Massager metal plate milling device
By constructing a CNC milling system that integrates multi-source sensor information fusion and adaptive adjustment, the problem of real-time monitoring and adaptive adjustment in existing technologies has been solved, achieving high-precision and stable machining of massager sheet metal parts.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGXI HEQIKANG ELECTRONICS CO LTD
- Filing Date
- 2026-03-20
- Publication Date
- 2026-04-28
AI Technical Summary
Existing CNC milling equipment cannot monitor and respond to multi-source information during the machining process in real time, resulting in excessive surface roughness and dimensional tolerances exceeding the allowable range, affecting product assembly accuracy and service life. Furthermore, it lacks adaptive adjustment capabilities, which can easily lead to equipment damage.
By introducing a workpiece condition assessment module, a cutting dynamics condition assessment module, and a machining accuracy confidence assessment module, a multi-source sensor information fusion and comprehensive quantitative assessment system is constructed to realize real-time monitoring of data such as suction fluctuation, workpiece temperature rise, vibration intensity, and cutting power, and to achieve adaptive adjustment through milling cutter speed adjustment and cutting depth adjustment.
It significantly improves the stability and precision of the processing, reduces scrap rate and safety risks, and increases production efficiency.
Smart Images
Figure CN121928403A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of CNC machine tool technology, and in particular relates to a sheet metal milling device for massagers. Background Technology
[0002] As a core component of daily health care equipment, massagers require precision milling to achieve complex shapes, accurate mounting holes, and highly flat assembly surfaces for their sheet metal components such as the shell and frame. These sheet metal parts generally have the physical characteristics of thinness, weak structural rigidity, and poor resistance to deformation. During milling, they are highly susceptible to high-frequency vibrations caused by cutting force disturbances. Furthermore, the vacuum adsorption clamping method is highly sensitive to air pressure stability; even slight fluctuations in adsorption force can cause workpiece displacement during processing. The continuous accumulation of cutting heat leads to a further increase in workpiece temperature, exacerbating material thermal deformation. Meanwhile, fluctuations in spindle load and speed tracking errors can easily induce cutting chatter. These factors interact, resulting in excessive surface roughness and dimensional tolerances exceeding allowable limits, severely impacting product assembly accuracy and lifespan.
[0003] While current mainstream CNC milling machines are equipped with a basic machining table, protective housing, milling spindle, and vacuum adsorption platform, their control systems can only execute preset fixed process parameters. They cannot perform multi-source information fusion and quantitative evaluation of the workpiece's evolving state (including adsorption force fluctuations, temperature trends, and vibration intensity of the mounting surface) and cutting dynamics (covering unit volume cutting power fluctuations, spindle load stability, and chatter spectrum characteristics) during machining. When the machining condition deteriorates, the machine lacks adaptive adjustment capabilities based on state assessment; it cannot dynamically correct the milling cutter speed to suppress chatter, nor can it adjust the cutting depth in a timely manner to compensate for workpiece deformation. More importantly, under abnormal conditions such as workpiece loosening or severe chatter, existing systems lack a graded response mechanism. Surface defects or dimensional deviations are often only discovered during post-machining quality inspection, leading to increased material scrap rates, extended production cycles, and even equipment damage risks due to continuous abnormal operation.
[0004] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention
[0005] The purpose of this invention is to provide a sheet metal milling device for massagers, which aims to solve the above-mentioned problems.
[0006] This invention is implemented as follows: a sheet metal milling device for a massager includes a processing table, a housing on the processing table, a milling machine inside the housing, an xy-axis drive assembly for driving the milling machine horizontally between the milling machine and the housing, and a vacuum adsorption platform below the milling machine. The device further includes a control system, which comprises: The workpiece condition assessment module has its input terminal electrically connected to a sensor group set on the vacuum adsorption platform and the workpiece mounting surface to receive the suction fluctuation signal of the vacuum adsorption platform, the workpiece temperature signal, and the vibration signal of the workpiece mounting surface in real time. Based on the processed standard deviation of suction fluctuation, the maximum temperature rise of the workpiece, and the vibration intensity of the workpiece mounting surface, the workpiece condition assessment model is constructed and the workpiece condition index is output. The cutting dynamics state assessment module is electrically connected to the spindle driver and spindle sensor of the milling machine to receive the cutting power signal, spindle load signal and spindle speed signal of the milling machine in real time. Based on the processed unit volume cutting power, spindle load fluctuation standard deviation and spindle speed following error, the cutting dynamics state assessment model is constructed and the dynamics state index is output. The machining accuracy confidence assessment module has its input terminal electrically connected to a vibration sensor and an acoustic emission sensor installed on the milling machine spindle or workpiece to receive spindle vibration signals and acoustic emission signals in real time. Based on the processed chatter main frequency band energy ratio, spindle vibration acceleration RMS value, and acoustic emission characteristic frequency band energy, a machining accuracy confidence assessment model is constructed, and the machining accuracy confidence is output. A milling cutter speed adjustment module is provided, wherein the input end of the milling accuracy confidence assessment module is connected to the output end of the machining accuracy confidence assessment module, and the output end is connected to the spindle driver control end of the milling machine. The milling cutter speed adjustment module is used to construct a spindle speed adjustment model based on the reference spindle speed, machining accuracy confidence, chatter main frequency band energy ratio, and chatter main frequency value when the machining accuracy confidence is lower than a preset threshold, and outputs the milling cutter speed adjustment amount to the spindle driver to control the milling cutter speed adjustment. A depth-of-cut adjustment module is provided, with its input terminals connected to the output terminals of the workpiece state evaluation module, the cutting dynamics state evaluation module, and the machining accuracy confidence evaluation module, respectively. Its output terminal is connected to the control terminal of the xy-axis drive assembly or the Z-axis drive mechanism of the milling machine. The depth-of-cut adjustment module is used to construct a depth-of-cut adjustment model based on the workpiece thickness, dynamics state index, workpiece state index, and reference depth of cut when the machining accuracy confidence is still lower than a preset threshold after the milling cutter speed is actively adjusted. The module then outputs a depth-of-cut adjustment amount to control the adjustment of the milling cutter's cutting depth. The equipment alarm system has its input terminal connected to the output terminal of the machining accuracy confidence assessment module, and its output terminal connected to the main control power supply or emergency stop circuit of the device. The equipment alarm system is used to respond and execute a shutdown operation when the machining accuracy confidence is still lower than a preset threshold after the cutting depth is adjusted.
[0007] Further technical solutions, in the workpiece condition assessment model: The normalized suction fluctuation standard deviation index, workpiece maximum temperature rise index, and mounting surface vibration intensity index are multiplied by their corresponding positive influence coefficients and summed. Then, the summation result is subtracted from 1 to obtain the workpiece condition index. Among them, the suction fluctuation standard deviation index, the workpiece maximum temperature rise index, and the installation surface vibration intensity index are obtained by substituting the real-time vacuum adsorption platform's suction fluctuation standard deviation, workpiece maximum temperature rise, and workpiece installation surface vibration intensity into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0 and less than 1, and the sum of the three is 1.
