Machining Accuracy Research Articles

From Small Data Modeling to Large Language Model Screening: A Dual-Strategy Framework for Materials Intelligent Design.

Small data in materials present significant challenges to constructing highly accurate machine learning models, severely hindering the widespread implementation of data-driven materials intelligent design. In this study, the Dual-Strategy Materials Intelligent Design Framework (DSMID) is introduced, which integrates two innovative methods. The Adversarial domain Adaptive Embedding Generative network (AAEG) transfers data between related property datasets, even with only 90 data points, enhancing material composition characterization and improving property prediction. Additionally, to address the challenge of screening and evaluating numerous alloy designs, the Automated Material Screening and Evaluation Pipeline (AMSEP) is implemented. This pipeline utilizes large language models with extensive domain knowledge to efficiently identify promising experimental candidates through self-retrieval and self-summarization. Experimental findings demonstrate that this approach effectively identifies and prepares new eutectic High Entropy Alloy (EHEA), notably Al14(CoCrFe)19Ni28, achieving an ultimate tensile strength of 1085 MPa and 24% elongation without heat treatment or extra processing. This demonstrates significantly greater plasticity and equivalent strength compared to the typical as-cast eutectic HEA AlCoCrFeNi2.1. The DSMID framework, combining AAEG and AMSEP, addresses the challenges of small data modeling and extensive candidate screening, contributing to cost reduction and enhanced efficiency of material design. This framework offers a promising avenue for intelligent material design, particularly in scenarios constrained by limited data availability.

Projecting Large Fires in the Western US With an Interpretable and Accurate Hybrid Machine Learning Method

AbstractMore frequent and widespread large fires are occurring in the western United States (US), yet reliable methods for predicting these fires, particularly with extended lead times and a high spatial resolution, remain challenging. In this study, we proposed an interpretable and accurate hybrid machine learning (ML) model, that explicitly represented the controls of fuel flammability, fuel availability, and human suppression effects on fires. The model demonstrated notable accuracy with a F1‐score of 0.846 ± 0.012, surpassing process‐driven fire danger indices and four commonly used ML models by up to 40% and 9%, respectively. More importantly, the ML model showed remarkably higher interpretability relative to other ML models. Specifically, by demystifying the “black box” of each ML model using the explainable AI techniques, we identified substantial structural differences across ML fire models, even among those with similar accuracy. The relationships between fires and their drivers, identified by our model, were aligned closer with established fire physical principles. The ML structural discrepancy led to diverse fire predictions and our model predictions exhibited greater consistency with actual fire occurrence. With the highly interpretable and accurate model, we revealed the strong compound effects from multiple climate variables related to evaporative demand, energy release component, temperature, and wind speed, on the dynamics of large fires and megafires in the western US. Our findings highlight the importance of assessing the structural integrity of models in addition to their accuracy. They also underscore the critical need to address the rise in compound climate extremes linked to large wildfires.

Study on new thermal error compensation method for complex surface five-axis machining based on new experimental strategy

Five-axis machine tools are well-suited for processing complex surface parts. However, during the machining process, thermal error can have a significant impact on the machining accuracy. To propose an accountable on-line compensation method for five-axis milling, this paper developed a new experimental device for in-situ measurements of thermal-induced displacements in both X- and Y-axis directions. With the help of this self-developed device, a five-axis milling experiment was carried out to collect the temperature change of key thermal points and thermal-induced error during continuous impeller milling. A thermal error prediction model was established with collected data and a thermal error compensation system was developed for the five-axis milling accordingly. The following validation experiment showed that the surface profile error of the blades of milled impeller can be stabilized within the range of 0.1 mm with the help of proposed the thermal error compensation system. The results of the validation experiment proved the effectiveness of the proposed method for the reduction of thermal error during the five-axis milling.

