Intelligent control method and system for silicon carbide processing lathe

Abstract

The invention discloses an intelligent control method and system for a silicon carbide machining lathe, and relates to the technical field of industrial equipment control, and the method comprises the steps: obtaining the position feedback data of a servo motor coding module, the cutting force data of a cutting force sensing module, and the temperature data of a miniature temperature sensing module; performing time sequence decomposition, and identifying an interference source type and an interference source contribution degree; determining a lathe control strategy; if the lathe control strategy is a thermal drift compensation strategy, a reverse compensation signal is generated according to the temperature data; if the lathe control strategy is a cutting force fluctuation damping strategy, a nonlinear adjusting signal is generated according to the cutting force data; according to the interference source type, the interference intensity, the cutter state and the target signal, cutting parameters are adjusted; and controlling the lathe according to the cutting parameters. According to the method, the cutting parameters can be adjusted by combining sensing data and interference information, so that lathe control is realized, and the control precision and the workpiece machining quality are improved.

Description

A method and system for intelligent control of a silicon carbide machining lathe Technical Field

[0001] This invention relates to the field of industrial equipment control technology, and in particular to an intelligent control method and system for a silicon carbide machining lathe. Background Technology

[0002] In the intelligent manufacturing equipment industry, silicon carbide materials are widely used in aerospace, new energy vehicles, and high-end electronic devices due to their superior physicochemical properties. Existing systems use fixed processes to process silicon carbide materials; however, during processing, problems such as rapid tool wear, drastic fluctuations in cutting force, and high sensitivity to changes in processing temperature may arise. These challenges make it difficult for existing systems to respond effectively and in real time to subtle changes in material properties and external environmental interference, resulting in low processing accuracy and poor workpiece quality.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this invention is to propose an intelligent control method and system for silicon carbide machining lathes, which can combine sensor data and interference information to adjust cutting parameters to achieve lathe control, thereby improving control accuracy and workpiece machining quality.

[0005] On one hand, embodiments of the present invention provide an intelligent control method for a silicon carbide machining lathe, comprising the following steps:

[0006] Acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module;

[0007] Based on the position feedback data, the cutting force data, and the temperature data, time series decomposition is performed to identify the type of interference source and the degree of contribution of the interference source;

[0008] The lathe control strategy is determined based on the type of interference source and the degree of contribution of the interference source;

[0009] If the lathe control strategy is a thermal drift compensation strategy, then a reverse compensation signal is generated based on the temperature data, and the reverse compensation signal is used to correct the position feedback deviation.

[0010] If the lathe control strategy is a cutting force fluctuation damping strategy, then a nonlinear adjustment signal is generated based on the cutting force data. The nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuations.

[0011] Based on the type of interference source, the intensity of interference, the tool state, and the target signal, the cutting parameters are adjusted, wherein the target signal includes the reverse compensation signal or the nonlinear adjustment signal.

[0012] The lathe is controlled according to the cutting parameters.

[0013] On the other hand, embodiments of the present invention provide an intelligent control system for a silicon carbide machining lathe, comprising:

[0014] The data acquisition module is used to acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module.

[0015] The time series decomposition module is used to perform time series decomposition based on the position feedback data, the cutting force data, and the temperature data to identify the type of interference source and the degree of contribution of the interference source;

[0016] The control strategy determination module is used to determine the lathe control strategy based on the type of interference source and the degree of contribution of the interference source;

[0017] The compensation signal generation module is used to generate a reverse compensation signal based on the temperature data if the lathe control strategy is a thermal drift compensation strategy. The reverse compensation signal is used to correct the position feedback deviation.

[0018] The adjustment signal generation module is used to generate a nonlinear adjustment signal based on the cutting force data if the lathe control strategy is a cutting force fluctuation damping strategy. The nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuation.

[0019] The cutting parameter adjustment module is used to adjust the cutting parameters according to the type of interference source, the intensity of interference, the tool state and the target signal, wherein the target signal includes the reverse compensation signal or the nonlinear adjustment signal;

[0020] The lathe control module is used to control the lathe according to the cutting parameters.

[0021] The embodiments of this application include at least the following beneficial effects: First, the embodiments of this application acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module, and perform time series decomposition to identify the type and degree of interference source contribution. Then, the lathe control strategy is determined. If the lathe control strategy is a thermal drift compensation strategy, a reverse compensation signal is generated to correct the position feedback deviation. If the lathe control strategy is a cutting force fluctuation damping strategy, a nonlinear adjustment signal is generated to nonlinearly adjust the feed axis acceleration command. Then, the cutting parameters are adjusted according to the type of interference source, interference intensity, tool state, and target signal. Finally, the lathe is controlled according to the cutting parameters. This allows for the adjustment of cutting parameters by combining sensor data and interference information to achieve lathe control, thereby improving control accuracy and workpiece machining quality.

[0022] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and the drawings. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0024] Figure 1 is a flowchart of an intelligent control method for a silicon carbide machining lathe according to an embodiment of the present invention;

[0025] Figure 2 is a schematic diagram of the structure of an intelligent control system for a silicon carbide machining lathe according to an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0027] In the intelligent manufacturing equipment industry, silicon carbide (SiC) materials are widely used in aerospace, new energy vehicles, and high-end electronic devices due to their superior physicochemical properties, such as extremely high hardness, excellent high-temperature resistance, and semiconductor characteristics. However, these properties also make the processing of SiC materials a highly challenging task. Existing systems use fixed processes to process SiC materials; however, during processing, problems such as rapid tool wear, drastic fluctuations in cutting force, and high sensitivity to changes in processing temperature may arise. These challenges make it difficult for existing systems to respond effectively and in real time to subtle changes in material properties and interference from the external environment, resulting in low processing accuracy and low workpiece quality. Meanwhile, in the industrial internet environment, how to efficiently integrate and analyze heterogeneous data from various sources such as vibration sensors, temperature sensors, and cutting parameter monitoring, and transform this data into an effective basis for equipment adaptive adjustment, is a pressing technical challenge that needs to be addressed.

[0028] To address these challenges, existing intelligent control systems acquire real-time feedback information from multiple sensors, with precise position data provided by servo motor encoders being crucial. Utilizing this encoder feedback, the intelligent control system can precisely control the spindle speed and tool feed rate, thereby ensuring the stability of the cutting process and the final machining accuracy. This absolute reliance on position feedback forms the foundation of the entire precision machining control strategy.

[0029] However, during prolonged continuous machining, cutting silicon carbide generates a significant amount of heat. Although lathes are equipped with cooling and lubrication systems, some of this heat is still conducted through the mechanical structure to adjacent components, including the servo motors driving the spindle and feed axes and their associated encoders. When the electronic components inside the encoder are exposed to elevated operating temperatures for extended periods, a phenomenon known as thermal drift gradually occurs. This phenomenon refers to the slight and slow change in the electrical characteristics of electronic components (such as resistance, capacitance, or the properties of semiconductor junctions) with temperature variations. For example, the performance of the photoelectric conversion elements or signal processing circuitry inside the encoder may slightly deviate due to increased temperature, resulting in a small, slowly changing deviation in its output position reading that is temperature-dependent. This deviation is not a sudden malfunction but gradually becomes apparent with prolonged machining time and accumulated temperature. Its magnitude is usually very small and insufficient to trigger the lathe's hardware-level fault alarm systems, as these systems are typically configured to detect more significant and sudden anomalies.

[0030] This minute position reading deviation caused by thermal drift results in a slight, non-linear error in the servo motor encoder's output position feedback signal, especially after prolonged continuous cutting operations. This means that the information received by the control system regarding the tool's trajectory is subtly inconsistent with the tool's actual trajectory in physical space. For example, when the encoder reports the tool at 100.000 mm on the X-axis, the actual tool position might be 100.002 mm or 99.998 mm, and this deviation fluctuates slowly with temperature changes. Because this error is extremely small, it does not trigger the lathe's conventional fault diagnosis mechanisms. However, for silicon carbide machining requiring micron- or even submicron-level precision, this persistent, unrecognized deviation affects machining quality.