[0008] Further technical solutions, in the cutting dynamics state evaluation model: The normalized cutting power index is directly multiplied by the corresponding preset first influence coefficient; The normalized spindle load fluctuation standard deviation index is transformed by the natural logarithm of (1 plus the index) and then multiplied by the corresponding preset second influence coefficient. The normalized spindle speed following error index is directly multiplied by the corresponding preset third influence coefficient; Summing the product of the above three terms and then subtracting the summation from 1 yields the dynamic state index; Among them, the cutting power index, the spindle load fluctuation standard deviation index, and the spindle speed following error index are obtained by substituting the real-time milling machine's unit volume cutting power, spindle load fluctuation standard deviation, and spindle speed following error into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0 and less than 1, and the sum of the three is 1.
[0009] Further technical solutions, in the confidence evaluation model for machining accuracy: The normalized flutter main frequency band energy index, spindle vibration acceleration index and acoustic emission characteristic frequency band energy index are multiplied by their corresponding positive influence coefficients and summed. Then, the negative exponent of the summation result is taken to obtain the machining accuracy confidence level. Among them, the flutter main frequency band energy index, the spindle vibration acceleration index, and the acoustic emission characteristic frequency band energy index are obtained by substituting the flutter main frequency band energy ratio, the spindle vibration acceleration RMS value, and the acoustic emission characteristic frequency band energy into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0, and the sum of the three is 1.
[0010] Further technical solutions, in the spindle speed adjustment model: First, calculate the ratio of the chatter frequency to the frequency per revolution of the reference spindle speed to obtain the relative chatter frequency ratio; Secondly, a continuous adjustment direction function is constructed using the difference between this ratio and its rounded-to-integer value; Then, the product of the preset direction sensitivity coefficient and the continuous adjustment direction function is used as the input of the hyperbolic tangent function to obtain the basic quantities of the speed adjustment direction and amplitude; Finally, multiply this basic quantity by the adjustment ratio coefficient, the reference spindle speed, the deviation of the machining accuracy confidence level from 1, and the chatter frequency band energy index to obtain the milling cutter speed adjustment amount.
[0011] A further technical solution, in the depth-of-cut adjustment model: Multiply the dynamic state index, workpiece state index, and machining accuracy confidence level by their respective deviations from 1, take the cube root of the product, multiply it by the reference cutting depth, and take the negative value to obtain the cutting depth adjustment amount.
[0012] A further technical solution is to use the maximum-minimum normalization formula as: (input value - minimum value) / (maximum value - minimum value), where the maximum or minimum value is determined by the system's historical data or theoretical safety threshold, ensuring that the normalized calculation result is between 0 and 1.
[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention provides a sheet metal milling device for massagers. By introducing a workpiece condition assessment module, a cutting dynamics condition assessment module, and a machining accuracy confidence assessment module, it achieves multi-source sensor information fusion and comprehensive quantitative evaluation of the machining process. The system can simultaneously monitor data from multiple dimensions, including suction fluctuations, workpiece temperature rise, vibration intensity, cutting power, spindle load fluctuations, speed following error, chatter energy, spindle vibration, and acoustic emission, and convert these into quantifiable condition indices and confidence levels. This comprehensive real-time evaluation capability is generally lacking in existing technologies, enabling the system to accurately identify the deterioration trend of the machining condition, rather than relying solely on a single or lagging indicator.
[0014] This invention provides a sheet metal milling device for massagers, which constructs a graded, progressive adaptive adjustment mechanism. When the machining accuracy confidence level first decreases, the milling cutter speed adjustment module can actively adjust the milling cutter speed based on key information such as the chatter frequency to optimize cutting conditions and suppress chatter. This contrasts sharply with the limitations of traditional solutions that have fixed parameters and cannot dynamically respond to chatter. When speed adjustment is still insufficient to restore accuracy, the depth-of-cut adjustment module further intervenes, adjusting the depth of cut to reduce the cutting load and thus stabilize the machining process. This two-stage adjustment strategy enables the system to take more refined and effective intervention measures according to the severity and nature of the problem, significantly improving the stability and adaptability of the machining process.
[0015] The present invention provides a sheet metal milling device for massagers. By constructing a set of sheet metal milling devices for massagers that integrates real-time status assessment, hierarchical adaptive adjustment and intelligent alarm, it effectively solves the problems of existing technologies that cannot perceive the processing status in real time, lack adaptive adjustment capabilities, and lack hierarchical intervention and automatic alarm functions. It significantly improves the processing accuracy, stability and production efficiency of sheet metal parts for massagers, and reduces scrap rate and safety risks. Attached Figure Description
[0016] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a schematic diagram of the structure in this invention; Figure 3 This is a schematic diagram illustrating the interaction between the various functional modules in this invention.
[0017] In the attached diagram: 1. Machining table; 2. Housing; 3. Milling machine; 4. Vacuum adsorption platform; 5. XY axis drive assembly. Detailed Implementation
[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.
[0019] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.
