Collaborative optimization method for static minimal quantity lubrication and grinding process parameters
By constructing multiple prediction models and related influencing factors, the electrostatic micro-lubrication and grinding process parameters were optimized, solving the problem of large errors in lubrication performance evaluation and achieving efficient lubrication cooling and improved machining quality.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NINGBO POLYTECHNIC
- Filing Date
- 2026-01-13
- Publication Date
- 2026-04-24
AI Technical Summary
In existing technologies, the synergistic optimization of electrostatic micro-lubrication technology and grinding process parameters has not yet reached its optimal level, resulting in large errors in lubrication performance evaluation results, which affects processing efficiency and quality.
By establishing models for charged liquid phase penetration into capillaries, particle size threshold prediction, and lubrication layer distribution prediction, and combining hardness influence coefficient and temperature rise correlation influence factor, electrostatic micro-lubrication parameters and grinding process parameters are optimized to achieve dynamic adjustment and synergistic optimization.
It improves the accuracy of lubrication and cooling performance evaluation and the overall efficiency of the machining process, avoids local optima caused by single parameter optimization, and improves machining quality and efficiency.
Smart Images
Figure CN121920604A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of grinding technology, and in particular to a method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters. Background Technology
[0002] The quality of grinding processes (such as surface roughness, residual stress, grinding temperature, and material removal rate) is influenced by a combination of process parameters, primarily including: wheel speed, workpiece feed rate, depth of cut (radial / axial), and cooling / lubrication conditions. These parameters are not independent but rather coupled and interact with each other. While MQL (Mini-Low Lubrication) and electrostatic atomization technologies have shown great potential, how to scientifically and synergistically optimize electrostatic micro-lubrication technology with specific grinding process parameters to fully leverage their advantages in improving grinding performance remains a key issue that urgently needs to be addressed in the field of green manufacturing.
[0003] Currently, the most direct and effective way to reduce the hazards of cutting fluid is to adopt green cutting technologies with less or no cutting fluid. Dry cutting generates high cutting temperatures, and the intense friction between the tool, workpiece, and chips leads to insufficient workpiece precision and short tool life. Key technical issues to be addressed include tool materials, tool coatings, tool structure, and cutting temperature control. Improving the efficiency of electrostatic micro-lubrication penetration also requires considering the variations in droplet surface tension, contact angle, and other characteristics, which can cause changes in droplet size and thus affect lubrication distribution and penetration efficiency. Furthermore, these factors are susceptible to interference from external environmental factors, leading to errors in subsequent lubrication performance evaluations and reducing the efficiency of co-optimization management of electrostatic micro-lubrication and grinding process parameters. Summary of the Invention
[0004] To overcome the shortcomings of the prior art, this application provides a method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters.
[0005] This application provides a method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters, the method comprising:
[0006] Step S1: Based on the surface tension, contact angle, and viscosity of the charged cutting fluid droplets in the historical test data of electrostatic micro-lubrication and grinding process parameters, a model for the penetration of charged liquid phase into capillaries is established. Based on the correlation and influence characteristics between the surface tension, contact angle, and viscosity of the charged cutting fluid droplets and the voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the ground workpiece, a first correlation influence factor, a second correlation influence factor, and a third correlation influence factor are obtained. Based on the process parameters to be tested in the historical test data and the surface tension, contact angle, and viscosity of the charged cutting fluid droplets, a particle size threshold prediction model is established. Based on the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid in the historical test data, a lubrication layer distribution prediction model is established.
[0007] Step S2: Based on the correlation and influence characteristics of the distribution characteristics of the lubricating layer to be tested on the surface microhardness of the workpiece in the historical test data, the hardness influence coefficient is obtained. The temperature rise data generated by the process parameters to be tested and the distribution characteristics of the lubricating layer to be tested under the operating state in the historical test data is extracted. Based on the correlation and influence characteristics between the temperature rise data to be tested and the degree of performance damage caused by the temperature rise data to be tested, the correlation influence damage coefficient is obtained.
[0008] Step S3: Based on the first, second, and third correlation influence factors, the pre-treatment tension data, pre-treatment contact angle data, and pre-treatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication in the current period are corrected for correlation influence changes to obtain initial tension data, initial contact angle data, and initial viscosity data. Based on the correlation change influence test between the initial tension data, initial contact angle data, initial viscosity data, charged liquid phase penetration into capillary model, lubrication layer distribution prediction model, hardness influence coefficient, and correlation influence damage coefficient, the overall performance index evaluation result of lubrication and cooling processing performance assessment is obtained. Based on the overall performance index evaluation result, the electrostatic micro-lubrication parameters and grinding process parameters are synergistically optimized and controlled to obtain parameter synergistic optimization results.
[0009] Preferably, historical test data on the coordination of electrostatic micro-lubrication and grinding process parameters are obtained, and the surface tension, contact angle and viscosity of the charged cutting fluid droplets are extracted from the historical test data to obtain the tension data, contact angle data and viscosity data to be measured;
[0010] Grinding process parameters, electrostatic parameters of the applied electrostatic field, and characteristic parameters of the cutting fluid using charged cutting fluid are extracted from historical test data to obtain the process parameters to be tested, electrostatic parameters to be tested, and cutting fluid data to be tested.
