool utilization and overall production line efficiency.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application discloses a method for optimizing cutting parameters of CNC lathes, which aims to solve the problem of decreased production efficiency and stability caused by information not being shared in multi-machine collaborative machining scenarios of existing CNC lathes.

[0007] The technical solution of this application is as follows:

[0008] In a first aspect, this application discloses a method for optimizing cutting parameters of a CNC lathe, specifically including:

[0009] Receive machining tasks for batches of workpieces, divide the machining tasks into machining batches, and determine the detection machine tools and detection machining steps for the batches of workpieces;

[0010] During the detection and machining process, the cutting process signals of the detection machine tool and its corresponding machine tool machining parameters are collected in real time.

[0011] Based on the cutting process signal and its corresponding machine tool processing parameters, the expected stable signal range is dynamically established to identify and quantify transient impact events caused by hard points inside the batch workpiece material.

[0012] Statistical and normalization processing is performed on transient impact events to generate risk assessment information for batch workpiece materials;

[0013] Share risk assessment information with other machine tools within the processing batch;

[0014] Based on the risk assessment information, the cutting parameters for subsequent processing steps are pre-adjusted, and after the pre-adjustment is completed, the batch of workpieces is processed.

[0015] Furthermore, identify transient impact events caused by hard points within the batch workpiece material, including:

[0016] Before the detection and processing steps, obtain the reference signal of the detection machine tool performing no-load motion;

[0017] Based on the reference signal, a background noise envelope characterizing the noise characteristics of the detection machine tool itself is generated;

[0018] The cutting process signal is compared with the background noise envelope to remove background noise and obtain a purified cutting signal.

[0019] Based on the purified cutting signal and machine tool processing parameters, combined with the expected stable signal range, abnormal over-limit sections and the start and end of events are determined, and transient impact events caused by hard points inside the batch workpiece material are identified.

[0020] Furthermore, the transient impact events caused by hard particles within the batch workpiece material are quantified, including:

[0021] The transient change characteristics of the purification cutting signal are extracted within the preset analysis time window. The transient change characteristics include the signal rise rate, the duration of the signal peak, and the signal energy concentration.

[0022] Based on the machine tool processing parameters, the signal characteristic pattern of the transient impact event is determined. The signal characteristic pattern includes the range of expected signal rise rate, the range of expected signal peak duration, and the range of expected signal energy concentration.

[0023] The transient change characteristics are compared with the signal characteristic patterns. When the transient change characteristics match the signal characteristic patterns, the corresponding signal waveform in the purification cutting signal is confirmed to be a transient impact event.

[0024] The peak intensity of a transient impact event is quantified by the difference between the actual peak value of the transient impact event and the expected peak value of the signal characteristic pattern.

[0025] Furthermore, quantifying transient impact events caused by hard particles within the material of a batch of workpieces also includes:

[0026] Real-time monitoring of the purification cutting signal;

[0027] The transient impact event identification window is determined based on the machine tool processing parameters;

[0028] When multiple consecutive or superimposed transient impact waveforms of the purification cutting signal are detected within the transient impact event identification window, a transient impact waveform fusion mechanism is performed to obtain the fused transient impact waveform.

[0029] Based on the peak intensity and duration of each transient impact event, the deviation of the fused transient impact waveform from the upper bound of the expected stable signal range within the transient impact event identification window is integrated to obtain the integration area, and the integration area is determined as the cumulative impact load intensity.

[0030] The length of the transient impact event identification window is dynamically adjusted based on the frequency of occurrence and average duration of transient impact events.

[0031] Furthermore, after processing a batch of workpieces, the process also includes:

[0032] Based on real-time acquired cutting process signals, the risk assessment information is iteratively updated;

[0033] When there is a discrepancy between the updated risk assessment information and the shared risk assessment information within the processing batch, different machine tools will perform weighted correction on the shared risk assessment information according to their own machine tool processing parameters, and dynamically adjust the adjustment range of cutting parameters based on the corrected risk assessment information to generate cutting parameter adjustment results adapted to each machine tool.

[0034] Furthermore, based on the frequency and average duration of transient impact events, the length of the transient impact event identification window is dynamically adjusted, including:

[0035] Real-time calculation of the frequency of occurrence and average duration of transient impact events within the current transient impact event identification window;

[0036] Based on the machine tool processing parameters, dynamically set the impact event intensity threshold that matches the current processing conditions;

[0037] When the frequency or average duration of an impact event exceeds the impact event density threshold, shorten the length of the transient impact event identification window.

[0038] When the frequency and average duration of the event are consistently below the impact event density threshold, the length of the transient impact event identification window is extended.

[0039] Furthermore, the frequency and average duration of transient impact events within the current transient impact event identification window are calculated in real time, including:

[0040] Based on the cleaned cutting signal, the start time, end time and duration of the identified transient impact events are recorded;

[0041] Based on the machine tool machining parameters, the length of the cutting path segment corresponding to the current machining state is dynamically determined, and the length of the cutting path segment is used to limit the length of the transient impact event identification window.

[0042] The number of transient impact events within the cutting path segment is counted, and the durations of the transient impact events are summed to obtain the total duration.

[0043] Divide the number of transient impact events by the length of the cutting path segment to obtain the frequency of transient impact events.

[0044] Divide the total duration by the number of transient impact events to obtain the average duration of the transient impact events.

[0045] Furthermore, based on the machine tool machining parameters, the length of the cutting path segment corresponding to the current machining state is dynamically determined. The machine tool machining parameters include cutting speed, specifically:

[0046] Real-time monitoring of the current cutting speed of the detection machine tool;

[0047] Based on the current cutting speed, predict the trend of the cutting speed change of the detection machine tool within a future preset time window;

[0048] Based on the cutting speed change trend and combined with the current cutting speed of the machine tool, the length of the cutting path segment that matches the cutting speed change trend is dynamically determined. Specifically, when the cutting speed change trend indicates that the cutting speed will continue to increase, the length of the cutting path segment is increased; when the cutting speed change trend indicates that the cutting speed will continue to decrease, the length of the cutting path segment is decreased; and when the cutting speed change trend indicates that the cutting speed will remain stable, the length of the cutting path segment is kept unchanged.

[0049] Furthermore, predicting the cutting speed change trend of the detection machine tool within a future preset time window includes:

[0050] Real-time monitoring of cutting process signals, including cutting force signals and vibration signals;

[0051] Based on the machine tool processing parameters and with reference to the expected stable signal range, the dynamic threshold range of the cutting force signal and the dynamic threshold range of the vibration signal are dynamically determined.

[0052] Spectral analysis is performed on the cutting force signal and vibration signal to identify the first periodic interference component or the first quasi-periodic interference component in the cutting force signal that overlaps with the natural frequency or harmonic frequency of the machine tool's own components, and to identify the second periodic interference component or the second quasi-periodic interference component in the vibration signal that overlaps with the natural frequency or harmonic frequency of the machine tool's own components.

[0053] When the cutting force signal exceeds the dynamic threshold range of the cutting force signal, and the excess portion does not match the first periodic interference component or the first quasi-periodic interference component, it is determined that there is transient interference in the cutting process signal or abnormal sensor data; and

[0054] When the vibration signal exceeds the dynamic threshold range of the vibration signal, and the excess part does not match the second periodic interference component or the second quasi-periodic interference component, it is determined that there is instantaneous interference in the cutting process signal or abnormal sensor data.

[0055] When there is instantaneous interference in the cutting process signal or abnormal sensor data, the cutting speed data is corrected by using historical cutting speed data or cutting speed data at adjacent time points to obtain corrected cutting speed data.

[0056] Based on the corrected cutting speed data, the trend of cutting speed change of the detection machine tool within a future preset time window is predicted.

[0057] Secondly, this application also discloses a CNC lathe cutting parameter optimization system, specifically including:

[0058] The order receiving module is used to receive processing tasks for batches of workpieces, divide the processing tasks into processing batches, and determine the detection machine tool and its detection processing steps for the batch of workpieces.

