Ultra-precision Machining Research Articles

Explainable and generalizable AI-driven multiscale informatics for dynamic system modelling

Ultra-precision machining requires system modelling that both satisfies explainability and conforms to data fidelity. Existing modelling approaches, whether based on data-driven methods in present artificial intelligence (AI) or on first-principle knowledge, fall short of these qualities in high-demanding industrial applications. Therefore, this paper develops an explainable and generalizable ‘grey-box’ AI informatics method for real-world dynamic system modelling. Such a grey-box model serves as a multiscale ‘world model’ by integrating the first principles of the system in a white-box architecture with data-fitting black boxes for varying hyperparameters of the white box. The physical principles serve as an explainable global meta-structure of the real-world system driven by physical knowledge, while the black boxes enhance local fitting accuracy driven by training data. The grey-box model thus encapsulates implicit variables and relationships that a standalone white-box model or black-box model fails to capture. Case study on an industrial cleanroom high-precision temperature regulation system verifies that the grey-box method outperforms existing modelling methods and is suitable for varying operating conditions.

Effects of Ar ion bombardment on the structure and properties of β-quartz

β-quartz is an important component that affects the physical and chemical properties of glass-ceramics. The material removal process of ion beam machining is complicated as one of the ultra-precision machining forms. Therefore, it is important to understand the removal mechanism of ion beam machining materials. The methods of molecular dynamics (MD) and first-principles were adopted to investigate the microscopic mechanism of ion beam bombardment of the β-quartz structure. The damage evolution of structures and the evolution of material properties caused by defects respectively from the atomic and electronic levels. Through MD analysis, we find that the random collision between incident ions and matrix atoms can lead to the formation of sputtered atoms. The incidence of ions will increase the disorder degree of the matrix, and the influence on the disorder degree at different depths of the matrix is different. The degree of disorder increases obviously in the range of 50–70 Å from the surface of the matrix. It has a significant impact on the surface topology of the matrix, the number of defects, surface roughness, and matrix density. The results of first-principle calculation show that the refraction, absorption and other characteristics of the system are significantly different with different doping forms, and the energy loss is most affected by substituted doping. And the substituted doping defects have the most significant effects on the structural energy loss. Among, the Ar-substituted O doping structure leads to higher energy loss, and the peak energy loss intensity reaches 5.146. In contrast, the Ar-substituted Si doping structure causes the least energy loss, the intensity is 2.543. This study provides a theoretical basis for the selection of parameters and the realization of ultra-smooth surface in actual ion beam machining.

Investigation of the error averaging effect on design and assembly of external slider closed hydrostatic guideways

Hydrostatic guideways are widely used in ultra-precision machine tools. However, in most cases, assemblers must rely on their experience to adjust the manufacturing errors of the guide rail and slider, which indirectly controls the error motion. To ensure assembly accuracy and improve efficiency, an error motion calculation model is established based on the external slider closed hydrostatic guideway. The accuracy of the model is verified through experiments, and the main factors that affect the error motion are analysed. Finally, a traversal algorithm is used to calculate the distribution of the error averaging coefficient. Once the error motion index is clarified, the design and assembly process of the guideway can be constrained by the error averaging coefficient.

Bidirectional long-short term memory predictor for material removal rate in computer-controlled optical surfacing

In Computer-Controlled Optical Surfacing technology, the precision of the removal function directly affects the accuracy of computer-aided processing software predictions, which in turn influences subsequent polishing machine processing. A key parameter for constructing the removal function is the material removal rate, which is often challenging to obtain accurately. Currently, the Preston equation is widely used to describe the principles of material removal. However, as a linear equation that omits many factors, it struggles to accurately model the removal function in complex machining scenarios. Therefore, this paper proposes a hybrid neural network model combining Convolutional Neural Networks and Bidirectional Long Short-Term Memory to predict the material removal rate. The model's parameters are optimized using an improved Grey Wolf Optimization algorithm, ultimately establishing a removal function closely consistent with an actual removal function. We first tested our method on the PHM2016 Data Challenge dataset, achieving a mean squared error of 6.19 and an R2 of 0.9949, outperforming other mainstream neural network prediction models developed in recent years. Additionally, we further validated the performance of the neural network using a small grinding head polishing dataset, achieving MSE and R2 values of 1.9035 and 0.99902, respectively. Finally, we applied this method to construct the removal function on an actual small grinding head production line. Compared to the traditional Preston equation-based removal function, the predicted residual surface's PV and RMS errors were reduced from 28.24 % to 35.58 %–4.563 % and 4.86 %, respectively. These validation results demonstrate that the proposed method not only facilitates easier acquisition of the removal function model but also significantly enhances the accuracy of computer-aided processing software predictions, thereby better guiding ultra-precision machining processes.

