Explore the vast range of titles available.

Chatter is one form of severe self-excited vibration in machining process which leads to many machining problems. In this paper, a new method of chatter identification is proposed. During the machining process, the acceleration signal of vibration is obtained and the time domain root mean square value of the acceleration is calculated every proper segment, through which the real-time acceleration root mean square (RMS) sequence is obtained. Then, the coefficient of variation (i.e., the ratio of the standard deviation to the mean, CV) of the RMS sequence is defined as the indicator for chatter identification. The milling experiment shows that CV can well distinguish the state (stable or chatter) of the machining process. The proposed method has a quantitative and dimensionless indicator, which works for different machining materials and machining parameters, and even can be expected to work in a wider range condition, such as different machine tool and cutting method. This paper also designs a fast algorithm of CV, making it an ideal candidate for online monitoring system.

Titanium alloy is a typical difficult-to-cut material due to its high strength and high stiffness. To solve the problem of the low efficiency and poor surface quality in the milling of titanium alloy, this paper proposes a novel longitudinal-torsional ultrasonic vibration milling (LTUVM) process. An ultrasonic horn with spiral slots was designed to convert the longitudinal vibration into longitudinal-torsional vibration. The tooltip trajectory was modeled, and the finite elements analysis was used to analyze the cutting mechanism of LTUVM. The simulation results indicate a kind of separation cutting characteristics in every vibration cycle, which is beneficial to reduce the cutting force and improve the surface finish compared with the single longitudinal ultrasonic vibration milling (SLUVM). Then, cutting tests were conducted on Ti-6Al-4V to evaluate the performance of LTUVM. Experimental results demonstrated that the LTUVM could reduce the cutting force by 46–86% compared with the conventional milling (CM) and the SLUVM due to its separation cutting characteristics. Moreover, the surface morphology was analyzed, and a fractal dimension (FD) method was proposed to characterize the regularity and fragmental property of the machined surfaces. The surface morphology analysis results showed that the LTUVM can be used as a novel high-efficiency and high-quality surface texturing method for Ti-6Al-4V. The textured surface of the LTUVM has the superiority of high integrity and periodicity, which could be applied to effectively tune the tribological property of surface.

The main challenges in reconstruction-based anomaly detection include the breakdown of the generalization gap due to improved fitting capabilities and the overfitting problem arising from simulated defects. To overcome this, we propose a new method called PRFF-AD, which utilizes progressive reconstruction and hierarchical feature fusion. It consists of a reconstructive sub-network and a discriminative sub-network. The former achieves anomaly-free reconstruction while maintaining nominal patterns, and the latter locates defects based on pre- and post-reconstruction information. Given defective samples, we find that adopting a progressive reconstruction approach leads to higher-quality reconstructions without compromising the assumption of a generalization gap. Meanwhile, to alleviate the network’s overfitting of synthetic defects and address the issue of reconstruction errors, we fuse hierarchical features as guidance for discriminating defects. Moreover, with the help of an attention mechanism, the network achieves higher classification and localization accuracy. In addition, we construct a large dataset for packaging chips, named GTanoIC, with 1750 real non-defective samples and 470 real defective samples, and we provide their pixel-level annotations. Evaluation results demonstrate that our method outperforms other reconstruction-based methods on two challenging datasets: MVTec AD and GTanoIC.

Ultrasonic elliptical vibration cutting (UEVC) is a superior machining method for difficult-to-cut materials. The shape of the elliptical tool trajectory crucially affects the integrity of the machined surface. However, the existing designs of UEVC tool typically suffer from difficult motion decoupling, resulting in the low controllability of elliptical trajectory. This study presents a new optimization design method of UEVC tool with dual longitudinal generators to create well-decoupled elliptical trajectory. A theoretical model which establishes the effects of key design variables on the tool resonant characteristics was adopted to optimize the configuration angle (the included angle of two longitudinal generators) as 90°. The structural parameters of the connection blocks of the two generators were also optimized to achieve resonance matching in both the normal (depth-of-cut) and tangential (cutting) directions. The optimized tool has a stable resonant frequency of 18 kHz, a working space of 5.1 μm × 5.3 μm, and a small trajectory error of less than 0.75 μm. The vibration measurement results showed that the optimized design can enable the high controllability of tool trajectory. Grooving experiments were conducted to verify the improved cutting performance of the designed tool due to the excellent control of tool trajectory.

High-volume fraction silicon carbide-reinforced aluminum matrix (SiCp/Al) composites are widely used in many industrial fields due to their excellent material properties. However, these composites are regarded as one of the most difficult-to-machine materials, owing to the presence of many hard and brittle SiC reinforcements. Rotary ultrasonic machining (RUM) is an effective processing method for SiCp/Al composites. The material removal mechanism in RUM of SiCp/Al composites was investigated by comparing the deformation characteristics of the composites in ultrasonic vibration-assisted scratch (UVAS) tests and conventional scratch (CS) tests which were performed on a rotary ultrasonic machine. The influence of ultrasonic vibration on the machining process was analyzed. Furthermore, the morphologies of the scratching surfaces, scratching forces, and material removal process were evaluated in detail. The theoretical and experimental results revealed that ultrasonic vibration changes the interaction between the cutting tool and the workpiece. The vibration enhanced the Al matrix and facilitated SiC reinforcements removal by increasing the cracks in them. Therefore, the scratching forces in UVAS were smaller and more stable than those in CS. The coefficient of friction (COF) was also smaller than that of CS and hence, the adhesion effect of Al matrix during the scratching process was weakened. This study shows that the removal mode of SiC reinforcements plays a decisive role in the formation of the machined surface. These results can serve as a guide for selecting appropriate processing parameters to obtain improved machining quality of SiCp/Al composites.