[0059] The machining monitoring module is used to collect the cutting process signals of the detection machine tool and its corresponding machine tool machining parameters in real time during the detection machining steps;

[0060] The impact event identification and quantification module is used to dynamically establish the expected stable signal range based on the cutting process signal and its corresponding machine tool processing parameters, and to identify and quantify transient impact events caused by hard points inside the batch workpiece material.

[0061] The risk assessment module is used to perform statistical and normalization processing on transient impact events and generate risk assessment information for batch workpiece materials.

[0062] The information sharing module is used to share risk assessment information with other machine tools within the scope of a processing batch;

[0063] The parameter adjustment module is used to pre-adjust the cutting parameters of subsequent processing steps based on risk assessment information, and then perform batch processing of workpieces after the pre-adjustment is completed.

[0064] Beneficial Effects: The CNC lathe cutting parameter optimization method disclosed in this application achieves targeted optimization for specific batches of workpieces by receiving machining tasks, dividing machining batches, determining the detection machine tool and detection machining steps. During the detection machining steps, cutting process signals and machine tool machining parameters are acquired in real time, providing accurate raw data for subsequent analysis. Based on this data, a dynamic expected stable signal range is established, and transient impact events caused by hard points inside the batch workpiece material are identified and quantified. This innovative step effectively captures sudden tool wear risks that are difficult to detect using traditional methods, solving the problem of abnormal tool wear caused by the inhomogeneity of the workpiece material's microstructure. Subsequently, the transient impact events are statistically and normally processed to generate risk assessment information for the batch workpiece material, transforming discrete impact events into quantifiable risk indicators. More importantly, this method shares risk assessment information with other machine tools within the processing batch, breaking the information silos of traditional single-machine local assessments. This enables multiple machine tools to work together to address common material risks. This application can more accurately predict and prevent sudden tool wear, avoid efficiency decline caused by conservative strategies, and significantly improve the stability and economic benefits of CNC lathes in the processing of complex workpieces. Attached Figure Description

[0065] Figure 1 This is a flowchart illustrating a method for optimizing cutting parameters on a CNC lathe, as provided in this application.

[0066] Figure 2 A flowchart of a CNC lathe cutting parameter optimization system provided in this application.

[0067] In the diagram: 1. Order receiving module; 2. Processing monitoring module; 3. Impact event identification and quantification module; 4. Risk assessment module; 5. Information sharing module; 6. Parameter adjustment module. Detailed Implementation

[0068] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0069] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0070] Reference Figure 1 This application proposes a method for optimizing cutting parameters of a CNC lathe, including:

[0071] S1000: Receives a processing task for a batch of workpieces, divides the processing task into processing batches, and determines the detection machine tool and its detection processing steps for the batch of workpieces.

[0072] S2000: During the detection and machining process, the cutting process signal of the detection machine tool and its corresponding machine tool machining parameters are collected in real time;

[0073] S3000: Based on the cutting process signal and its corresponding machine tool processing parameters, dynamically establish the expected stable signal range, identify and quantify transient impact events caused by hard points inside the batch workpiece material;

[0074] S4000: Performs statistical and normalization processing on transient impact events to generate risk assessment information for batch workpiece materials;

[0075] S5000: Shares risk assessment information with other machine tools within the scope of the processing batch;

[0076] S6000: Based on risk assessment information, the cutting parameters for subsequent machining steps are pre-adjusted, and the batch of workpieces is machined after the pre-adjustment is completed.

[0077] Specifically, a batch of workpieces refers to a group of workpieces with the same material, design requirements, and processing technology within a single production task. Unified management and optimization of batch workpieces helps improve production efficiency and product consistency. "Probing machine tool" refers to the CNC lathe selected to perform the probing machining step within the processing batch. This machine tool is responsible for collecting key data during the actual cutting process, providing a basis for risk assessment of the entire batch. "Probing machining step" refers to the pre-processing stage performed on the probing machine tool before the formal processing of the batch workpieces, used to collect cutting process signals and machine tool processing parameters. This step aims to simulate actual processing conditions to identify potential material problems. "Cutting process signals" refer to signals reflecting the cutting state, such as cutting force, vibration, and acoustic emission, collected in real time by sensors during the cutting process. These signals are key data for identifying transient impact events caused by hard points in the material. "Machine tool processing parameters" refer to various process parameters set during CNC lathe machining, such as spindle speed, feed rate, depth of cut, and tool type. These parameters directly affect the cutting process and signal characteristics. The "expected stable signal range" refers to the stable range that the cutting process signal should be within under normal cutting conditions without hard particle impact. This range serves as the benchmark for judging transient impact events. Transient impact events refer to phenomena caused by hard particles within the workpiece material, resulting in abnormal spikes or violent fluctuations in the cutting process signal within a very short time. Identifying and quantifying such events is the core of this application. Risk assessment information refers to comprehensive information reflecting the distribution and impact degree of hard particles within the batch of workpiece material, obtained through statistical and normalization processing of transient impact events. This information is used to guide the adjustment of subsequent cutting parameters.

[0078] Specifically, upon receiving a machining task for a batch of workpieces, the first step is to divide the machining task into machining batches and determine the probing machine tools and probing machining steps for each batch. For example, machining batches can be divided based on the workpiece's material batch number, production order number, or similarity in machining processes. Within each machining batch, one or more CNC lathes can be designated as probing machine tools, and one or more probing machining steps can be planned for them. Probing machining steps can be a representative operation in the actual machining process or a specially designed testing operation.

[0079] During the probing and machining process, it is necessary to acquire the cutting process signals and corresponding machining parameters of the probing machine tool in real time. These cutting process signals can be obtained through various sensors installed on the probing machine tool. For example, piezoelectric sensors can be used to acquire cutting force signals, accelerometers to acquire vibration signals, or acoustic emission sensors to acquire acoustic emission signals. These sensors transmit real-time data to the data acquisition system. Simultaneously, machining parameters, such as spindle speed, feed rate, and depth of cut, can be obtained from the CNC system or machine tool controller.

[0080] Based on the acquired cutting process signals and their corresponding machine tool machining parameters, a dynamic expected stable signal range is established, and transient impact events caused by hard particles within the batch workpiece material are identified and quantified. The expected stable signal range can be established by statistically analyzing the cutting process signals in areas without hard particles, such as calculating the mean and standard deviation, and setting a confidence interval. The identification of transient impact events can be achieved through signal processing techniques. For example, when the cutting process signal exceeds the upper or lower bound of the expected stable signal range within a very short time and exhibits spikes or violent fluctuations, it can be preliminarily identified as a transient impact event. Quantifying transient impact events can include evaluating parameters such as peak intensity, duration, and energy.

[0081] Statistical and normalization processing is performed on identified and quantified transient impact events to generate risk assessment information for batches of workpiece materials. Statistical processing may include calculating the frequency of transient impact events, average peak intensity, and cumulative impact load. Normalization processing can unify impact event parameters of different types or dimensions to a comparable scale; for example, normalizing peak intensity to the range of 0-1. The final risk assessment information can be a comprehensive index reflecting the distribution density of hard particles within the batch of workpiece material, its hardness grade, and its potential impact on tool wear.

[0082] Within the processing batch, the generated risk assessment information is shared with other machine tools. Information sharing can be achieved in several ways. For example, risk assessment information can be uploaded to a central server via a local area network, making it accessible to all CNC lathes involved in processing that batch of workpieces; or it can be transmitted directly between machine tools point-to-point via wireless communication modules. The purpose of sharing is to ensure that all relevant machine tools can promptly obtain the risk characteristics of the workpiece material in that batch.

[0083] Based on shared risk assessment information, cutting parameters for subsequent machining steps are pre-adjusted. For example, if the risk assessment indicates that the batch of workpiece material has many hard particles and high impact strength, the cutting speed, feed rate, or depth of cut in subsequent machining steps can be reduced in advance to decrease the risk of tool wear. Alternatively, a more wear-resistant tool material or tool geometry can be selected. After pre-adjustment, the batch of workpieces can be machined.