A non-iterative compensation method for machining errors of thin-walled parts considering coupling effect of tool-workpiece deformation

Thin-walled parts have significant application value in the aerospace industry due to their high strength-to-weight ratio. However, their low rigidity makes them susceptible to cutting force-induced error, which may seriously affect their machining accuracy. Error compensation is an effective method for addressing this issue. However, the commonly used mirror compensation method can cause residual errors due to the coupling effect of the tool-workpiece deformation. The influence mechanism of this coupling effect on error compensation is analyzed through an iterative compensation method. This method efficiently reduces residual errors. However, its computation efficiency is insufficient to meet the requirements of real-time compensation. Therefore, a non-iterative compensation method is proposed to directly calculate the compensation values considering the coupling effect of the tool-workpiece deformation. Through the approximate invariance of the overall cutting coefficient matrix and the pre-given system parameter, the proposed method avoids the repeated calculation of cutting forces and improves the computation efficiency. Experiment results of milling thin-walled blades show that after compensation using the proposed method, the machining accuracy of the thin-walled blade has seen a further increase of 18.1% in comparison to the mirror compensation method. Moreover, the proposed method achieves comparable compensation accuracy to the iterative method with a 66% reduction in computation time. The proposed method has significant potential for real-time compensation in the machining of complex 5-axis thin-walled parts.

Evaluation of a finite state machine algorithm to measure stepping with ankle accelerometry: performance across a range of gait speeds, tasks, and individual walking ability.

Wearable sensors, including accelerometers, are a widely accepted tool to assess gait in clinical and free-living environments. Methods to identify phases and subphases of the gait cycle are necessary for comprehensive assessment of pathological gait. The current study evaluated the accuracy of a finite state machine (FSM) algorithm to detect strides by identifying gait cycle subphases from ankle-worn accelerometry. Algorithm performance was challenged across a range of speeds (0.4-2.6 m/s), task conditions (e.g., single- and dual-task walking), and individual characteristics. Specifically, the study included a range of treadmill speeds in young adults and overground walking conditions in older adults with neurological disease. Manually counted and algorithm-derived stride detection from acceleration data were evaluated using error analysis and Bland-Altman plots for visualization. Overall, the algorithm successfully detected strides (>96% accuracy) across gait speed ranges and tasks, for young and older adults. The accuracy of an FSM algorithm combined with ankle-worn accelerometers, provides an analytical approach with affordable and portable tools that permits comprehensive assessment of gait unbounded by setting and proves to perform well in in walking tasks characterized by variable walking. These algorithm capabilities and advancements are critical for identifying phase dependent gait impairments in clinical and free-living assessment.

An exact solution of the lubrication equations for the Oldroyd-B model in a hyperbolic pipe

An exact analytical solution of the lubrication equations for the steady, isothermal, incompressible flow of a viscoelastic Oldroyd-B fluid in a hyperbolic cylindrical contracting pipe is derived. The solution is valid for values of the Deborah number, De, up to order unity (De is defined as the ratio of the longest relaxation time of the polymer to the characteristic residence time of the fluid in the pipe), all values of the ratio of the polymer viscosity to the total viscosity of the fluid, η, and typical values of the contraction ratio, Λ, encountered in experiments and practical applications. It is provided in terms of the streamfunction only and is used in the momentum balance to derive a strongly non-linear ordinary differential equation of second order with unknown a function which corresponds to a modified fluid velocity along the main flow direction. The final equation is solved semi-numerically using a fully spectral (Legendre)-Galerkin approach to resolve the unknown function almost down to machine accuracy. The exact solution for the polymer extra-stresses, which is emphasized is not the full solution of the complete lubrication equations, allows for the derivation of a variety of theoretical expressions for the average pressure-drop along the pipe. In all cases, a decrease in the pressure drop compared to the Newtonian value with increasing De, η and/or Λ is predicted. The differences between the corresponding analytical solution for the planar geometrical configuration are also identified and discussed.

Precision polishing force control technology of abrasive belt-wheel polishing servo control system based on HGO-SMC

Due to cylinder friction, valve dead zone effect, flow nonlinearity, and measurement noise exist in pneumatic system. In the actual machining, it is difficult to achieve precise control of the pneumatic servo polishing system using the traditional PID algorithm, so the machining accuracy of the blades cannot be guaranteed. A sliding mode control (SMC) method based on high gain observer (HGO) is proposed. The HGO observes the output signal of the system and feeds the estimated signal back to the SMC to ensure that the observation error is uniform ultimate boundedness. Simulation and experiments show that compared with the PID algorithm, the steady-state error of HGO-SMC can be reduced by 66.4%. Compared with SMC(sgn), the high-frequency chattering of HGO-SMC can be reduced by 72.5%. Moreover, the polishing processing experiment shows that HGO-SMC can improve the form and position accuracy by 52.6% and reduce the surface roughness by 55.6%, compared with the original PID control algorithm of the polishing machine tool.

