Solid-liquid two-phase nanofluid coolant optimization method for dynamic cooling of machine tool spindle

Abstract

A kind of solid-liquid two-phase nanofluid cooling fluid optimization method for machine tool spindle dynamic cooling, first carry out the sensitivity / correlation analysis of parameter variable, find out the nanofluid parameter that is strongly related to convective heat transfer characteristics;Then describe the percolation structure of nanoparticle aggregation, based on thermal conductivity model and effective viscosity model into finite element simulation;Then analyze and determine the key parameters of nanofluid in spindle cooling, qualitative selection of the optimization combination of key parameters;According to the "two-step method", nanofluid is prepared, and convective heat transfer experiment is carried out for nanofluid, and the experimental results of nanofluid are compared transversely, and the nanofluid with maximum average convective heat transfer coefficient is selected quantitatively;The present application greatly improves the convective heat transfer rate of the cooling fluid in the machine tool spindle, makes the precision machine tool spindle thermal error in high-speed machining long-term stable, and then eliminates the influence of spindle thermal error on machining precision of machine tool through tool.

Description

Technical Field

[0001] This invention relates to the field of precision machine tool technology, and more specifically to an optimization method for a solid-liquid two-phase nanofluid coolant for dynamic cooling of machine tool spindles. Background Technology

[0002] Precision machine tools typically require a considerable preheating time to reach thermal equilibrium before machining begins. Tool setting is used to adjust the relative position between the workpiece coordinate system and the machine tool coordinate system, compensating for the stable portion of the spindle thermal error. Therefore, during machining, the fluctuation of spindle thermal error is the main factor affecting end-effector accuracy. Although existing active spindle cooling strategies can significantly reduce thermal error, they are difficult to maintain long-term stability of thermal error. The limitations of current active spindle cooling methods can be summarized as follows: ① Active cooling based on isothermal or offline model control is difficult to adapt to dynamically changing spindle thermal characteristics and cannot maintain thermal error stability over a long period; ② Real-time dissipation of heat generated by various heat sources can minimize thermal deformation of the spindle structure, but it leads to high cooling energy consumption and additional costs such as complex flow channel structures and multi-loop cooling systems; ③ Additional cooling devices may alter the spindle's dynamic characteristics.

[0003] Therefore, a strategy (1. Mohan Lei, Feng Gao, Yan Li, et al. Feedback control–based active cooling with pre-estimated reliability for stabilizing the thermal error of a precision mechanical spindle[J].The International Journal of Advanced Manufacturing Technology,2022,121:2023–2040. 2. Mohan Lei, Yang J, Gao F, et al. Closed-loop thermal error control with a physical-based ensemble model for the precision spindle of a machine tool.The International Journal of Advanced Manufacturing Technology,2023,125:1859–1877.) treats active cooling and precision spindle thermal error as a closed-loop control system, and considers changes in ambient temperature and heat generation from heat source components as disturbances. Utilizing the ability of the closed-loop system to adjust back to its original equilibrium state after disturbance, the spindle thermal error can be kept stable over a long period. By compensating for the stable portion of the thermal error through tool setting, the impact of spindle thermal error on the machining accuracy of the machine tool end effector is limited to a very small range. The aforementioned strategy aims to maintain long-term stability of the spindle's thermal error, without minimizing the overall thermal deformation of the spindle structure, thus resulting in low cooling energy consumption. It can be achieved using single-loop cooling without modifying the flow channel structure or adding additional cooling devices, and without altering the spindle's dynamic characteristics. This closed-loop thermal error control system uses the cooling system as the actuator, with cooling fluid temperature, pressure, and flow rate as control variables, and thermal error as the controlled variable. Real-time feedback is provided through online temperature monitoring and thermal error data model output. However, during machining, the spindle is affected by heat generated by the motor, bearings under load, and cutting, leading to drastic increases and changes in thermal error and more complex thermal characteristics. The existing maximum cooling rate (saturation limit) is insufficient to significantly reduce the spindle's thermal error, and the saturation problem of the cooling actuator is very serious. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the present invention aims to provide an optimization method for solid-liquid two-phase nanofluid coolant for dynamic cooling of machine tool spindle, so as to significantly improve the convective heat transfer rate of coolant in machine tool spindle, thereby stabilizing the thermal error of precision machine tool spindle in high-speed machining for a long time, and then eliminating the influence of spindle thermal error on machine tool machining accuracy through tool setting.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] An optimization method for solid-liquid two-phase nanofluid coolant for dynamic cooling of machine tool spindles includes the following steps:

[0007] Step 1: Using publicly available and experimentally measured nanofluid convective heat transfer data, perform sensitivity / correlation analysis of parameter variables using clustering methods to identify nanofluid parameters that are strongly correlated with convective heat transfer characteristics.