[0031] In precision machining scenarios involving silicon carbide, where high precision is required, material properties are complex, and the material is susceptible to thermal effects, it is necessary to accurately identify and decouple two distinct types of combined interference: the minute positional feedback deviation caused by thermal drift from long-term operation of the servo motor encoder, and the instantaneous, small-amplitude, high-frequency cutting force fluctuation caused by differences in the internal residual stress distribution of new batches of silicon carbide workpieces. This is to prevent the system from misjudging and overcompensating for these minute and persistent interferences, thereby suppressing the formation of minute oscillation trajectories of the tool on the workpiece surface, and ensuring the normal service life of the tool and the micron-level machining quality of the workpiece.

[0032] The embodiments of this application will be explained in detail below with reference to the accompanying drawings:

[0033] Figure 1 is an optional flowchart of an intelligent control method for a silicon carbide machining lathe provided in an embodiment of this application. The method in Figure 1 may include, but is not limited to, steps S101 to S107.

[0034] Step S101: Obtain the position feedback data from the servo motor encoder module, the cutting force data from the cutting force sensor module, and the temperature data from the micro temperature sensor module;

[0035] Step S102: Based on the position feedback data, cutting force data, and temperature data, perform time series decomposition to identify the type of interference source and the degree of contribution of the interference source;

[0036] Step S103: Determine the lathe control strategy based on the type and contribution level of the interference source;

[0037] Step S104: If the lathe control strategy is a thermal drift compensation strategy, then a reverse compensation signal is generated based on the temperature data. The reverse compensation signal is used to correct the position feedback deviation.

[0038] Step S105: If the lathe control strategy is a cutting force fluctuation damping strategy, then a nonlinear adjustment signal is generated based on the cutting force data. The nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuation.

[0039] Step S106: Adjust the cutting parameters according to the type of interference source, interference intensity, tool status and target signal. The target signal includes reverse compensation signal or nonlinear adjustment signal.

[0040] Step S107: Control the lathe according to the cutting parameters.

[0041] Steps S101 to S107 shown in the embodiments of this application can combine sensor data and interference information to adjust cutting parameters in order to achieve lathe control, thereby improving control accuracy and workpiece machining quality.

[0042] In some embodiments, steps S101-S107 can first acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensing module, and temperature data from the micro temperature sensing module. For example, a high-precision photoelectric encoder or magnetic encoder can be installed on the servo motor as the servo motor encoder module to acquire its output pulse signals or digital signals in real time, thereby obtaining position feedback data. A piezoelectric sensor or resistance strain gauge sensor can be installed on the tool holder or workpiece fixture as the cutting force sensing module to measure the force components in three directions generated during the cutting process in real time and convert them into electrical signals for output, thereby obtaining cutting force data. The micro temperature sensing module can use thermocouples or thermistors and is placed near the cutting area, at key heat-generating components such as the spindle bearing or feed screw to monitor temperature changes in real time and obtain temperature data. These sensor data can be transmitted to the lathe's CNC system or a separate industrial computer for processing via a data acquisition card or industrial Ethernet.

[0043] Then, based on the position feedback data, cutting force data, and temperature data, time series decomposition is performed to identify the type and contribution level of interference sources. Empirical Mode Decomposition (EMD), wavelet decomposition, or Seasonal-Trend Decomposition (STL) algorithms can be used to process the data. These decomposition methods can break down the original signal into different components, such as trend components, periodic components, and residual components. By analyzing the characteristics of these components, specific patterns related to thermal drift, internal material stress fluctuations, etc., can be identified. For example, slowly changing trend components may be related to thermal drift, while high-frequency residual components may be related to instantaneous cutting force fluctuations. After identifying the different signal components, feature extraction and pattern recognition techniques can be used to match these components with a pre-defined interference source model to determine the type of interference source. For example, if a slow position deviation highly correlated with temperature changes is identified, it may be determined as thermal drift interference. Simultaneously, the influence of different interference source components on the overall machining process, i.e., the contribution level of the interference source, can be assessed by calculating indicators such as the energy, amplitude, or variance of the components. Understandably, time series decomposition is a data analysis technique that aims to break down time series data into trend, periodic, and random components in order to better understand the data’s internal structure and identify anomalies.

[0044] Then, based on the type and contribution level of the interference source, the lathe control strategy is determined. If the main interference source is thermal drift and its contribution is high, a thermal drift compensation strategy can be prioritized as the lathe control strategy. If the main interference source is cutting force fluctuation and its contribution is high, a cutting force fluctuation damping strategy can be prioritized as the lathe control strategy. In some cases, if there are multiple interference sources and their contributions are not negligible, a composite control strategy may be necessary.

[0045] If the lathe control strategy is a thermal drift compensation strategy, a reverse compensation signal is generated based on the temperature data. This reverse compensation signal is used to correct the position feedback deviation. A mapping model between temperature and position deviation can be established, which can be obtained through offline experiments or online learning. When the temperature data changes, the model is used to predict the position deviation caused by thermal drift, and then a reverse compensation signal with the same magnitude but opposite direction to this deviation is generated. This reverse compensation signal can be directly superimposed on the servo motor's position command or input as a feedforward signal into the position control loop, thereby correcting the position feedback deviation and offsetting the impact of thermal drift on machining accuracy.

[0046] If the lathe control strategy is a cutting force fluctuation damping strategy, a nonlinear adjustment signal is generated based on the cutting force data. This nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuations. A nonlinear controller can be designed, which takes the cutting force data as input and outputs a nonlinear adjustment signal. This nonlinear adjustment signal can be used to nonlinearly adjust the feed axis acceleration command. For example, when the cutting force suddenly increases, the nonlinear adjustment signal can instantaneously reduce the feed axis acceleration command to reduce the cutting load and thus suppress cutting force fluctuations. This nonlinear adjustment can be implemented using methods such as fuzzy control or sliding mode control to better adapt to the complexity and uncertainty of cutting force fluctuations. It is understood that the feed axis acceleration command is a command that controls the rate of change of the lathe's feed axis speed.

[0047] Based on the type and intensity of the interference source, tool condition, and target signal, cutting parameters are adjusted. The target signal includes either a reverse compensation signal or a nonlinear adjustment signal. If the interference source is thermal drift and a reverse compensation signal has been generated, the feed rate or spindle speed can be fine-tuned based on the compensated position feedback data and tool wear conditions (e.g., monitored via acoustic emission sensors or vision systems) to further optimize the machining process. If the interference source is cutting force fluctuation and a nonlinear adjustment signal has been generated, the depth of cut or spindle speed can be dynamically adjusted based on the adjusted feed axis acceleration command and tool condition to maintain machining efficiency while suppressing fluctuations. The adjustment of cutting parameters can employ biomimetic optimization algorithms, such as genetic algorithms or particle swarm optimization algorithms, to maximize machining efficiency or minimize tool wear while meeting machining accuracy and surface quality requirements. It is understood that cutting parameters include key process parameters affecting the cutting process, such as spindle speed, feed rate, and depth of cut.

[0048] Finally, the lathe is controlled according to the cutting parameters. The adjusted cutting parameters are sent to the various actuators of the lathe, such as the servo motor drivers and the spindle frequency converter, through the CNC system. The servo motor drivers control the movement of the feed axis according to the new feed rate and acceleration commands, and the spindle frequency converter controls the spindle rotation according to the new spindle speed commands. Through this control, the lathe can respond to various disturbances in the machining process in real time and dynamically adjust its working state, thereby achieving high-precision and high-efficiency machining of silicon carbide materials.

[0049] Through the above technical solution, this embodiment integrates multi-source sensor data and utilizes advanced time-series decomposition technology to accurately identify composite interference sources such as thermal drift and cutting force fluctuations during the machining process, as well as their contribution levels. This overcomes the limitations of traditional control systems in distinguishing and effectively responding to interferences of different natures. By dynamically determining control strategies such as thermal drift compensation or cutting force fluctuation damping, and generating corresponding reverse compensation signals or nonlinear adjustment signals, this embodiment can specifically correct position feedback deviations and suppress cutting force fluctuations. This avoids overcompensation and endogenous tool micro-oscillations caused by misjudgment in traditional systems, thereby effectively reducing abnormal tool wear and significantly improving the surface quality and accuracy of silicon carbide workpieces. This embodiment achieves refined and adaptive control of the silicon carbide machining process, significantly improving machining stability and reliability, and providing solid technical support for the application of silicon carbide materials in high-end fields.