Interference-enhanced micro-vision-based single-shot imaging of five degrees-of-freedom error motions for ultra-precision rotary axes

The measurement of five degrees-of-freedom (5-DOF) error motions, including radial, axial, and tilt motions, is crucial for ultra-precision rotary axes, which are key components of ultra-precision machine tools and instrumentation. In this study, we propose an interference-enhanced micro-vision technique to concurrently derive the 5-DOF error motions from a single-shot two-dimensional image, which was captured by a standard industrial camera equipped with an interference objective lens. By consolidating the essential features into a single optical path, the interference-enhanced micro-vision technique ingeniously merges machine micro-vision and modified white-light interference to detect in-plane and out-of-plane motions. Numerical simulations demonstrated, the basic principle for deriving the 5-DOF error motions, and the magnification of objective lens had inconsistent effects on the measurement accuracy for the radial and tilt motions, i.e. higher magnification led to higher radial accuracy but lower tilt accuracy. As practical application, the error motion detection capability was demonstrated by simultaneously measuring the 5-DOF synchronous and asynchronous error motions for a typical air bearing spindle at rotation speeds of 8.33, 108.33, and 308.33 rpm. The synchronous errors were nearly identical at various spindle speeds. However, because of system dynamics, increased vibrations were observed to be superimposed on the basic tilt error motions as the spindle speeds increased, which were verified by the vibration marks imprinted on the turned surfaces. For the 5-DOF motion measurements, the least-square fitting using large-volume edge and greyscale data of the captured image enabled super-high resolutions, despite using a camera with a relatively large pixel size and low bit depth. These results demonstrate that the proposed interference-enhanced micro-vision technique is a simple and effective tool for measuring spatial error motions in ultra-precision rotary axes.

Molecular dynamics simulation and experimental study of the material machinability characteristics and crack propagation mechanisms for fused silica double nanoscratches

Fused silica is a high-purity, hard and brittle, difficult-to-machine amorphous material that is widely used in the aerospace, optical instruments, semiconductors, microelectronics and other important fields. However, the influence of adjacent scratches on material deformation and subsurface damage during ultraprecision machining by abrasive particles is not well understood. This study aims to investigate the influence of micron-spaced double scratches on the deformation and damage for fused silica through molecular dynamics simulations and double nanoscratch experiments. An amorphous model of fused silica is created using a two-step simulated annealing method with three-directional periodic boundary conditions. Variable loaded double scratches with different nanometer distances are simulated on the surface of the model, and variable loaded double scratch experiments with varying micron-sized distances between scratches are conducted for fused silica wafers. The simulation and experimental results reveal that the deformation of fused silica during nanoscale double scratching involves both elastic and plastic deformation, and the surface and subsurface damage primarily consists of radial, lateral, Hertzian, and median cracks. The distance between the double scratches affects the load variation, surface and subsurface crack propagation, extent of material damage and removal rate, and surface roughness. Examination of the material subsurfaces using the FIB-TEM sample preparation and testing technique shows that the distance between the double scratches also influences the subsurface damage to the material.