[0008] Step 2: Describe the percolation structure of nanoparticle polymerization based on the Prasher cluster model, and base the thermal conductivity k on the Koo-Kleinstreuer model. nf Model and Masoumi effective viscosity μ nf The model is then substituted into the finite element simulation.

[0009] Step 3: Using FLUENT software and its UDF (User Defined Function), a solid-liquid two-phase fluid numerical simulation based on the Lagrangian–Eulerian model was carried out to analyze the influence of the working parameters and preparation parameters of the nanofluid on heat conduction and convective heat transfer, and to determine the key parameters of the nanofluid in spindle cooling.

[0010] Step 4: Using ANSYS software, conduct orthogonal simulation experiments on the key parameters determined in Step 3, qualitatively analyze the optimal combination of parameters that makes the convective heat transfer coefficient h have a significant increasing trend, and thus qualitatively select several optimal combinations of key parameters of nanofluids.

[0011] Step 5: Based on the optimized combination of key nanofluid parameters in Step 4, prepare nanofluids using the "two-step method";

[0012] Step 6: Conduct convective heat transfer experiments on the nanofluids prepared in Step 5, compare the experimental results of the nanofluids laterally, and quantitatively select the nanofluid with the largest average convective heat transfer coefficient h.

[0013] The operating parameters in step 3 include coolant temperature and flow rate, and the preparation parameters include nanoparticle material properties, size, volume fraction, and substrate liquid properties.

[0014] The "two-step method" for preparing nanofluids in step 5 specifically involves:

[0015] 5.1) Add metal or metal compound nanoparticles to the base solution according to volume fraction;

[0016] 5.2) Add dispersant, stir the nanoparticle suspension with a magnetic stirrer, and use an ultrasonic pulse oscillation device with a power of 110W and a frequency of 40±5kHz to ultrasonically oscillate the suspension for more than 30 minutes to make the nanoparticles uniformly dispersed.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] The coupling effect of nanofluid parameters and operating conditions on its convective heat transfer rate is quite complex. This invention adopts a method of qualitative analysis of data and finite element method and quantitative research of convective heat transfer experiment to effectively optimize the parameters of nanofluid to improve its heat transfer efficiency in the main shaft channel, thereby significantly improving the actuator saturation problem in the main shaft closed-loop thermal error control. Therefore, it has the advantages of high efficiency, accuracy and practicality. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention.

[0020] Figure 2 This is a schematic diagram of a nanofluid convection heat transfer experiment according to an embodiment of the present invention, where points 1, 2, 3, 4, 5, and 6 are the installation positions of the magnetic temperature sensor.

[0021] Figure 3 The following is a simulation diagram illustrating the improvement in spindle cooling efficiency after optimizing the nanofluid in this embodiment of the invention; wherein, (a) is a thermal deformation field cloud map of the spindle; (b) shows the effect of applying maximum cooling rate (i.e., minimum coolant temperature 5°C) to the spindle within one control cycle (300 seconds) after 30 minutes of constant temperature preheating; (c) shows the effect and comparison of reducing spindle thermal error by maximum cooling rate within one control cycle when using nanofluid and single-phase coolant respectively. Detailed Implementation

[0022] The technical solution of the present invention will now be clearly and completely described in conjunction with the accompanying drawings and embodiments.

[0023] Reference Figure 1 An optimization method for solid-liquid two-phase nanofluid coolant for dynamic cooling of machine tool spindles includes the following steps:

[0024] Step 1: Using publicly available and experimentally measured nanofluid convective heat transfer data, perform data cleaning, and use clustering methods to conduct sensitivity / correlation analysis of parameter variables to identify nanofluid parameters that are strongly correlated with convective heat transfer characteristics.

[0025] This embodiment removes outliers from experimental data using BSOD (Booting-based Outlier Detection) and employs nearest neighbor imputation to fill missing data, thus cleaning the experimental data. For key parameters of nanofluids, such as solid particle material properties, size, morphology and volume fraction, substrate liquid properties, and dispersant / additive type, sensitivity / correlation analysis of the parameter variables is performed using Random Forest Regression (RFR) and Fuzzy Clustering methods to identify nanofluid parameters that are strongly correlated with convective heat transfer characteristics.

[0026] Step 2: Describe the percolation structure of nanoparticle polymerization based on the Prasher cluster model, and base the thermal conductivity k on the Koo-Kleinstreuer model. nf Model (reflecting Brownian motion of particles) and Masoumi effective viscosity μ nf The model is then substituted into the finite element simulation.

