Flexible Pressing Material CNC Machining Method Based on Data Processing

Abstract

The present invention specifically relates to a flexible pressing CNC machining method based on data processing. First, a torque sensor, a vibration sensor and a temperature sensor are integrated in a CNC machine tool, and a machine learning model is trained using historical machining data. The machine learning model includes the relationship between machining parameters, tool wear status, material characteristic factors and machining residual shape, surface roughness and bonding strength. The machine learning model is used to predict the optimal machining parameter combination under different conditions. According to the prediction results of the machine learning model, the machining path is fine-tuned to ensure that the peak and trough shape of the machining residue remains consistent in each area. In the narrow end area, path planning is used to avoid local stress concentration and excessive cutting. The machining is started, and the sensor monitoring and adaptive control unit are turned on at the same time. The machining parameters are dynamically adjusted according to real-time data feedback to ensure the stability of the machining process.

Description

Technical Field

[0001] The invention belongs to the field of numerical control, and in particular relates to a flexible pressing material numerical control processing method based on data processing. Background Art

[0002] In the current manufacturing environment, especially in the production process of precision molds and complex parts, CNC machining technology has become the key to improving production efficiency, ensuring machining accuracy and product consistency. However, for specific application areas such as the manufacture of flexible pressing devices, traditional CNC machining methods face many challenges. Especially in the processing of the casting surface of the metal base, due to its high requirements for the residual shape of the processing (such as the peak and trough structure), the traditional method is difficult to control accurately, and puts forward high requirements on the setting of processing parameters, the selection of tools, and the dynamic response capability of the machine tool. If improperly controlled, it may lead to inconsistent residual shapes of the processing, affecting the subsequent bonding strength. Summary of the invention

[0003] The purpose of the present invention is to provide a flexible pressing material numerical control processing method based on data processing to solve the problems raised in the above background technology.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] A flexible pressing material numerical control processing method based on data processing, comprising the steps of

[0006] First, in CNC machine tools, torque sensors, vibration sensors, and temperature sensors are integrated to monitor the cutting force, vibration frequency, and tool or workpiece temperature during machining in real time;

[0007] Establish real-time data processing algorithms to instantly analyze sensor feedback data and identify abnormalities during processing;

[0008] Based on the real-time data analysis results, an adaptive control unit is built. The adaptive control unit dynamically adjusts the processing parameters according to the current processing status to ensure the stability and consistency of the processing process;

[0009] Use historical processing data to train a machine learning model that includes the relationship between processing parameters, tool wear status, material characteristics, and processing residual shape, surface roughness, and bonding strength. Use the machine learning model to predict the best combination of processing parameters under different conditions.

[0010] Based on the prediction results of the machine learning model, the processing path is fine-tuned to ensure that the peak and trough shapes of the processing residues remain consistent in all areas. In the narrow end area, path planning is used to avoid local stress concentration and excessive cutting. The processing is started, and the sensor monitoring and adaptive control unit are turned on at the same time. The processing parameters are dynamically adjusted according to real-time data feedback to ensure the stability of the processing process.

[0011] Furthermore, abnormal conditions during machining include sudden changes in cutting force, abnormal vibration frequency, or excessive temperature.

[0012] Furthermore, the establishment of a real-time data processing algorithm includes: first, collecting raw data from high-precision torque sensors, vibration sensors, and intelligent temperature monitoring systems; the data is filtered and processed, as well as signal smoothing, and data normalization is performed;

[0013] By using Fourier transform, frequency domain features are extracted from the cutting force signal, and high-frequency components or energy surges that deviate from the normal cutting mode are identified as signs of sudden changes in cutting force.

[0014] By performing fast Fourier transform (FFT) on the vibration signal, analyzing its spectral characteristics, and setting thresholds to identify vibration frequencies and amplitudes that exceed the normal range;

[0015] A temperature rise model based on historical data is established to predict the temperature change trend under normal processing. When the actual measured value deviates significantly from the predicted value, an abnormal temperature alarm is issued.