Modeling and simulation of temperature field in high-shear and low-pressure grinding with liquid-body-armor-like grinding wheel

High-shear and low-pressure grinding using liquid-body-armor-like grinding wheel is a new grinding process. It has a great potential in the field of precision and ultra-precision machining. To clarify the unclear grinding heat characteristics, heat generation mechanism, and heat source distribution of the liquid-body-armor-like grinding wheel, a temperature field model was developed for high-shear and low-pressure grinding with ABAQUS. The thermal characteristics of the new grinding wheel were investigated by studying the heat flux and convective loads under different grinding parameters. The effectiveness of the heat model was validated through high-shear and low-pressure grinding experiments. The simulation results closely matched with the experimental results, with an average error of 5.84 % and a minimum error of 0.94 %. The grinding temperature decreased gradually from the centre of the heat source towards the edge of the workpiece in both horizontal and vertical directions. The temperature raised with the increase of the grinding velocity, ranging from 39.66 °C at 4 m/s to 77.82 °C at 14 m/s. It decreased with the workpiece feed rate, dropping from 39.66 °C at 500 mm/min to 32.00 °C at 3000 mm/min. As the normal grinding force increased, it gradually increases from 30.80 °C to 39.66 °C. The grinding temperature of the liquid-body-armor-like grinding wheel was lower than that of the traditional grinding wheel.

Anisotropy Dependence of Material Deformation Mechanisms in Nanoscratching Monocrystalline BaF2: Experiments and Atomic Simulations.

Monocrystalline barium fluoride (BaF2), known for its exceptional optical properties in the infrared spectrum, exhibits anisotropy that influences surface quality and material removal efficiency during ultraprecision machining. This research explores the impact of anisotropy on the deformation and removal mechanisms of monocrystalline BaF2 by integrating nanoscratch tests with molecular dynamics (MD) simulations. Nanoscratch tests conducted on variously oriented monocrystalline BaF2 surfaces using a ramp loading mode facilitated the identification of surface cracks and a systematic description of material removal behaviors. This study elucidates the effect of crystal orientation on the ductile-brittle transition (DBT) of monocrystalline BaF2, further developing a critical depth prediction model for DBT on the (111) crystal plane to reveal the underlying anisotropy mechanisms. Moreover, nanofriction and wear behaviors in monocrystalline BaF2 are found to be predominantly influenced by scratch direction, crystal surface, and applied load, with the (110) and (100) planes showing pronounced frictional and wear anisotropy. A coefficient of friction model, accounting for the material's elastic recovery, establishes the intrinsic relationship between anisotropic friction and wear behaviors, the size effect, and scratch direction. Lastly, MD modeling of nanoscratched monocrystalline BaF2 reveals the diversity of dislocations and strain distributions along the (111) [-110] and [-1-12] crystal directions, offering atomic scale insights into the origins of BaF2 anisotropy. Thus, this study provides a theoretical foundation for the efficient processing of fluorine-based infrared optic materials exhibiting anisotropy.

Study on the elastic impact deformation behavior and damage mechanism during the polishing process of PS@CeO2 core-shell abrasive

Discrete element simulations were carried out to investigate the elastic impact damage mechanism of PS@CeO2 core-shell abrasive (CSCAP). Aiming at understanding the deformation behavior and damage mechanism of CSCAP during the polishing process, effects of the impact velocity, impact angle on the rates of deformation and recovery deformation, stress distribution, the number of microcracks were systematically investigated. Tensile cracks formed owing to the CSCAP deformation was gradually increasing when impacting the workpiece downwards. After the CSCAP reached the maximum deformation, macroscopic cracks at the interface were extended gradually. Owing to the redistribution of the compressive force, the interface was observed to predominantly occur in shear cracks, while the shell layer was found to be distributed with tensile cracks. The number of microcracks increase with the increase in impact speed, the impact angle was within the range of 15° to 45°. This investigation enhances the comprehension of nanoscale deformation damage mechanism in CSCAP during ultra-precision machining processes.

Influence of water on the machinability of calcium fluoride crystal

Calcium fluoride (CaF2) optical components are produced by ultra-precision machining, but the often-neglected ambient water may potentially affect its processing. This work provides a systematic study on the role of water in the machinability of single-crystal CaF2. Micro-scratching experiments revealed poor machinability and reduced deformability under water adsorption. Atomic scratching simulations reproduced the mechanical deformation and identified the enhancement in dislocation nucleation with water, which led to the reduction in deformability due to dense cross-linking of dislocations. Atomic scale cutting simulations also identified easier formation of subsurface cracks under water adsorption. With the aid of mechanical schematic models based on the simplification of the scratching process, this work revealed that while water enhances the plastic flow of the material during machining, the accumulation of dislocations creates stress conditions sufficient to induce material fracture, which is a unique influence of water adsorption on the reduction in machinability of CaF2.