[0050] In some embodiments, in step S102, time series decomposition is performed based on position feedback data, cutting force data, and temperature data to identify the type of interference source and the degree of contribution of the interference source. This may include, but is not limited to, the following steps:

[0051] Step S201: Perform multi-scale decomposition on the location feedback data to obtain location signal components at different time scales;

[0052] Step S202: Identify non-periodic transient micro-jump signals based on position signal components;

[0053] Step S203: Perform time-frequency analysis on the cutting force data to identify broadband cutting force pulses;

[0054] Step S204: Identify the interference source of the transient micro-jump signal based on the amplitude, first duration, and synchronization with the broadband cutting force pulse of the transient micro-jump signal;

[0055] Step S205: If the transient micro-jump signal interference source is a micrometer-level stepped fracture event, then perform micro-jump signal separation on the position feedback data;

[0056] Step S206: Perform low-pass filtering and trend analysis on the position feedback data after separating the micro-jump signal to obtain the position deviation component;

[0057] Step S207: Identify the internal stress fluctuation components of the material based on the broadband cutting force pulse and cutting force data;

[0058] Step S208: Based on the temperature data, perform temperature correlation analysis on the position deviation components to obtain the temperature correlation analysis results;

[0059] Step S209: Identify the type of interference source based on the analysis results of the internal stress fluctuation composition and temperature correlation of the material;

[0060] Step S210: Assess the contribution level of the interference source based on the type of interference source.

[0061] In some embodiments, the position feedback data can first be decomposed into multiple scales to obtain position signal components at different time scales. Multiple scale decomposition refers to breaking down the original signal into components with different characteristics at different time scales, such as through wavelet transform or empirical mode decomposition, thereby enabling comprehensive analysis of the signal from macroscopic trends to microscopic details. This yields position signal components at different time scales, which help reveal motion characteristics within different frequency ranges. Based on these position signal components, non-periodic transient micro-jumping signals can be identified. Transient micro-jumping signals typically manifest as short-duration, high-amplitude abnormal fluctuations, possibly caused by microscopic fractures within materials or minor vibrations in equipment.

[0062] Then, time-frequency analysis is performed on the cutting force data to identify broadband cutting force pulses. For example, broadband cutting force pulses can be identified through short-time Fourier transform or wavelet packet decomposition. Broadband cutting force pulses are usually associated with impact or fracture events during material removal and have a wide frequency range. The source of interference in the transient micro-jump signal is identified based on the amplitude, initial duration, and synchronicity with the broadband cutting force pulse. The initial duration refers to the length of time from the start to the end of the micro-jump signal. Synchronicity analysis helps determine whether the positional micro-jump occurs simultaneously with the cutting force pulse, thereby inferring their common physical origin.

[0063] If the transient micro-jump signal interference source is a micrometer-level stepped fracture event, then micro-jump signal separation is performed on the position feedback data to remove the influence of these transient interferences and obtain smoother, more trend-oriented position data. Low-pass filtering and trend analysis are then performed on the position feedback data after micro-jump signal separation to obtain the position deviation component. Low-pass filtering removes high-frequency noise, and trend analysis extracts the long-term trend of the signal, thus accurately characterizing the actual position deviation of the lathe during machining.

[0064] Then, based on broadband cutting force pulses and cutting force data, the internal stress fluctuation components of the material are identified. This can be achieved by analyzing specific frequencies or patterns in the cutting force signal that are related to changes in the material's internal structure. Simultaneously, based on temperature data, temperature correlation analysis is performed on the position deviation components to obtain temperature correlation analysis results. The aim is to quantify the degree of influence of temperature changes on position deviation, for example, by establishing a mathematical model between temperature and position deviation.

[0065] Finally, based on the correlation analysis of internal stress fluctuations and temperature within the material, the type of interference source is identified. For example, it can be determined whether the interference is caused by thermal drift, stress release due to internal material defects, or micro-fracture caused by tool wear. The contribution of the interference source is then assessed based on its type. Further evaluation of the interference source's impact on the overall machining error is possible.

[0066] Through the above technical solutions, this embodiment, via multi-scale decomposition, time-frequency analysis, signal separation, and multi-dimensional correlation analysis, can more accurately identify non-periodic transient micro-jugation signals, broadband cutting force pulses, internal material stress fluctuation components, and temperature-related positional deviation components. It can further determine the specific type of interference source (e.g., micrometer-level stepped fracture events, thermal drift, etc.) and its contribution level. This refined identification capability allows for more targeted subsequent lathe control strategies, effectively suppressing various interferences during machining and significantly improving machining accuracy, surface quality, and tool life. Its advantages are particularly evident in the precision machining of difficult-to-machine materials such as silicon carbide.

[0067] In some embodiments, step S207, identifying the internal stress fluctuation components of the material based on broadband cutting force pulses and cutting force data, may include, but is not limited to, the following steps:

[0068] The broadband cutting force pulse interference source is identified based on the peak energy, second duration, and synchronization with the transient micro-jump signal of the broadband cutting force pulse.

[0069] If the broadband cutting force pulse interference source is a micron-level stepped fracture event, then the cutting force data is separated into cutting force pulses.

[0070] High-pass filtering and spectrum analysis were performed on the cutting force data after cutting force pulse separation to obtain the internal stress fluctuation components of the material.

[0071] In some embodiments, the broadband cutting force pulse interference source can be identified first based on the peak energy, second duration, and synchronicity with the transient micro-jump signal of the broadband cutting force pulse. The peak energy of the broadband cutting force pulse refers to the maximum energy value reached by the cutting force signal within a short period during the cutting process, reflecting the intensity of the instantaneous impact. The second duration refers to the length of time the broadband cutting force pulse lasts from start to finish, characterizing the persistence of the impact event. Synchronicity with the transient micro-jump signal refers to the degree of correlation between the occurrence time of the broadband cutting force pulse and the occurrence time of the transient micro-jump signal on the time axis. By analyzing whether they occur simultaneously or approximately simultaneously, it can be determined whether they originate from the same physical event. By comprehensively analyzing these characteristics, potential interference sources of the broadband cutting force pulse can be identified more accurately.

[0072] If the broadband cutting force pulse interference source is a micrometer-scale stepped fracture event, it indicates that microscale brittle fracture of the material occurred during the machining of silicon carbide, leading to a significant stepped change in the cutting force signal. In this case, to accurately extract the internal stress fluctuation components of the material, cutting force pulse separation is required. Cutting force pulse separation is a signal processing procedure aimed at separating the instantaneous pulse signal caused by the micrometer-scale stepped fracture event from the original cutting force data to eliminate its interference with subsequent analysis. In practical applications, cutting force pulse separation can be achieved through various signal processing techniques, such as wavelet transform, empirical mode decomposition (EMD), or adaptive filtering, to effectively separate the pulse component from the background signal.

[0073] The cutting force data after cutting force pulse separation is then subjected to high-pass filtering and spectral analysis to obtain the internal stress fluctuation components of the material. The purpose of high-pass filtering is to remove potential low-frequency trends and DC components from the cutting force data. These low-frequency components are usually related to macroscopic factors such as tool wear and machine tool vibration, rather than the microscopic stress fluctuations within the material. High-pass filtering highlights high-frequency signals, which are more likely to reflect stress fluctuations reflecting changes in the material's internal microstructure. Spectral analysis involves processing the high-pass filtered data using Fourier transform and other methods to obtain the energy distribution of the signal at different frequencies, thereby identifying specific frequency components related to internal stress fluctuations. Its aim is to accurately extract stress fluctuation information caused by changes in the material's internal microstructure (such as lattice defects and grain boundary stress) from complex cutting force signals.

[0074] Through the above technical solution, this embodiment can more accurately identify the changes in cutting force caused by internal stress fluctuations during the processing of silicon carbide materials, rather than simply attributing all broadband cutting force pulses to internal stress. This refined identification method enables the system to distinguish different types of interference sources, especially to identify and process cutting force pulses generated by micron-level stepped fracture events, thereby avoiding confusion caused by external interference or macroscopic phenomena in the identification of internal stress fluctuations. This improves the accuracy and reliability of monitoring the processing status of silicon carbide materials, providing a more precise basis for subsequent lathe control strategy formulation.