[0011] Based on the process parameters to be tested, electrostatic parameters to be tested, cutting fluid data to be tested, contact angle data to be tested, and viscosity data to be tested, a model for the penetration of charged liquid phase into capillary is established.
[0012] The voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the workpiece to be ground are extracted from historical test data to obtain the voltage data to be measured, the ambient temperature value to be measured, and the surface roughness to be measured.
[0013] Preferably, the voltage data to be measured, the ambient temperature value to be measured, and the roughness to be measured are combined into a feature dataset.
[0014] Based on the correlation and change characteristics between the tension data to be measured and the first feature dataset, the first correlation influence factor between the tension data to be measured and the first feature dataset is obtained.
[0015] The first feature dataset and the tension data to be measured are combined to form the second feature dataset.
[0016] Based on the correlation and change characteristics between the contact angle data to be measured and the second feature dataset, a second correlation influence factor between the contact angle data to be measured and the second feature dataset is obtained.
[0017] Based on the correlation and influence characteristics between the viscosity data, voltage data, and ambient temperature value to be measured, a third correlation influence factor between the viscosity data and ambient temperature value to be measured is obtained.
[0018] Preferably, the particle size characteristic data to be tested is extracted from historical test data based on the process parameters to be tested, the tension data to be tested, the contact angle data to be tested, and the viscosity data to be tested.
[0019] Based on the process parameters to be measured, the tension data to be measured, the contact angle data to be measured, the viscosity data to be measured, and the particle size characteristic data to be measured, a particle size threshold prediction model is established;
[0020] The permeability values of the charged cutting fluid, as well as the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid, are extracted from historical test data to obtain the permeability values and key parameters to be tested.
[0021] Based on the particle size characteristic data, permeability value, and key parameters to be measured, a lubrication layer distribution prediction model is established.
[0022] Preferably, the distribution characteristics of the lubricating layer to be tested, the material characteristic parameters of the workpiece to be tested, and the degree of influence of the distribution characteristics of the lubricating layer to be tested on the microhardness of the workpiece material surface caused by the changes in the microhardness of the workpiece material surface under the influence of the distribution characteristics of the lubricating layer to be tested are extracted from historical test data.
[0023] Based on the correlation and variation characteristics between the distribution characteristics of the lubricating layer to be tested, the material characteristic parameters of the workpiece to be tested, and the degree of influence of the hardness to be tested, the hardness influence coefficient is obtained.
[0024] Extract the distribution characteristics of the lubricating layer to be tested and the temperature rise amplitude induced by frictional heat generated by the process parameters to be tested under the working condition from historical test data to obtain the temperature rise data to be tested.
[0025] Based on the correlation and influence characteristics between the distribution characteristics of the lubricating layer to be tested, the process parameters to be tested, and the temperature rise data to be tested, an additional temperature rise correlation coefficient is obtained.
[0026] The degree of performance impairment caused by the temperature rise data under test to lubrication and cooling performance and processing performance is extracted from historical test data to obtain the degree of impairment index.
[0027] Based on the correlation and change characteristics between the measured temperature rise data and the measured damage index, the correlation and damage coefficient between the measured temperature rise data and the measured damage index is obtained.
[0028] Preferably, the pretreatment tension data, pretreatment contact angle data, and pretreatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication in the current period are obtained, and the current electrostatic parameters and current cutting fluid data of the applied electrostatic field in the current period are also obtained. The pretreatment tension data, pretreatment contact angle data, pretreatment viscosity data, current electrostatic parameters, and current cutting fluid data are input into the charged liquid phase penetration capillary model for testing to obtain pretreatment permeability characteristic data.
[0029] Based on feature dataset one, the preprocessed dataset one belonging to the current period is statistically analyzed, and the initial tension data is obtained based on the preprocessed tension data, preprocessed dataset one, and the first correlation influence factor.
[0030] Based on feature dataset two, the preprocessed dataset two to which the current period belongs is statistically analyzed. Based on the preprocessed contact angle data, preprocessed dataset two, and the second correlation influencing factor, the initial measured contact angle data is obtained.
[0031] The ambient temperature value of the current period is detected to obtain the initial temperature value. Based on the preprocessed viscosity data, the initial temperature value, and the third correlation influence factor, the initial viscosity data is obtained.
[0032] Preferably, based on the process parameters to be tested, the current process parameters are obtained, and the current process parameters, initial tension data, initial contact angle data, and initial viscosity data are input into the particle size threshold prediction model for testing to obtain the current particle size prediction threshold.
[0033] Based on the key parameters to be tested, the current key parameters are statistically determined. The preprocessed permeability characteristic data, the current particle size prediction threshold, and the current key parameters are then input into the lubrication layer distribution prediction model for testing to obtain the current lubrication layer predicted distribution characteristics.
[0034] Obtain the material characteristic parameters of the current workpiece to which the current test workpiece belongs, and obtain the current hardness influence value based on the current workpiece material characteristic parameters, the current lubrication layer predicted distribution characteristics, and the hardness influence coefficient;
[0035] Based on historical test data and the material characteristics of the workpiece to be tested, a prediction model for lubrication and cooling processing performance is established.
[0036] Preferably, the current predicted distribution characteristics of the lubrication layer, the current process parameters, the current workpiece material characteristic parameters, and the current hardness influence value are input into the lubrication and cooling processing performance prediction model for testing, and the lubrication and cooling processing performance evaluation index is obtained.