[0084] This application's method for optimizing CNC lathe cutting parameters lays the foundation for subsequent data acquisition and analysis by receiving machining tasks, dividing machining batches, and determining the detection machine tool and machining steps. During the detection machining steps, cutting process signals and machine tool machining parameters are acquired in real time, providing raw data for identifying transient impact events caused by hard points within the material. Based on this data, a dynamic expectation of a stable signal range is established, and transient impact events are identified and quantified, accurately capturing sudden tool wear causes that are difficult to detect using traditional methods. Subsequently, the transient impact events are statistically and normally processed to generate risk assessment information for the batch of workpiece materials, transforming discrete impact events into quantifiable risk indicators. Crucially, this risk assessment information is shared with other machine tools within the machining batch, breaking the limitations of traditional single-machine local optimization and achieving collaborative optimization among multiple machine tools. Finally, based on the shared risk assessment information, the cutting parameters for subsequent machining steps are pre-adjusted, and machining is performed after adjustment, effectively avoiding the risk of sudden tool wear caused by material inhomogeneity and improving machining efficiency and tool life.

[0085] Furthermore, identify transient impact events caused by hard points within the batch workpiece material, including:

[0086] S3100: Identifies transient impact events caused by hard particles within the material of a batch of workpieces, including:

[0087] S3200: Before the detection and machining step, acquire the reference signal for the detection machine tool to perform no-load motion;

[0088] S3300: Based on the reference signal, generate a background noise envelope characterizing the noise characteristics of the detection machine tool itself;

[0089] S3400: Compares the cutting process signal with the background noise envelope, removes background noise, and obtains a purified cutting signal;

[0090] S3500: Based on the cleaned cutting signal and machine tool processing parameters, combined with the expected stable signal range, it is used to determine the abnormal over-limit section and the start and end of the event, and to identify transient impact events caused by hard points inside the batch workpiece material.

[0091] Specifically, before the probing process begins, the probing machine tool needs to be run unloaded to obtain a reference signal under no-cutting load conditions. This reference signal is mainly used to capture background interference such as mechanical vibration and electrical noise during normal operation of the probing machine tool, providing a benchmark for subsequent noise stripping. For example, vibration, force, or sound signals during unloaded operation can be collected using accelerometers, force sensors, or acoustic sensors installed on the probing machine tool.

[0092] Based on the acquired no-load motion reference signal, a background noise envelope characterizing the inherent noise properties of the detection machine tool can be generated. This envelope is a quantitative representation of the inherent noise level of the detection machine tool, reflecting the amplitude and frequency distribution of the background noise generated by the detection machine tool system when there is no workpiece cutting. Generation methods may include filtering, rectifying, and smoothing the reference signal, or using statistical methods (such as mean, standard deviation, and RMS value) to construct a dynamic noise envelope.

[0093] The real-time acquired cutting process signal is compared with the aforementioned background noise envelope. By treating the portion of the cutting process signal that overlaps with or falls below the background noise envelope as background noise and removing it, a purified cutting signal is obtained. This step aims to eliminate the interference of the machine tool's own noise on the cutting process signal, making the subsequent identification of transient impact events more accurate. For example, adaptive filtering, wavelet denoising, or threshold-based signal truncation methods can be used to remove background noise.

[0094] Based on the cleaned cutting signal, machine tool machining parameters, and the expected stable signal range dynamically established in the above steps, abnormal out-of-bounds sections in the signal can be identified, and the start and end points of transient impact events can be determined. The expected stable signal range represents the signal fluctuation range under normal cutting conditions. When the cleaned cutting signal significantly exceeds this expected stable signal range within a specific time period, it is identified as an abnormal out-of-bounds section. Combining machine tool machining parameters (such as feed rate, spindle speed, depth of cut, etc.), it can be further confirmed whether these abnormal out-of-bounds sections conform to the characteristics of transient impact events caused by hard particles inside the batch workpiece material, thereby identifying transient impact events.

[0095] The proposed solution first acquires a reference signal of the no-load motion of the probing machine tool before the machining step, and then generates a background noise envelope based on this reference signal, providing a clean noise benchmark for subsequent signal processing. During the actual cutting process, comparing the real-time acquired cutting process signal with this background noise envelope effectively removes background noise, resulting in a more realistic purified cutting signal. This process ensures that subsequent analysis of transient impact events is not interfered with by the inherent noise of the probing machine tool itself. Furthermore, based on the purified cutting signal, machine tool machining parameters, and a pre-established expected stable signal range, by determining the abnormal out-of-bounds segments of the signal and identifying the start and end points of the event, transient impact events caused by hard points inside the batch workpiece material can be accurately identified. This step-by-step processing and multi-dimensional judgment method significantly improves the accuracy and reliability of transient impact event identification.

[0096] This application further proposes a method for quantifying transient impact events caused by hard particles inside batch workpiece materials, specifically including:

[0097] S3600: Extracts transient change characteristics of the clean cutting signal within a preset analysis time window. The transient change characteristics include signal rise rate, signal peak duration, and signal energy concentration.

[0098] S3700: Based on the machine tool processing parameters, determine the signal characteristic pattern of the transient impact event. The signal characteristic pattern includes the expected signal rise rate range, the expected signal peak duration range, and the expected signal energy concentration range.

[0099] S3800: Compare transient change characteristics with signal characteristic patterns. When the transient change characteristics match the signal characteristic patterns, confirm that the corresponding signal waveform in the purification cutting signal is a transient impact event.

[0100] S3900: Quantify the peak intensity of a transient impact event based on the difference between the actual peak value of the transient impact event and the expected peak value of the signal characteristic pattern.

[0101] Specifically, the preset analysis time window refers to a continuous time interval defined in the cutting process signal, used for focused analysis of possible transient impact events. The length of this window can be dynamically adjusted according to the specific machining material, cutting speed, and machine tool characteristics to ensure that the complete waveform of the transient impact event can be effectively captured.

[0102] Among them, transient change characteristics describe the rapid change characteristics exhibited by the purification cutting signal within a preset analysis time window. Signal rise rate refers to the time or slope required for the signal to rise rapidly from the baseline level to the peak, reflecting the suddenness of the impact event; signal peak duration refers to the length of time the signal remains at a high level after reaching the peak, reflecting the persistence of the impact event; signal energy concentration refers to the distribution of signal energy within a specific frequency range, reflecting the energy intensity of the impact event. These characteristics can be extracted using digital signal processing techniques, such as wavelet transform, short-time Fourier transform, or envelope demodulation.

[0103] In practical applications, signal characteristic patterns are pre-defined based on historical processing data, material property databases, or expert experience, and are used to characterize the signal characteristic range of typical transient impact events. For example, the expected signal rise rate range can be defined as a time interval on the order of milliseconds, the expected signal peak duration range can be defined as a time interval on the order of microseconds to milliseconds, and the expected signal energy concentration range can be defined as a certain energy threshold range. The determination of these ranges aims to distinguish real impact events caused by hard particles from background noise or normal cutting fluctuations.

[0104] The purpose of comparing transient change characteristics with signal characteristic patterns is to accurately identify transient impact events caused by hard points inside the material of a batch of workpieces through pattern matching. When the extracted transient change characteristics (such as the actual signal rise rate, peak duration, and energy concentration) fall within the preset signal characteristic pattern (such as the expected range), the signal waveform can be confirmed as a transient impact event.

[0105] Quantifying the peak intensity of a transient impact event is achieved by calculating the difference between the actual peak value of the signal and the expected peak value of the signal's characteristic pattern. For example, the ratio of the actual peak value to the expected peak value, or the absolute difference between the actual and expected peak values, can be calculated. This quantification method provides an intuitive indicator reflecting the severity of a hard-point impact, offering more refined data support for subsequent risk assessments.