[0075] In some embodiments, in step S103, determining the lathe control strategy based on the type and contribution level of the interference source may include, but is not limited to, the following steps:

[0076] Step S301: Obtain the position deviation feedforward correction module and the instantaneous cutting force damping control module;

[0077] Step S302: Determine the module priority and effect strength based on the type and contribution level of the interference source;

[0078] Step S303: Determine the superposition order of motion commands for the position deviation feedforward correction module according to the module priority and action intensity;

[0079] Step S304: According to the superposition order of motion commands, perform deviation correction simulation on the position deviation feedforward correction module to obtain the deviation correction simulation result;

[0080] Step S305: Determine the adjustment range of the instantaneous cutting force damping control module based on the module priority and intensity of action;

[0081] Step S306: Based on the adjustment range, perform a nonlinear adjustment simulation on the instantaneous cutting force damping control module to obtain the nonlinear adjustment simulation results;

[0082] Step S307: Based on the deviation correction simulation results and the nonlinear adjustment simulation results, evaluate the tool oscillation information, workpiece machining quality, and tool wear condition;

[0083] Step S308: Determine the lathe control strategy based on the tool oscillation information, workpiece machining quality, and tool wear condition.

[0084] In some embodiments, since strategy selection is based solely on the type and contribution level of the interference source, it may not adequately consider the synergistic effects between different control modules, potential conflicts, and the combined impact on machining quality and tool life. This could lead to poor control performance or even trigger new machining problems in complex and variable environments. To address this, a position deviation feedforward correction module and an instantaneous cutting force damping control module can be developed. The position deviation feedforward correction module aims to eliminate or reduce position deviations caused by factors such as thermal drift through pre-calculation and compensation. It is an active correction mechanism used to predictively correct potential position errors before the actual motion command is issued. The instantaneous cutting force damping control module focuses on suppressing cutting force fluctuations generated during the cutting process. Its purpose is to maintain the stability of the cutting force by dynamically adjusting the acceleration command of the feed axis, thereby avoiding a decrease in machining quality or accelerated tool wear caused by drastic changes in cutting force.

[0085] Then, based on the type and contribution level of the interference source, the module priority and intensity are determined. Module priority refers to the execution order or weight of the position deviation feedforward correction module and the instantaneous cutting force damping control module in the control strategy, while intensity refers to the correction or adjustment magnitude of each module when performing its function. For example, if thermal drift is the main interference source and its contribution is high, the priority of the position deviation feedforward correction module will be increased, and its intensity will be set to a higher level. Conversely, if cutting force fluctuation is the main problem, the priority and intensity of the instantaneous cutting force damping control module will be increased accordingly. Simultaneously, based on the module priority and intensity, the motion command superposition order of the position deviation feedforward correction module is determined. This means that when generating the final motion command, how the deviation correction signal is fused with the original command, and at what stage it is superimposed, will be precisely planned. Based on the motion command superposition order, deviation correction simulation is performed on the position deviation feedforward correction module to obtain the deviation correction simulation results. The correction effect of the position deviation under a specific superposition order and intensity can be predicted, thus obtaining the deviation correction simulation results.

[0086] Based on module priority and intensity, the adjustment range of the instantaneous cutting force damping control module is determined. This adjustment range defines the maximum amplitude and frequency of nonlinear adjustment of the feed axis acceleration command when suppressing cutting force fluctuations. Based on this adjustment range, a nonlinear adjustment simulation of the instantaneous cutting force damping control module is performed to obtain the simulation results. This is used to evaluate its ability to suppress cutting force fluctuations and its impact on the system's dynamic response, thus yielding the nonlinear adjustment simulation results.

[0087] Finally, based on the simulation results of deviation correction and nonlinear adjustment, the tool oscillation information, workpiece machining quality, and tool wear state are evaluated. Tool oscillation information reflects the stability of the machining process, workpiece machining quality is a key indicator for measuring machining accuracy, and tool wear state directly affects machining cost and efficiency. Through these multi-dimensional evaluations, the potential machining effects under different control strategy combinations can be fully understood. Furthermore, based on the tool oscillation information, workpiece machining quality, and tool wear state, the most suitable lathe control strategy for the current disturbance conditions and machining requirements is determined.

[0088] Through the above technical solution, this embodiment introduces a position deviation feedforward correction module and an instantaneous cutting force damping control module, and dynamically determines and simulates their priority, intensity, motion command superposition sequence, and adjustment range, significantly improving the adaptability and effectiveness of the control strategy. This embodiment can more accurately predict the impact of different control combinations on tool oscillation, workpiece machining quality, and tool wear, thereby avoiding machining defects caused by single or uncoordinated control strategies. Therefore, it not only improves machining accuracy and surface quality but also extends tool life, reduces production costs, and achieves efficient and high-quality machining of silicon carbide materials.

[0089] In some embodiments, in step S302, determining the module priority and effect strength based on the type and contribution level of the interference source may include, but is not limited to, the following steps:

[0090] Step S401: Generate oscillation signal characteristics through the controlled micro-oscillation induction module;

[0091] Step S402: Match the position feedback data, cutting force data and oscillation signal characteristics to obtain the actively introduced signal;

[0092] Step S403: Based on the type of interference source and the degree of contribution of the interference source, assess the degree of impact of the actively introduced signal on the processing state;

[0093] Step S404: Determine the module priority and effect intensity based on the degree of impact.

[0094] In some embodiments, relying solely on passive identification of the type and contribution level of interference sources may make it difficult to accurately capture the actual response characteristics and optimal operating parameters of lathe control modules (e.g., position deviation feedforward correction modules and instantaneous cutting force damping control modules) under different interference scenarios. This passive identification method may result in insufficient refinement in determining module priorities and operating strengths, thus affecting the optimization effect of the overall control strategy. Therefore, oscillation signal characteristics can be generated first through a controlled micro-oscillation induction module. A controlled micro-oscillation induction module is a device capable of actively introducing small oscillation signals of specific frequency and amplitude into the lathe system. Its purpose is to detect the dynamic response characteristics of the lathe system under the influence of different interference sources through active excitation. For example, the controlled micro-oscillation induction module can be a piezoelectric actuator or electromagnetic exciter, configured to apply preset small oscillations to the tool or workpiece during machining to simulate or amplify certain potential interference effects. The oscillation signal characteristics refer to signals with specific frequency, amplitude, phase, and waveform. These characteristics are designed to induce an observable response in the system without significantly affecting normal machining, thereby revealing the system's sensitivity to specific interferences.

[0095] Then, the position feedback data, cutting force data, and oscillation signal characteristics are matched to obtain the actively introduced signal. The actively introduced signal refers to the signal obtained by matching the oscillation signal characteristics generated by the controlled micro-oscillation induction module with the actual position feedback data and cutting force data acquired by the lathe. This matching process aims to identify the system response caused by actively introduced oscillations from the actual data, thereby distinguishing the active excitation from the system's inherent response. For example, the highly correlated portion of the active oscillation signal characteristics can be extracted from background noise and normal machining signals using methods such as signal superposition, correlation analysis, or frequency domain analysis, and used as the actively introduced signal.

[0096] Next, based on the type and contribution level of the interference source, the impact of actively introduced signals on the machining state is assessed. Given the current main interference sources and their contribution ratios, the effects of actively introduced micro-oscillation signals on the current machining state (e.g., tool oscillation, workpiece surface roughness, cutting force fluctuations, etc.) are analyzed. The aim is to quantify the potential effects of different control modules under specific interference scenarios. For example, if actively introduced signals significantly exacerbate a certain type of oscillation, it indicates that the damping control module for that oscillation may require higher priority or stronger action.

[0097] Finally, the module priorities and their intensity are determined based on the degree of influence. Based on a quantitative analysis of the impact of the actively introduced signal, the relative importance (i.e., priority) and specific parameter settings (i.e., intensity) of the position deviation feedforward correction module and the instantaneous cutting force damping control module are dynamically adjusted. For example, if the evaluation results show that the actively introduced signal has a significant impact on the position deviation, the priority of the position deviation feedforward correction module may be increased, and its correction parameters may be adjusted to provide stronger correction capabilities; conversely, if the impact on cutting force fluctuations is significant, the priority and adjustment range of the instantaneous cutting force damping control module will be adjusted accordingly.