[0016] Furthermore, based on the real-time data analysis results, an adaptive control unit is constructed specifically to: dynamically adjust the thresholds of cutting force, vibration and temperature according to the real-time monitored processing conditions;

[0017] When the processing status is detected to deviate from the normal range, the adaptive control unit adjusts the processing parameters, or adjusts the speed, step distance, or stops the processing completely;

[0018] The adaptive control unit also automatically adjusts machining parameters according to the degree of tool wear, increasing the rotation speed to maintain cutting efficiency in the early stage of tool wear, and reducing the cutting depth and feed rate in the later stage of tool wear;

[0019] For areas with peak and trough machining residues, the tool path feed angle and sequence are dynamically adjusted.

[0020] Furthermore, using historical processing data to train machine learning models specifically includes:

[0021] A gradient boosting tree model was selected, and the data set was divided into training set, validation set and test set. The model was trained on the training set using a cross-validation strategy. For new machining tasks, material properties, expected machining residual shape requirements and bonding strength targets were input, and the model automatically predicted the optimal machining parameter combination, including cutting speed, feed rate, tool type and wear tolerance.

[0022] Furthermore, in the narrow end area, path planning is used to avoid local stress concentration and over-cutting, and tool path smoothing, including Bezier curve or spline curve interpolation, is used to make the path transition smoothly when entering and exiting the narrow area.

[0023] The Bezier curve or spline curve interpolation specifically adopts cubic Bezier curve or quintic spline curve interpolation. The Bezier curve determines the shape of the curve through control points. The quintic spline curve is composed of quintic polynomials in each interval, which ensures the continuity of the first-order and second-order derivatives on the entire path, further reducing the change of acceleration.

[0024] Beneficial effects: The specific technical effects of this application are reflected in the following aspects:

[0025] Improved machining accuracy and consistency: This method monitors key parameters such as cutting force, vibration frequency and temperature in real time, and uses adaptive control technology to dynamically adjust machining parameters, effectively avoiding abnormal situations during the machining process, ensuring that the residual shape of the peak and trough structure is highly consistent, and improving the geometric accuracy and dimensional stability of the product.

[0026] Optimization of production efficiency: Through integrated tool wear monitoring and dynamic adjustment strategies, not only the tool life is extended, but also high cutting efficiency is maintained by timely adjustment of processing parameters, reducing downtime caused by tool replacement and reprocessing, thereby significantly improving production efficiency.

[0027] Material utilization and cost savings: Accurate machining parameter prediction and adaptive path planning effectively reduce over-cutting and scrap rates, especially in critical areas such as narrow ends. Smooth tool path design reduces material waste and avoids workpiece damage caused by local stress concentration, which helps reduce production costs in the long run.

[0028] Enhanced process adaptability and flexibility: The multi-objective optimization machine learning model can automatically predict the optimal combination of processing parameters based on different material properties and processing requirements, enabling the technical solution to flexibly respond to the processing challenges of various complex workpieces and improve the adaptability and flexibility of the processing process.

[0029] Intelligent fault prevention and safety management: Real-time data processing algorithms and abnormality identification mechanisms can detect and respond to potential problems in the processing process in advance, such as sudden changes in cutting force, abnormal vibration or excessive temperature, effectively preventing equipment damage and safety accidents and ensuring production safety.

[0030] To sum up, the technical solution of the present application not only solves the difficulties of traditional CNC machining in the manufacturing of complex parts, especially flexible pressing devices, through intelligent and adaptive processing strategies, but also promotes the development of the manufacturing industry towards higher precision, higher efficiency and greater intelligence, bringing innovative technological progress to the field of precision manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a method flow chart of the flexible pressing material numerical control processing method based on data processing of the present application. DETAILED DESCRIPTION

[0032] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] The present application discloses a flexible pressing material numerical control processing method based on data processing, such as Figure 1 As shown, the detailed technical solution includes the following steps:

[0034] First, it is necessary to upgrade the CNC machine tools and integrate high-precision torque sensors, vibration sensors and temperature sensors to monitor the cutting force, vibration frequency and tool / workpiece temperature in real time during the machining process. Specific high-precision torque sensors: select torque sensors with high resolution and fast response capabilities, and install them at the interface between the spindle and the tool to accurately measure the changes in cutting force during the machining process. This sensor should be able to feedback the slight fluctuations in cutting force in real time, help the control system to promptly identify changes in cutting conditions, and prevent machining inconsistencies caused by overload or underload.

[0035] Specific high-performance vibration sensors: Install high-frequency response vibration sensors on key parts of machine tools (such as spindle head and bed) to monitor the vibration frequency and amplitude during the machining process. Vibration data can reflect the tool status, workpiece clamping stability and dynamic characteristics during the machining process.