[0020] like Figures 1-3As shown, a sheet metal milling device for a massager provided in one embodiment of the present invention includes a processing table (1), a housing (2) disposed on the processing table (1), a milling machine (3) disposed in the housing (2), an xy-axis drive assembly (5) for driving the milling machine (3) to move horizontally between the milling machine (3) and the housing (2), and a vacuum adsorption platform (4) disposed below the milling machine (3). The device is characterized in that it further includes a control system, the control system comprising: The workpiece status assessment module is electrically connected to the sensor group set on the vacuum adsorption platform (4) and the workpiece mounting surface to receive the suction fluctuation signal, workpiece temperature signal and workpiece mounting surface vibration signal of the vacuum adsorption platform (4) in real time, and constructs a workpiece status assessment model based on the processed suction fluctuation standard deviation, maximum temperature rise of the workpiece and vibration intensity of the workpiece mounting surface, and outputs the workpiece status index. The cutting dynamics state evaluation module is electrically connected to the spindle driver and spindle sensor of the milling machine (3) to receive the cutting power signal, spindle load signal and spindle speed signal of the milling machine (3) in real time, and constructs a cutting dynamics state evaluation model based on the processed unit volume cutting power, spindle load fluctuation standard deviation and spindle speed following error, and outputs the dynamics state index. The machining accuracy confidence assessment module is electrically connected to the vibration sensor and acoustic emission sensor set on the spindle or workpiece of the milling machine (3) to receive the spindle vibration signal and acoustic emission signal in real time, and construct the machining accuracy confidence assessment model based on the processed chatter main frequency band energy ratio, spindle vibration acceleration RMS value and acoustic emission characteristic frequency band energy, and output the machining accuracy confidence. The milling cutter speed adjustment module has its input end connected to the output end of the machining accuracy confidence assessment module, and its output end connected to the spindle driver control end of the milling machine (3). The milling cutter speed adjustment module is used to construct a spindle speed adjustment model based on the reference spindle speed, machining accuracy confidence, chatter main frequency band energy ratio and chatter main frequency value when the machining accuracy confidence is lower than a preset threshold, and output the milling cutter speed adjustment amount to the spindle driver to control the milling cutter speed adjustment. The depth of cut adjustment module has its input end connected to the output end of the workpiece state evaluation module, the cutting dynamics state evaluation module, and the machining accuracy confidence evaluation module, respectively, and its output end connected to the control end of the Z-axis drive mechanism of the xy-axis drive assembly (5) or the milling machine (3). The depth of cut adjustment module is used to construct a depth of cut adjustment model based on the workpiece thickness, dynamics state index, workpiece state index, and reference depth of cut when the machining accuracy confidence is still lower than a preset threshold after the milling cutter speed is actively adjusted, and outputs the depth of cut adjustment amount to control the milling cutter cutting depth to be adjusted. The equipment alarm system has its input terminal connected to the output terminal of the machining accuracy confidence assessment module, and its output terminal connected to the main control power supply or emergency stop circuit of the device. The equipment alarm system is used to respond and execute a shutdown operation when the machining accuracy confidence is still lower than a preset threshold after the cutting depth is adjusted.
[0021] In this embodiment, the machining table 1 serves as the basic support structure of the entire machining device, bearing and fixing other components to ensure the stability of the machining process. The outer casing 2 is mounted on the machining table 1 to house and protect core machining components such as the milling machine 3, while also providing a certain level of safety protection. The milling machine 3 is the core component that performs milling operations, removing material from the workpiece using a high-speed rotating milling cutter. The xy-axis drive assembly 5 is configured to precisely control the movement of the milling machine 3 in the horizontal plane, thereby enabling the machining of complex contours of the workpiece. The vacuum suction platform 4 is positioned below the milling machine 3, generating vacuum suction to firmly fix the sheet metal workpiece in the machining position, preventing displacement or vibration of the workpiece during cutting.
[0022] The workpiece condition assessment module is designed to monitor and quantify the physical state of the workpiece in real time during the processing, such as its fixation stability, degree of thermal deformation, and vibration.
[0023] The cutting dynamics state assessment module is designed to monitor and quantify dynamic characteristics during milling processes in real time, such as changes in cutting forces, fluctuations in spindle load, and stability of rotational speed.
[0024] The machining accuracy confidence assessment module is designed to evaluate the reliability of the accuracy level achievable in the current machining process in real time, and provides a quantitative judgment by analyzing a variety of accuracy-related indicators.
[0025] The milling cutter speed adjustment module is designed to dynamically adjust the spindle speed of the milling machine 3 based on the evaluation results of the machining conditions, in order to optimize cutting conditions and suppress undesirable phenomena.
[0026] The depth of cut adjustment module is designed to further adjust the depth of cut of the milling machine 3 when the speed adjustment is insufficient to improve the machining condition, so as to reduce the cutting load and stabilize the machining process.
[0027] Equipment alarm systems are designed to issue timely alarms and execute shutdown operations when the processing condition deteriorates severely and cannot be restored by adjustment measures, in order to prevent equipment damage or safety accidents.
[0028] The workpiece condition assessment module is configured to monitor the stability of the workpiece in real time during the milling process. Specifically, this module can collect suction fluctuation data from the vacuum adsorption platform 4, workpiece temperature data, and vibration data from the workpiece mounting surface. For example, multiple independent sensors can be set to measure suction, temperature, and vibration separately, with a safety range set for each parameter. When any parameter exceeds its preset range, the system will issue a warning. Alternatively, a simple weighted averaging method can be used to linearly combine these raw measurements to generate a preliminary workpiece condition index.
[0029] The cutting dynamics state assessment module is configured to monitor the dynamic characteristics of the milling process in real time. This module can acquire the unit volume cutting power of the milling machine 3, the fluctuation of the spindle load, and the spindle speed tracking error. For example, by using a pre-established empirical rule base, different combinations of cutting power, load fluctuations, and speed errors can be mapped to different dynamic state levels to determine the stability of the current cutting process. Alternatively, a simple arithmetic average of these parameters can be performed to obtain a rough dynamic state assessment value.
[0030] The machining accuracy confidence assessment module is configured to evaluate the reliability of the achievable accuracy level of the current machining process in real time. This module can collect the energy ratio of the main frequency band of chatter, the RMS value of the spindle vibration acceleration, and the energy of the acoustic emission characteristic frequency band. For example, a threshold-based judgment mechanism can be used, where the machining accuracy confidence is considered to have decreased when any of the chatter energy, spindle vibration, or acoustic emission energy exceeds its respective critical value. Alternatively, a simple linear superposition method can be used to directly add these normalized indicators to obtain a preliminary confidence score.
[0031] The milling cutter speed adjustment module is configured to actively adjust the milling cutter speed when the machining accuracy confidence level is below a preset threshold. For example, a proportional control-based strategy can be used to adjust the spindle speed proportionally to the difference between the machining accuracy confidence level and the preset threshold. When the confidence level decreases, the speed is reduced accordingly; when the confidence level increases, the speed is increased accordingly. Alternatively, several fixed speed adjustment levels can be preset, and the corresponding level can be selected for speed adjustment when the confidence level falls within different decreasing ranges.
[0032] The depth-of-cut adjustment module is configured to further adjust the depth of cutter when the machining accuracy confidence level fails to recover above a preset threshold after active adjustment of the cutter speed. For example, a simple fixed-step adjustment method can be used, decreasing the depth of cut by a preset fixed amount each time until the machining accuracy confidence level recovers. Alternatively, a suitable depth-of-cut adjustment amount can be determined by consulting a preset experience table based on the workpiece thickness, the current dynamic state index, and the workpiece state index.
[0033] The equipment alarm system is configured to perform a shutdown operation when the milling cutter speed and depth of cut have been adjusted, but the machining accuracy confidence level is still lower than a preset threshold. For example, a simple delay mechanism can be set up so that the system automatically triggers a shutdown after the machining accuracy confidence level remains below the threshold for a certain period of time. Alternatively, a manual emergency stop button can be configured for operator intervention upon observing a serious abnormality.