[0106] This application's solution addresses the problem of merely identifying transient impact events without assessing their severity by introducing a refined quantification mechanism. Specifically, by extracting the transient change characteristics of the cleaned cutting signal within a preset analysis time window and comparing them with signal characteristic patterns determined based on machine tool machining parameters, it is possible to effectively distinguish real transient impact events caused by hard particles from other signal fluctuations. The accurate capture of key characteristics such as signal rise rate, peak duration, and energy concentration makes the identification of transient impact events more accurate. Furthermore, by calculating the difference between the actual peak value of the transient impact event and the expected peak value of the signal characteristic pattern, the peak intensity of the impact event can be directly quantified, providing a specific and operable quantitative indicator for subsequent risk assessment. This quantification method elevates the assessment of the hazard level of hard particles within batch workpiece materials from qualitative identification to quantitative analysis, significantly enhancing the accuracy and reliability of risk assessment.

[0107] In some preferred embodiments, the following specific example illustrates the situation:

[0108] Suppose that during the CNC machining of a batch of aero-engine blades, the machine tool acquires cutting process signals in real time during the machining detection step. After purifying the cutting signals, the system extracts a signal waveform within a preset analysis time window, such as a 50-millisecond window. This waveform has a signal rise rate of 0.5V / ms, a peak duration of 20 microseconds, and a signal energy concentration of 80%. Simultaneously, based on the current machine tool machining parameters (such as cutting speed, feed rate, and depth of cut), the system determines the signal characteristic pattern of a transient impact event, with an expected signal rise rate range of 0.4-0.6V / ms, an expected peak duration range of 15-25 microseconds, and an expected signal energy concentration range of 75-85%. Comparison reveals that the extracted transient change characteristics perfectly match the signal characteristic pattern, thus confirming the signal waveform as a transient impact event. Furthermore, the actual peak value of this transient impact event is 1.2V, while the expected peak value of the signal characteristic pattern is 1.0V. Based on the difference between the actual signal peak value and the expected peak value (1.2V - 1.0V = 0.2V), the system quantifies the peak intensity of the transient impact event as 0.2V. This quantized value is then used to update the risk assessment information for the batch of workpiece materials, guiding more precise adjustments to the cutting parameters of subsequent machining steps. For example, for areas with high peak intensity, the cutting speed or feed rate can be appropriately reduced to avoid tool damage.

[0109] This application further proposes a method for quantifying transient impact events caused by hard particles inside batch workpiece materials, which also includes:

[0110] S31000: Real-time monitoring and purification cutting signals;

[0111] S31100: Determine the transient impact event identification window based on the machine tool processing parameters;

[0112] S31200: When multiple consecutive or superimposed transient impact waveforms of the purification cutting signal are detected within the transient impact event identification window, a transient impact waveform fusion mechanism is performed to obtain the fused transient impact waveform.

[0113] S31300: Based on the peak intensity and duration of each transient impact event, the deviation of the fused transient impact waveform from the upper bound of the expected stable signal range within the transient impact event identification window is integrated to obtain the integration area, and the integration area is determined as the cumulative impact load intensity.

[0114] S31400: Dynamically adjusts the length of the transient impact event identification window based on the frequency and average duration of transient impact events.

[0115] Specifically, real-time monitoring and purification of cutting signals refers to continuously acquiring the cutting signal data stream after background noise has been removed using sensors, in order to promptly capture any transient impact events that may occur. This monitoring process can be implemented using a high-speed data acquisition system to ensure signal integrity and real-time performance.

[0116] Determining the transient impact event identification window based on machine tool machining parameters refers to setting a suitable time or space range based on current machine tool machining parameters such as cutting speed, feed rate, and depth of cut, for the centralized analysis and identification of transient impact events. For example, during high-speed cutting, the window length may be shorter to accommodate rapidly changing cutting conditions; during low-speed cutting, the window length may be longer to ensure that all potential impact events are captured.

[0117] In practical applications, when multiple consecutive or superimposed transient impact waveforms of the purification cutting signal are detected within the transient impact event identification window, a transient impact waveform fusion mechanism is implemented to obtain a fused transient impact waveform. This means that when multiple impact events are very close in time or partially overlap, the system does not treat them as independent events and simply superimpose them. Instead, it uses a specific algorithm (e.g., envelope fusion, weighted averaging, or energy-based fusion) to merge them into a fused waveform that better represents the overall impact effect. The aim is to more accurately assess the actual load under complex impact scenarios.

[0118] Based on the peak intensity and duration of each transient impact event, the deviation of the fused transient impact waveform from the upper bound of the expected stable signal interval within the transient impact event identification window is integrated to obtain the integration area, which is then determined as the cumulative impact load intensity. This means that it not only focuses on the peak value of a single impact, but also comprehensively considers the impact intensity, duration, and cumulative effect above the expected stable signal interval through integral calculation, thereby obtaining a quantitative index that better reflects the overall influence of hard particles in the material. The upper bound of the expected stable signal interval can be understood as the upper threshold of the cutting signal under normal cutting conditions.

[0119] The length of the transient impact event identification window is dynamically adjusted based on the frequency and average duration of transient impact events. The aim is to enable the identification window to adaptively match the distribution characteristics of impact events during actual processing. For example, when impact events occur frequently or last for a long time, the window length may need to be shortened to improve response speed and local accuracy; when impact events are sparse, the window length can be extended to ensure that all events are captured and a more macroscopic analysis is performed.

[0120] The proposed solution ensures continuous awareness of the cutting process by real-time monitoring and purification of cutting signals. By determining the transient impact event identification window based on machine tool machining parameters, the analysis range can adapt to different machining conditions. When multiple continuous or superimposed transient impact waveforms are detected within the identification window, a transient impact waveform fusion mechanism integrates these complex impact signals into a unified fused waveform, avoiding errors that may result from isolated analysis of individual events. Subsequently, the cumulative impact load intensity is calculated by integrating the deviation of the fused transient impact waveform from the upper bound of the expected stable signal range. This indicator comprehensively considers both the impact intensity and duration, thus more comprehensively quantifying the overall impact of material hard particles on the tool and workpiece. Finally, the length of the identification window is dynamically adjusted according to the occurrence frequency and average duration of transient impact events, enabling the system to adaptively optimize the detection and analysis efficiency of impact events and ensure accurate evaluation results under different impact densities.

[0121] In some preferred embodiments, the following specific example illustrates the situation:

[0122] Suppose that during the CNC turning of a batch of aero-engine blades, the machine tool collects the purification cutting signal in real time during the cutting process. At a certain stage of the machining process, due to the presence of multiple tiny hard particles inside the blade material, the purification cutting signal exhibits three transient impact waveforms with similar amplitudes but partial overlap within a short period of time (e.g., 50 milliseconds).

[0123] First, the system determines a transient impact event recognition window based on current machine tool machining parameters such as cutting speed and feed rate, for example, set to 100 milliseconds. When three consecutive or superimposed transient impact waveforms are detected within this 100-millisecond window, the system activates a transient impact waveform fusion mechanism. For example, a signal envelope extraction algorithm can be used to fuse these three waveforms into a single, smoother fused transient impact waveform, which can reflect the overall energy and duration of the three impact events.

[0124] Next, the system integrates the deviation of the fused transient impact waveform from the upper bound of the expected stable signal interval within the identification window, based on the peak intensity and duration of each of the three transient impact events. For example, if the upper bound of the expected stable signal interval is 100 units, and the fused waveform reaches 150 units at a certain moment and persists for a period of time, then this 50-unit deviation and its duration will be integrated. The resulting integration area, for example, 2000 units * milliseconds, will be determined as the cumulative impact load intensity within that time period.