[0098] To illustrate this technical solution more clearly, a specific example is used below. Assume that during silicon carbide machining, the time-series decomposition module identifies the primary interference source as micrometer-level step-like fracture events, and that these events contribute significantly. To more accurately determine the priority and intensity of the position deviation feedforward correction module and the instantaneous cutting force damping control module, the controlled micro-oscillation induction module is activated. It applies a preset, low-amplitude sweep frequency oscillation signal to the tool to simulate or amplify the vibration modes associated with the micrometer-level step-like fracture events. During the application of the oscillation signal, the data acquisition module continuously acquires position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro-temperature sensor module. Subsequently, these real-time data are matched with the oscillation signal characteristics generated by the controlled micro-oscillation induction module. Frequency domain analysis is used to identify the system response caused by the active oscillation, thus obtaining the actively introduced signal. For example, if at a specific frequency, components appear in the position feedback data and cutting force data that are synchronized with the active oscillation frequency and have significantly increased amplitude, it indicates that the system is highly sensitive to oscillations at that frequency.

[0099] Next, based on the identified micron-level stepped fracture event as a source of interference and its high contribution, the impact of actively introduced signals on the machining state is assessed. For example, if actively introduced oscillations lead to a momentary increase in position feedback deviation and an intensification of cutting force pulses, it indicates that the current control strategy may be insufficient in suppressing such interference. Based on this quantitative assessment, the system dynamically adjusts module priorities and their intensity. Specifically, if the increase in position deviation is more significant, the priority of the position deviation feedforward correction module is increased, and the gain of its reverse compensation signal is appropriately increased; if the intensification of cutting force pulsation is more prominent, the priority of the instantaneous cutting force damping control module is increased, and the adjustment range of its nonlinear adjustment signal is widened to more effectively suppress cutting force fluctuations. Through this active diagnostic and feedback mechanism, the lathe control strategy can more accurately adapt to complex machining environments, thereby optimizing machining performance.

[0100] Through the above technical solution, this embodiment introduces a controlled micro-oscillation induction module and evaluates its impact on the machining state, enabling the system to actively detect and quantify its sensitivity to specific disturbances. Consequently, the parameter adjustments of the position deviation feedforward correction module and the instantaneous cutting force damping control module will more accurately match the current machining state and disturbance characteristics, significantly improving the real-time adaptability and robustness of the control strategy. This proactive diagnosis and optimization mechanism helps to effectively suppress various complex disturbances in the precision machining of difficult-to-machine materials such as silicon carbide, thereby improving machining accuracy, surface quality, and tool life.

[0101] In some embodiments, step S404, determining the module priority and effect intensity based on the degree of influence, may include, but is not limited to, the following steps:

[0102] Step S501: Adjust the sensitivity of the instantaneous cutting force damping control module according to the actively introduced signal;

[0103] Step S502: After adjusting the instantaneous cutting force damping control module, monitor the acoustic emission data;

[0104] Step S503: Based on the acoustic emission data, identify abnormal acoustic emission events as external interference signals induced by resonance;

[0105] Step S504: Adjust the oscillation parameters of the controlled micro-oscillation induction module according to the signal strength and frequency of the external interference signal. The oscillation parameters include the oscillation frequency and the oscillation amplitude.

[0106] Step S505: Determine the module priority and effect intensity based on the degree of influence, oscillation parameters, signal strength and frequency of external interference signals.

[0107] In some embodiments, relying solely on the influence of actively introduced signals to determine the priority and intensity of the control module may fail to adequately capture microscopic damage or surface defects caused by resonance effects. This is especially true in complex dynamic cutting environments, where potential resonance-induced external interference signals may be ignored, thus affecting the accuracy and robustness of the control strategy. Failure to address these issues may result in insufficient adaptability of the control strategy to actual machining conditions, thereby impacting machining quality and tool life.

[0108] Therefore, the sensitivity of the instantaneous cutting force damping control module can be adjusted based on the actively introduced signal. The sensitivity of the instantaneous cutting force damping control module refers to its responsiveness to cutting force fluctuations. By adjusting this sensitivity according to the actively introduced signal, the response characteristics of the damping control module can be matched with the potential interference modes under the current machining state. For example, when the actively introduced signal indicates the presence of a cutting force fluctuation risk at a specific frequency, the sensitivity of the damping control module within that frequency range can be increased to enhance its suppression capability.

[0109] After adjusting the instantaneous cutting force damping control module, acoustic emission data is monitored. Acoustic emission data refers to the transient elastic wave signal generated by stress wave release during material deformation or fracture. Information on microscopic events during processing can be acquired in real time to obtain acoustic emission data, such as the initiation and propagation of microcracks, friction and wear, etc. These events are often accompanied by specific acoustic emission signal characteristics.

[0110] Then, based on the acoustic emission data, abnormal acoustic emission events are identified as external interference signals induced by resonance. Abnormal acoustic emission events refer to events in which the amplitude, frequency, duration, and other characteristics of the acoustic emission signal deviate significantly from the baseline or preset threshold under normal machining conditions. By analyzing the acoustic emission data, such as using time-frequency analysis and pattern recognition techniques, specific acoustic emission signals related to resonance phenomena caused by lathe structure resonance, tool vibration, or internal material defects can be identified and used as external interference signals induced by resonance.

[0111] Then, based on the signal strength and frequency of the external interference signal, the oscillation parameters of the controlled micro-oscillation induction module are adjusted. These oscillation parameters include the oscillation frequency and amplitude. The controlled micro-oscillation induction module is used to actively introduce minute oscillations during the processing to detect the dynamic response of the system. When a resonance-induced external interference signal is identified, the oscillation parameters of the controlled micro-oscillation induction module can be dynamically adjusted according to its signal strength (e.g., energy level) and signal frequency (e.g., dominant frequency component). For example, if the external interference signal exhibits resonance at a specific frequency, the oscillation frequency of the induction module can be adjusted to avoid or cancel that resonance frequency, or the oscillation amplitude can be adjusted to actively suppress the resonance effect.

[0112] Finally, based on the degree of influence, oscillation parameters, and the signal strength and frequency of external interference signals, the priority and intensity of each module are determined. By comprehensively considering the impact of actively introduced signals on the processing state, dynamically adjusted oscillation parameters, and the intensity and frequency of resonance-induced external interference signals monitored in real time, the complexity of the current processing environment can be assessed more comprehensively and accurately. This allows for the dynamic determination of the priority and intensity of each control module to achieve optimal control performance.

[0113] Through the above technical solution, this embodiment, by introducing acoustic emission monitoring and resonance-induced signal identification, can detect and quantify resonance risks and micro-damage events during the machining process earlier and more accurately. Dynamically adjusting the oscillation parameters of the controlled micro-oscillation induction module allows the system to actively intervene and suppress harmful resonances, thereby avoiding or mitigating their adverse effects on machining quality and tool life. Consequently, the determined module priorities and intensity of action are more aligned with actual machining requirements, more effectively suppressing cutting force fluctuations and correcting position feedback deviations, ultimately achieving higher machining accuracy, better surface quality, and longer tool life. Its advantages are particularly pronounced when processing silicon carbide materials susceptible to resonance.

[0114] In some embodiments, after adjusting the oscillation parameters of the controlled micro-oscillation induction module according to the signal strength and frequency of the external interference signal in step S504, the method may also include, but is not limited to, the following steps:

[0115] Step S601: Analyze the microstructure of silicon carbide material to obtain crystal structure parameters and defect distribution information;

[0116] Step S602: Perform vibration modal analysis on the mechanical structure of the lathe to obtain the natural frequency spectrum of the lathe;

[0117] Step S603: Compare the oscillation parameters with the crystal structure parameters, defect distribution information, and lathe natural frequency spectrum to determine whether the oscillation parameters overlap with the secondary resonance mode of the material or the natural frequency of the lathe, and obtain the overlap risk analysis results.