[0034] In a preferred embodiment of the present invention, the workpiece condition evaluation model includes: ; in This is the coefficient for the influence of suction fluctuation. The coefficient representing the influence of workpiece temperature rise. The vibration influence coefficient of the mounting surface. , , , ; The standard deviation index of suction fluctuation. The maximum temperature rise index of the workpiece. The vibration intensity index of the mounting surface. This refers to the workpiece condition index; , as well as The method of obtaining the suction fluctuation standard deviation, the maximum temperature rise of the workpiece and the vibration intensity of the workpiece mounting surface of the vacuum adsorption platform (4) are successively substituted into the maximum-minimum value normalization formula for normalization processing, and the suction fluctuation standard deviation index, the maximum temperature rise index of the workpiece and the vibration intensity index of the mounting surface are obtained in sequence.
[0035] In this embodiment, the workpiece condition assessment model is a mathematical model used to quantitatively assess the stability and health status of a workpiece during processing. This model can be deployed in the industrial control computer of the massager sheet metal milling machine, running as a software module to receive sensor data and perform calculations in real time. Alternatively, the model can be implemented using dedicated hardware circuitry, such as integration into a programmable logic controller or digital signal processor, to achieve faster response times and higher computational efficiency.
[0036] Suction fluctuation influence coefficient Workpiece temperature rise influence coefficient and the vibration influence coefficient of the mounting surface These are weighted parameters used to measure the degree of influence of different factors on the workpiece's condition. These coefficients can be determined through statistical analysis and machine learning training on a large amount of historical processing data, such as learning the relationship between each factor and the final processing quality through regression analysis or neural network models. Alternatively, these coefficients can be set based on expert experience and material properties, through experimental verification and iterative optimization.
[0037] Suction fluctuation standard deviation index Maximum temperature rise index of the workpiece Vibration intensity index of the mounting surface These are the results after standardizing the raw sensor data. Obtaining these indices requires real-time data acquisition using pressure sensors mounted on the vacuum adsorption platform 4, thermocouples or infrared temperature sensors mounted on or near the workpiece, and acceleration sensors mounted on the workpiece mounting surface. This raw data is then fed into the processing unit for calculation to obtain the corresponding standard deviation, maximum temperature rise, and vibration intensity.
[0038] Workpiece condition index This is the final output of the workpiece condition assessment model. It is a comprehensive numerical value that reflects the overall stability of the current workpiece. This index can be used as a direct input to subsequent decision-making modules, such as triggering milling cutter speed adjustment, depth of cut adjustment, or equipment alarm systems.
[0039] The solution proposed in this application addresses the problem of accurately quantifying workpiece condition by constructing a specific workpiece condition assessment model. This model integrates three key influencing factors: the standard deviation of suction fluctuation in the vacuum adsorption platform 4, the maximum temperature rise of the workpiece, and the vibration intensity of the workpiece mounting surface, and outputs a unified workpiece condition index. During operation, the system collects raw data in real time, including the standard deviation of suction fluctuations on the vacuum adsorption platform 4, the maximum temperature rise of the workpiece, and the vibration intensity of the workpiece mounting surface. This raw data is then sent to the processing unit and standardized using a maximum-minimum normalization formula to obtain the standard deviation index of the suction fluctuations. Maximum temperature rise index of the workpiece Vibration intensity index of the mounting surface This normalization process ensures that data with different physical dimensions can be compared and calculated uniformly. Then, these normalized exponents are compared with the preset suction fluctuation influence coefficient. Workpiece temperature rise influence coefficient and the vibration influence coefficient of the mounting surface These values are then substituted into the workpiece condition assessment model formula for calculation. Since the thermal deformation of the workpiece material and the temperature rise typically exhibit a non-linear relationship... The square root transform can more accurately map the negative impact of thermal deformation on the workpiece condition; the destructive effect of vibration intensity on processing stability increases sharply with its amplitude. Using square operations can effectively amplify the deterioration trend under high vibration intensity, making the assessment results more sensitive.
[0040] in, , , The sum of all factors is 1, and all are within the range (0,1). This allows the model to flexibly adjust according to the importance of each factor's influence on the workpiece state, avoiding the excessive dominance of a single factor. Ultimately, the model outputs a workpiece state index. It can accurately and in real time reflect the overall stability of the workpiece. When the workpiece condition index... A value close to 1 indicates a good workpiece condition; a decrease in this value indicates a deteriorating workpiece condition. This quantitative evaluation mechanism provides the massager sheet metal milling device with reliable workpiece condition perception capabilities, enabling it to provide a solid data foundation for subsequent cutting dynamics condition assessment, machining accuracy confidence assessment, and adaptive control strategies such as milling cutter speed adjustment and depth of cut adjustment based on accurate workpiece condition information. This effectively addresses problems such as unstable workpiece clamping and accumulated thermal deformation that are prone to occur during sheet metal processing.
[0041] In a preferred embodiment of the present invention, the cutting dynamics state evaluation model includes: ; in The cutting power influence coefficient is... The load fluctuation impact coefficient. The influence coefficient of spindle speed. , , , ; This refers to the cutting power index. The standard deviation index of spindle load fluctuation. The spindle speed following error index. It is the dynamic state index; , as well as The method of obtaining is as follows: the unit volume cutting power, spindle load fluctuation standard deviation and spindle speed following error of the real-time milling machine (3) are successively substituted into the maximum-minimum value normalization formula for normalization processing, and the cutting power index, spindle load fluctuation standard deviation index and spindle speed following error index are obtained in sequence.
[0042] In this embodiment, the cutting dynamics state assessment model aims to quantitatively evaluate the stability and health of cutting dynamics during milling. It provides a unified dynamics state index by comprehensively considering multiple key parameters to facilitate subsequent decision-making and adjustments. This model can be integrated into the device's control system, for example, through an embedded processor or industrial PC.
[0043] The above mathematical expression defines the dynamic state index. The calculation method uses a weighted summation to calculate the normalized cutting power index. Spindle load fluctuation standard deviation index (After logarithmic processing) and spindle speed following error index To synthesize. The subtraction operation and the introduction of 1 make... The value can intuitively reflect the "health" of the dynamic state; the higher the value, the better the state.
[0044] in, The cutting power influence coefficient is... The load fluctuation impact coefficient. These are the spindle speed influence coefficients, used to adjust the relative importance or weight of each input parameter in the dynamic state evaluation model. They can be determined based on empirical knowledge, expert systems, or through machine learning algorithms (e.g., trained based on historical machining data and machining quality feedback). For example, in some machining scenarios, the stability of cutting power may be more critical than the spindle speed following error. Higher values can be assigned to these coefficients. These coefficients are limited to between 0 and 1 and sum to 1, ensuring the reasonableness of the influence coefficients and the stability of the model. This allows the model to perform effective weighted averaging, avoiding an excessive or unreasonable impact of a single parameter on the overall evaluation results, thus guaranteeing the objectivity and interpretability of the evaluation results.