[0125] Simultaneously, the system calculates the occurrence frequency (e.g., 3 events / 100 milliseconds) and average duration of these three impact events in real time. Based on this information, the system dynamically adjusts the length of the transient impact event recognition window. For example, if the impact events occur frequently, the system may shorten the recognition window to 80 milliseconds to more precisely capture and analyze locally high-density impact events; if the impact events are sparse, the window may be extended to 150 milliseconds to ensure broader coverage. In this way, the system can more accurately assess the cumulative impact of material hard particles on the machining process and provide more refined guidance for subsequent adjustments to cutting parameters.

[0126] Furthermore, in another embodiment of this application, after processing a batch of workpieces, the process further includes:

[0127] S7000: Based on real-time acquired cutting process signals, risk assessment information is iteratively updated;

[0128] S8000: When there is a discrepancy between the updated risk assessment information and the shared risk assessment information within the processing batch, different machine tools will perform weighted correction on the shared risk assessment information according to their own machine tool processing parameters, and dynamically adjust the adjustment range of the cutting parameters based on the corrected risk assessment information to generate cutting parameter adjustment results adapted to each machine tool.

[0129] Specifically, real-time acquired cutting process signals can include, but are not limited to, cutting force signals, vibration signals, acoustic emission signals, or spindle current signals. These signals directly reflect the current cutting state and the actual response of the workpiece material. By analyzing these real-time acquired cutting process signals, more accurate information can be obtained regarding the distribution of hard particles in the material, the intensity and frequency of transient impact events, thereby correcting and improving the initially generated risk assessment information. Iterative updates refer to periodically or event-driven recalculation or adjustment of the risk assessment information based on the latest cutting process signal data, enabling it to dynamically reflect changes in material properties during machining.

[0130] Furthermore, when the updated risk assessment information deviates from the shared risk assessment information within the machining batch scope—for example, when the updated risk assessment information shows a higher transient impact risk or more frequent transient impact events than the initial assessment—correction is required. In this case, different machine tools will weight and correct the shared risk assessment information based on their respective machine tool machining parameters. Machine tool machining parameters may include machine tool model, service life, maintenance records, tool wear status, current load, spindle speed, feed rate, etc. These parameters reflect the individual characteristics and current operating status of each machine tool. Through weighted correction, the shared risk assessment information can be better adapted to the actual situation of each machine tool. For example, for a machine tool with severe tool wear, its risk assessment information may have a higher weight, leading to more conservative adjustments to cutting parameters.

[0131] Therefore, based on the revised risk assessment information, the adjustment range of cutting parameters can be dynamically adjusted. For example, if the revised risk assessment information shows a significant increase in the risk of hard spots in a certain area, the downward adjustment range of cutting parameters (such as feed rate and depth of cut) can be increased accordingly, or the cutting path can be adjusted to avoid high-risk areas. Ultimately, cutting parameter adjustment results adapted to each machine tool are generated, ensuring that each machine tool can perform machining under optimized cutting parameters, thereby improving overall machining efficiency and quality.

[0132] Specifically, after batch workpiece machining, the system continuously collects cutting process signals in real time. These signals contain the latest information on workpiece material properties and machine tool status during actual machining. Based on this real-time data, the risk assessment information is iteratively updated. This means the system can dynamically learn and adapt to the actual distribution and impact of hard particles within the material, rather than relying solely on initial detection results. When the updated risk assessment information deviates from the initially shared information, it indicates a difference between the actual machining situation and expectations. At this point, by introducing machine tool machining parameters for weighted correction, the unique performance and current state of each machine tool can be fully considered. For example, a machine tool with poor performance or severe tool wear will be more sensitive to risk information, thus adopting a more cautious strategy when adjusting cutting parameters. This personalized weighted correction mechanism allows the shared risk assessment information to be "localized" and "customized," ensuring that the adjustment of cutting parameters is not only dynamic but also targeted, thereby avoiding low machining efficiency or quality problems caused by a one-size-fits-all approach.

[0133] In some preferred embodiments, the following specific example illustrates the situation:

[0134] Suppose a batch of 100 identical aero-engine blades made of a novel high-temperature alloy material known to contain randomly distributed hard particles. First, the first batch of blades is probing and machined using a probing machine tool, generating initial risk assessment information. Based on this, the cutting parameters of all machine tools are pre-adjusted. During subsequent batch processing, the system continuously collects cutting process signals from each machine tool in real time. For example, during the machining of the 50th blade, the cutting force and vibration signals of a certain machine tool show more frequent and stronger transient impact events than during the initial probing. This causes its risk assessment information to be iteratively updated, showing a significant deviation from the initial shared information. At this point, the machine tool will weight and correct the shared risk assessment information based on its own machining parameters (e.g., the machine tool has been operating continuously for 8 hours, and the tool wear has reached a warning value). Specifically, due to tool wear, the machine tool may assign higher weights to the risk assessment information, thus selecting smaller feed rates and depths of cut adjustments when adjusting cutting parameters to reduce the risk of further tool wear and ensure machining quality. Meanwhile, if another newly commissioned machine tool shows good status in its machining parameters, it can make minor weighted adjustments to the same risk assessment information, allowing it to adopt relatively aggressive cutting parameters while ensuring safety, in order to maintain high production efficiency. In this way, each machine tool can obtain the most suitable cutting parameter adjustment results based on its unique machining conditions and status.

[0135] Furthermore, in another embodiment of this application, S31400 includes:

[0136] S31410: Real-time calculation of the occurrence frequency and average duration of transient impact events within the current transient impact event identification window;

[0137] S31420: Dynamically set the impact event intensity threshold that matches the current processing conditions based on the machine tool processing parameters;

[0138] S31430: When the frequency or average duration of an event exceeds the impact event density threshold, shorten the length of the transient impact event identification window.

[0139] S31440: When the frequency and average duration of occurrence are consistently below the impact event density threshold, extend the length of the transient impact event identification window.

[0140] The real-time calculation of the frequency and average duration of transient impact events aims to obtain real-time statistical information on transient impact events caused by hard particles within the batch of workpiece material during the current cutting process. The frequency of occurrence can be understood as the number of times a transient impact event occurs within a specific time or cutting path segment, while the average duration reflects the average duration of a single transient impact event. These parameters are key indicators for assessing the density and potential hazards of hard particles within the batch of workpiece material.

[0141] Furthermore, dynamically setting the impact event intensity threshold based on machine tool machining parameters is to ensure that the threshold can adapt to different machining conditions and batch workpiece material characteristics. For example, the impact event intensity threshold can be appropriately relaxed when rough machining or machining materials with high hardness; while it should be tightened when finishing or machining materials with high surface quality requirements. This impact event intensity threshold can be dynamically adjusted based on empirical data, material databases, or through machine learning models to ensure that it matches the current machining status and expected risk level.

[0142] Specifically, when the frequency or average duration of transient impact events calculated in real time exceeds a preset impact event density threshold, it indicates that the distribution of hard particles in the current cutting area is relatively dense or the impact duration caused by a single hard particle is relatively long, posing a high cutting risk. In this case, shortening the length of the transient impact event identification window can improve the system's response speed and identification accuracy to locally dense impact events, avoiding the averaging or missed identification of multiple independent transient impact events due to an excessively long window. Conversely, when the frequency and average duration are consistently below the impact event density threshold, it indicates that the distribution of hard particles in the current cutting area is sparse, and the cutting process is relatively stable. In this case, extending the length of the transient impact event identification window helps accumulate sufficient data over a longer time or cutting path segment, thereby more stably and reliably identifying occasional transient impact events, while reducing potential misjudgments due to an excessively short window.

[0143] Furthermore, in another embodiment, this application proposes that S31410 includes:

[0144] S31411: Based on the clean cutting signal, record the start time, end time and duration of the identified transient impact event;

[0145] S31412: Based on the machine tool machining parameters, dynamically determine the length of the cutting path segment corresponding to the current machining state, and use the length of the cutting path segment to limit the length of the transient impact event identification window;

[0146] S31413: Count the number of transient impact events within the cutting path segment, and sum the durations of the transient impact events to obtain the total duration;

[0147] S31414: Divide the number of transient impact events by the length of the cutting path segment to obtain the frequency of transient impact events;

[0148] S31415: Divide the total duration by the number of transient impact events to obtain the average duration of the transient impact events.

