Machines

This paper proposes a novel magnetic abrasive finishing (MAF) process that uses an auxiliary magnetic machining tool for the internal surface finishing of a thick-walled tube. The auxiliary magnetic machining tool and external poles form a closed magnetic field circuit. Thus, a stronger magnetic force can be generated during the process. In the current study, we focus on analyzing the distribution of the magnetic field and magnetic flux density and investigating the finishing characteristics of a mixed magnetic abrasive finishing process and speed of relative revolutions. Based on the finishing characteristics, we also conduct a stage-by-stage finishing process by changing the combinations of the mixed magnetic abrasive finishing process. The finishing quality of the internal surface was mainly evaluated by the measured roundness and surface roughness. The experimental results show that the roundness and surface roughness Ra are affected when the total amount of WA abrasive and iron powder is too much; a better surface roughness could be obtained when the difference in the speed of relative revolutions is considerable, but the roundness is the worst. Furthermore, the original roundness measurement of 270 µm can reach 10 µm, and the surface roughness Ra can increase from an original surface roughness of 4.1 µm to reach 10 nm after 105 min of the stage-by-stage finishing process. Full article

In this paper, active disturbance rejection control (ADRC) is applied to the vibration control of the all-clamped plate structure with an inertial actuator. Knowing that modeling uncertainties, dynamic nonlinearities and multivariable couplings are often the major causes of a downgrading performance and instability, a cascade ADRC controller is, hence, utilized to mitigate the effects of these issues. The dynamics regarding the all-clamped plate structure and inertial actuator are obtained through theoretical analysis and experimental testing. Furthermore, the real-time control experimental verification is carried out on the hardware-in-the-loop platform based on the NI PCIe-6343 data acquisition card. The comparative experimental results show that the proposed cascade ADRC controller has a better vibration suppression performance, disturbance rejection performance and decoupling ability. Full article

The tire is the only part that connects the vehicle and the road surface. Many important properties of vehicles are related to the mechanical properties of tires, such as handling stability, braking safety, vertical vibration characteristics, and so on. Although a great deal of research on tire dynamics has been completed, mainly focusing on steady-state tire force and moment characteristics, as well as linear unsteady force characteristics, less research has been conducted on nonlinear unsteady characteristics, especially when the vertical load changes dynamically. Therefore, the main purpose of this paper is to improve the tire unsteady-state model and verify it by experiment. To achieve this goal, we first study the nonlinear unsteady tire cornering theoretical model and obtain clear force and torque frequency response functions. Then, based on the results of the theoretical model, a high-precision and high-efficiency semi-physical model is developed. Finally, model identification and accuracy verification are carried out based on the bench test data. The model developed in this paper has high accuracy, and it significantly improves the expression of the aligning torque, which helps to improve the virtual simulation of transient conditions, such as vehicle handling and dynamic load conditions. Full article

Fused filament fabrication (FFF), an additive manufacturing process, is an emerging technology with issues in the uncertainty of mechanical properties and quality of printed parts. The consideration of all main and interaction effects when changing print parameters is not efficiently feasible, due to existing stochastic dependencies. To address this issue, a machine learning method is developed to increase reliability by optimizing input parameters and predicting system responses. A structure of artificial neural networks (ANN) is proposed that predicts a system response based on input parameters and observations of the system and similar systems. In this way, significant input parameters for a reliable system can be determined. The ANN structure is part of physics-informed machine learning and is pretrained with domain knowledge (DK) to require fewer observations for full training. This includes theoretical knowledge of idealized systems and measured data. New predictions for a system response can be made without retraining but by using further observations from the predicted system. Therefore, the predictions are available in real time, which is a precondition for the use in industrial environments. Finally, the application of the developed method to print bed adhesion in FFF and the increase in system reliability are discussed and evaluated. Full article

Online monitors of the running gears systems of high-speed trains play critical roles in ensuring operational safety and reliability. Status signals collected from high-speed train running gears are very complex regarding working environments, random noises and many other real-world constraints. This paper proposed fault detection (FD) models using canonical correlation analysis (CCA) and just-in-time learning (JITL) to process scalar signals of high-speed train gears, named as CCA-JITL. After data preprocessing and normalization, CCA transforms covariance matrices of high-dimension historical data into low-dimension subspaces and maximizes correlations between the most important latent dimensions. Then, JITL components formulate local FD models which utilize subsets of testing samples with larger Euclidean distances to training data. A case study introduced a novel system design of an online FD architecture and demonstrated that CCA-JITL FD models significantly outperformed traditional CCA models. The approach is applicable to other dimension reduction FD models such as PCA and PLS. Full article