[0118] Step S604: If the overlap risk analysis result indicates that there is an overlap risk, then update the oscillation parameters according to the preset safety interval;

[0119] Step S605: Identify abnormal signals in the target data that are related to the instantaneous changes in the local hardness or toughness of the material. The target data includes cutting force data or acoustic emission data.

[0120] Step S606: Adjust the acceleration / deceleration curve of the feed axis and the spindle speed according to the updated oscillation parameters and abnormal signals.

[0121] In some embodiments, adjusting oscillation parameters solely based on external interference signals may fail to adequately consider the microstructural characteristics of the silicon carbide material itself and the inherent vibration modes of the lathe's mechanical structure. If these issues are not addressed, the induced oscillation parameters may overlap with the material's secondary resonance modes or the lathe's natural frequency, leading to resonance, decreased machining accuracy, accelerated tool wear, and even damage to the workpiece or equipment.

[0122] Therefore, the microstructure of silicon carbide materials can be analyzed first to obtain crystal structure parameters and defect distribution information. Crystal structure parameters of silicon carbide materials, such as lattice constant, grain size, and grain boundary distribution, as well as defect distribution information, such as dislocations, vacancies, and stacking faults, can be obtained through techniques such as X-ray diffraction, scanning electron microscopy, or transmission electron microscopy. This information helps to understand the material's response characteristics under vibrations at different frequencies.

[0123] Then, vibration modal analysis is performed on the lathe's mechanical structure to obtain its natural frequency spectrum. Vibration modes and natural frequency spectra of the lathe at different frequencies can be determined using methods such as finite element analysis (FEA) or experimental modal analysis (EMA). This includes identifying the resonant frequencies of key components such as the spindle, feed axis, and tool post to avoid machining near these frequencies.

[0124] The oscillation parameters are then compared with crystal structure parameters, defect distribution information, and the lathe's natural frequency spectrum to determine whether the oscillation parameters overlap with the material's secondary resonance mode or the lathe's natural frequency, thus obtaining the overlap risk analysis results. The purpose is to assess whether the currently set oscillation parameters have the risk of overlapping with the material's secondary resonance mode or the lathe's natural frequency. For example, if the oscillation frequency is close to a certain lattice vibration frequency of the material or a certain natural frequency of the lathe, then an overlap risk is considered to exist.

[0125] If the overlap risk analysis indicates the presence of overlap risk, the oscillation parameters are updated according to a preset safety interval. The preset safety interval is a frequency or amplitude range used to ensure sufficient margin between the updated oscillation parameters and the potential resonance frequency to avoid resonance. For example, the oscillation frequency can be adjusted upwards or downwards by a percentage indicated by the preset safety interval to move it away from the resonance point. The preset safety interval can be calculated by statistically analyzing a large amount of historical data (which may include oscillation parameters, crystal structure parameters, defect distribution information, and the lathe's natural frequency spectrum).

[0126] Finally, abnormal signals related to instantaneous changes in the local hardness or toughness of the material are identified in the target data. This target data includes cutting force data or acoustic emission data, designed to detect potential local inhomogeneities within the silicon carbide material during machining, such as hard spots, soft areas, or microcrack propagation. These changes can cause instantaneous anomalies in the cutting force or acoustic emission signals. Based on the updated oscillation parameters and abnormal signals, the acceleration / deceleration curves of the feed axis and the spindle speed are adjusted. This means the system can dynamically adjust the lathe's kinematic parameters to adapt to material changes, ensuring machining stability and quality, by using optimized oscillation parameters to avoid resonance and combining real-time sensing of the material's local properties.

[0127] Through the above technical solution, this embodiment can significantly reduce the risk of resonance caused by the overlap of oscillation parameters with the natural frequencies of the material or lathe during silicon carbide processing, thereby effectively avoiding abnormal tool wear, decreased workpiece processing quality, and potential structural damage to the equipment. This embodiment, through in-depth analysis of the microstructure of silicon carbide and the mechanical structure of the lathe, makes the adjustment of oscillation parameters more scientific and precise, ensuring that the induced micro-oscillations not only effectively suppress external interference but also coexist harmoniously with the physical properties of the processing system itself. Furthermore, by real-time identification of abnormal signals indicating instantaneous changes in local hardness or toughness of the material, this embodiment improves the adaptability and robustness of the lathe control system to complex and non-uniform silicon carbide materials, thereby significantly improving the stability and reliability of the processing while ensuring high-precision machining.

[0128] In some embodiments, step S605, identifying abnormal signals in the target data related to instantaneous changes in local hardness or toughness of the material, may include, but is not limited to, the following steps:

[0129] Multi-feature extraction is performed on the target data to obtain multi-dimensional features, including instantaneous amplitude, frequency distribution, kurtosis, skewness, and waveform.

[0130] Based on the signal characteristics and fluctuation range under normal processing conditions, a real-time baseline model of the target data is established;

[0131] The target data is compared with a real-time baseline model to identify deviation signals;

[0132] Analyze the duration and magnitude of the deviation based on the deviation signal;

[0133] Based on the interrelationships between deviation duration, deviation magnitude, and multi-dimensional features, an abnormal signal feature vector is constructed.

[0134] The abnormal signal feature vector is matched with the signal feature template to identify the abnormal signal. The signal feature template includes the local hardness feature template and the instantaneous change feature template of toughness.

[0135] In some embodiments, instantaneous changes in local hardness or toughness of a material may exhibit a variety of complex signal characteristics, such as different amplitudes, frequencies, durations, and waveforms. If only a single or simple threshold method is used for identification, it may lead to false alarms or missed alarms, thereby affecting the accuracy and timeliness of subsequent feed axis acceleration / deceleration curves and spindle speed adjustments, and consequently affecting machining quality and tool life.

[0136] Therefore, multi-feature extraction can be performed on the target data to obtain multi-dimensional features, including instantaneous amplitude, frequency distribution, kurtosis, skewness, and waveform. Multiple statistical or time-frequency domain features that comprehensively reflect the signal characteristics can be extracted from the target data. Instantaneous amplitude reflects the instantaneous intensity of the signal; frequency distribution reveals the energy distribution of the signal at different frequencies; kurtosis measures the sharpness or impulsiveness of the signal; skewness describes the symmetry of the signal distribution; and waveform directly reflects the time-domain morphology of the signal. The extraction of these multi-dimensional features aims to provide rich and comprehensive information for subsequent anomalous signal analysis.

[0137] Then, based on the signal characteristics and fluctuation range under normal processing conditions, a real-time baseline model for the target data is established. This model can be built by performing statistical analysis and machine learning training on a large amount of target data under normal processing conditions. The model can capture the typical characteristics, trends, and permissible fluctuation range of signals under normal processing conditions. The purpose of establishing a real-time baseline model is to provide a reliable reference standard for identifying abnormal signals, ensuring that any signal deviating from the normal state can be effectively detected.

[0138] The target data is then compared with the real-time baseline model to identify deviation signals. Various statistical methods or machine learning algorithms can be employed, such as distance-based anomaly detection, density-based anomaly detection, or residual analysis based on predictive models. By comparing the real-time acquired target data with the pre-established real-time baseline model, signals that significantly differ from the normal processing state can be effectively identified; these difference signals are the deviation signals.

[0139] Based on the deviation signal, analyze the deviation duration and deviation magnitude. Deviation duration refers to the length of time the signal deviates from the baseline model, while deviation magnitude refers to the degree to which the signal deviates from the baseline model. Analyzing these parameters helps to preliminarily determine the nature and severity of the abnormal event. For example, a short-duration, high-amplitude deviation may indicate a transient impact, while a long-duration, low-amplitude deviation may indicate progressive wear.

[0140] An anomalous signal feature vector is constructed based on the interrelationships between deviation duration, deviation amplitude, and multi-dimensional features. The deviation duration, deviation amplitude, and extracted multi-dimensional features (such as instantaneous amplitude, frequency distribution, kurtosis, skewness, and waveform) can be combined to form a high-dimensional feature representation. This combination aims to comprehensively capture the intrinsic characteristics of the anomalous signal, providing richer information for subsequent accurate identification.