[0045] This refers to the cutting power index. The standard deviation index of spindle load fluctuation. These are the spindle speed following error exponents. These exponents are dimensionless values of the original physical quantities after standardization, used to unify the scale of different physical quantities in the model. Cutting power exponent. The standard deviation index of spindle load fluctuation reflects the energy required for cutting per unit volume. The stability of the spindle load was quantified, while the spindle speed following error index was... This indicates the degree of deviation between the actual spindle speed and the set speed.
[0046] The dynamic state index is the final output of the cutting dynamic state evaluation model. Its value is typically between 0 and 1, used to visually represent the quality of the current cutting dynamic state. For example, a value close to 1 indicates a good dynamic state, while a value close to 0 indicates a poor dynamic state or potential problems.
[0047] , as well as The data is obtained by normalizing the raw data acquired in real time. During actual machining, a power sensor mounted on the milling machine's spindle monitors the cutting power per unit volume in real time, and a spindle speed and load information are acquired via a spindle encoder or current sensor. This allows for the calculation of the spindle load fluctuation standard deviation and the spindle speed following error. The purpose of normalization is to eliminate dimensional and numerical range differences between different physical quantities, making them comparable within the model.
[0048] This application proposes a sophisticated cutting dynamics state evaluation model to quantitatively assess the dynamic performance of the milling machine 3 in the sheet metal milling device for massagers. The core of this model lies in transforming three key parameters—real-time monitored cutting power per unit volume, spindle load fluctuation standard deviation, and spindle speed following error—into a unified dynamics state index using rigorous mathematical methods. First, the raw data of the milling machine 3's unit volume cutting power, spindle load fluctuation standard deviation, and spindle speed following error are collected in real time. To eliminate the inherent dimensional and numerical range differences between these physical quantities, these raw data are sequentially substituted into the maximum-minimum normalization formula for processing, thereby obtaining the dimensionless cutting power index. Spindle load fluctuation standard deviation index and spindle speed following error index Among them, the standard deviation index of spindle load fluctuation. Performing logarithmic processing effectively smooths out fluctuations and enhances the model's robustness to outliers. These normalized exponents are then input into a pre-defined mathematical model.
[0049] In this model, the cutting power influence coefficient Load fluctuation impact coefficient Influence coefficient of spindle speed As weighting factors, these coefficients adjust the degree of influence of each index. These coefficients are limited to between 0 and 1 and sum to 1, ensuring that the contribution ratio of each factor in the model is reasonable and adjustable, avoiding the excessive dominance of a single factor. Ultimately, the model outputs a dynamic state index. The closer the value is to 1, the more stable and healthy the cutting dynamics; conversely, a value closer to 1 indicates potential dynamic problems. This comprehensive evaluation mechanism enables the device to accurately and reliably sense the dynamic changes in the cutting process, providing precise decision-making basis for subsequent milling cutter speed adjustment, cutting depth adjustment, and even equipment alarm systems. This effectively solves the problems of inaccurate parameter quantification and unreasonable weight allocation in traditional evaluation methods, significantly improving the stability and adaptability of the machining process.
[0050] In a preferred embodiment of the present invention, the machining accuracy confidence evaluation model includes: ; in This is the flutter effect coefficient. The influence coefficient of spindle vibration. The characteristic influence coefficient of acoustic emission. , , , ; The energy index of the flutter main frequency band. The main shaft vibration acceleration index, The energy index of the characteristic frequency band of acoustic emission. For the confidence level of machining accuracy; , as well as The method for obtaining the flutter main frequency band energy ratio, spindle vibration acceleration RMS value and acoustic emission characteristic frequency band energy is as follows: substitute the flutter main frequency band energy ratio, spindle vibration acceleration RMS value and acoustic emission characteristic frequency band energy into the maximum-minimum value normalization formula for normalization processing, and obtain the flutter main frequency band energy index, spindle vibration acceleration index and acoustic emission characteristic frequency band energy index in sequence.
[0051] In this embodiment, the machining accuracy confidence assessment model aims to quantify the reliability of the massager sheet metal parts reaching a preset accuracy standard under the current machining state. This model can be based on machine learning methods, trained on a large amount of historical machining data (including various sensor data and final product accuracy test results) to predict the current machining accuracy confidence level; alternatively, the model can be based on expert systems or fuzzy logic, using a series of preset rules and membership functions to comprehensively judge the machining accuracy confidence level. This model provides a unified, quantitative indicator for the entire machining process, transforming multi-source heterogeneous sensor data into an understandable measure of machining quality reliability.
[0052] Exponential function model The purpose of this model is to weight and sum multiple factors affecting machining accuracy (such as chatter, spindle vibration, and acoustic emission), and then map them to a confidence interval of 0-1 using an exponential function. This mathematical model can be computed in software using a control program running on a programmable logic controller (PLC) or an industrial personal computer, or it can be computed in high-speed parallel using a dedicated hardware accelerator (such as a field-programmable gate array (FPGA)). Its exponential decay characteristic ensures that a significant increase in any negative factor will lead to a rapid decrease in confidence, consistent with the trend of accuracy degradation in actual machining.
[0053] , , These coefficients represent the relative importance or degree of influence of chatter, spindle vibration, and acoustic emission characteristics on the confidence level of machining accuracy. Their weights can be determined through experimental data analysis using statistical methods such as regression analysis and principal component analysis; alternatively, they can be initially set based on expert experience or process knowledge, and then verified and optimized through actual machining. These coefficients ensure that the influence of different factors on machining accuracy can be reasonably quantified and weighted, reflecting their importance in actual machining.
[0054] Flutter main frequency band energy index This index quantifies the energy intensity of flutter within a specific frequency range, reflecting its severity. It can be obtained by performing a Fourier transform on signals acquired by accelerometers or force sensors, analyzing their spectrum, extracting the energy of the dominant flutter frequency band, and then calculating the flutter dominant frequency band energy index. Alternatively, time-frequency analysis methods such as wavelet transform can be used to locate the flutter components in the time-frequency domain and calculate their energy to obtain the flutter dominant frequency band energy ratio. As a key input for assessing the confidence level of machining accuracy, this index directly reflects the impact of flutter on surface quality and dimensional accuracy during machining.
[0055] Spindle vibration acceleration index This index is the root mean square (RMS) value of the spindle vibration acceleration, reflecting the severity of the overall spindle vibration. It can be obtained by collecting vibration signals in real time using an accelerometer mounted on the milling machine spindle or spindle box and calculating its RMS value; alternatively, it can be obtained by measuring spindle runout using a non-contact displacement sensor and indirectly estimating the vibration acceleration. This index reflects the stability of the spindle system itself. Excessive spindle vibration directly affects the relative motion accuracy between the tool and the workpiece, thus impacting machining accuracy.