Methods for Corrosion Detection in Pipes Using Thermography: A Case Study on Synthetic Datasets

This study reviews advanced methods for corrosion detection and characterization in pipes using thermography, with a focus on addressing the limitations posed by small datasets. Thermography captures temperature distributions on the surface of pipes to identify subsurface defects. The challenges of sequential data processing, neural network performance, feature extraction, and dataset size are discussed, with proposed solutions such as advanced algorithms, feature selection techniques, and data augmentation. Given the significant gap in the current literature, there is a need for larger, more diverse datasets to train more robust and accurate machine learning models. A case study combining experimental data with Finite Element Method (FEM) simulations demonstrates that augmenting datasets with synthetic data significantly improves defect detection accuracy. These findings highlight the potential of integrating thermography with machine learning to enhance defect detection, providing insights for future research and practical applications.

Machine Learning Application to Classify Asteroids Based on Orbital Parameters

Abstract A machine learning application was developed to detect Potentially Hazardous Asteroid and mitigate asteroid collision risk with the Earth by applying three classifiers: the K-Nearest Neighbors, Naïve Bayes, and Random Forest. The study determined the most effective classifier for developing an asteroid classification program based on orbital motion. The machine learning classifier was then evaluated by its precision, accuracy, F1-score, and recall in determining Potentially Hazardous Asteroids and non-Potentially Hazardous Asteroids. The result presented Random Forest as the most appropriate classifier with the highest accuracy score of 99.53%, followed by the Naive Bayes classifier with an accuracy score of 92.00%, and the KNN classifier with an accuracy score of 84.45%. The study provided information on the most accurate machine learning classifier with the impact parameters for asteroid classification in an early warning system. By improving an embedded real-time detection system for Potentially Hazardous Asteroids, the study contributes to more effective strategies for mitigating the risk of asteroid impacts and enhancing planetary defence.

Study on micro electrochemical milling with programmable dynamic eccentric rotating electrode

Micro electrochemical machining (ECM) has the advantages of non-contact machining, no tool loss and no residual stress, and has great development potential in the field of microstructure manufacturing. However, the micron-scale machining gap is difficult to renew and discharge the electrolyte and electrolytic products in time. In this paper, a novel micro ECM with programmable dynamic eccentric rotating electrode (DER-ECM) is proposed, which can effectively improve the discharge of electrolytic products in the machining area. According to the process characteristics, a disc flexure hinge structure with one-stage amplification was designed, which was driven by a piezo-ceramic actuator, and the programmable eccentricity ranging from 0 to 20 μm. The multi-physics model of flow field and electric field of DER-ECM, micro ECM with eccentric rotating electrode (ER-ECM) and micro ECM with rotating electrode (R-ECM) were established. The theoretical simulation results showed that the flow rate generated by DER-ECM at the bottom of the electrode is 97 times that of ER-ECM. The current density generated by DER-ECM showed periodic pulsation, and the pulsation period was determined by the driving frequency of the piezo-ceramic actuator. The experimental results showed that DER-ECM could effectively eliminate the surface spike of the workpiece in the bottom machining area and effectively improve the machining accuracy. The micro-groove widths obtained under DER-ECM, ER-ECM and R-ECM were 418 μm, 446 μm and 468 μm, respectively. In addition, the influence of three types of dynamic eccentric rotation electrode trajectories on DER-ECM was studied. The experimental results showed that the sawtooth dynamic eccentric rotation electrode had better machining accuracy. Theoretical simulation and experimental results showed that DER-ECM could improve the flow field and achieve higher machining efficiency and machining localization.