[0037] Based on the current predicted distribution characteristics of the lubrication layer, the current process parameters, and the additional temperature rise correlation coefficient, the current predicted temperature rise value is obtained;
[0038] Based on the current predicted temperature rise value and the associated impact damage coefficient, the performance index impact value is obtained;
[0039] Based on the current temperature rise prediction value, error compensation is performed on the lubrication and cooling processing performance evaluation index to obtain the overall performance index evaluation result;
[0040] Based on the overall performance index evaluation results, the electrostatic micro-lubrication parameters and grinding process parameters are synergistically optimized and controlled to obtain the parameter synergistic optimization results.
[0041] Compared with the prior art, the present invention has the following characteristics and beneficial effects:
[0042] By analyzing historical test data characteristics of the synergy between electrostatic micro-lubrication and grinding process parameters, and the influence of correlation changes among various test data points in the historical data, a predictive model was constructed to facilitate subsequent evaluation of electrostatic micro-lubrication cooling performance and workpiece machining performance. Several predictive models were established, such as a charged liquid phase penetration capillary model, a particle size threshold prediction model, and a lubrication layer distribution prediction model. These models can accurately predict changes in relevant parameters, providing a reliable basis for subsequent optimization. The models consider the correlation changes among multiple factors, making them more closely reflect actual working conditions and improving prediction accuracy. Preliminary detection of the initial surface tension, contact angle, and viscosity of the charged cutting fluid within the applied electrostatic field region is performed, but this is affected by environmental voltage, ambient temperature, and workpiece surface roughness. The same degree of influence can produce different degrees of detection errors, leading to inaccuracies in subsequent evaluation data. By introducing a hardness influence coefficient and a correlation influence damage coefficient, the impact of lubrication layer distribution on workpiece hardness and the damage of temperature rise to machining performance are comprehensively considered. This allows for a more comprehensive evaluation of lubrication and cooling machining performance. Finally, by combining various characteristic data obtained from current testing and testing multiple prediction models, a total performance index evaluation result is obtained. The total performance index evaluation result can intuitively reflect the overall performance of the machining process, facilitating timely problem identification and optimization. Parameters are dynamically adjusted based on real-time performance index evaluation results, achieving synergistic optimization and control of electrostatic micro-lubrication parameters and grinding process parameters. This avoids local optima that may result from optimizing a single parameter, thereby improving the overall efficiency and quality of the machining process. Attached Figure Description
[0043] Figure 1 This embodiment is a flowchart illustrating the steps of a method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters. Detailed Implementation
[0044] The present invention will be further described in detail below with reference to the following embodiments.
[0045] Reference Figure 1 A method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters, comprising the following steps:
[0046] Step S1: Based on the historical test data of the surface tension, contact angle, and viscosity of the charged cutting fluid droplets in the synergy of electrostatic micro-lubrication and grinding process parameters, a model for the penetration of the charged liquid phase into the capillary is established. Based on the correlation and influence characteristics between the surface tension, contact angle, and viscosity of the charged cutting fluid droplets and the voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the ground workpiece, the first correlation influence factor, the second correlation influence factor, and the third correlation influence factor are obtained. Based on the process parameters to be tested in the historical test data and the surface tension, contact angle, and viscosity of the charged cutting fluid droplets, a particle size threshold prediction model is established. Based on the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid in the historical test data, a lubrication layer distribution prediction model is established.
[0047] Step S2: Based on the correlation and influence characteristics of the distribution characteristics of the lubricating layer to be tested on the surface microhardness of the workpiece in the historical test data, the hardness influence coefficient is obtained. The temperature rise data generated by the process parameters to be tested and the distribution characteristics of the lubricating layer to be tested under the operating state in the historical test data is extracted. Based on the correlation and influence characteristics between the temperature rise data to be tested and the degree of performance damage caused by the temperature rise data to be tested, the correlation influence damage coefficient is obtained.
[0048] Step S3: Based on the first, second, and third correlation influence factors, the pre-treatment tension data, pre-treatment contact angle data, and pre-treatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication in the current period are corrected for correlation influence changes to obtain initial tension data, initial contact angle data, and initial viscosity data. Based on the correlation change influence test between the initial tension data, initial contact angle data, initial viscosity data, charged liquid phase penetration into capillary model, lubrication layer distribution prediction model, hardness influence coefficient, and correlation influence damage coefficient, the overall performance index evaluation result of lubrication and cooling processing performance assessment is obtained. Based on the overall performance index evaluation result, the electrostatic micro-lubrication parameters and grinding process parameters are synergistically optimized and controlled to obtain the parameter synergistic optimization result.