Inertial Measurement Unit-based methods have great potential in capturing motion in large-scale and complex environments with many people. Sparse Inertial Measurement Unit-based methods have more research value due to their simplicity and flexibility. However, improving the computational efficiency and reducing latency in such methods are challenging. In this paper, we propose Fast Inertial Poser, which is a full body motion estimation deep neural network based on 6 inertial measurement units considering body parameters. We design a network architecture based on recurrent neural networks according to the kinematics tree. This method introduces human body shape information by the causality of observations and eliminates the dependence on future frames. During the estimation of joint positions, the upper body and lower body are estimated using separate network modules independently. Then the joint rotation is obtained through a well-designed single-frame kinematics inverse solver. Experiments show that the method can greatly improve the inference speed and reduce the latency while ensuring the reconstruction accuracy compared with previous methods. Fast Inertial Poser runs at 65 fps with 15 ms latency on an embedded computer, demonstrating the efficiency of the model. Inertial Measurement Units-based motion capture effective application in large scale and complex environments depends on improved efficiency and reduced latency. Here, authors propose a full body motion estimation deep neural network based on 6 IMUs, which runs at 65 fps with 15 ms latency on an embedded computer.

On-machine inspection (OMI) systems have been widely used for the automatic setting of workpieces and the determination of the kinematic errors of a machine tool. Touch-trigger probes are commonly used in OMI systems due to their high reliability and low cost. However, the associated pre-travel error, which is inherent by nature, often severely affects the measurement accuracy and must be compensated during the measuring process in order to ensure the required measurement precision. This paper proposes a compensation method for probe errors using a 3D error map. To construct the 3D error map of the probe, a weighted least squares (WLS) method based on the probe’s mechanical model was first employed to accurately fit the reference sphere center; subsequently, the errors of the calibrated directions could be obtained. Finally, the bicubic Coons patch (BCP) interpolation method was used to calculate the errors of the uncalibrated directions. To validate the proposed method, calibration tests were conducted on an OMI system mounted with a strain gauge probe. The compensation results indicated an increase of 57.40% in the accuracy compared to that of the existing method.

Touch-trigger probes are widely used in on-machine inspection (OMI) systems, which have become increasingly important in an interprocess inspection. The pretravel errors are often compensated by data-driven calibration. However, its accuracy is significantly affected by its internal mechanical structure and trigger mechanism. In this paper, we propose a rigid-flexible coupling pretravel error model, considering the bending of the stylus and the lifting of the stylus carrier in the triggering process. This paper presents a method to identify the acceleration of machine tool by fitting the measured results at different speeds into quadratic curves and establishes the response error model. The trigger force of 73 direction probes is identified by a three-directional force sensor. The trigger force identified by the experiment is analyzed in the rigid-flexible simulation model, and the trigger resistance threshold of the probe is obtained. The simulation model of trigger error is established. Through the response error model and trigger error simulation model proposed in this paper, the probe pretravel error compensation accuracy can be improved by 39.6%.

Under complex alternating loads, the bearing hole chamfer is susceptible to fatigue cracks as this is a common concentration detail in aircraft structure. However, due to the small size and small inclination angle, it is difficult to strengthen the hole chamfer by general strengthening methods, and it is usually accompanied by poor dimensional accuracy. This paper proposes a high-precision, high-coverage ultrasonic peening process to improve the surface performance of the hole edge chamfer. Kinematic analysis for ultrasonic peening of hole chamfer using a theoretical model was developed to study the surface generation mechanism of the strengthening process under different strengthening parameters. In addition, the effects of preload depth, spindle speed, and coverage times on surface integrity, including surface roughness, plastic deformation layer depth, and residual stress, were also investigated and compared with kinematic analysis. The best reinforcement parameters were selected according to the surface integrity results for the edge chamfer of 45 steel holes. The results show that the surface roughness of the hole edge chamfer increased significantly, while the plastic deformation layer depth and the residual stress are mainly influenced by preload depth.

Machining distortion is a major issue in the machining of thin-walled monolithic components, which are now widely used in the aerospace industry. The evolution of the stress field within the workpiece during machining (due to material removal and cutting loads) is the main cause of global machining distortion for thin-walled parts. This paper presents a new modeling method for machining distortion that can represent this stress evolution process. The finite element method (FEM) is used to model the material removal and mechanical loads during roughing. Theoretical modeling is used to model the influence of finishing on stress redistribution within surface material. Finally, FEM is used to calculate the machining distortion. This method can be implemented on small servers and personal computers. An application case of a thin-walled component was considered. The simulation result with the new method showed superior accuracy compared with traditional simulation method. The method presented could be used to predict the machining distortion for parts, which would allow further improvement in machining techniques.