Inverse design of high-strength medium-Mn steel using a machine learning-aided genetic algorithm approach

To develop medium-Mn steels with an ultimate tensile strength (UTS) exceeding 2 GPa and excellent ductility, we created a highly accurate UTS prediction machine learning (ML) model using a boosted decision tree model and 1520 dataset of tensile properties of medium-Mn steels with micro-alloying elements. We also optimized the hyper-parameters of a genetic algorithm (GA) using the Shannon diversity index to enhance search efficiency while retaining diversity. In a high-dimensional search space with millions of potential combinations, the ML-GA approach efficiently identified diverse chemical compositions and austenitizing conditions to achieve UTSs above 2 GPa. The k-means clustering method then grouped them into five distinct specimens based on similarities. These five specimens, fabricated using inversly designed chemical compositions and austenitizing temperatures, successfully exhibited UTSs exceeding 2 GPa and greater ductility compared to hot-stamped C steels. These excellent tensile properties were attributed to grain refinement resulting from the low austenitizing temperature and the pinning effect of micro-alloying element carbides, such as TiC and VC.

Comparative analysis of machine learning algorithms for ECG-based heart attack prediction: A study using Bangladeshi patient data

This study aims to identify the most accurate machine learning algorithm for predicting heart attacks using demographic data, physiological measurements, and electrocardiogram (ECG) results. We utilized a dataset of 4,000 patient records, combining data from DMCH and Kaggle. Our methodology involved comprehensive data preprocessing, including ECG noise removal and feature selection using the Brouta algorithm. We implemented and compared six machine learning algorithms: Decision Tree, Random Forest, Logistic Regression, Support Vector Machine, XGBoost, and K-Nearest Neighbors. The results demonstrate that our proposed method can accurately predict heart attacks with high sensitivity and specificity. Among the tested algorithms, Random Forest achieved the highest accuracy of 87%, with well-balanced precision (0.86), recall (0.85), and F1-score (0.87). K-Nearest Neighbors and XGBoost also showed strong performance, with accuracies of 81% and 80% respectively. This study contributes to the field by utilizing a large, diverse dataset and providing a comprehensive comparison of multiple algorithms. Our findings suggest the potential for integrating machine learning, particularly Random Forest models, into clinical practice for early heart attack risk assessment, representing a significant step towards improving cardiovascular care through advanced data analysis techniques.

Study on the 2D equivalent nonlinear dynamics simulation model related to high speed precision bearing and spindle

The stiffness and contact stress of the bearing spindle system are the key factors affecting its machining accuracy and service life under the condition of high-speed service, which will be affected by different structural parameters and service conditions. It is essential to establish an efficient and accurate computational simulation model to understand the influence of complex factors on the nonlinear dynamic characteristics of the bearing spindle system. Firstly, in the paper, a 2D axisymmetric finite element model of the bearing, aims at the relationship between stiffness and contact stress of the bearing under high-speed service, has been built based on a classical dynamic analysis model and the reversed method of material parameters of equivalent rolling balls. Additionally, for the BT30 spindle, the 2D axisymmetric finite element model of the bearing spindle system also has been built and applied in mechanical analysis of spindle under different conditions of assembly and service, based on the bearing model. The results show that the axial force of bearings decreases as the rotational speed increases, and an augmentation in speed will result in a reduction in the axial stiffness of the BT30 spindle. In addition, the maximum contact stress exhibits a slight decline as the rotational speed increases. Furthermore, with an escalating preload, the stiffness and contact stress of the spindle undergo substantial increments, however, these parameters will cease to alter once a certain threshold is reached.

Intelligent Evaluation and Dynamic Prediction of Oyster Freshness with Electronic Nose Based on the Distribution of Volatile Compounds Using GC-MS Analysis.

The quality of oysters is reflected by volatile organic components. To rapidly assess the freshness level of oysters and elucidate the changes in flavor substances during storage, the volatile compounds of oysters stored at 4, 12, 20, and 28 °C over varying durations were analyzed using GC-MS and an electronic nose. Data from both GC-MS and electronic nose analyses revealed that alcohols, acids, and aldehydes are the primary contributors to the rancidity of oysters. Notably, the relative and absolute contents of Cis-2-(2-Pentenyl) furan and other heterocyclic compounds exhibited an upward trend. This observation suggests the potential for developing a simpler test for oyster freshness based on these compounds. Linear Discriminant Analysis (LDA) demonstrated superior performance compared to Principal Component Analysis (PCA) in differentiating oyster samples at various storage times. At 4 °C, the classification accuracy of the optimal support vector machine (SVM) and random forest (RF) models exceeded 90%. At 12 °C, 20 °C, and 28 °C, the classification accuracy of the best SVM and RF models surpassed 95%. Pearson correlation analysis of the concentrations of various volatile compounds and characteristic markers with the sensor response values indicated that the selected sensors were more aligned with the volatiles emitted by oysters. Consequently, the volatile compounds in oysters during storage can be predicted based on the response information from the sensors in the detection system. This study also demonstrates that the detection system serves as a viable alternative to GC-MS for evaluating oysters of varying freshness grades.