[0036] Intelligent temperature monitoring system: A composite monitoring system that integrates contact and non-contact temperature sensors. The contact sensor is directly installed on the tool shank or workpiece clamping area to monitor local overheating; the non-contact infrared thermometer is used to scan the entire processing area to capture the temperature rise of the workpiece surface and cutting area.

[0037] The flexible pressing CNC machining method based on data processing also includes establishing a real-time data processing algorithm, which can instantly analyze the data fed back by the sensor and identify abnormal conditions during the machining process, such as sudden changes in cutting force, abnormal vibration frequency or excessive temperature, which are precursors to inconsistent machining residual shapes.

[0038] First, raw data is collected from high-precision torque sensors, vibration sensors, and intelligent temperature monitoring systems. The data is filtered, such as using a low-pass filter to remove high-frequency noise, and signal smoothing to ensure data accuracy. At the same time, data normalization is performed to facilitate comparison and analysis of data in different dimensions.

[0039] Cutting force characteristic analysis: Use time series analysis methods, such as Fourier transform, to extract frequency domain features from the cutting force signal and identify high-frequency components or energy surges that deviate from the normal cutting mode as signs of sudden changes in cutting force.

[0040] Vibration signature analysis: By performing a fast Fourier transform (FFT) on the vibration signal, analyzing its spectral characteristics, and setting thresholds to identify vibration frequencies and amplitudes that are out of the normal range, especially those specific frequency bands associated with tool failure or workpiece instability.

[0041] Temperature monitoring and early warning: Establish a temperature rise model based on historical data to predict the temperature change trend under normal processing. When the actual measured value deviates significantly from the predicted value, an abnormal temperature alarm is issued.

[0042] It is also possible to further use ensemble learning algorithms (such as random forests, support vector machines or neural networks) to build a classification model, using the extracted features as input to determine in real time whether the current processing state is abnormal. Once an abnormal situation is identified (such as a sudden change in cutting force, abnormal vibration or excessive temperature), the algorithm immediately triggers an alarm and takes corresponding measures through the adaptive control unit.

[0043] The flexible pressing CNC machining method based on data processing also includes building an adaptive control unit based on the real-time data analysis results. The adaptive control unit can dynamically adjust the machining parameters according to the current machining state to ensure the stability and consistency of the machining process. Specifically: dynamically adjust the thresholds of cutting force, vibration and temperature according to the machining conditions monitored in real time. These thresholds are not fixed, but are dynamically optimized based on historical data learning and current working conditions to more accurately identify machining anomalies.

[0044] When the monitoring processing status deviates from the normal range, the adaptive control unit implements a multi-level response strategy. The primary response is only to slightly adjust the processing parameters, such as slightly reducing the feed rate; the intermediate response includes adjusting the speed and step distance to adapt to the current processing conditions; in case of serious abnormalities, emergency measures are taken, such as completely stopping the processing to avoid tool damage or workpiece scrapping.

[0045] The adaptive control unit also integrates tool wear monitoring and automatically adjusts machining parameters according to the degree of tool wear. For example, in the initial stage of tool wear, the rotation speed is appropriately increased to maintain cutting efficiency, while in the later stage, the cutting depth and feed speed are reduced to extend the tool life and ensure machining quality.

[0046] For complex surfaces, the adaptive control unit can fine-tune the machining path based on the current machining status, especially for areas with peak and trough machining residues, dynamically adjusting the feed angle and sequence of the tool path to avoid concentrated cutting in high-stress areas and reduce the risk of local overload.

[0047] The logic framework of expressing the functions of the adaptive control unit in code form is:

[0048]

[0049]

[0050] This code describes the workflow of an adaptive control unit, which continuously receives data from sensors, analyzes whether the machining status is normal, and takes corresponding adjustment measures according to different abnormal levels. At the same time, it considers the dynamic adjustment strategy for tool wear, as well as path optimization for complex surfaces to reduce local stress concentration and overload risks. The whole process embodies the core idea of ​​dynamic optimization based on real-time data feedback and historical learning to ensure the stability and consistency of the machining process.

[0051] The flexible pressing CNC machining method based on data processing also includes using historical machining data to train a machine learning model, which includes the relationship between machining parameters, tool wear status, material characteristic factors and machining residual shape, surface roughness and bonding strength. Through this machine learning model, the best machining parameter combination under different conditions is predicted.