Impact load is a kind of aperiodic excitation with a short action time and large amplitude, it had more significant effect on the structure than static load. The reconstruction (or identification namely) of impact load is of great importance for validating the structural strength. The aim of this article was to reconstruct the impact load accurately. An impact load identification method based on impulse response theory (IRT) and BP (Back Propagation) neural network is proposed. The excitation and response signals were transformed to the same length by extracting the peak value (amplitude of sine wave) in the rising oscillation period of the response. First, we deduced that there was an approximate linear relationship between the discrete-time integral of impact load and the amplitude of the oscillation period of the response. Secondly, a BP neural network was used to establish a linear relationship between the discrete-time integral of the impact load and the peak value in the rising oscillation period of the response. Thirdly, the network was trained and verified. The error between the actual maximum amplitude of impact load and the identification value was 2.22%. The error between the actual equivalent impulse and the identification value was 0.67%. The results showed that this method had high accuracy and application potential. Full article

In the production and manufacturing industry, factors such as rolling equipment and processes may cause various defects on the surface of the steel plate, which greatly affect the performance and subsequent machining accuracy. Therefore, it is essential to identify defects in time and improve the quality of production. An intelligent detection system was constructed, and some improved algorithms such as dataset enhancement, annotation and lightweight convolution neural network are proposed in this paper. (1) Compared with the original YOLOV5 (You Only Look Once), the precision is 0.924, and the inference time is 29.8 ms, which is 13.8 ms faster than the original model. Additionally, the parameters and calculations are also far less than YOLOV5. (2) Ablation experiments were designed to verify the effectiveness of the proposed algorithms. The overall accuracy was improved by 0.062; meanwhile, the inference time was reduced by 21.7 ms. (3) Compared with other detection models, although RetinaNet has the highest accuracy, it takes the longest time. The overall performance of the proposed method is better than other methods. This research can better meet the requirements of the industry for precision and real-time performance. It can also provide ideas for industrial detection and lay the foundation for industrial automation. Full article

Linear sliders are linear actuators using linear motors. It is used in many applications, such as factory lines and linear motor cars. In recent years, the demand for smaller semiconductor devices has been increasing due to the proliferation of smartphones. High-precision positioning of linear motors is needed because manufacturing semiconductor devices uses the stage with linear motors. However, linear motors have nonlinearity due to the position dependence of interlinkage flux. It affects precise positioning. In this study, the nonlinear characteristics due to the position dependence of the flux are expressed as a mathematical model by using a distributed constant magnetic circuit. A method compensating it using an operator-based feedback controller with the obtained mathematical model is proposed. The effectiveness of the proposed method is confirmed by simulating and experimenting with the reference following disturbance elimination. Full article

It is always an important and challenging issue to achieve an effective fault diagnosis in rotating machinery in industries. In recent years, deep learning proved to be a high-accuracy and reliable method for data-based fault detection. However, the training of deep learning algorithms requires a large number of real data, which is generally expensive and time-consuming. To cope with this, we proposed a Resnet classifier with model-based data augmentation, which is applied for bearing fault detection. To this end, a dynamic model was first established to describe the bearing system by adjusting model parameters, such as speed, load, fault size, and the different fault types. Large amounts of data under various operation conditions can then be generated. The training dataset was constructed by the simulated data, which was then applied to train the Resnet classifier. In addition, in order to reduce the gap between the simulation data and the real data, the envelop signals were used instead of the original signals in the training process. Finally, the effectiveness of the proposed method was demonstrated by the real bearing experimental data. It is remarkable that the application of the proposed method can be further extended to other mechatronic systems with a deterministic dynamic model. Full article

The returnability of the deep-sea mining pump has been a key issue restricting further development of deep-sea mining technologies. Although many research studies have been conducted on mining pump, the mining process still faces challenges. Particularly, the reflux capacities of externally developed mine pumps are often insufficient, resulting in blockages in the flow channels. In this study, we determine that the blade wrap angle is one of the key factors affecting the reflux of the ore pump, which is also based on earlier research. Therefore, a numerical simulation of the ore pump was performed using computational fluid dynamics–discrete element method, and it was determined to be beneficial to the reflux of particles. The hydraulic performance and reflux ability were studied via experiments. Full article