Vibration energy-based indicators for multi-target condition monitoring in milling operations

The demand for intelligent process monitoring is increasing in aerospace manufacturing to ensure tight tolerances and high surface quality. Real-time monitoring in machining is crucial for machined accuracy and process reliability, reducing production times and costs, and enhancing automation of the manufacturing process. This study presents a robust multi-target condition monitoring method based on the vibration signals. Firstly, three new energy ratio indicators with dimensionless characteristics were defined for tool wear, breakage, and chatter monitoring. Secondly, the vibration energy loss from the tool tip to the tool holder, and spindle housing was measured and compared, and the rules of vibration loss from the tool tip to the spindle housing were revealed. Using force signals as a reference, the monitoring performance of industrially acceptable acceleration and sound signals in multi-target condition monitoring was quantitatively analyzed. Finally, the performance of the proposed vibration energy-based indicators was experimentally illustrated and quantitatively evaluated. It is shown that these indicators can be used to discriminate between tool breakage and chatter, as well as to assess tool wear. The new monitoring method can also minimize the costs of process monitoring by reducing the use of expensive sensors or overusing multiple sensors in a smart manufacturing system.

Analysis of existing techniques for image-based recognition of agricultural crops diseases

An analysis of existing methods for processing and identifying the disease of agricultural crops was carried out. Other methods for identifying informative regions, including Fourier transformation, k-means clustering, Histogram Equalization, Scale-Invariant Feature Transform (SIFT), Local Binary Patterns (LBP), Histogram of Oriented Gradients (HOG) algorithms, as well as their combinations were reviewed. The studied approaches demonstrated high accuracy in the identification and classification of various plant diseases. The study examined hybrid models, such as, for example, logistic regression with decision trees and Extreme Learning Machines (ELM). The accuracy of the Support Vector Machine (SVM), ELM, and Decision Trees algorithms were compared, convinced, that the importance of choosing the right parameters and fine-tuning to improve accuracy is important. The methods of deep learning such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM) and their usage in the scope of the image recognition, which are also used for disease recognition, were examined. The recognition accuracy of several CNN models was compared: DenseNet121, MobileNetV2, NASNetMobile and EfficientNetB0, and the last demonstrated the best results. The modification of ready-made architectures, including the architecture of the EfficientNetB0 neural network, has been analyzed as a way of adapting existing models to specific recognition requirements.

FBSM: Foveabox-based boundary-aware segmentation method for green apples in natural orchards

Image segmentation of target fruits is an essential part of machine vision systems, aiming to facilitate more accurate early fruit measurement and machine harvesting in natural orchard environments. Given that instance-level pooling and down-sampling operations in conventional segmentation models often lead to the loss of detailed information, resulting in coarse partitioning masks, we endeavor to restore the boundary information of the masks. To achieve high-quality fruit segmentation with clear boundaries and refined masks, a Foveabox-based boundary-aware segmentation model (FBSM) is constructed by adding a multi-stage mask prediction head incorporating fine-grained features to the anchor-free detection model FoveaBox to realize accurate segmentation of green apple fruits. At each stage, a bi-layer fusion structure (BFS) is employed to fully fuse the fine-grained features in a double-fusion manner to guide the subsequent instance mask prediction. Finally, the boundary recovery module (BRM) is leveraged to recover the lost boundary detail information and obtain more accurate boundaries for the continuous optimization process of the fruit instance mask. Experimental results demonstrate that the proposed FBSM model achieves a mean average precision (mAP) of 60.7 % for green apple instance segmentation in unstructured natural orchard environments, surpassing traditional instance segmentation models. Moreover, it enhances base model detection accuracy by 1.3 % while striking a better balance between detection and segmentation performance.