In-situ TEM investigation of dislocation healing and recrystallization in nanoscratched silicon at elevated temperatures up to 800 °C

Nanoscratching introduces detrimental surface and subsurface defects like amorphous regions, dislocations, and stacking faults in monocrystalline silicon, hindering its application in microelectronics and high-performance optics. This study leverages in-situ transmission electron microscopy to unveil the thermal evolution of these defects in atomic scale. A key finding is the amorphous phase recrystallization starting at ∼500 °C. Epitaxial growth from the crystalline-amorphous boundary, guided by adjacent crystal planes, restores the original diamond structure phase. By 700–800 °C, almost complete recrystallization occurs, maintaining similar interplanar spacing despite residual crystal distortions and dislocations. Notably, heating above 600 °C results in the gradual vanishing of stacking faults, suggesting a dynamic thermal evolution of the crystal defects induced by surface nanoscratching. This work demonstrates thermal annealing as a promising strategy to mitigate nanoscratch-induced defects, paving the way for defect-free-surface of silicon components in ultra-precision machining processes. It offers valuable insights into the interplay between nanoscratching, temperature, and defect evolution, laying the groundwork for surface and subsurface defects elimination in silicon under thermal fields.

Scanning Electrochemical Probe Lithography for Ultra-Precision Machining of Micro-Optical Elements with Freeform Curved Surface.

Two challenges should be overcome for the ultra-precision machining of micro-optical element with freeform curved surface: one is the intricate geometry, the other is the hard-to-machining optical materials due to their hardness, brittleness or flexibility. Here scanning electrochemical probe lithography (SECPL) is developed, not only to meet the machining need of intricate geometry by 3D direct writing, but also to overcome the above mentioned mechanical properties by an electrochemical material removal mode. Through the electrochemical probe a localized anodic voltage is applied to drive the localized corrosion of GaAs. The material removal rate is obtained as a function of applied voltage, motion rate, scan segment, etc. Based on the material removal function, an arbitrary geometry can be converted to a spatially distributed voltage. Thus, a series of micro-optical element are fabricated with a machining accuracy in the scale of 100 s of nanometers. Notably, the spiral phase plate shows an excellent performance to transfer parallel light to vortex beam. SECPL demonstrates its excellent controllability and accuracy for the ultra-precision machining of micro-optical devices with freeform curved surface, providing an alternative chemical approach besides the physical and mechanical techniques.

Dynamic simulation and active vibration control design of an ultra-precision fly-cutting machine tool

Dynamic simulation and active vibration control design of an ultra-precision fly-cutting machine tool

Interfacial Heterogeneous Microstructure and Subsequent Machinability of 316L/CuSn10 Bimetallic Structure Manufactured by Laser Powder Bed Fusion

Additively manufactured bimetallic structures combine the advantages of dissimilar materials and can achieve localized properties through a customized composition distribution. However, additively manufactured parts may still lack the dimensional accuracy and surface integrity essential for precision mechanical assemblies that the post-machining process can address. Therefore, this study aims to systematically investigate the microstructure and machinability of 316L/CuSn10 bimetallic structures fabricated using laser powder bed fusion. The results show that the fusion zone of the bimetallic structure had refined grains of microscale size owing to the mixture of the primary elements of the bimetals, which resulted in the highest microhardness of 3.4 GPa. The difference in microstructure and microhardness between the single-material and fusion zones also causes significant differences in the cutting response during the ultraprecision process. The 316L stainless steel side exhibited the highest cutting force and more severe material accumulation in the chips. The cutting force drops when cutting through the fusion zone, with an observable fracture in the chips and separation of dissimilar materials on the machined grooves, indicating that the heterogeneous properties of additively manufactured 316L/CuSn10 bimetallic structures pose challenges to the improvement of surface quality. The simulation results also showed that stress accumulation occurred in the tool path through the fusion zone owing to the higher yield strength and hardness of stainless steel, indicating that lower cutting speeds and depths of cut are favorable for reducing cutting force and improving surface quality. This study provides deep insight into the microstructure evolution mechanism and a theoretical basis for improving the surface quality of additively manufactured bimetallic structures using an ultraprecision machining process.