[0052] The specific implementation steps are as follows:

[0053] First, the machining instance data is extracted from the historical database, including successful machining parameter configurations, tool types and wear conditions, material property information, and the corresponding machining residual shapes, surface roughness measurements, and actual bonding strength test results.

[0054] Clean the data, remove outliers and missing data, and standardize or normalize processing parameters and measurement indicators to eliminate dimensional effects.

[0055] Perform feature engineering to extract features such as the relationship between tool life and cutting forces, and the correlation between material hardness and surface roughness.

[0056] Select the gradient boosting tree model, divide the data set into training set, validation set and test set, use the cross-validation strategy to train the model on the training set to avoid overfitting, and adjust the model parameters on the validation set to optimize the prediction performance.

[0057] When training the model, multi-objective optimization is considered, which not only pursues the precise control of the machining residual shape, but also maximizes the bonding strength and surface quality. It is necessary to adopt a multi-objective optimization algorithm such as NSGA-II or Pareto frontier analysis.

[0058] The test set is used to evaluate the generalization ability of the model, focusing on the prediction accuracy, recall, and the deviation between the prediction results and the actual measured values.

[0059] Through sensitivity analysis, we can identify the features that have the greatest impact on the prediction results, introduce new success or failure case data, and continuously optimize model performance.

[0060] For a new machining task, the material properties, expected machining residual shape requirements and bonding strength targets are input, and the model automatically predicts the optimal machining parameter combination, including but not limited to cutting speed, feed rate, tool type and wear tolerance.

[0061] Through the above-mentioned refinement steps, the technical solution of the present application can not only accurately predict the optimal combination of processing parameters under different conditions, but also can continue to evolve with the accumulation of time and data to adapt to a wider range of processing needs and material properties, and ultimately achieve the purpose of improving the processing quality and production efficiency of the flexible pressing device.

[0062] According to the prediction results of the machine learning model, the processing path is fine-tuned to ensure that the peak and trough shape of the processing residue remains consistent in all areas, especially in the key area of ​​the narrow end. Path planning is used to avoid local stress concentration and excessive cutting. The processing is started, and the sensor monitoring and adaptive control unit are turned on at the same time. The processing parameters are dynamically adjusted according to real-time data feedback to ensure the stability of the processing process. In the narrow end area, tool path smoothing, such as Bezier curve or spline curve interpolation, is used to make the path transition smoothly when entering and exiting the narrow area, reducing stress concentration caused by sharp changes in the path. At the same time, the corner processing in the tool path is optimized, and arc transition is used instead of right angles to reduce local stress peaks. Because its control points can be flexibly adjusted, the Bezier curve can create a smooth path that meets specific curvature requirements.

[0063] In order to achieve a smooth transition of the tool path, cubic Bezier curves or quintic spline curves are used for interpolation. The Bezier curve determines the shape of the curve through control points to ensure the continuity and smoothness of the curve at the starting point, end point and control points. The quintic spline curve is composed of quintic polynomials in each interval to ensure the continuity of the first and second order derivatives on the entire path, further reduce the change of acceleration, and reduce the impact during dynamic cutting.

[0064] Corner processing optimization: In path planning, when encountering a right-angle or sharp-angle turning point, arc transition is used instead of a right-angle turn. Specifically, by calculating the tangent direction of the path on both sides of the corner, determining the center position and radius, and generating one or more small-radius arcs to connect the two straight line segments, the continuity and smoothness of the tool trajectory during the cutting process are ensured, and the sudden change of tool load and workpiece stress concentration caused by sharp turns are reduced.

[0065] It should be noted that when training the machine learning model, in order to simultaneously pursue the precise control of the residual shape of the processing, maximize the bonding strength and surface quality, we use NSGA-II (non-dominated sorting genetic algorithm version 2) or multi-objective Pareto frontier analysis method for multi-objective optimization. The following are the detailed implementation steps of the process:

[0066] Objective function definition:

[0067] Objective 1: Minimize the machining residual shape error: Define an error function to measure the deviation between the actual machining residual shape and the ideal peak and trough shape. By calculating indicators such as the residual sum of squares or contour matching, the goal is to make this error as small as possible.