[0056] Acoustic emission characteristic band energy index This index refers to the acoustic emission signal energy generated by phenomena such as microcrack propagation, friction, and plastic deformation during the cutting process within a specific frequency range. It can be obtained by acquiring high-frequency elastic wave signals using an acoustic emission sensor (usually a piezoelectric sensor), followed by filtering, amplification, and energy calculation; alternatively, it can be obtained by using an ultrasonic sensor to detect the acoustic characteristics of the cutting area and extract the energy of the characteristic frequency band. This index provides a sensitive indicator of the microscopic state of the cutting process, enabling early warning of problems such as tool wear and abnormal chip formation, all of which can affect machining accuracy.
[0057] The maximum-minimum normalization formula aims to scale raw data of different dimensions and ranges (such as flutter main frequency band energy ratio, principal shaft vibration acceleration RMS value, and acoustic emission characteristic frequency band energy) to a dimensionless range of 0-1. This formula can be calculated programmatically in the data preprocessing module to normalize real-time sensor data; alternatively, it can be converted into a normalization exponent using a lookup table or a preset mapping function. This normalization method ensures that different physical quantities can be effectively weighted and compared, avoiding the influence of dimensional differences on the evaluation results and improving the robustness and accuracy of the model.
[0058] The solution in this application uses a machining accuracy confidence assessment module to obtain in real time the chatter dominant frequency band energy ratio, spindle vibration acceleration RMS value, and acoustic emission characteristic frequency band energy of the sheet metal milling device for massagers. These raw data are first processed using a maximum-minimum normalization formula to convert them into a dimensionless chatter dominant frequency band energy index. Spindle vibration acceleration index Harmony emission characteristic band energy index This normalization process ensures that data from different sources and with different dimensions can be standardized to a uniform range, facilitating subsequent weighted calculations. These normalized indices are then substituted into the processing accuracy confidence assessment model, which employs an exponential function. To calculate the confidence level of machining accuracy Influence coefficients in the model , , The indices are weighted, and their sum equals 1, ensuring a reasonable allocation of weights for each factor. The exponential decay mechanism means that a significant increase in any negative factor will lead to a decrease in confidence level. The rapid decline in accuracy allows for a sensitive reflection of the deterioration trend in machining precision. Ultimately, the module outputs a quantified confidence level of machining precision. This provides a basis for decision-making in subsequent milling cutter speed adjustment and depth-of-cut adjustment modules. This detailed mathematical model and parameter normalization process enable the massager sheet metal milling device to accurately and comprehensively quantify the processing status, effectively distinguishing the coupled influence of different factors on processing quality, thus solving the problem of inaccurate or incomplete evaluation results in traditional methods.
[0059] In a preferred embodiment of the present invention, the spindle speed adjustment model includes: , ; ; in The relative frequency ratio of flutter. Dimensionless The dominant flutter frequency, The reference spindle speed, The direction function is continuously adjusted. For input values Rounding to the nearest integer; To adjust the proportional coefficient, , The direction sensitivity coefficient, , To determine the confidence level for machining accuracy. The energy index of the flutter main frequency band is tanh( () is the hyperbolic tangent function. This is the milling cutter speed adjustment amount.
[0060] In this embodiment, the flutter relative frequency ratio It is the flutter main frequency With spindle frequency The ratio of 60 to 1 / 60 quantifies the dynamic relationship between chatter and spindle speed, which is crucial for identifying the chatter generation mechanism and selecting appropriate suppression strategies. Chatter dominant frequency Vibration signals from the 3rd spindle of the milling machine or the workpiece mounting surface can be obtained in real time through spectrum analysis.
[0061] Reference spindle speed It can be the actual rotation speed of the current milling machine 3, or it can be a preset process parameter.
[0062] Continuous adjustment of direction function Used to determine the relative frequency ratio of flutter The decimal part determines the direction and initial adjustment amount of the spindle speed, aiming to guide the spindle speed away from the chatter and instability region. This function calculates... Compared to its rounded integer value The deviation between them is mapped to an adjustment direction and intensity to ensure the smoothness and effectiveness of the adjustment.
[0063] Milling cutter speed adjustment This is the final output value of the spindle speed adjustment model, indicating the required adjustment to the current reference spindle speed. The incremental or decremental adjustments are designed to achieve precise, adaptive control of the milling cutter speed to suppress chatter and optimize the machining process.
[0064] Adjustment ratio coefficient It is a constant greater than zero, used to control the milling cutter speed adjustment. The overall size or the degree of radicalness of adjustment, by adjusting The value of can balance the adjustment of response speed and system stability. The value can be set as a fixed value based on experience during the system design phase, and then dynamically adjusted based on historical processing data or real-time feedback (such as the changing trend of processing accuracy confidence).
[0065] Directional sensitivity coefficient It is a positive constant used to adjust the hyperbolic tangent function tanh( ) for continuously adjusting direction function The sensitivity of the modulatory effect affects the degree of regulation. The changing shape of the curve makes it possible to... Even at smaller sizes, effective adjustments can be made. Machining accuracy confidence level. It is a quantitative indicator output by the machining accuracy confidence assessment module, which reflects the accuracy reliability of the current machining process. The lower the value, the less reliable the machining accuracy, prompting a larger speed adjustment.
[0066] Flutter main frequency band energy index It is another quantitative indicator output by the machining accuracy confidence assessment module, which represents the energy intensity of the frequency band where the main frequency of chatter is located. The higher the value, the more severe the chatter, prompting a greater speed adjustment.
[0067] hyperbolic tangent function tanh( () is a sigmoid function whose output value ranges from -1 to 1. In this model, it is used to... The value is mapped to a smooth and bounded range, thereby ensuring the milling cutter speed adjustment amount The changes are gradual and controlled.
[0068] Rounding function Used to round a real number to the nearest integer, in this model, it is used to calculate the flutter relative frequency ratio. The integer part, which in turn helps determine the continuous adjustment direction function. .
[0069] The milling cutter speed adjustment module of this application plays a crucial role in the sheet metal milling device for massagers. When the machining accuracy confidence assessment module detects the machining accuracy confidence level... When the speed drops below a preset threshold, it indicates that chatter or other unstable factors affecting accuracy may exist during the machining process. In this case, the milling cutter speed adjustment module is activated. This module first acquires the current dominant chatter frequency in real time. and reference spindle speed Based on this data, the flutter relative frequency ratio was calculated. This ratio directly reflects the coupling relationship between the chatter frequency and the spindle speed. Subsequently, the direction function is continuously adjusted... ,according to The deviation from its nearest integer determines an initial adjustment direction and intensity. This function is designed to guide the spindle speed away from the chatter instability region. Based on this, the milling cutter speed adjustment module further incorporates machining accuracy confidence. and flutter main frequency band energy index To finely adjust the amount. When lower and At higher speeds, the model will calculate a larger cutter speed adjustment. Among them, the hyperbolic tangent function tanh( The introduction of the proportional gain makes the adjustment variable smoother and more controlled when it changes, avoiding system oscillations caused by sudden parameter changes. and direction sensitivity coefficient This provides flexible parameter configuration to adapt to different machining conditions and material properties. Ultimately, the calculated milling cutter speed adjustment... The data is sent to the control system of milling machine 3 to adjust the milling cutter's rotational speed in real time. Through this dynamic and adaptive adjustment mechanism, the device can actively suppress chatter generated during machining, thereby stabilizing machining accuracy. This adjustment mechanism is closely integrated with the machining accuracy confidence assessment module to form a closed-loop control system, enabling the device to intelligently adjust process parameters according to changes in real-time machining status, significantly improving machining stability and reliability.