Subsurface damage evolution of β-Ga2O3 (010) substrates during lapping and chemical mechanical polishing

For semiconductor single crystal substrates, subsurface damage (SSD) is an important factor for device failure. For β-Ga2O3 with two cleavage planes, the SSD is the key issue that needs to be addressed. In this work, β-Ga2O3 (010) substrates were lapped and polished precisely. The formation and evolution mechanisms of SSD were analyzed by cross-section transmission electron microscopy and wet chemical etching techniques. The linear distribution of scratches was found on the substrates after lapping. After chemical mechanical polishing (CMP), the surface scratches were eliminated. However, the SSD layer may still exist under the surface. Chemical etching was developed to identify the "invisible damage" to the CMP-finished (010) substrate. The SSD including nanocrystalline, tilt boundary, twist boundary, and high-density dislocations in β-Ga2O3 was first reported. Distinctively, there was no slip occurrence in the SSD of the (010) plane different from the (100) and (001) planes. The Schmid factor of multiple slip systems was calculated to analyze the cause of no slip on the subsurface. The formation mechanism of SSD was believed that when the slip system could not be activated, the stress was released through rotation and distortion of the crystal plane. The accumulation of grain boundary dislocations at twist grain boundary junctions provided a driving force for the formation of nanovoids. The relationship of substrate damage evolution and crystal strain with different machining parameters was established by high-resolution X-ray diffraction and Raman spectra. A non-damaging β-Ga2O3 (010) substrate with a surface roughness of less than 0.2 nm was obtained by the optimized CMP process. The research results had guiding significance for damage control of β-Ga2O3 and other brittle crystals in ultra-precision machining.

Digital-twin-driven intelligent tracking error compensation of ultra-precision machining

In ultra-precision machining (UPM), linear axis tracking affects contour accuracy and final machining quality. Traditional error modeling is complicated by the identification of numerous unknown parameters linked to nonlinear characteristics in the linear feed axes. To fill this gap, this study proposed a digital-twin-driven framework integrating the developed G-code interpreter and the deep learning model to achieve real-time tracking error compensation for UPM. To enhance the prediction accuracy of the tracking error of each axis of UPM machines, Bayesian hyperparameter optimization and feature importance analysis were conducted in the proposed TCN-BiLSTM model using high-quality training datasets from well-designed experiments. Ultimately, validation of the proposed system on a three-axis ultra-precision milling machine demonstrated its excellent performance. The experimental results showed that the optimized TCN-BiLSTM model exhibited an excellent capacity to predict the tracking error of the X-axis and Y-axis with minimal mean absolute error values of 0.000009 and 0.000023, respectively. Implementing the customized application reduced X-axis and Y-axis tracking errors by approximately 45–75% and 40–70%, respectively. This study first validates the feasibility of deep learning to improve accuracy in the UPM field, which will provide significant insight into speeding up the digitalization and intellectualization of the UPM scenario.

A Novel Approach to Surface Roughness Virtual Sample Generation to Address the Small Sample Size Problem in Ultra-Precision Machining.

Surface roughness is one of the main bases for measuring the surface quality of machined parts. A large amount of training data can effectively improve model prediction accuracy. However, obtaining a large and complete surface roughness sample dataset during the ultra-precision machining process is a challenging task. In this article, a novel virtual sample generation scheme (PSOVSGBLS) for surface roughness is designed to address the small sample problem in ultra-precision machining, which utilizes a particle swarm optimization algorithm combined with a broad learning system to generate virtual samples, enriching the diversity of samples by filling the information gaps between the original small samples. Finally, a set of ultra-precision micro-groove cutting experiments was carried out to verify the feasibility of the proposed virtual sample generation scheme, and the results show that the prediction error of the surface roughness prediction model was significantly reduced after adding virtual samples.