Design of a mixed robotic machining system and its application in support removal from metal additive manufactured thin-wall parts

Robotic machining could provide a solution for removing supports from metal additive manufactured workpieces, replacing labor-intensive work. However, the robot’s intrinsic weaknesses of low positioning accuracy and structural rigidity primarily restrict its applications. Improving the accuracy of robotic machining remains an unresolved issue. A mixed solution is proposed, in which a portable CNC machine with the capability of visual feature recognition is equipped with a universal industrial robot. The robot implements positioning motions in a large space, while the portable CNC fulfills accurate machining motions on a local feature of the workpiece. A sizeable weight of the portable CNC exerts a moderate load on the industrial robot’s joints, increasing joint stiffness. The mixed machining system exhibits high accuracy and stiffness when milling a steel/titanium alloy workpiece, achieving tolerances up to ±0.04 mm on a 60×80 mm U-shaped path without exciting any structural vibration modes. When the dimension of the workpiece exceeds the machining range of the portable CNC, a combined algorithm of coarse-fine registration based visual identification and robot error compensation is designed to align the spatial coordinates of the machining motion with that of the positioning motion, thereby extending the machining range with high accuracy. Through the proposed mixed robot machining method, experiments of doubling the machining range have been done to verify that the mixed machining robotic system is able to slot a 550 mm-long path with accuracy of ±0.1 mm. Furthermore, the mixed robotic machining system is applied to recognize and remove multiple supports of lattices, grids and ribs from a titanium-alloy additive manufactured thin-wall workpiece with high accuracy and high efficiency.

Diagnosis of ophthalmologic diseases in canines based on images using neural networks for image segmentation

The primary challenge in diagnosing ocular diseases in canines based on images lies in developing an accurate and reliable machine learning method capable of effectively segmenting and diagnosing these conditions through image analysis. Addressing this challenge, the study focuses on developing and rigorously evaluating a machine learning model for diagnosing ocular diseases in canines, employing the U-Net neural network architecture as a foundational element of this investigation.Through this extensive evaluation, the authors identified a model that exhibited good reliability, achieving prediction scores with an Intersection over Union (IoU) exceeding 80 %, as measured by the Jaccard index. The research methodology encompassed a systematic exploration of various neural network backbones (VGG, ResNet, Inception, EfficientNet) and the U-Net model, combined with an extensive model selection process and an in-depth analysis of a custom training dataset consisting of historical images of different medical symptoms and diseases in dog eyes.The results indicate a fairly high degree of accuracy in the segmentation and diagnosis of ocular diseases in canines, demonstrating the model's effectiveness in real-world applications. In conclusion, this potentially makes a significant contribution to the field by utilizing advanced machine-learning techniques to develop image-based diagnostic routines in veterinary ophthalmology. This model's successful development and validation offer a promising new tool for veterinarians and pet owners, enhancing early disease detection and improving health outcomes for canine patients.

A Stand‐Off Laser‐Induced Breakdown Spectroscopy (LIBS) System for Remote Bacteria Identification

ABSTRACTBacteria are the primary cause of infectious diseases, making rapid and accurate identification crucial for timely pathogen diagnosis and disease control. However, traditional identification techniques such as polymerase chain reaction and loop‐mediated isothermal amplification are complex, time‐consuming, and pose infection risks. This study explores remote (~3 m) bacterial identification using laser‐induced breakdown spectroscopy (LIBS) with a Cassegrain reflective telescope. Principal component analysis (PCA) was employed to reduce the dimensionality of the LIBS spectral data, and the accuracy of support vector machine (SVM) and Random Forest (RF) algorithms was compared. Multiple repeated experiments showed that the RF model achieved a classification accuracy, recall, precision, and F1‐score of 99.81%, 99.80%, 99.79%, and 0.9979, respectively, outperforming the SVM model and providing more accurate remote bacterial identification. The method based on laser‐induced plasma spectroscopy and machine learning has broad application prospects, supporting noncontact disease diagnosis, improving public health, and advancing medical research and technological development.