[0027] This embodiment is based on the thermal conductivity k of Koo-Kleinstreuer. nf Model and Masoumi effective viscosity μ nf The models are as follows:

[0028]

[0029]

[0030] In equation (1), the first term represents the enhanced heat conduction based on the permeation structure, and the second term represents the influence of the Brownian motion of the particles; in the above equation, k a The thermal conductivity of the particulate polymer structure (calculated based on the Nan model), d p Where is the diameter of the nanoparticle, v B The average Brownian motion speed of the particles: μ b and k b These represent the viscosity and thermal conductivity of the substrate fluid, ρ. p With ρ b These are the material densities of the metal particles and the substrate fluid, respectively. β and c represent the volume fractions of nanoparticles, particle aggregates, and liquid, respectively. v For specific heat capacity, C 0～4 These are empirical or undetermined constant parameters, where T is temperature and P is pressure;

[0031] Step 3: Using FLUENT software and its UDF (User Defined Function), a numerical simulation analysis of solid-liquid two-phase fluid based on the Lagrangian–Eulerian model is conducted to analyze the influence of heat conduction and convective heat transfer on the working parameters (coolant temperature, flow rate, etc.) and preparation parameters (nanoparticle material properties, size, volume fraction, substrate liquid properties, etc.) of the nanofluid, and to determine the key parameters of the nanofluid in spindle cooling.

[0032] Step 4: The coupling effect of nanofluid parameters on its convective heat transfer characteristics is complex, including the fluid thermal conductivity k. nf The larger the viscosity μ nf The smaller the value, the greater the convective heat transfer coefficient h between it and the wall; however, changing key parameters of nanofluids, such as increasing the volume fraction of nanoparticles, may decrease the thermal conductivity k. nf and viscosity μ nf Since the convective heat transfer coefficient h increases simultaneously, it is impossible to determine whether it increases or decreases. Therefore, based on the two-phase fluid empirical model and simulation method in steps 2 and 3, orthogonal simulation experiments of the key parameters determined in step 3 are carried out using ANSYS and Fluent software under the typical working conditions of the spindle in machine tool processing. The optimal combination of parameters that makes the convective heat transfer coefficient h have a significant increasing trend is qualitatively analyzed, and thus several optimal combinations of key parameters of nanofluids are qualitatively selected.

[0033] This embodiment uses the Mechanical and Transient Thermal modules in ANSYS to simulate and calculate the transient thermal load on the spindle; numerical simulation of nanofluids within the spindle flow channel is performed using FLUENT software; and the thermal conductivity k is substituted according to the actual structure and dimensions of the spindle flow channel. nf and effective viscosity μ nf The model (determined in step 2) uses the Mixture "solid-liquid" two-phase flow model to reflect the effects of frictional collision, Brownian motion, diffusion, etc. between particles and between particles and the wall. Orthogonal simulation experiments are carried out on the key parameters of the nanofluid in the main axis flow channel (determined in step 3) to analyze the optimized combination of several key parameters that may cause the convective heat transfer coefficient h to have a significant increasing trend.

[0034] Step 5: Based on several optimized combinations of key nanofluid parameters from Step 4, prepare nanofluids using a two-step method, specifically:

[0035] 5.1) Add metal or metal compound nanoparticles to the base solution according to volume fraction;

[0036] 5.2) Add dispersant, use a magnetic stirrer to fully stir the nanoparticle suspension, and use a high-capacity ultrasonic pulse oscillation device with a power of 110W and a frequency of 40±5kHz to ultrasonically oscillate the suspension for more than 30 minutes to make the nanoparticles uniformly dispersed and ensure the dispersion stability of the nanofluid.

[0037] Step 6: Conduct convective heat transfer experiments on the nanofluids prepared in Step 5, compare the experimental results of the nanofluids laterally, and quantitatively select the nanofluid with the largest average convective heat transfer coefficient h.

[0038] This embodiment is specifically as follows:

[0039] 6.1) Measure the fluid temperature at the inlet and outlet of the spiral coil using a threaded temperature sensor; install (by welding) a magnetic temperature sensor at equidistant axial positions at the top and bottom of the copper spiral coil.

[0040] 6.2) Measure the temperature difference between the inlet and outlet of the spiral coil and calculate the change in heat energy of the coolant flowing into and out of the spiral coil; combine the measured temperature gradient of the outer wall of the spiral coil to obtain the average convective heat transfer coefficient between the coolant and the tube wall; compare the experimental results of nanofluids laterally and select the nanofluid with the largest average convective heat transfer coefficient.

[0041] like Figure 2 As shown, in this embodiment, a threaded temperature sensor is integrated into the cooling circuit to measure the inlet and outlet temperatures; a magnetic temperature sensor is installed (by welding) at equidistant axial positions at the top and bottom of the copper spiral coil.