[0068] Objective 2: Maximize bonding strength: Based on material science principles and experimental data, a relationship model between bonding strength, processing parameters, and surface roughness is established. The goal is to improve the theoretically predicted bonding strength value by optimizing processing parameters.

[0069] Goal 3: Surface quality optimization: measured by surface roughness Ra value (arithmetic mean deviation) or other related indicators, the goal is to reduce surface roughness and improve surface finish.

[0070] Population initialization and encoding strategy: According to the range and constraints of the processing parameters, a population containing multiple individuals (i.e., processing parameter combinations) is initialized. Each individual represents a set of processing parameters (such as cutting speed, feed rate, tool type, etc.) through a specific encoding strategy (such as real number encoding or binary encoding).

[0071] Fitness evaluation: For each individual in the population, the scores of the three objectives are calculated using the established multi-objective evaluation system. This usually involves simulating the machining process or using historical data for approximate calculations to obtain the machining residual shape error, predicted bonding strength value and surface roughness index.

[0072] Non-dominated sorting and crowding distance calculation: The NSGA-II algorithm is used to perform non-dominated sorting on the population and divide the Pareto frontier into different levels. For individuals of the same level, the crowding distance indicator is used to further distinguish the quality of individuals, and those individuals with more dispersed distribution on the Pareto frontier are preferentially retained to promote diversity.

[0073] Selection, crossover and mutation: Based on the results of non-dominated sorting and crowding distance, excellent individuals are selected for reproduction, and a new generation of population is generated through the operation of genetic algorithms (such as roulette wheel selection, single-point crossover or simulated binary crossover, mutation).

[0074] Termination conditions and result processing: Repeat the above process until the predetermined algebraic or convergence criteria are reached. Finally, a solution is selected from the Pareto front solution set, which reflects the different trade-offs of processing parameters. Users can choose the appropriate combination of processing parameters according to specific needs (such as whether processing quality or production efficiency is more important).

[0075] Visualization and interpretation: Use the Pareto frontier diagram to show the trade-offs between different processing parameter combinations and help understand how to optimize specific objectives or comprehensively consider all objectives by adjusting processing parameters.

[0076] Through the above detailed steps, the multi-objective optimization algorithm not only ensures the precise control of the residual shape after machining, but also maximizes the bonding strength and surface quality, and provides scientific and reasonable machining parameter recommendations for the CNC machining of the flexible pressing device, effectively solving the limitations of the single parameter optimization in the traditional method, and further improving the overall efficiency of the machining process and product quality.

[0077] The technical solution proposed in this application - a flexible pressing CNC machining method based on data processing, realizes comprehensive intelligent management of the machining process by deeply integrating advanced sensing technology, real-time data analysis, adaptive control strategy and multi-objective optimization machine learning model, and significantly improves machining quality and production efficiency. The specific technical effects are reflected in the following aspects:

[0078] Improved machining accuracy and consistency: This method monitors key parameters such as cutting force, vibration frequency and temperature in real time, and uses adaptive control technology to dynamically adjust machining parameters, effectively avoiding abnormal situations during the machining process, ensuring that the residual shape of the peak and trough structure is highly consistent, and improving the geometric accuracy and dimensional stability of the product.

[0079] Optimization of production efficiency: Through integrated tool wear monitoring and dynamic adjustment strategies, not only the tool life is extended, but also high cutting efficiency is maintained by timely adjustment of processing parameters, reducing downtime caused by tool replacement and reprocessing, thereby significantly improving production efficiency.

[0080] Material utilization and cost savings: Accurate machining parameter prediction and adaptive path planning effectively reduce over-cutting and scrap rates, especially in critical areas such as narrow ends. Smooth tool path design reduces material waste and avoids workpiece damage caused by local stress concentration, which helps reduce production costs in the long run.

[0081] Enhanced process adaptability and flexibility: The multi-objective optimization machine learning model can automatically predict the optimal combination of processing parameters based on different material properties and processing requirements, enabling the technical solution to flexibly respond to the processing challenges of various complex workpieces and improve the adaptability and flexibility of the processing process.

[0082] Intelligent fault prevention and safety management: Real-time data processing algorithms and abnormality identification mechanisms can detect and respond to potential problems in the processing process in advance, such as sudden changes in cutting force, abnormal vibration or excessive temperature, effectively preventing equipment damage and safety accidents and ensuring production safety.