[0149] Specifically, based on the cleaned cutting signal, the system records the start time, end time, and duration of identified transient impact events. This means that after a transient impact event is identified, the system accurately records the start and end points of the event on the timeline, as well as its duration. This time information is the foundational data for subsequent calculations of occurrence frequency and average duration, ensuring a comprehensive understanding of the event's temporal characteristics.

[0150] Specifically, based on machine tool machining parameters, the length of the cutting path segment corresponding to the current machining state is dynamically determined. This cutting path segment length is then used to define the length of the transient impact event identification window. In other words, the system intelligently calculates a suitable cutting path segment length based on the current machine tool machining parameters (such as cutting speed and feed rate). This cutting path segment length is not fixed but dynamically adjusted according to actual machining conditions, and this length serves as the effective range of the transient impact event identification window. The purpose is to make the statistics of impact events more closely reflect the actual machining process, avoiding statistical deviations caused by a fixed window length, thereby improving the accuracy of the statistical results.

[0151] In practical applications, counting the number of transient impact events within the cutting path segment and summing their durations to obtain the total duration means that within the dynamically determined cutting path segment length, the system counts all identified transient impact events and sums their respective durations to obtain the total duration of all transient impact events within that segment. This step provides the necessary raw data for subsequent calculations of frequency and average duration.

[0152] Furthermore, dividing the number of transient impact events by the length of the cutting path segment yields the frequency of transient impact events. The purpose is to quantify the density of transient impact events within a unit cutting path length, which intuitively reflects the distribution density of hard particles in the material.

[0153] In addition, the total duration is divided by the number of transient impact events to obtain the average duration of the transient impact events. The purpose of this method is to reflect the average duration of a single transient impact event, which helps to assess the severity of a single impact event and the potential scope of its impact on the processing.

[0154] This application's solution provides reliable foundational data for subsequent quantitative analysis by accurately recording the start time, end time, and duration of each transient impact event. By dynamically determining the cutting path segment length based on machine tool machining parameters, the matching between the statistical window and the actual machining state is ensured, avoiding statistical errors that may arise from a fixed window and making the statistical results more representative. Based on this, by statistically analyzing the number and total duration of transient impact events within a defined cutting path segment, and further calculating the occurrence frequency and average duration, a quantitative assessment of the density of transient impact events and the sustained impact of individual events is achieved. These precisely calculated occurrence frequencies and average durations can serve as reliable inputs for dynamically adjusting the length of the transient impact event identification window, making the window adjustment more sensitive and accurate, better adapting to the distribution characteristics of hard points under different machining conditions, and improving the system's ability to perceive machining risks.

[0155] Optionally, this application further proposes that the machine tool machining parameters include cutting speed, specifically S31412 including:

[0156] A1000: Real-time monitoring and detection of the machine tool's current cutting speed;

[0157] A2000: Based on the current cutting speed, predict the trend of cutting speed change of the detection machine tool within a future preset time window;

[0158] A3000: Based on the cutting speed change trend and combined with the current cutting speed of the detected machine tool, dynamically determine the length of the cutting path segment that matches the cutting speed change trend;

[0159] A4000: When the cutting speed trend indicates that the cutting speed will continue to increase, increase the length of the cutting path segment;

[0160] A5000: When the cutting speed trend indicates that the cutting speed will continue to decrease, reduce the length of the cutting path segment;

[0161] A6000: When the cutting speed trend indicates that the cutting speed will remain stable, keep the cutting path segment length unchanged.

[0162] Specifically, machine tool machining parameters can be understood as various process parameters that affect the cutting process, such as feed rate, depth of cut, and spindle speed. In this implementation, particular attention is paid to cutting speed. Cutting speed refers to the linear velocity of the tool relative to the workpiece, and it is a key parameter affecting cutting force, cutting temperature, and tool wear. The current cutting speed of the machine tool can be monitored and detected in real time. This can be done by directly reading the data from the machine tool control system or by measuring it using sensors installed on the machine tool, such as encoders or laser tachometers.

[0163] Furthermore, predicting the cutting speed trend of the probe machine tool within a future preset time window based on the current cutting speed can be achieved through various prediction models. For example, time series analysis models (such as ARIMA and LSTM neural networks) can be used to learn and predict historical cutting speed data, or predictions can be made based on the instruction sequence of the current machining program. The future preset time window can be set according to actual machining needs and the performance of the prediction model; for example, it can be a time range from a few seconds to tens of seconds in the future.

[0164] Therefore, based on the cutting speed variation trend and combined with the current cutting speed of the machine tool, the cutting path segment length matching the cutting speed variation trend is dynamically determined. Specifically, when the cutting speed variation trend indicates that the cutting speed will continue to increase, the cutting path segment length is increased to more promptly capture transient impact events that may be exacerbated by high-speed cutting. This ensures that statistics are performed on a longer cutting path, thereby obtaining more representative impact event frequency and average duration. Conversely, when the cutting speed variation trend indicates that the cutting speed will continue to decrease, the cutting path segment length is reduced to avoid statistical distortion due to sparse impact events caused by low-speed cutting, focusing on local features at the current speed. When the cutting speed variation trend indicates that the cutting speed will remain stable, the cutting path segment length is kept constant to maintain statistical stability.

[0165] This application's solution addresses the problem of fixed or untimely adjustments to the cutting path segment length in traditional methods by closely linking the dynamic adjustment of the cutting path segment length to the detection of the machine tool's cutting speed and its changing trend. When the cutting speed changes, the frequency and duration of transient impact events triggered by hard particles within the workpiece material may also change. For example, during high-speed cutting, the collision between the tool and hard particles may be more intense, resulting in more frequent or shorter-lasting impact events; while during low-speed cutting, impact events may be relatively sparse. By monitoring the cutting speed in real time and predicting its changing trend, the system can anticipate changes in the cutting environment and adjust the cutting path segment length accordingly. This adaptive adjustment allows the statistical analysis window for transient impact events to better match the current cutting state, thereby ensuring that the calculated frequency and average duration more accurately reflect the material risk under the current machining conditions. For example, when the cutting speed is expected to increase, increasing the cutting path length helps to collect enough impact event samples over a longer machining distance to smooth out the instantaneous fluctuations that may be caused by high-speed cutting and obtain more stable statistical results; while when the cutting speed is expected to decrease, decreasing the cutting path length helps to avoid invalid statistics in low-speed areas with sparse impact events, thereby improving the real-time performance and effectiveness of the statistics.

[0166] In some preferred embodiments:

[0167] Suppose a batch of workpieces is being machined on a CNC lathe, including a variable-speed cutting path. At the start of machining, the machine tool's current cutting speed is monitored in real time as 100 meters per minute. Based on historical data and the current machining program, the system predicts that the cutting speed will gradually increase to 150 meters per minute over the next 10 seconds. Based on this prediction, the system dynamically increases the length of the cutting path segment used for statistical analysis of transient impact events, for example, from the default 50 mm to 80 mm. In this way, as the cutting speed gradually increases, the system can collect more cutting process signals along the longer cutting path, thus more accurately calculating the frequency and average duration of transient impact events, avoiding the problem of insufficient or unrepresentative statistical samples due to speed variations.

[0168] For example, in another machining stage, the current cutting speed of the detection machine tool is monitored as 80 meters per minute, and it is predicted that the cutting speed will continue to decrease to 50 meters per minute within the next 5 seconds. At this time, the system will dynamically reduce the length of the cutting path segment, for example, from the default 50 mm to 30 mm. This adjustment allows the system to focus more on the characteristics of local impact events at the current lower cutting speed, avoiding excessively long statistical analysis in low-speed areas where impact events may be sparse, thereby improving the real-time performance and response speed of risk assessment.