[0049] Specifically, by analyzing the characteristics of historical test data on the synergy between electrostatic micro-lubrication and grinding process parameters, as well as the influence of correlation changes among various test data points in the historical data, a predictive model was constructed to facilitate subsequent evaluation of the cooling performance and workpiece machining performance of electrostatic micro-lubrication. Several predictive models were established, such as a charged liquid phase penetration capillary model, a particle size threshold prediction model, and a lubrication layer distribution prediction model. These models can accurately predict changes in relevant parameters, providing a reliable basis for subsequent optimization. The models consider the correlation changes among various factors, making them more closely reflect actual working conditions and improving prediction accuracy. Preliminary measurements were taken of the initial surface tension, contact angle, and viscosity of the charged cutting fluid within the applied electrostatic field region. However, these parameters are affected by environmental voltage, ambient temperature, and workpiece surface roughness. Different degrees of influence can lead to varying degrees of detection errors, resulting in inaccurate subsequent evaluation data. By introducing a hardness influence coefficient and a correlation influence damage coefficient, the impact of lubrication layer distribution on workpiece hardness and the damage of temperature rise to machining performance are comprehensively considered. This allows for a more comprehensive evaluation of lubrication and cooling machining performance. Finally, by combining various characteristic data obtained from current testing and testing multiple prediction models, a total performance index evaluation result is obtained. The total performance index evaluation result can intuitively reflect the overall performance of the machining process, facilitating timely problem identification and optimization. Parameters are dynamically adjusted based on real-time performance index evaluation results, achieving synergistic optimization and control of electrostatic micro-lubrication parameters and grinding process parameters. This avoids local optima that may result from optimizing a single parameter, thereby improving the overall efficiency and quality of the machining process.
[0050] The specific step S1 includes the following sub-steps:
[0051] Historical test data on the coordination of electrostatic micro-lubrication and grinding process parameters were obtained. The surface tension, contact angle and viscosity of the charged cutting fluid droplets were extracted from the historical test data to obtain the tension data, contact angle data and viscosity data to be measured.
[0052] Grinding process parameters, electrostatic parameters of the applied electrostatic field, and characteristic parameters of the cutting fluid using charged cutting fluid are extracted from historical test data to obtain the process parameters to be tested, electrostatic parameters to be tested, and cutting fluid data to be tested.
[0053] A model for the penetration of charged liquid phase into the capillary is established based on the process parameters to be tested, the electrostatic parameters to be tested, the cutting fluid data to be tested, the contact angle data to be tested, and the viscosity data to be tested.
[0054] The voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the workpiece to be ground are extracted from historical test data to obtain the voltage data to be measured, the ambient temperature value to be measured, and the surface roughness to be measured.
[0055] The measured voltage data, measured ambient temperature value, and measured roughness are combined to form a feature dataset.
[0056] Based on the correlation and change characteristics between the tension data to be measured and the feature dataset 1, the first correlation influence factor between the tension data to be measured and the feature dataset 1 is obtained.
[0057] Feature dataset one and the tension data to be measured are combined to form feature dataset two.
[0058] Based on the correlation and influence characteristics between the contact angle data to be measured and feature dataset two, a second correlation influence factor between the contact angle data to be measured and feature dataset two is obtained.
[0059] Based on the correlation and influence characteristics between the viscosity data, voltage data, and ambient temperature value to be measured, a third correlation influence factor between the viscosity data and ambient temperature value to be measured is obtained.
[0060] Based on the process parameters to be tested, the tension data to be tested, the contact angle data to be tested, and the viscosity data to be tested, the particle size characteristic data to be tested are extracted from historical test data.
[0061] A particle size threshold prediction model is established based on the process parameters to be measured, tension data to be measured, contact angle data to be measured, viscosity data to be measured, and particle size characteristic data to be measured.
[0062] The permeability values of the charged cutting fluid, as well as the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid, are extracted from historical test data to obtain the permeability values and key parameters to be tested.
[0063] Based on the particle size characteristics, permeability values, and key parameters to be measured, a prediction model for the distribution of the lubricating layer is established.
[0064] Specifically, data such as the tension data, contact angle data, and viscosity data to be measured (tension data can be measured using a surface tension meter (by measuring the maximum force required to pull a platinum ring away from the liquid surface to calculate surface tension), contact angle data can be measured using an optical contact angle meter (using a CCD camera to capture the droplet profile and fitting the contact angle using the Young-Laplace equation), viscosity data can be measured using a capillary viscometer (based on Poiseuille's law, measuring the time it takes for the liquid to flow through a standard capillary)), and electrostatic parameters to be measured (applied voltage, jetting current)). Characteristic data such as flow / charge density), charged liquid phase penetration into capillary model (e.g., based on the mapping relationship of "process parameters - cutting fluid characteristics - penetration effect", the process parameters to be measured include grinding wheel speed, workpiece feed speed, cooling