[0083] To sum up, the technical solution of the present application not only solves the difficulties of traditional CNC machining in the manufacturing of complex parts, especially flexible pressing devices, through intelligent and adaptive processing strategies, but also promotes the development of the manufacturing industry towards higher precision, higher efficiency and greater intelligence, bringing innovative technological progress to the field of precision manufacturing.

[0084] It should be noted that each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments. The device and system embodiments described above are merely schematic, in which the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.

Claims

1. A flexible pressing material numerical control processing method based on data processing, characterized in that: Included steps First, in CNC machine tools, torque sensors, vibration sensors, and temperature sensors are integrated to monitor the cutting force, vibration frequency, and tool or workpiece temperature during machining in real time; Establish real-time data processing algorithms to instantly analyze sensor feedback data and identify abnormalities during processing; Based on the real-time data analysis results, an adaptive control unit is built. The adaptive control unit dynamically adjusts the processing parameters according to the current processing status to ensure the stability and consistency of the processing process; Use historical processing data to train a machine learning model that includes the relationship between processing parameters, tool wear status, material characteristics, and processing residual shape, surface roughness, and bonding strength. Use the machine learning model to predict the best combination of processing parameters under different conditions. According to the prediction results of the machine learning model, the processing path is fine-tuned to ensure that the peak and trough shape of the residual processing remains consistent in all areas. In the narrow end area, path planning is used to avoid local stress concentration and excessive cutting. The processing is started, and the sensor monitoring and adaptive control unit are turned on at the same time. The processing parameters are dynamically adjusted according to real-time data feedback to ensure the stability of the processing process; Based on the real-time data analysis results, the adaptive control unit is constructed to: dynamically adjust the thresholds of cutting force, vibration and temperature according to the real-time monitored processing conditions; When the processing status is detected to deviate from the normal range, the adaptive control unit adjusts the processing parameters, or adjusts the speed, step distance, or stops the processing completely; The adaptive control unit also automatically adjusts machining parameters according to the degree of tool wear, increasing the rotation speed to maintain cutting efficiency in the early stage of tool wear, and reducing the cutting depth and feed rate in the later stage of tool wear; For areas with peak and trough machining residues, the tool path feed angle and sequence are dynamically adjusted; Using historical processing data to train machine learning models specifically includes: Select the gradient boosting tree model, divide the data set into training set, validation set and test set, and use the cross-validation strategy to train the model on the training set; for new processing tasks, input material properties, expected processing residual shape requirements and bonding strength targets, and the model automatically predicts the optimal processing parameter combination, including cutting speed, feed rate, tool type and wear tolerance.

2. The flexible pressing material numerical control processing method based on data processing according to claim 1 is characterized in that: Abnormal conditions during machining include sudden changes in cutting forces, abnormal vibration frequencies or excessive temperatures.

3. The flexible pressing material numerical control processing method based on data processing according to claim 1 is characterized in that: Establishing a real-time data processing algorithm includes: first, collecting raw data from high-precision torque sensors, vibration sensors, and intelligent temperature monitoring systems; filtering the data, as well as signal smoothing, and performing data normalization; By using Fourier transform, frequency domain features are extracted from the cutting force signal, and high-frequency components or energy surges that deviate from the normal cutting mode are identified as signs of sudden changes in cutting force. By performing fast Fourier transform (FFT) on the vibration signal, analyzing its spectral characteristics, and setting thresholds to identify vibration frequencies and amplitudes that exceed the normal range; A temperature rise model based on historical data is established to predict the temperature change trend under normal processing. When the actual measured value deviates significantly from the predicted value, an abnormal temperature alarm is issued.

4. The flexible pressing material numerical control processing method based on data processing according to claim 1 is characterized in that: In the narrow end area, path planning is used to avoid local stress concentration and over-cutting, and tool path smoothing, including Bezier curve or spline curve interpolation, is used to make the path transition smoothly when entering and exiting the narrow area.

5. The flexible pressing material numerical control processing method based on data processing according to claim 4 is characterized in that: The Bezier curve or spline curve interpolation specifically adopts cubic Bezier curve or quintic spline curve interpolation. The Bezier curve determines the shape of the curve through control points. The quintic spline curve is composed of quintic polynomials in each interval, which ensures the continuity of the first-order and second-order derivatives on the entire path, further reducing the change of acceleration.

Images