[0169] Through this dynamic adjustment, regardless of changes in cutting speed, the system can ensure that the statistical data of transient impact events used for risk assessment are highly accurate and timely, providing a reliable basis for subsequent optimization and adjustment of cutting parameters.

[0170] In this regard, this application further proposes that A2000 includes:

[0171] A2100: Real-time monitoring of cutting process signals, including cutting force signals and vibration signals;

[0172] A2200: Based on the machine tool processing parameters and with reference to the expected stable signal range, dynamically determine the dynamic threshold range of the cutting force signal and the dynamic threshold range of the vibration signal;

[0173] A2300: Performs spectrum analysis on cutting force signals and vibration signals to identify the first periodic interference component or the first quasi-periodic interference component in the cutting force signal that overlaps with the natural frequency or harmonic frequency of the machine tool's own components, and to identify the second periodic interference component or the second quasi-periodic interference component in the vibration signal that overlaps with the natural frequency or harmonic frequency of the machine tool's own components.

[0174] A2400: When the cutting force signal exceeds the dynamic threshold range of the cutting force signal, and the excess part does not match the first periodic interference component or the first quasi-periodic interference component, it is determined that there is instantaneous interference in the cutting process signal or abnormal sensor data.

[0175] A2500: When the vibration signal exceeds the dynamic threshold range of the vibration signal, and the excess part does not match the second periodic interference component or the second quasi-periodic interference component, it is determined that there is instantaneous interference in the cutting process signal or abnormal sensor data.

[0176] A2600: When there is instantaneous interference in the cutting process signal or abnormal sensor data, the cutting speed data is corrected by using historical cutting speed data or cutting speed data at adjacent time points to obtain corrected cutting speed data;

[0177] A2700: Based on the corrected cutting speed data, predict the trend of cutting speed change of the detection machine tool within a future preset time window.

[0178] Specifically, real-time monitoring of cutting process signals is crucial. These signals can be understood as physical quantities that directly reflect the cutting state, such as cutting force signals and vibration signals. Cutting force signals directly reflect the intensity of the interaction between the tool and the workpiece, while vibration signals reveal the dynamic stability during the cutting process. Both signals are highly sensitive for assessing abnormal states in the cutting process.

[0179] Specifically, based on machine tool machining parameters and referencing the expected stable signal range, the dynamic threshold ranges of the cutting force signal and vibration signal are dynamically determined. The dynamic threshold ranges are set to adapt to different machining conditions and material properties, ensuring that signal fluctuations under normal cutting conditions are not misjudged as abnormal. The expected stable signal range provides a benchmark under normal cutting conditions, allowing the thresholds to more accurately reflect the signal characteristics under the current machining environment.

[0180] In practical applications, the purpose of spectral analysis of cutting force and vibration signals is to identify periodic or quasi-periodic interference components that overlap with the natural frequencies or harmonic frequencies of the machine tool's own components. These interference components usually originate from the normal operation of the machine tool, such as spindle rotation and feed system movement. They are not caused by hard particles inside the workpiece material, and therefore need to be identified and eliminated to avoid misjudging transient impact events.

[0181] When the cutting force signal exceeds its dynamic threshold range, and the excess portion does not match the first periodic interference component or the first quasi-periodic interference component, it can be determined that there is transient interference or abnormal sensor data in the cutting process signal. Similarly, when the vibration signal exceeds its dynamic threshold range, and the excess portion does not match the second periodic interference component or the second quasi-periodic interference component, it is also determined that there is transient interference or abnormal sensor data. This judgment mechanism can effectively distinguish between normal fluctuations caused by the machine tool's own characteristics and abnormal signals caused by external factors or sensor malfunctions.

[0182] When there are transient disturbances in the cutting process signal or abnormal sensor data, the cutting speed data is corrected using historical cutting speed data or cutting speed data from adjacent time points. This correction method aims to eliminate the influence of abnormal data on the prediction results, ensuring that the cutting speed data used for prediction is reliable and accurate. For example, correction can be performed through interpolation, smoothing, or substitution based on historical trends.

[0183] This application's solution effectively improves the quality of cutting speed data by introducing real-time monitoring of cutting process signals, dynamic threshold setting, spectrum analysis, and anomaly signal identification and correction mechanisms. By distinguishing between inherent machine tool interference and actual instantaneous interference or sensor anomalies, and correcting abnormal data, it ensures that the data used to predict cutting speed trends is pure and reliable. It is precisely this strict control and optimization of the data source that enables subsequent cutting speed trend predictions to more accurately reflect the actual machining state.

[0184] Reference Figure 2 The specific embodiments of this application also disclose a CNC lathe cutting parameter optimization system, including:

[0185] The order receiving module 1 is used to receive processing tasks for batches of workpieces, divide the processing tasks into processing batches, and determine the detection machine tool and its detection processing steps for the batch of workpieces.

[0186] The machining monitoring module 2 is used to collect the cutting process signals of the detection machine tool and its corresponding machine tool machining parameters in real time during the detection machining step;

[0187] The impact event identification and quantification module 3 is used to dynamically establish the expected stable signal range based on the cutting process signal and its corresponding machine tool processing parameters, and to identify and quantify transient impact events caused by hard points inside the batch workpiece material.

[0188] Risk assessment module 4 is used to perform statistical and normalization processing on transient impact events and generate risk assessment information for batch workpiece materials;

[0189] Information sharing module 5 is used to share risk assessment information with other machine tools within the scope of processing batches;

[0190] The parameter adjustment module 6 is used to pre-adjust the cutting parameters of subsequent processing steps based on risk assessment information, and to perform batch processing of workpieces after the pre-adjustment is completed.

[0191] This application proposes a CNC lathe cutting parameter optimization system to address the challenges of unpredictable tool wear due to internal material inhomogeneities in traditional CNC lathe machining, as well as the reduced production efficiency and stability caused by information silos in multi-machine collaborative machining. Through modular design, the system enables the reception and management of batch workpiece machining tasks, real-time monitoring of the cutting process, intelligent identification and quantification of transient impact events, assessment and sharing of batch material risks, and dynamic adjustment of cutting parameters. The collaborative operation of these modules forms a closed-loop optimization mechanism, ensuring proactive adjustment of machining strategies when facing complex material properties, thereby effectively improving machining quality, extending tool life, and increasing overall production efficiency.

[0192] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for optimizing cutting parameters on a CNC lathe, characterized in that, include: Receive a processing task for a batch of workpieces, divide the processing task into processing batches, and determine the detection machine tool and its detection processing steps for the batch of workpieces; During the detection and processing step, the cutting process signals of the detection machine tool and its corresponding machine tool processing parameters are acquired in real time. Based on the cutting process signal and its corresponding machine tool processing parameters, a dynamic stable signal range is established to identify and quantify transient impact events caused by hard points inside the batch workpiece material. The transient impact events are statistically analyzed and normalized to generate risk assessment information for batch workpiece materials. The risk assessment information shall be shared with other machine tools within the scope of the processing batch. Based on the risk assessment information, the cutting parameters for subsequent processing steps are pre-adjusted, and after the pre-adjustment is completed, the processing of the batch of workpieces is performed; The machine tool machining parameters include cutting speed. Based on the machine tool machining parameters, the length of the cutting path segment corresponding to the current machining state is dynamically determined, including: Real-time monitoring of the current cutting speed of the probe machine tool; Based on the current cutting speed, predict the trend of the cutting speed change of the detection machine tool within a future preset time window; Based on the cutting speed change trend and combined with the current cutting speed of the detection machine tool, the length of the cutting path segment that matches the cutting speed change trend is dynamically determined. Specifically, when the cutting speed change trend indicates that the cutting speed will continue to increase, the length of the cutting path segment is increased; when the cutting speed change trend indicates that the cutting speed will continue to decrease, the length of the cutting path segment is decreased; and when the cutting speed change trend indicates that the cutting speed will remain stable, the length of the cutting path segment remains unchanged. The prediction of the cutting speed change trend of the detection machine tool within a future preset time window includes: The cutting process signals, including cutting force signals and vibration signals, are monitored in real time. Based on the machine tool processing parameters and with reference to the expected stable signal range, the dynamic threshold range of the cutting force signal and the dynamic threshold range of the vibration signal are dynamically determined. Spectral analysis is performed on the cutting force signal and the vibration signal to identify a first periodic interference component or a first quasi-periodic interference component in the cutting force signal that overlaps with the natural frequency or harmonic frequency of the detection machine tool's own components, and to identify a second periodic interference component or a second quasi-periodic interference component in the vibration signal that overlaps with the natural frequency or harmonic frequency of the detection machine tool's own components. When there is instantaneous interference or abnormal sensor data in the cutting process signal, the cutting speed data is corrected using historical cutting speed data or cutting speed data at adjacent time points to obtain corrected cutting speed data. Based on the corrected cutting speed data, the trend of cutting speed change of the detection machine tool within a future preset time window is predicted.