[0042] The beneficial effects of this embodiment:

[0043] This invention first performs data analysis and sets the thermal conductivity k... nf and viscosity μ nf The empirical model was substituted into Fluent two-phase fluid finite element simulation to qualitatively analyze which parameters of the nanofluid have the most significant impact on the convective heat transfer rate and what parameter configurations lead to a significant increase in the convective heat transfer coefficient of the nanofluid. This allowed for the selection of optimal combinations of nanofluid parameters that could potentially achieve the desired convective heat transfer coefficient. Subsequently, convective heat transfer experiments were conducted to quantitatively compare the actual convective heat transfer efficiency of nanofluids with different parameter optimization combinations, thereby selecting the optimal parameter combination for preparing nanofluids used in spindle cooling.

[0044] Reference Figure 3 In the early stages, finite element simulation was used ( Figure 3 (a) The effect of nanofluids on the cooling effect of the spindle during machining was analyzed; Figure 3In (b), after preheating the 16°C constant temperature coolant for 30 minutes, the spindle is cooled at the maximum rate (i.e., the lowest coolant temperature that a typical liquid chiller can reach is 5°C) for 300 seconds, which is one control cycle. Figure 3 (c) demonstrates the effect of cooling the spindle at maximum rate for 300 seconds. Compared to single-phase water, the nanofluid coolant reduces the convective heat transfer rate within the spindle channel from 5233 W / m. 2 • Temperature increased to 7745W / m 2 At ℃, under spindle operating conditions of 8000 rpm, 10000 rpm and 12000 rpm, within one control cycle (300 seconds), nanofluid cooling reduced thermal errors by 1.59 μm, 1.3 μm and 0.79 μm, respectively; in comparison, single-phase fluid (water) cooling reduced thermal errors by 0.71 μm, 0.40 μm and -0.10 μm, respectively.

[0045] The above results reflect that the present invention can effectively improve the spindle cooling efficiency, thereby significantly improving the actuator saturation problem in the spindle closed-loop thermal error control and enhancing its effectiveness in improving machine tool machining accuracy.

Claims

1. A method for optimizing a solid-liquid two-phase nanofluid coolant for dynamic cooling of a machine tool spindle, characterized in that, Includes the following steps: Step 1: Using publicly available and experimentally measured nanofluid convective heat transfer data, perform sensitivity / correlation analysis of parameter variables using clustering methods to identify nanofluid parameters that are strongly correlated with convective heat transfer characteristics. Step 2: Describe the percolation structure of nanoparticle polymerization based on the Prasher cluster model, and base the thermal conductivity k on the Koo-Kleinstreuer model. nf Model and Masoumi effective viscosity μ nf The model is then substituted into the finite element simulation. Step 3: Using FLUENT software and its UDF (User Defined Function), a solid-liquid two-phase fluid numerical simulation based on the Lagrangian–Eulerian model is conducted to analyze the influence of the operating parameters and preparation parameters of the nanofluid on heat conduction and convection heat transfer, and to determine the key parameters of the nanofluid in spindle cooling. The operating parameters include coolant temperature and flow rate, and the preparation parameters include nanoparticle material properties, size, volume fraction, and substrate liquid properties. Step 4: Using ANSYS software, conduct orthogonal simulation experiments on the key parameters determined in Step 3, qualitatively analyze the optimal combination of parameters that makes the convective heat transfer coefficient h have a significant increasing trend, and thus qualitatively select several optimal combinations of key parameters of nanofluids. Step 5: Based on the optimized combination of key nanofluid parameters from Step 4, prepare nanofluids using a "two-step method"; the specific details of the "two-step method" for preparing nanofluids in Step 5 are as follows: 5.1) Add metal or metal compound nanoparticles to the base solution according to volume fraction; 5.2) Add dispersant, stir the nanoparticle suspension with a magnetic stirrer, and use an ultrasonic pulse oscillation device with a power of 110W and a frequency of 40±5kHz to ultrasonically oscillate the suspension for more than 30 minutes to make the nanoparticles uniformly dispersed. Step 6: Conduct convective heat transfer experiments on the nanofluids prepared in Step 5, compare the experimental results of the nanofluids laterally, and quantitatively select the nanofluid with the largest average convective heat transfer coefficient h.

2. The method according to claim 1, characterized in that, Step 6 specifically includes: 6.1) Measure the fluid temperature at the inlet and outlet of the spiral coil using a threaded temperature sensor; install a magnetic temperature sensor at equidistant axial positions at the top and bottom of the copper spiral coil. 6.2) Measure the temperature difference between the inlet and outlet of the spiral coil and calculate the change in heat energy of the coolant flowing into and out of the spiral coil; combine the measured temperature gradient of the outer wall of the spiral coil to obtain the average convective heat transfer coefficient between the coolant and the tube wall; compare the experimental results of nanofluids laterally and select the nanofluid with the largest average convective heat transfer coefficient.

Images