[0141] Finally, the abnormal signal feature vectors are matched with signal feature templates to identify the abnormal signals. These templates include material local hardness feature templates and instantaneous toughness change feature templates. The signal feature templates are typical signal characteristic patterns established in advance through experimental simulations for different types of material local hardness and instantaneous toughness changes. For example, an increase in material local hardness may cause a high-amplitude pulse at a specific frequency in the cutting force signal, while an instantaneous change in toughness may cause a sudden energy change in the acoustic emission signal with a specific waveform. By matching the constructed abnormal signal feature vectors with these templates, the specific type of abnormal signal can be accurately identified, thus providing an accurate basis for subsequent control strategy adjustments.

[0142] Through the above technical solution, this embodiment significantly improves the accuracy and reliability of abnormal signal identification by comprehensively analyzing multi-dimensional features such as instantaneous amplitude, frequency distribution, kurtosis, skewness, and waveform, and combining real-time baseline models and feature template matching. This refined identification capability helps avoid false alarms and missed alarms, ensuring that subsequent feed axis acceleration / deceleration curves and spindle speed adjustments can more accurately respond to the actual state of the material, thereby effectively suppressing unstable factors in the machining process, improving machining quality, extending tool life, and optimizing overall machining efficiency.

[0143] In some embodiments, in step S606, adjusting the acceleration / deceleration curve of the feed axis and the spindle speed based on the updated oscillation parameters and abnormal signals may include, but is not limited to, the following steps:

[0144] Monitor tool vibration data;

[0145] Based on cutting force data and tool vibration data, analyze the micro-geometric variation characteristics of the tool-workpiece contact area;

[0146] Based on the characteristics of microscopic geometric changes, predict the frequency and amplitude of transient impacts of cutting forces;

[0147] The frequency of the transient impact of the cutting force is compared with the frequency of the abnormal signal to obtain the first comparison result;

[0148] The amplitude of the transient impact of the cutting force is compared with the amplitude of the abnormal signal to obtain the second comparison result;

[0149] Based on the updated oscillation parameters, the first comparison result, and the second comparison result, adjust the acceleration / deceleration curve of the feed axis and the spindle speed.

[0150] In some embodiments, the micro-geometric changes in the tool-workpiece contact area and the resulting transient impacts of cutting forces are highly dynamic and complex. Adjusting solely based on macroscopic oscillation parameters and abnormal signals may not fully capture and effectively address these transient, local dynamic characteristics, thereby affecting machining accuracy and surface quality.

[0151] Therefore, tool vibration data can be monitored first. This can be achieved in real time using high-precision accelerometers mounted on the tool clamping system or spindle. These sensors can capture the minute vibrations generated by the tool during cutting, including their frequency, amplitude, and phase information. The purpose is to provide basic data for subsequent analysis of the microscopic dynamics of the tool-workpiece contact area.

[0152] Then, based on cutting force data and tool vibration data, the microscopic geometric changes in the tool-workpiece contact area are analyzed. This includes the interaction patterns between the tool cutting edge and the workpiece material at the microscale, such as microcrack propagation during chip formation, material spalling, and geometric changes caused by tool wear. By combining cutting force data from the cutting force sensing module with real-time monitored tool vibration data, signal fusion or pattern recognition can be used to extract and analyze features of these complex microscopic geometric changes. For example, abrupt changes or periodic fluctuations in cutting force data may be related to the formation of microcracks, while specific frequency components of tool vibration data may reflect the impact or friction state between the tool and the workpiece. The aim is to gain a deeper understanding of the transient physical phenomena in the cutting process.

[0153] Then, based on the microscopic geometric change characteristics, the frequency and amplitude of transient cutting force impacts are predicted. Once the microscopic geometric change characteristics are identified, potential transient cutting force impacts can be predicted based on a pre-established data-driven model. For example, when a specific type of microcrack propagation characteristic is detected, the high-frequency, high-amplitude transient impacts that it may cause can be predicted. The predicted frequency and amplitude are key parameters for subsequent precise adjustments. The data-driven model can be trained using a large amount of historical data (including microscopic geometric change characteristics and corresponding transient cutting force impacts).

[0154] The first comparison result is obtained by comparing the frequency of the transient impact of the cutting force with the frequency of the abnormal signal. Methods such as spectrum analysis and correlation analysis can be used to determine the degree of matching or difference between the two in the frequency domain. The purpose is to assess whether the abnormal signal matches the actual transient impact frequency. Simultaneously, the second comparison result is obtained by comparing the amplitude of the transient impact of the cutting force with the amplitude of the abnormal signal. Indicators such as amplitude difference and relative error can be used to quantify the degree of matching between the two in terms of energy or intensity. The purpose is to assess whether the abnormal signal matches the actual transient impact amplitude.

[0155] Finally, based on the updated oscillation parameters and the first and second comparison results, the acceleration / deceleration curves of the feed axis and the spindle speed are adjusted. Based on the updated oscillation parameters, and combined with the first and second comparison results, the lathe's motion commands can be more precisely corrected. For example, if the comparison results show that the predicted transient impact highly matches the abnormal signal, the acceleration or deceleration of the feed axis can be fine-tuned before or during the impact, or the spindle speed can be instantaneously changed to avoid the resonant frequency or reduce the impact energy. The aim is to effectively suppress transient impacts and improve machining stability by dynamically adjusting motion parameters.

[0156] Through the above technical solution, this embodiment achieves a deeper understanding of the cutting process by real-time monitoring of tool vibration and analysis of microscopic geometric changes, thereby enabling more accurate prediction of the frequency and amplitude of transient cutting force impacts. This refined prediction and abnormal signal comparison mechanism makes the adjustment of feed axis acceleration / deceleration curves and spindle speed more targeted and real-time, significantly improving the lathe's ability to suppress transient impacts. Consequently, it can effectively reduce tool vibration and workpiece surface defects, extend tool life, and ultimately improve the machining quality and efficiency of silicon carbide materials, especially when processing materials with transient changes in local hardness or toughness, where its advantages are even more pronounced.

[0157] The beneficial effects of implementing the embodiments of the present invention include: First, the embodiments of this application acquire the position feedback data of the servo motor encoder module, the cutting force data of the cutting force sensor module, and the temperature data of the micro temperature sensor module, and perform time series decomposition to identify the type and degree of interference source contribution. Then, the lathe control strategy is determined. If the lathe control strategy is a thermal drift compensation strategy, a reverse compensation signal is generated to correct the position feedback deviation. If the lathe control strategy is a cutting force fluctuation damping strategy, a nonlinear adjustment signal is generated to nonlinearly adjust the feed axis acceleration command. Then, the cutting parameters are adjusted according to the type of interference source, the interference intensity, the tool state, and the target signal. Finally, the lathe is controlled according to the cutting parameters. Thus, the cutting parameters can be adjusted by combining sensor data and interference information to achieve lathe control, thereby improving control accuracy and workpiece machining quality.

[0158] As shown in Figure 2, this embodiment of the invention also provides an intelligent control system for a silicon carbide machining lathe, comprising:

[0159] The data acquisition module 701 is used to acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module.

[0160] The time series decomposition module 702 is used to perform time series decomposition based on position feedback data, cutting force data, and temperature data to identify the type of interference source and the degree of contribution of the interference source.

[0161] The control strategy determination module 703 is used to determine the lathe control strategy based on the type of interference source and the degree of contribution of the interference source;

[0162] The compensation signal generation module 704 is used to generate a reverse compensation signal based on temperature data if the lathe control strategy is a thermal drift compensation strategy. The reverse compensation signal is used to correct the position feedback deviation.

[0163] The adjustment signal generation module 705 is used to generate a nonlinear adjustment signal based on the cutting force data if the lathe control strategy is a cutting force fluctuation damping strategy. The nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuation.

[0164] The cutting parameter adjustment module 706 is used to adjust the cutting parameters according to the type of interference source, interference intensity, tool status and target signal. The target signal includes a reverse compensation signal or a nonlinear adjustment signal.

[0165] The lathe control module 707 is used to control the lathe according to the cutting parameters.