In the case of simultaneous localization and mapping, route planning and navigation are based on data captured by multiple sensors, including built-in cameras. Nowadays, mobile devices frequently have more than one camera with overlapping fields of view, leading to solutions where depth information can also be gathered along with ordinary RGB color data. Using these RGB-D sensors, two- and three-dimensional point clouds can be recorded from the mobile devices, which provide additional information for localization and mapping. The method of matching point clouds during the movement of the device is essential: reducing noise while having an acceptable processing time is crucial for a real-life application. In this paper, we present a novel ISVD-based method for displacement estimation, using key points detected by SURF and ORB feature detectors. The ISVD algorithm is a fitting procedure based on SVD resolution, which removes outliers from the point clouds to be fitted in several steps. The developed method removes these outlying points in several steps, in each iteration examining the relative error of the point pairs and then progressively reducing the maximum error for the next matching step. An advantage over relevant methods is that this method always gives the same result, as no random steps are included. Full article

In recent years, flexible continuum robots have been substantially developed. Absolute nodal coordinates formulation (ANCF) gives a feasible path for simulating the behavior of flexible robots. However, the model of finger-shaped robots is often regarded as a cylinder and characterized by a beam element. Obviously, this is short of characterizing the geometrical feature of fingers in detail, especially under bending conditions. Additionally, for the lower-order plate element, it is hard to characterize the bending behavior of the flexible finger due to fewer nodes; a higher-order plate element often requires an extremely long computing time. In this work, an improved ANCF lower-order plate element is used to increase the accuracy of the Yeoh model and characterize the geometrical structure of silicone rubber fingers, taking into particular consideration the effect of volume locks and multi-body system constraints. Since it is a kind of lower-order plate element, essentially, the computing time is nearly the same as that of conventional lower-order plate elements. The validity of this model was verified by comparing it with the results of the published reference. The flexible finger, manufactured using silicone rubber, is characterized by the novel ANCF lower-order plate element, whereby its mechanical deformation and bending behavior are simulated both efficiently and accurately. Compared to the ANCF beam element, conventional lower-order plate element, and higher-order plate element, the novel plate element in this paper characterizes the external contour of the finger better, reflects bending behavior more realistically, and converges in less computing time. Full article

Point cloud is a versatile geometric representation that could be applied in computer vision tasks. On account of the disorder of point cloud, it is challenging to design a deep neural network used in point cloud analysis. Furthermore, most existing frameworks for point cloud processing either hardly consider the local neighboring information or ignore context-aware and spatially-aware features. To deal with the above problems, we propose a novel point cloud processing architecture named TR-Net, which is based on transformer. This architecture reformulates the point cloud processing task as a set-to-set translation problem. TR-Net directly operates on raw point clouds without any data transformation or annotation, which reduces the consumption of computing resources and memory usage. Firstly, a neighborhood embedding backbone is designed to effectively extract the local neighboring information from point cloud. Then, an attention-based sub-network is constructed to better learn a semantically abundant and discriminatory representation from embedded features. Finally, effective global features are yielded through feeding the features extracted by attention-based sub-network into a residual backbone. For different downstream tasks, we build different decoders. Extensive experiments on the public datasets illustrate that our approach outperforms other state-of-the-art methods. For example, our TR-Net performs 93.1% overall accuracy on the ModelNet40 dataset and the TR-Net archives a mIou of 85.3% on the ShapeNet dataset for part segmentation. Full article

The fault quantification of a tapered roller bearing (TRB) can provide a reliable guide for predictive maintenance. Currently, damage size estimation based on vibration signals has been proposed and developed. However, most approaches are focused on the theory of ball bearings. Unlike the point contact of a ball bearing, the contact between the tapered roller and the raceway is a line contact. So, the current practices based on the micro-motion theory of ball bearings are limited when estimating the TRB’s fault. To accurately estimate the TRB’s damage size, a dynamic model of non-penetrating damaged TRB was established to research the vibration response mechanism and explain the influence. The model takes the deflection factor of a tapered roller into consideration and uses the elastohydrodynamic lubrication model to simulate the influence of the lubrication factor. Then, a revised formula for estimating the TRBs’ fault size is proposed by uncovering the relationship between the impact time and the damage location. Simulation analysis and experimental analysis prove the correctness of the dynamic model and the effectiveness of the size estimation formula. Full article