pressure, and electrostatic field parameters (such as voltage, charge density, because the electrostatic field will change the surface tension of the cutting fluid, thus affecting the contact angle), cutting fluid characteristic parameters (including the mass of the cutting fluid (obtained by cutting fluid quality testing equipment), average flow rate (obtained by cutting fluid average flow rate testing equipment), driving force caused by external pressure difference = (P1-P2)*π*(r squared), electrostatic attraction of the spatial electrostatic field on the charged cutting fluid = E*q, etc., where p refers to the pressure at both ends of the capillary, π and r are constants, E is the electric field strength, q (This refers to the charge), using existing technology: Bosanquet equation (where the driving force caused by the external pressure difference is added to the electrostatic attraction of the space electrostatic field on the charged cutting fluid), Bosanquet equation is the standard dynamic model in the field of capillary flow), the voltage data to be measured, the ambient temperature value to be measured, and the roughness to be measured (the voltage data to be measured can be observed by an electrostatic voltmeter, the ambient temperature value to be measured can be detected by a temperature sensor, and the roughness to be measured can be detected by a surface roughness profiler), the first correlation influence factor (such as Z=a0+a1*V+a2*W+a3*Cu+ε, where Z refers to the tension data to be measured, a0 refers to the intercept term, a1 refers to the correlation influence factor between the tension data to be measured and the voltage data to be measured, a2 refers to the correlation influence factor between the tension data to be measured and the voltage data to be measured). The correlation factors between ambient temperature values, a3 refers to the correlation factor between the measured tension data and the measured roughness, V refers to the measured voltage data, W refers to the measured ambient temperature value, Cu refers to the measured roughness, and ε refers to the random error term), the second correlation factor, and the third correlation factor (based on the first correlation factor, and so on). The surface tension, contact angle, and viscosity of the charged cutting fluid droplet will undergo unsteady fluctuations due to electrostatic field voltage (changing charge density), ambient temperature (changing fluid properties), and workpiece surface roughness (changing actual contact area), resulting in deviations between the theoretical model and actual working conditions. By predicting the changing trends of the measured tension data, measured contact angle data, and measured viscosity data, process parameters (such as voltage amplitude and nozzle position) can be adjusted in advance.The particle size threshold prediction model (the process parameters to be measured also include nozzle size and flow rate) uses dimensionless numbers (such as Oh number) to significantly influence the predicted droplet aerosol particle size based on the measured tension data, contact angle data, viscosity data, and process parameters. The predicted particle size threshold is then compared with the measured particle size characteristic data to statistically determine the error value. The Oh number is then compensated for using this error value. A historical Oh number-particle size threshold matching table is used (e.g., for water: Oh number ~ 0.017, predicted particle size range 60-80 μm, Oh number ~ 0.017). The larger the number, the less likely it is to break, and the larger the average particle size), thus obtaining the particle size threshold prediction model. Key parameters to be measured include: evaporation rate (e.g., using an evaporation rate tester or hygrometer. These devices can measure the evaporation rate of a liquid under certain conditions, thus obtaining the evaporation rate of the cutting fluid), surface energy (e.g., using a contact angle meter. By measuring the contact angle of different liquids on a solid surface, the surface energy of the solid can be calculated), substrate properties (depending on the specific properties to be detected, different devices may be needed; for example, a hardness tester is used to measure the hardness of a material, a roughness meter is used to measure surface roughness, and an X-ray diffractometer (XRD) is used to analyze crystal structure, etc.), and ambient temperature characteristic data. A lubrication layer distribution prediction model is also available (e.g., using Gaussian process regression (GPR). Input features include: particle size, penetration rate, surface energy, evaporation rate, temperature, substrate hardness; output: lubrication film thickness distribution field, i.e., subsequent lubrication layer prediction distribution characteristics).
[0065] The specific step S2 includes the following sub-steps:
[0066] Extract the distribution characteristics of the lubricating layer to be tested, the material characteristic parameters of the workpiece to be tested, and the degree of influence of the distribution characteristics of the lubricating layer to be tested on the microhardness of the workpiece material surface caused by the changes in the microhardness of the workpiece material surface.
[0067] The hardness influence coefficient is obtained based on the correlation and variation characteristics between the distribution characteristics of the lubricating layer to be tested, the material characteristic parameters of the workpiece to be tested, and the degree of influence of the hardness to be tested.
[0068] Extract the distribution characteristics of the lubricating layer to be tested and the temperature rise amplitude induced by frictional heat generated by the process parameters to be tested under the working condition from historical test data to obtain the temperature rise data to be tested.
[0069] Based on the characteristics of the distribution of the lubricating layer under test, the influence of the correlation changes among the process parameters under test and the temperature rise data under test, an additional temperature rise correlation coefficient is obtained.
[0070] The degree of performance impairment caused by the temperature rise data under test to lubrication and cooling performance and processing performance is extracted from historical test data to obtain the degree of impairment index.
[0071] Based on the correlation and change characteristics between the measured temperature rise data and the measured damage index, the correlation and damage coefficient between the measured temperature rise data and the measured damage index is obtained.