[0166] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0167] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

Claims

1. A method for intelligent control of a silicon carbide machining lathe, characterized in that, Includes the following steps: The lathe control strategy is determined by acquiring position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module; based on the position feedback data, the cutting force data, and the temperature data, time series decomposition is performed to identify the type and degree of interference source contribution; based on the type and degree of interference source contribution, the lathe control strategy is determined. If the lathe control strategy is a thermal drift compensation strategy, a reverse compensation signal is generated based on the temperature data. This reverse compensation signal is used to correct position feedback deviations. If the lathe control strategy is a cutting force fluctuation damping strategy, a nonlinear adjustment signal is generated based on the cutting force data. This nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuations. Cutting parameters are adjusted based on the interference source type, interference intensity, tool state, and target signal. The target signal includes either the reverse compensation signal or the nonlinear adjustment signal. The lathe is controlled based on these cutting parameters.

2. The method according to claim 1, characterized in that, The step of performing time series decomposition based on the position feedback data, the cutting force data, and the temperature data to identify the type and degree of interference source contribution includes: performing multi-scale decomposition on the position feedback data to obtain position signal components at different time scales; identifying non-periodic transient micro-jumping signals based on the position signal components; performing time-frequency analysis on the cutting force data to identify broadband cutting force pulses; identifying transient micro-jumping signal interference sources based on the amplitude, first duration, and synchronization with the broadband cutting force pulses of the transient micro-jumping signal; and if the transient micro-jumping signal interferes with... If the disturbance source is a micrometer-level stepped fracture event, then the position feedback data is subjected to micro-jump signal separation; the position feedback data after micro-jump signal separation is subjected to low-pass filtering and trend analysis to obtain the position deviation component; based on the broadband cutting force pulse and the cutting force data, the internal stress fluctuation component of the material is identified; based on the temperature data, temperature correlation analysis is performed on the position deviation component to obtain the temperature correlation analysis result; based on the internal stress fluctuation component of the material and the temperature correlation analysis result, the type of disturbance source is identified; based on the type of disturbance source, the contribution degree of the disturbance source is evaluated.

3. The method according to claim 2, characterized in that, The step of identifying the internal stress fluctuation components of the material based on the broadband cutting force pulse and the cutting force data includes: identifying the broadband cutting force pulse interference source based on the peak energy, second duration, and synchronization with the transient micro-jump signal of the broadband cutting force pulse; if the broadband cutting force pulse interference source is a micrometer-level stepped fracture event, then performing cutting force pulse separation on the cutting force data; and performing high-pass filtering and spectrum analysis on the cutting force data after cutting force pulse separation to obtain the internal stress fluctuation components of the material.

4. The method according to claim 1, characterized in that, The process of determining a lathe control strategy based on the type and contribution level of the interference source includes: acquiring a position deviation feedforward correction module and an instantaneous cutting force damping control module; determining the module priority and intensity based on the type and contribution level of the interference source; determining the motion command superposition order of the position deviation feedforward correction module based on the module priority and intensity; performing deviation correction simulation on the position deviation feedforward correction module according to the motion command superposition order to obtain deviation correction simulation results; determining the adjustment range of the instantaneous cutting force damping control module based on the module priority and intensity; performing nonlinear adjustment simulation on the instantaneous cutting force damping control module according to the adjustment range to obtain nonlinear adjustment simulation results; evaluating tool oscillation information, workpiece machining quality, and tool wear state based on the deviation correction simulation results and the nonlinear adjustment simulation results; and determining the lathe control strategy based on the tool oscillation information, workpiece machining quality, and tool wear state.

5. The method according to claim 4, characterized in that, The step of determining the module priority and intensity based on the type and contribution of the interference source includes: generating oscillation signal characteristics through a controlled micro-oscillation induction module; matching the position feedback data, the cutting force data, and the oscillation signal characteristics to obtain an actively introduced signal; evaluating the influence of the actively introduced signal on the machining state based on the type and contribution of the interference source; and determining the module priority and intensity based on the degree of influence.

6. The method according to claim 5, characterized in that, The step of determining the priority and intensity of the module based on the degree of influence includes: adjusting the sensitivity of the instantaneous cutting force damping control module based on the actively introduced signal; monitoring acoustic emission data after adjusting the instantaneous cutting force damping control module; identifying abnormal acoustic emission events as resonance-induced external interference signals based on the acoustic emission data; adjusting the oscillation parameters of the controlled micro-oscillation induction module based on the signal strength and frequency of the external interference signal, wherein the oscillation parameters include oscillation frequency and oscillation amplitude; and determining the priority and intensity of the module based on the degree of influence, the oscillation parameters, and the signal strength and frequency of the external interference signal.

7. The method according to claim 6, characterized in that, After adjusting the oscillation parameters of the controlled micro-oscillation induction module based on the signal strength and frequency of the external interference signal, the method further includes: analyzing the microstructure of the silicon carbide material to obtain crystal structure parameters and defect distribution information; performing vibration mode analysis on the mechanical structure of the lathe to obtain the lathe's natural frequency spectrum; comparing the oscillation parameters with the crystal structure parameters, the defect distribution information, and the lathe's natural frequency spectrum to determine whether the oscillation parameters overlap with the material's secondary resonance mode or the lathe's natural frequency, and obtaining an overlap risk analysis result; if the overlap risk analysis result indicates an overlap risk, updating the oscillation parameters according to a preset safety interval; identifying abnormal signals in the target data related to instantaneous changes in the material's local hardness or toughness, the target data including the cutting force data or the acoustic emission data; and adjusting the acceleration / deceleration curve of the feed axis and the spindle speed based on the updated oscillation parameters and the abnormal signals.

8. The method according to claim 7, characterized in that, The identification of abnormal signals related to instantaneous changes in local hardness or toughness of the target data includes: extracting multiple features from the target data to obtain multi-dimensional features, including instantaneous amplitude, frequency distribution, kurtosis, skewness, and waveform; establishing a real-time baseline model of the target data based on the signal characteristics and fluctuation range under normal processing conditions; comparing the target data with the real-time baseline model to identify deviation signals; analyzing the deviation duration and deviation amplitude based on the deviation signals; constructing an abnormal signal feature vector based on the interrelationship between the deviation duration, the deviation amplitude, and the multi-dimensional features; and matching the abnormal signal feature vector with a signal feature template to identify the abnormal signal, wherein the signal feature template includes a material local hardness feature template and a toughness instantaneous change feature template.

9. The method according to claim 7, characterized in that, The step of adjusting the acceleration / deceleration curve of the feed axis and the spindle speed based on the updated oscillation parameters and the abnormal signal includes: monitoring tool vibration data; analyzing the microscopic geometric change characteristics of the tool-workpiece contact area based on the cutting force data and the tool vibration data; predicting the frequency and amplitude of transient cutting force impacts based on the microscopic geometric change characteristics; comparing the frequency of the transient cutting force impacts with the frequency of the abnormal signal to obtain a first comparison result; comparing the amplitude of the transient cutting force impacts with the amplitude of the abnormal signal to obtain a second comparison result; and adjusting the acceleration / deceleration curve of the feed axis and the spindle speed based on the updated oscillation parameters, the first comparison result, and the second comparison result.

10. An intelligent control system for a silicon carbide machining lathe, characterized in that, include: The data acquisition module is used to acquire position feedback data from the servo motor encoder module, cutting force data from the cutting force sensor module, and temperature data from the micro temperature sensor module. The time series decomposition module is used to perform time series decomposition based on the position feedback data, the cutting force data, and the temperature data to identify the type and degree of interference source contribution; the control strategy determination module is used to determine the lathe control strategy based on the type and degree of interference source contribution. The compensation signal generation module is used to generate a reverse compensation signal based on the temperature data if the lathe control strategy is a thermal drift compensation strategy. The reverse compensation signal is used to correct the position feedback deviation. The adjustment signal generation module is used to generate a nonlinear adjustment signal based on the cutting force data if the lathe control strategy is a cutting force fluctuation damping strategy. The nonlinear adjustment signal is used to nonlinearly adjust the feed axis acceleration command to suppress cutting force fluctuation. The cutting parameter adjustment module is used to adjust the cutting parameters according to the type of interference source, the intensity of interference, the tool state, and the target signal, wherein the target signal includes the reverse compensation signal or the nonlinear adjustment signal; the lathe control module is used to control the lathe according to the cutting parameters.

Images