Fault diagnosis is essential for assuring the safety and dependability of rotating machinery systems. Several emerging techniques, especially artificial intelligence-based technologies, are used to overcome the difficulties in this field. In most engineering scenarios, machines perform in normal conditions, which implies that fault data may be hard to acquire and limited. Therefore, the data imbalance and the deficiency of labels are practical challenges in the fault diagnosis of machinery bearings. Among the mainstream methods, transfer learning-based fault diagnosis is highly effective, as it transfers the results of previous studies and integrates existing resources. The knowledge from the source domain is transferred via Domain Adversarial Training of Neural Networks (DANN) while the dataset of the target domain is partially labeled. A semi-supervised framework based on uncertainty-aware pseudo-label selection (UPS) is adopted in parallel to improve the model performance by utilizing abundant unlabeled data. Through experiments on two bearing datasets, the accuracy of bearing fault classification surpassed the independent approaches. Full article

More hemiplegia patients tend to use equipment for rehabilitation training due to the lack of physical therapists and the low effect of manual training. Nowadays, lower limb rehabilitation training devices for patients in grade 2 of the Medical Research Council (MRC-2) scale are still scarce and have some issues of poor autonomy and cannot relieve the muscle weakness of patients. To address these problems, a prototype based on gravity balance was designed with the combination of springs and linkages to enable patients to passively experience the rehabilitation training in the state of balancing the gravity of lower limbs. The motion of the mechanism was analyzed to obtain the functional relation between the motor rotation angle and the joints’ angle. Based on the principle of constant potential energy, a gravity balance mathematical model of the device was established, analyzed, and simulated. Moreover, through the training experiment, the results show that when subjects in three different weights were trained under the rehabilitation device with and without gravity balance, the required torques of the motor and EMG signal strength of the knee and hip joints decreased by a degree of significance, which verified the effectiveness of the device’s gravity balancing characteristics for MRC-2 patients. Full article

This work addresses the end-effector trajectory-tracking force and motion control of a three-dimensional three-link robot considering measurement noises. The last two links of the manipulator are considered as structurally flexible. An absolute coordinate approach is used while obtaining the dynamic equations to avoid complex dynamic equations. In this approach, each link is modeled as if there is no connection between the links. Then, joint connections are expressed as constraint equations. After that, these constraint equations are used in dynamic equations to decrease the number of equations. Then, the resulting dynamic equations are transformed into a form which is suitable for controller design. Furthermore, the dynamic equations are divided as pseudostatic equilibrium and deviation equations. The control torques resulting from the pseudostatic equilibrium and the elastic deflections are obtained easily as the solution of algebraic equations. On the other hand, the control torques corresponding to the deviations are obtained without any linearization. Encoders, strain gauges, position sensors and force and moment sensors are required for measurements. Low pass filters are considered for the sensors. For the crossover frequencies of the sensors, low and high values are chosen to observe the filtering effect on the robot output. Full article

Scientifically and accurately predicting the state of health (SOH) and remaining useful life (RUL) of batteries is the key technology of automotive battery management systems. The selection of the health indicator (HI) that characterizes battery aging affects the accuracy of the prediction model construction, which in turn affects the accuracy of SOH and RUL estimation. Therefore, this paper analyzes the current status of HI selection for lithium-ion batteries by systematically reviewing the existing literature on the selection of HIs. According to the relationship between HI and battery aging, battery HI can be divided into two categories: direct HI and indirect HI. The capacity and internal resistance of the battery can directly represent the aging degree of the battery and are the direct HIs of the battery. Indirect HIs refer to characteristic parameters extracted from battery charge and discharge data that can characterize the degree of battery aging. This paper analyzes and summarizes the advantages and disadvantages of various HIs and indirect HIs commonly used in current research, providing useful support and reference for future researchers in selecting HIs to characterize battery aging. Finally, in view of the capacity regeneration phenomenon in the aging process of the battery, the selection direction of future HI is proposed. Full article

Bolt connection is an important component in mechanical structure which significantly affects the dynamic property of the whole structure. In this paper, a three-dimensional transition interface model which contains geometric and physical parameters is proposed to model the bolted joint based on the contact analysis. The geometric parameters and the physical parameters are used to characterize the influence of contact area and contact pressure which are related to connection parameters such as material, roughness, connection thickness, and tightening force, respectively. After that, the geometric parameter identification method is proposed, and the geometric parameter database of bolt joints for machine tools is constructed based on the Kriging interpolation method. Then, the model updating method based on the combination of modal parameters and frequency response function is proposed to identify the physical parameters and thickness of the three-dimensional transition interface model. The database of the transition interface model is constructed after verifying the validity of the proposed model. Finally, an engineering example of an engraving machine tool is used to check the practicability of the proposed transition interface model and the usage of the constructed parameter database. Full article