[0072] Specifically, the characteristics of the lubricating layer to be tested (referring to the thickness distribution of the lubricating layer), the degree of influence of the hardness to be tested (e.g., the difference between the hardness value of the workpiece before electrostatic micro-lubrication and the hardness value of the workpiece during electrostatic micro-lubrication, which yields the degree of influence of the hardness to be tested), the material characteristic parameters of the workpiece to be tested (lubricating layer distribution characteristics: such as represented by parameter L (e.g., average thickness or coverage), such as the carbon content (C), chromium content (Cr), initial hardness (H0), and workpiece size (D) (diameter or thickness) of Cr12 mold steel), carbon content C (approximately 1.4-1.6%), and the degree of influence of hardness: denoted as Hi, indicating the addition of... The post-processing hardness change is represented by the formula Hi = k * f(L, C, Cr, H0, D) + ϵ, where f(*) is a nonlinear function (e.g., a product term or a power function), ϵ is the error term, and k is the hardness influence coefficient to be determined. The measured temperature rise data (e.g., the initial ambient temperature is the measured ambient temperature value; during electrostatic micro-lubrication: the measured lubrication layer distribution characteristics and the measured process parameters generate frictional heat under workpiece operation, affecting the ambient temperature rise; the difference between the measured ambient temperature value and the measured ambient temperature value is obtained, i.e., the measured temperature rise data) is calculated. An additional temperature rise correlation coefficient (based on the hardness influence coefficient, and so on) is also included. Frictional heat is a key factor in the grinding process. Excessively high temperatures can lead to workpiece surface burns and decreased dimensional accuracy. By predicting the temperature rise value, the potential thermal conditions can be understood before processing. When the temperature rises, the viscosity of the lubricant may change, leading to a decrease in the stability of the lubrication cooling layer. From the perspective of processing accuracy, temperature changes cause thermal expansion of the workpiece and grinding tool. Uneven thermal expansion can lead to dimensional deviations in machining, affecting the shape accuracy of the workpiece. (After assessing the extent of these effects, corresponding measures can be taken, such as adjusting grinding process parameters to control temperature and thus improve machining performance.) The measured damage index includes the damage values for lubrication and cooling efficiency and machining performance. Lubrication and cooling performance is quantified through cooling efficiency, such as estimation based on specific heat capacity and mass change. Machining performance is quantified by constructing a weighted objective function (P=w1*Ramax / Ra+w2*Fmax / F+w3*Tmax). / T), where Ra refers to surface roughness, F refers to grinding force (e.g., through a piezoelectric triaxial dynamic force gauge), T refers to grinding temperature (e.g., through an infrared thermal imager), max refers to the maximum value (all known data), w1, w2, w3 refer to weighting coefficients (preset by the user (e.g., determined by the Taguchi method)), and the correlation influence damage coefficient (e.g., taking the damage degree index to be measured as the value on the y-axis and the temperature rise data to be measured as the value on the x-axis, statistically analyzing the correlation change influence characteristic curve between the two, and obtaining the slope (calculating the average value), which is the correlation influence damage coefficient).
[0073] The specific step S3 includes the following sub-steps:
[0074] The system acquires the pretreatment tension, pretreatment contact angle, and pretreatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication during the current period. It also acquires the current electrostatic parameters and current cutting fluid data of the applied electrostatic field during the current period. The pretreatment tension, pretreatment contact angle, pretreatment viscosity data, current electrostatic parameters, and current cutting fluid data are input into the charged liquid phase penetration capillary model for testing to obtain pretreatment permeability characteristic data.
[0075] Based on feature dataset one, the preprocessed dataset one belonging to the current period is statistically analyzed. Based on the preprocessed tension data, preprocessed dataset one, and the first correlation influence factor, the initial tension data is obtained.
[0076] Based on feature dataset two, the preprocessed dataset two to which the current period belongs is statistically analyzed. Based on the preprocessed contact angle data, preprocessed dataset two, and the second correlation influencing factor, the initial measured contact angle data is obtained.
[0077] The ambient temperature value of the current period is detected to obtain the initial temperature value. Based on the preprocessed viscosity data, the initial temperature value, and the third correlation influence factor, the initial viscosity data is obtained.
[0078] Based on the process parameters to be tested, the current process parameters are obtained. The current process parameters, initial tension data, initial contact angle data, and initial viscosity data are then input into the particle size threshold prediction model for testing to obtain the current particle size prediction threshold.
[0079] Based on the key parameters to be tested, the current key parameters are statistically determined. The preprocessed permeability characteristic data, the current particle size prediction threshold, and the current key parameters are then input into the lubrication layer distribution prediction model for testing, and the current lubrication layer predicted distribution characteristics are obtained.
[0080] Obtain the material characteristic parameters of the current workpiece to which the current test workpiece belongs. Based on the material characteristic parameters of the current workpiece, the predicted distribution characteristics of the current lubrication layer, and the hardness influence coefficient, obtain the current hardness influence value.
[0081] Based on historical test data and the material characteristics of the workpiece to be tested, a prediction model for lubrication and cooling processing performance is established.
[0082] The current predicted distribution characteristics of the lubrication layer, current process parameters, current workpiece material characteristic parameters, and current hardness influence value are input into the lubrication and cooling processing performance prediction model for testing, and the lubrication and cooling processing performance evaluation index is obtained.
[0083] Based on the current predicted distribution characteristics of the lubrication layer, the current process parameters, and the additional temperature rise correlation coefficient, the current predicted temperature rise value is obtained.
[0084] Based on the current temperature rise forecast and the associated impact damage coefficient, the performance index impact value is obtained.
[0085] Based on the current temperature rise prediction, error compensation is applied to the lubrication and cooling processing performance evaluation index to obtain the overall performance index evaluation result.
[0086] Based on the overall performance index evaluation results, the electrostatic micro-lubrication parameters and grinding process parameters were synergistically optimized and controlled to obtain the parameter synergistic optimization results.