Efficient nanoscale characterization of wafer surfaces using intelligent sampling and Bayesian optimization

The geometrical attributes of a wafer, such as thickness, uniformity, and local curvature, serve as eminent determinants of its quality. Therefore, measurements that are both rapid and accurate are paramount as multi-stage fabrication processes hinge on them to ensure high product reliability, proper process control, and maximum efficiency. The manual wafer profiling measurement approach in use today is, unfortunately, a time-consuming process. This creates a need for fast and prompt analysis of wafer attributes. This research thus proposes an iterative sampling method that boasts reducing the number of samplings down to a minimum without the need to discard a sufficient level of accuracy for the purpose of reliable attribute estimation. Subsequently, we suggest a series of methods to improve the monitoring and evaluation of the surface topography, all within the context of ultra-precision/nanoscale manufacturing processes, such as ultra-precision machining (UPM) and chemical–mechanical planarization (CMP). For the purpose of both real-time and exception-based detection of anomalies in the UPM process, we constructed a process-machine interaction (PMI) model with Bayesian learning, which amalgamates the received data across heterogeneous sensors such as force, vibration, and acoustic emission in situ. Similarly, the CMP process is quantitatively characterized by a multi-scaled multilayer nonlinear decomposition model with predictive capabilities. This model helps understand the physio-mechanical phenomena underlying both CMP and the UPM processes and describe and predict the nonlinear interactions, as evidenced by the experimentally collected vibration signal patterns.

Study on the mechanism of influence of water molecule cooling and vacuum cooling on the machining of single crystal copper nanos

Aiming to clarify the influence mechanism of water molecule cooling on the machining of single-crystal Cu nanocutting, this study conducts an in-depth investigation through molecular dynamics simulations, incorporating both qualitative and quantitative analyses. The focal points of the research include the distribution of Von Mises stress, shear strain, cutting temperature, and dislocation formation. In terms of Von Mises stress, water molecule cooling significantly reduces the average stress at a cutting speed of 10 Å, and exhibits a turning point in stress reduction at a speed of 300 m/s. The average stress under water molecule cooling is consistently lower than that under vacuum cooling, and the difference diminishes at a cutting speed of 20 Å. Shear strain analysis reveals that at a cutting speed of 10 Å, the chip height under water molecule cooling is notably lower than that under vacuum cooling, while at 15 Å, it shows a trend of first decreasing and then increasing. The average chip height under vacuum cooling is 1.1 times that of water molecule cooling. At a cutting depth of 15 Å, speeds of 200 m/s and 300 m/s both demonstrate superior cooling performance, with cutting temperatures significantly lower than vacuum cooling, with temperature differences of approximately 85K and 180K, respectively. Dislocation analysis indicates that under water molecule cooling, the number of dislocations at greater cutting depths is significantly higher than under vacuum cooling, predominantly consisting of 1/6<112> (Shockley) dislocations. Overall, water molecule cooling shows significant advantages in several aspects, significantly reducing stress, chip height, and cutting temperature, while increasing the number of dislocations. These results provide quantified empirical data for ultra-precision nanocutting machining, offering important references for engineering practice and positively contributing to the optimization of nanocutting process parameters.

Effect of crystal orientation on surface/subsurface damage characteristics of nano-cutting Ni-based single crystal superalloy

In ultra-precision machining, material anisotropy has an essential impact on surface generation and subsurface damage. In this work, the molecular dynamics method is applied to create the diamond nano-cutting models of Ni-based single crystal (NBSX) superalloy workpieces with different crystal orientations, and the surface/subsurface damage characteristics, crystal structure changes, cutting force, and atomic temperature and stress are discussed. The results show significant differences in surface/subsurface quality, crystal structure transformation, cutting force, atomic temperature and stress distributions and dislocation density obtained from nano-cutting workpieces along different crystal orientations. In four groups of workpieces, the surface/subsurface quality is used as an evaluation index, which shows that the [110](001) crystal orientation has the best cutting effect, and the [111](1‾ 10) crystal orientation is the worst cutting direction. This work reveals the damage generation mechanism of the machined surface/subsurface of NBSX superalloy on an atomic scale, which provides technical support for improving machining quality and optimizing machining parameters.