[0070] In a preferred embodiment of the present invention, the cutting depth adjustment model includes: ; in The dynamic state index, This is the workpiece condition index. To determine the confidence level for machining accuracy. As the reference cutting depth, This is the depth of cut adjustment amount.
[0071] In this embodiment, the depth-of-cut adjustment model is a mathematical framework used to calculate the necessary adjustment to the current depth of cut based on real-time evaluated machining state parameters. This model aims to provide a quantitative, adaptive depth-of-cut adjustment strategy to address various unstable factors that may arise during machining. (Depth-of-cut adjustment amount) This indicates the specific value at which the depth of cut needs to be changed under the current machining conditions. Its positive or negative sign indicates whether the depth of cut should be increased or decreased; a negative value indicates that the depth of cut should be decreased to reduce the cutting load and stabilize the machining process.
[0072] Reference cutting depth This refers to the currently used or preset depth of cut value, which serves as a reference point for calculating adjustments. This value can be the initially set process parameters or the actual depth of cut after prior adjustments. (Dynamic state index) It is a quantitative indicator reflecting the dynamic stability of the cutting process. This index is typically obtained by evaluating parameters such as the cutting power per unit volume of the milling machine, the standard deviation of spindle load fluctuation, and the spindle speed following error. When A low value indicates poor cutting dynamics and a significant risk of instability.
[0073] Workpiece condition index It is a quantitative indicator reflecting the stability of the workpiece during processing. This index is typically obtained by evaluating parameters such as the standard deviation of suction fluctuation on the vacuum adsorption platform 4, the maximum temperature rise of the workpiece, and the vibration intensity of the workpiece mounting surface. A low value indicates that the workpiece clamping or its own state is unstable, which can easily lead to processing problems.
[0074] Machining accuracy confidence level It is a quantitative indicator reflecting the likelihood that the current machining process will achieve the expected accuracy. This confidence level is typically obtained by evaluating parameters such as the energy ratio of the dominant frequency band of chatter, the RMS value of spindle vibration acceleration, and the energy of the characteristic frequency band of acoustic emission. A low value indicates a risk of decreased machining accuracy, requiring intervention.
[0075] The solution proposed in this application solves the problem of quantifying the adjustment amount by providing a specific depth-of-cut adjustment model, ensuring accurate and adaptive adjustment of the depth of cut even when the confidence level of machining accuracy is low. This model is based on a reference depth of cut. Generate cutting depth adjustment amount By using a benchmark value as a reference point, instability in processing can be avoided due to excessively large or small adjustments. The formula incorporates... , and The product of these factors comprehensively reflects the inadequacy of the dynamic state index, workpiece state index, and machining accuracy confidence level, ensuring that the adjustment can simultaneously respond to the influence of multiple factors such as workpiece clamping, cutting process dynamics, and accuracy risks. The cube root operation is used to smooth the product result, preventing interference from extreme values and making the adjustment more gradual and reliable. The negative sign design indicates a reduction in the depth of cut, directly addressing the need to reduce the load when conditions are poor. This is achieved through the definition... , , , and By integrating real-time evaluation data and other parameters, the model achieves quantitative calculations, improving the accuracy and adaptability of the adjustment. This depth-of-cut adjustment module is activated when the machining accuracy confidence level remains below a preset threshold after active adjustment of the milling cutter speed. As a secondary adjustment strategy, it forms a progressive adaptive control system with the preceding milling cutter speed adjustment module. When adjusting the milling cutter speed alone is insufficient to restore machining stability, the depth of cut is reduced to further decrease the cutting load, thereby effectively suppressing chatter, reducing workpiece deformation, and improving machining accuracy. This tiered adjustment mechanism allows the device to handle complex machining conditions more flexibly and robustly, avoiding the limitations of a single adjustment method.
[0076] As a preferred embodiment of the present invention, the maximum-minimum value normalization formula is: (input value - minimum value) / (maximum value - minimum value), where the maximum or minimum value is determined by the system's historical data or theoretical safety threshold, ensuring that the normalized calculation result is between 0 and 1.
[0077] In this embodiment, the maximum-minimum normalization formula is a standardization method that transforms data with different dimensions or ranges into a unified interval (usually [0, 1]). Its core function is to eliminate the influence of dimensions in the original data, making different features comparable and thus preventing certain features with large numerical ranges from occupying unreasonable weights in subsequent evaluation models. This formula can be implemented in the data preprocessing module through software programming, receiving raw sensor data as "input values" and calculating based on preset "minimum" and "maximum" values. Alternatively, it can be implemented through hardware logic circuits or dedicated signal processors, scaling and offsetting analog or digital signals accordingly to output the normalized result. The maximum or minimum value is determined by historical system data or theoretical safety thresholds, aiming to provide a stable and reliable benchmark for data transformation. The system can continuously collect and store large amounts of sensor data, determining the parameters used for normalization through statistical analysis (such as calculating the maximum and minimum values of historical data). For example, the highest and lowest observed values over a past period (such as a week or a month) can be used as normalization parameters. Alternatively, theoretical upper and lower limits for each parameter can be pre-set as theoretical safety thresholds based on the equipment's design specifications, material properties, process requirements, or safety standards. For example, for a temperature sensor, the highest temperature the material can withstand can be set as the maximum value, and the lowest ambient temperature as the minimum value. In some cases, the system can also combine historical data and theoretical thresholds, first setting initial maximum and minimum values based on the theoretical safety thresholds, and then dynamically adjusting or optimizing these thresholds based on historical data during actual operation to adapt to changes in actual working conditions. Ensuring that the normalized calculation result is between 0 and 1 is an inherent characteristic of the maximum-minimum normalization formula. As long as the input value is between the minimum and maximum values, the calculation result will naturally fall within the [0, 1] interval. If the input value exceeds this range, truncation is usually performed, for example, setting values less than 0 to 0 and values greater than 1 to 1, to force it to remain within the [0, 1] interval, ensuring that all normalized data falls within a uniform and finite interval, facilitating unified quantitative comparison and calculation by subsequent evaluation models.