[0087] Specifically, this includes initial tension data (by substituting the pre-processed tension data, pre-processed dataset 1, and the first correlation influence factor into Z=a0+a1*V+a2*W+a3*Cu+ε, the initial tension data is obtained), initial contact angle data and initial viscosity data (based on the initial tension data, the second correlation influence factor, the third correlation influence factor, and so on), current process parameters (process parameters include voltage, air pressure, cutting speed, cutting depth, feed rate, etc.), and pre-processed permeability characteristic data (usually including permeability, effective porosity, relative permeability, dynamic contact angle change, and the effect of electric field on wettability and capillary action). The influence of fine-grained rise), the current hardness influence value (e.g., substituting known current workpiece material characteristics, current lubrication layer predicted distribution characteristics, and hardness influence coefficient into Hi=k*f(L, C, Cr, H0, D)+ϵ to obtain the current hardness influence value), lubrication and cooling processing performance prediction models (e.g., MQL system parameters from historical test data, grinding process parameters (e.g., grinding wheel linear speed (m / s), workpiece feed speed (mm / min), axial or radial depth of cut (μm), etc.), and workpiece material characteristics can be obtained through existing technologies: random forest regression or gradient boosting tree (XGBoost or L The model is constructed using the following methods: current temperature rise prediction (based on the current hardness influence value, and so on); performance index influence value (e.g., multiplying the current temperature rise prediction value by the associated influence damage coefficient to obtain the performance index influence value); total performance index evaluation result (e.g., summing the performance index influence value and the lubrication and cooling processing performance evaluation index); and parameter co-optimization results (e.g., taking surface roughness included in the total performance index evaluation result as an example, such as different electrostatic micro-lubrication parameters (e.g., oil quantity, electric field strength, nozzle position, etc.) and grinding process parameters (e.g., grinding wheel speed, feed rate, depth of cut). Variables (such as...) are used to predict the impact on lubrication and cooling performance and machining performance. For example, assuming our goal is to minimize surface roughness, one or more parameters are changed each time, and the corresponding surface roughness values are recorded. Then, statistical methods are used to analyze the data to find the optimal combination of parameters that minimizes surface roughness. For example, if it is found that when the oil volume of electrostatic micro-lubrication is 5 ml / h, the electric field strength is 3 kV, the grinding wheel linear speed is 30 m / s, and the workpiece feed rate is 1.2 mm / s, the surface roughness reaches the minimum value Ra = 0.2 μm, then this parameter combination can be used as the optimized result.
[0088] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. A method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters, characterized in that, Includes the following steps: Step S1: Based on the surface tension, contact angle, and viscosity of the charged cutting fluid droplets in the historical test data of electrostatic micro-lubrication and grinding process parameters, a model for the penetration of charged liquid phase into capillaries is established. Based on the correlation and influence characteristics between the surface tension, contact angle, and viscosity of the charged cutting fluid droplets and the voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the ground workpiece, a first correlation influence factor, a second correlation influence factor, and a third correlation influence factor are obtained. Based on the process parameters to be tested in the historical test data and the surface tension, contact angle, and viscosity of the charged cutting fluid droplets, a particle size threshold prediction model is established. Based on the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid in the historical test data, a lubrication layer distribution prediction model is established. Step S2: Based on the correlation and influence characteristics of the distribution characteristics of the lubricating layer to be tested on the surface microhardness of the workpiece in the historical test data, the hardness influence coefficient is obtained. The temperature rise data generated by the process parameters to be tested and the distribution characteristics of the lubricating layer to be tested under the operating state in the historical test data is extracted. Based on the correlation and influence characteristics between the temperature rise data to be tested and the degree of performance damage caused by the temperature rise data to be tested, the correlation influence damage coefficient is obtained. Step S3: Based on the first, second, and third correlation influence factors, the pre-treatment tension data, pre-treatment contact angle data, and pre-treatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication in the current period are corrected for correlation influence changes to obtain initial tension data, initial contact angle data, and initial viscosity data. Based on the correlation change influence test between the initial tension data, initial contact angle data, initial viscosity data, charged liquid phase penetration into capillary model, lubrication layer distribution prediction model, hardness influence coefficient, and correlation influence damage coefficient, the overall performance index evaluation result of lubrication and cooling processing performance assessment is obtained. Based on the overall performance index evaluation result, the electrostatic micro-lubrication parameters and grinding process parameters are synergistically optimized and controlled to obtain parameter synergistic optimization results.
2. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 1, characterized in that, Step S1 includes: Historical test data on the coordination of electrostatic micro-lubrication and grinding process parameters are obtained. The surface tension, contact angle and viscosity of the charged cutting fluid droplets are extracted from the historical test data to obtain the tension data, contact angle data and viscosity data to be measured. Grinding process parameters, electrostatic parameters of the applied electrostatic field, and characteristic parameters of the cutting fluid using charged cutting fluid are extracted from historical test data to obtain the process parameters to be tested, electrostatic parameters to be tested, and cutting fluid data to be tested. Based on the process parameters to be tested, electrostatic parameters to be tested, cutting fluid data to be tested, contact angle data to be tested, and viscosity data to be tested, a model for the penetration of charged liquid phase into capillary is established. The voltage data of the applied electrostatic field, the ambient temperature data, and the surface roughness of the workpiece to be ground are extracted from historical test data to obtain the voltage data to be measured, the ambient temperature value to be measured, and the surface roughness to be measured.
3. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 2, characterized in that, Step S1 also includes: The measured voltage data, measured ambient temperature value, and measured roughness are combined to form a feature dataset. Based on the correlation and change characteristics between the tension data to be measured and the first feature dataset, the first correlation influence factor between the tension data to be measured and the first feature dataset is obtained. The first feature dataset and the tension data to be measured are combined to form the second feature dataset. Based on the correlation and change characteristics between the contact angle data to be measured and the second feature dataset, a second correlation influence factor between the contact angle data to be measured and the second feature dataset is obtained. Based on the correlation and influence characteristics between the viscosity data, voltage data, and ambient temperature value to be measured, a third correlation influence factor between the viscosity data and ambient temperature value to be measured is obtained.
4. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 3, characterized in that, Step S1 also includes: Based on the process parameters to be tested, the tension data to be tested, the contact angle data to be tested, and the viscosity data to be tested, the particle size characteristic data to be tested is extracted from the historical test data; Based on the process parameters to be measured, the tension data to be measured, the contact angle data to be measured, the viscosity data to be measured, and the particle size characteristic data to be measured, a particle size threshold prediction model is established; The permeability values of the charged cutting fluid, as well as the evaporation rate, surface energy, substrate properties, and ambient temperature characteristics of the charged cutting fluid, are extracted from historical test data to obtain the permeability values and key parameters to be tested. Based on the particle size characteristic data, permeability value, and key parameters to be measured, a lubrication layer distribution prediction model is established.
5. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 4, characterized in that, Step S2 includes: Extract the distribution characteristics of the lubricating layer under test, the material characteristic parameters of the workpiece under test, and the degree of influence of the distribution characteristics of the lubricating layer under test on the microhardness of the workpiece material surface under the influence of the microhardness of the workpiece material surface under the influence of the distribution characteristics of the lubricating layer under test from historical test data. Based on the correlation and variation characteristics between the distribution characteristics of the lubricating layer to be tested, the material characteristic parameters of the workpiece to be tested, and the degree of influence of the hardness to be tested, the hardness influence coefficient is obtained. Extract the distribution characteristics of the lubricating layer to be tested and the temperature rise amplitude induced by frictional heat generated by the process parameters to be tested under the working condition from historical test data to obtain the temperature rise data to be tested. Based on the correlation and influence characteristics between the distribution characteristics of the lubricating layer to be tested, the process parameters to be tested, and the temperature rise data to be tested, an additional temperature rise correlation coefficient is obtained. The degree of performance impairment caused by the temperature rise data under test to lubrication and cooling performance and processing performance is extracted from historical test data to obtain the degree of impairment index. Based on the correlation and change characteristics between the measured temperature rise data and the measured damage index, the correlation and damage coefficient between the measured temperature rise data and the measured damage index is obtained.
6. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 5, characterized in that, Step S3 includes: Acquire the pretreatment tension data, pretreatment contact angle data, and pretreatment viscosity data of the charged cutting fluid droplets undergoing electrostatic micro-lubrication in the current period, and acquire the current electrostatic parameters and current cutting fluid data of the applied electrostatic field in the current period. Input the pretreatment tension data, pretreatment contact angle data, pretreatment viscosity data, current electrostatic parameters, and current cutting fluid data into the charged liquid phase penetration into the capillary model for testing to obtain pretreatment permeability characteristic data. Based on feature dataset one, the preprocessed dataset one belonging to the current period is statistically analyzed, and the initial tension data is obtained based on the preprocessed tension data, preprocessed dataset one, and the first correlation influence factor. Based on feature dataset two, the preprocessed dataset two to which the current period belongs is statistically analyzed. Based on the preprocessed contact angle data, preprocessed dataset two, and the second correlation influencing factor, the initial measured contact angle data is obtained. The ambient temperature value of the current period is detected to obtain the initial temperature value. Based on the preprocessed viscosity data, the initial temperature value, and the third correlation influence factor, the initial viscosity data is obtained.
7. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 6, characterized in that, Step S3 also includes: Based on the process parameters to be tested, the current process parameters are obtained. The current process parameters, initial tension data, initial contact angle data, and initial viscosity data are input into the particle size threshold prediction model for testing to obtain the current particle size prediction threshold. Based on the key parameters to be tested, the current key parameters are statistically determined. The preprocessed permeability characteristic data, the current particle size prediction threshold, and the current key parameters are then input into the lubrication layer distribution prediction model for testing to obtain the current lubrication layer predicted distribution characteristics. Obtain the material characteristic parameters of the current workpiece to which the current test workpiece belongs, and obtain the current hardness influence value based on the current workpiece material characteristic parameters, the current lubrication layer predicted distribution characteristics, and the hardness influence coefficient; Based on historical test data and the material characteristics of the workpiece to be tested, a prediction model for lubrication and cooling processing performance is established.
8. The method for synergistic optimization of electrostatic micro-lubrication and grinding process parameters according to claim 7, characterized in that, Step S3 also includes: The current predicted distribution characteristics of the lubrication layer, current process parameters, current workpiece material characteristic parameters, and current hardness influence value are input into the lubrication and cooling processing performance prediction model for testing, and the lubrication and cooling processing performance evaluation index is obtained. Based on the current predicted distribution characteristics of the lubrication layer, the current process parameters, and the additional temperature rise correlation coefficient, the current predicted temperature rise value is obtained; Based on the current predicted temperature rise value and the associated impact damage coefficient, the performance index impact value is obtained; Based on the current temperature rise prediction value, error compensation is performed on the lubrication and cooling processing performance evaluation index to obtain the overall performance index evaluation result; Based on the overall performance index evaluation results, the electrostatic micro-lubrication parameters and grinding process parameters are synergistically optimized and controlled to obtain the parameter synergistic optimization results.