2. The method for optimizing cutting parameters of a CNC lathe according to claim 1, characterized in that, Identify transient impact events caused by hard particles within the material of a batch of workpieces, including: Before the detection and processing step, a reference signal is acquired for the detection machine tool to perform no-load motion; Based on the reference signal, a background noise envelope characterizing the noise characteristics of the detection machine tool itself is generated; The cutting process signal is compared with the background noise envelope to remove background noise and obtain a purified cutting signal. Based on the purification cutting signal and the machine tool processing parameters, combined with the expected stable signal range, abnormal over-limit sections and the start and end of events are determined, and transient impact events caused by hard points inside the batch workpiece material are identified.

3. The method for optimizing cutting parameters of a CNC lathe according to claim 2, characterized in that, Quantifying transient impact events caused by hard particles within the material of a batch of workpieces, including: The transient change characteristics of the purification cutting signal are extracted within a preset analysis time window. The transient change characteristics include the signal rise rate, the signal peak duration, and the signal energy concentration. Based on the machine tool processing parameters, the signal characteristic pattern of the transient impact event is determined, and the signal characteristic pattern includes the expected signal rise rate range, the expected signal peak duration range, and the expected signal energy concentration range. The transient change characteristics are compared with the signal characteristic patterns. When the transient change characteristics match the signal characteristic patterns, the corresponding signal waveform in the purification cutting signal is confirmed to be a transient impact event. The peak intensity of the transient impact event is quantified based on the difference between the actual peak value of the transient impact event and the expected peak value of the signal characteristic pattern.

4. The method for optimizing cutting parameters of a CNC lathe according to claim 3, characterized in that, Quantifying transient impact events caused by hard particles within the material of a batch of workpieces also includes: Real-time monitoring of the purification cutting signal; The transient impact event identification window is determined based on the machine tool processing parameters. When multiple consecutive or superimposed transient impact waveforms of the purification cutting signal are detected within the transient impact event identification window, a transient impact waveform fusion mechanism is performed to obtain the fused transient impact waveform. Based on the peak intensity and duration of each transient impact event, the deviation of the fused transient impact waveform from the upper bound of the expected stable signal range within the transient impact event identification window is integrated to obtain the integration area, and the integration area is determined as the cumulative impact load intensity. The length of the transient impact event identification window is dynamically adjusted based on the frequency of occurrence and average duration of the transient impact event.

5. The method for optimizing cutting parameters of a CNC lathe according to claim 1, characterized in that, After processing the batch of workpieces, the process further includes: The risk assessment information is iteratively updated based on the real-time acquired cutting process signals; When the updated risk assessment information deviates from the shared risk assessment information within the processing batch, different machine tools perform weighted correction on the shared risk assessment information according to their own machine tool processing parameters, and dynamically adjust the adjustment range of cutting parameters based on the corrected risk assessment information to generate cutting parameter adjustment results adapted to each machine tool.

6. The method for optimizing cutting parameters of a CNC lathe according to claim 4, characterized in that, The step of dynamically adjusting the length of the transient impact event identification window based on the occurrence frequency and average duration of the transient impact event includes: The frequency of occurrence and average duration of the transient impact event within the current transient impact event identification window are calculated in real time. Based on the machine tool processing parameters, dynamically set the impact event intensity threshold that matches the current processing conditions; When the occurrence frequency or the average duration exceeds the impact event density threshold, the length of the transient impact event identification window is shortened. When the occurrence frequency and the average duration are consistently below the impact event density threshold, the length of the transient impact event identification window is extended.

7. The method for optimizing cutting parameters of a CNC lathe according to claim 6, characterized in that, The real-time calculation of the occurrence frequency and average duration of the transient impact event within the current transient impact event identification window includes: Based on the purified cutting signal, the start time, end time and duration of the identified transient impact event are recorded; The length of the transient impact event recognition window is limited according to the length of the cutting path segment. The number of transient impact events is counted within the cutting path segment, and the durations of the transient impact events are summed to obtain the total duration; The frequency of occurrence of the transient impact events is obtained by dividing the number of transient impact events by the length of the cutting path segment. The average duration of the transient impact event is obtained by dividing the total duration by the number of transient impact events.

8. A CNC lathe cutting parameter optimization system, characterized in that, include: The order receiving module is used to receive processing tasks for batches of workpieces, divide the processing tasks into processing batches, and determine the detection machine tool and its detection processing steps for the batch of workpieces. The machining monitoring module is used to collect the cutting process signals of the detection machine tool and its corresponding machine tool machining parameters in real time during the detection machining step. The impact event identification and quantification module is used to dynamically establish an expected stable signal range based on the cutting process signal and the corresponding machine tool processing parameters, and to identify and quantify transient impact events caused by hard points inside the batch workpiece material. The risk assessment module is used to perform statistical and normalization processing on the transient impact events to generate risk assessment information for batch workpiece materials. The information sharing module is used to share the risk assessment information with other machine tools within the scope of the processing batch; The parameter adjustment module is used to pre-adjust the cutting parameters of subsequent processing steps based on the risk assessment information, and to perform the processing of the batch of workpieces after the pre-adjustment is completed. The machine tool machining parameters include cutting speed. Based on the machine tool machining parameters, the length of the cutting path segment corresponding to the current machining state is dynamically determined, including: Real-time monitoring of the current cutting speed of the probe machine tool; Based on the current cutting speed, predict the trend of the cutting speed change of the detection machine tool within a future preset time window; Based on the cutting speed change trend and combined with the current cutting speed of the detection machine tool, the length of the cutting path segment that matches the cutting speed change trend is dynamically determined. Specifically, when the cutting speed change trend indicates that the cutting speed will continue to increase, the length of the cutting path segment is increased; when the cutting speed change trend indicates that the cutting speed will continue to decrease, the length of the cutting path segment is decreased; and when the cutting speed change trend indicates that the cutting speed will remain stable, the length of the cutting path segment remains unchanged. The prediction of the cutting speed change trend of the detection machine tool within a future preset time window includes: The cutting process signals, including cutting force signals and vibration signals, are monitored in real time. Based on the machine tool processing parameters and with reference to the expected stable signal range, the dynamic threshold range of the cutting force signal and the dynamic threshold range of the vibration signal are dynamically determined. Spectral analysis is performed on the cutting force signal and the vibration signal to identify a first periodic interference component or a first quasi-periodic interference component in the cutting force signal that overlaps with the natural frequency or harmonic frequency of the detection machine tool's own components, and to identify a second periodic interference component or a second quasi-periodic interference component in the vibration signal that overlaps with the natural frequency or harmonic frequency of the detection machine tool's own components. When there is instantaneous interference or abnormal sensor data in the cutting process signal, the cutting speed data is corrected using historical cutting speed data or cutting speed data at adjacent time points to obtain corrected cutting speed data. Based on the corrected cutting speed data, the trend of cutting speed change of the detection machine tool within a future preset time window is predicted.

Images