[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A sheet metal milling device for a massager, comprising a processing table (1), a housing (2) disposed on the processing table (1), a milling machine (3) disposed in the housing (2), an xy-axis drive assembly (5) for driving the milling machine (3) to move horizontally disposed between the milling machine (3) and the housing (2), and a vacuum adsorption platform (4) disposed below the milling machine (3), characterized in that, The device further includes a control system, the control system comprising: The workpiece status assessment module is electrically connected to the sensor group set on the vacuum adsorption platform (4) and the workpiece mounting surface to receive the suction fluctuation signal, workpiece temperature signal and workpiece mounting surface vibration signal of the vacuum adsorption platform (4) in real time, and constructs a workpiece status assessment model based on the processed suction fluctuation standard deviation, maximum temperature rise of the workpiece and vibration intensity of the workpiece mounting surface, and outputs the workpiece status index. The cutting dynamics state evaluation module is electrically connected to the spindle driver and spindle sensor of the milling machine (3) to receive the cutting power signal, spindle load signal and spindle speed signal of the milling machine (3) in real time, and constructs a cutting dynamics state evaluation model based on the processed unit volume cutting power, spindle load fluctuation standard deviation and spindle speed following error, and outputs the dynamics state index. The machining accuracy confidence assessment module is electrically connected to the vibration sensor and acoustic emission sensor set on the spindle or workpiece of the milling machine (3) to receive the spindle vibration signal and acoustic emission signal in real time, and construct the machining accuracy confidence assessment model based on the processed chatter main frequency band energy ratio, spindle vibration acceleration RMS value and acoustic emission characteristic frequency band energy, and output the machining accuracy confidence. The milling cutter speed adjustment module has its input end connected to the output end of the machining accuracy confidence assessment module, and its output end connected to the spindle driver control end of the milling machine (3). The milling cutter speed adjustment module is used to construct a spindle speed adjustment model based on the reference spindle speed, machining accuracy confidence, chatter main frequency band energy ratio and chatter main frequency value when the machining accuracy confidence is lower than a preset threshold, and output the milling cutter speed adjustment amount to the spindle driver to control the milling cutter speed adjustment. The depth of cut adjustment module has its input end connected to the output end of the workpiece state evaluation module, the cutting dynamics state evaluation module, and the machining accuracy confidence evaluation module, respectively, and its output end connected to the control end of the Z-axis drive mechanism of the xy-axis drive assembly (5) or the milling machine (3). The depth of cut adjustment module is used to construct a depth of cut adjustment model based on the workpiece thickness, dynamics state index, workpiece state index, and reference depth of cut when the machining accuracy confidence is still lower than a preset threshold after the milling cutter speed is actively adjusted, and outputs the depth of cut adjustment amount to control the milling cutter cutting depth to be adjusted. The equipment alarm system has its input terminal connected to the output terminal of the machining accuracy confidence assessment module, and its output terminal connected to the main control power supply or emergency stop circuit of the device. The equipment alarm system is used to respond and execute a shutdown operation when the machining accuracy confidence is still lower than a preset threshold after the cutting depth is adjusted.
2. The sheet metal milling device for massagers according to claim 1, characterized in that, In the workpiece condition assessment model: The normalized suction fluctuation standard deviation index, the maximum temperature rise index of the workpiece, and the vibration intensity index of the mounting surface are multiplied by their respective positive influence coefficients and summed. Then, the summation result is subtracted from 1 to obtain the workpiece condition index. Among them, the suction fluctuation standard deviation index, the workpiece maximum temperature rise index and the installation surface vibration intensity index are obtained by substituting the suction fluctuation standard deviation, the workpiece maximum temperature rise and the workpiece installation surface vibration intensity of the vacuum adsorption platform (4) acquired in real time into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0 and less than 1, and the sum of the three is 1.
3. The sheet metal milling device for massagers according to claim 1, characterized in that, In the cutting dynamics state evaluation model: The normalized cutting power index is directly multiplied by the corresponding preset first influence coefficient; The normalized spindle load fluctuation standard deviation index is transformed by the natural logarithm of (1 plus the index) and then multiplied by the corresponding preset second influence coefficient. The normalized spindle speed following error index is directly multiplied by the corresponding preset third influence coefficient; Summing the product of the above three terms and then subtracting the summation from 1 yields the dynamic state index; Among them, the cutting power index, spindle load fluctuation standard deviation index and spindle speed following error index are obtained by substituting the unit volume cutting power, spindle load fluctuation standard deviation and spindle speed following error of the milling machine (3) acquired in real time into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0 and less than 1, and the sum of the three is 1.
4. The sheet metal milling device for massagers according to claim 1, characterized in that, In the confidence assessment model for machining accuracy: The normalized flutter main frequency band energy index, spindle vibration acceleration index and acoustic emission characteristic frequency band energy index are multiplied by their corresponding positive influence coefficients and summed. Then, the negative exponent of the summation result is taken to obtain the machining accuracy confidence level. Among them, the flutter main frequency band energy index, the spindle vibration acceleration index, and the acoustic emission characteristic frequency band energy index are obtained by substituting the flutter main frequency band energy ratio, the spindle vibration acceleration RMS value, and the acoustic emission characteristic frequency band energy into the maximum-minimum value normalization formula for normalization. All influence coefficients are greater than 0, and the sum of the three is 1.
5. The sheet metal milling apparatus for massagers according to claim 4, characterized in that, In the spindle speed adjustment model: First, calculate the ratio of the chatter frequency to the frequency per revolution of the reference spindle speed to obtain the relative chatter frequency ratio; Secondly, a continuous adjustment direction function is constructed using the difference between this ratio and its rounded-to-integer value; Then, the product of the preset direction sensitivity coefficient and the continuous adjustment direction function is used as the input of the hyperbolic tangent function to obtain the basic quantities of the speed adjustment direction and amplitude; Finally, multiply this basic quantity by the adjustment ratio coefficient, the reference spindle speed, the deviation of the machining accuracy confidence level from 1, and the chatter frequency band energy index to obtain the milling cutter speed adjustment amount.
6. The sheet metal milling apparatus for massagers according to claim 1, characterized in that, In the depth of cut adjustment model: Multiply the dynamic state index, the workpiece state index, and the deviation of the machining accuracy confidence level from 1, respectively, take the cube root of the product, multiply it by the reference cutting depth, and take the negative value to obtain the cutting depth adjustment amount.
7. The sheet metal milling apparatus for massagers according to claim 1, characterized in that, The maximum-minimum normalization formula is: (input value - minimum value) / (maximum value - minimum value), where the maximum or minimum value is determined by the system's historical data or theoretical safety threshold, ensuring that the normalized calculation result is between 0 and 1.