Search

Industrial radiography is a pivotal non-destructive testing (NDT) method that ensures quality and safety in a wide range of industrial sectors. Nevertheless, the conventional human-based approaches to carrying out industrial radiography are prone to challenges that negatively impact the accuracy and efficiency of defect detection. To solve this challenge, numerous computer-based alternatives have been developed, including Automated Defect Recognition (ADR) using deep learning algorithms. At the core of training, these ADR algorithms demand large volumes of qualitative data that should be representative of real-world cases to be expected during deployment. However, the availability of digital X-ray radiography data, especially for open research, is limited by non-disclosure contractual terms in the industry. In this study, we present a pipeline that is capable of modelling synthetic images based on real digital X-ray radiography images. This is achieved through a systematic analysis of the intensity distribution, considering grey value (GV) statistical uniqueness related to exposure conditions used during image acquisition, type of imaged component, material thickness variations, X-ray beam divergence, anode heel effect, scatter radiation, edge delineation, etc. The generated synthetic images were exclusively utilized to train a deep learning model, yielding an impressive model performance with mean intersection over union (IoU) of 0.93, and mean dice coefficient of 0.96 when tested on real unseen digital X-ray radiography images. The presented methodology is scalable and adaptable, making it suitable for diverse industrial applications.

In order to reduce the energy consumption of the welding robot and ensure the cooperative movement of the robot joints, a trajectory planning method with optimal energy consumption based on improved sparrow search algorithm is proposed. Firstly, the trajectory planning model with optimal energy consumption is established based on the joint torque and angular velocity of the robot. To make the velocity, acceleration and jerk of each joint of the robot be bounded and continuous, the joint space trajectory is constructed with seventh degree B-spline curve. The total energy consumption of the robot is calculated by combining kinematic and dynamic parameters. On the basis of improved sparrow search algorithm, the time series corresponding to the optimal energy consumption is solved by using elite reverse learning, non-dominated sorting and Gaussian-Cauchy variation strategy, and then the optimal continuous motion trajectory of energy consumption is planned. The simulation results show that the proposed method can not only achieve continuous smooth control objective, but also effectively reduce energy consumption.

Intelligent robots are often used to explore various areas instead of humans. However, when the driving joint is damaged, the actuated robot degenerates to an underactuated robot, and the traditional control method is not suitable for the underactuated robot. In this work, a two-stage control approach for a planar prismatic–rotational (PR) underactuated robot is introduced. Firstly, we establish the dynamic model and describe the underactuated constraint between an underactuated rotational joint and active prismatic joint. Secondly, the trajectory with multiple parameters is planned to ensure that the two joints reach the target position. Based on underactuated constraints and the evaluation function, the differential evolution algorithm (DEA) is used to optimize these parameters. After that, in stage 1, we design the controller to move the active prismatic joint to the desired position. Meanwhile, the underactuated rotational joint is rotating freely. In stage 2, we design the controller for the active prismatic joint to track the planned trajectory. By means of this strategy, both joints reach their target locations simultaneously. The final simulation result demonstrates that this strategy is effective.

For the planar robot with underactuated constraints, a stable control method is presented on the foundation of motion planning method and intelligent optimization, which includes two stages. (1) Designing the controllers to control the actuated manipulators to given target states. (2) Planning the motion trajectory combined with the underactuated constraints between all links, using the intelligent algorithm to find the adaptable trajectory parameters, and tracking such planned trajectories to control full manipulators to the given states simultaneously. At last, multigroup simulations demonstrate the validity of the proposed method.

Milk fat synthesis in cows mainly includes the synthesis of short- and medium-chain fatty acids, the uptake, transport, and activation of long-chain fatty acids (LCFAs), the synthesis of triglycerides, and the synthesis of the genes, transcription factors, and signaling pathways involved. Although the various stages of milk fat synthesis have been outlined in previous research, only partial processes have been revealed. CLA consists of an aggregation of positional and geometric isomers of linoleic fatty acid, and the accumulated evidence suggests that the two isomers of the active forms of CLA (cis-9, trans-11 conjugated linoleic acid and trans-10, cis-12 conjugated linoleic acid, abbreviated as c9, t11-CLA and t10, c12-CLA) can reduce the fat content in milk by regulating lipogenesis, fatty acid (FA) uptake, oxidation, and fat synthesis. However, the mechanism through which CLA inhibits milk fat synthesis is unique, with most studies focusing only on the effects of CLA on one of the genes, transcription factors, or signaling pathways involved. In this study, we summarized the structure and function of classic genes and pathways (mTOR, SREBP, AMPK, and PPARG) and new genes or pathways (THRSP, METTL3, ELOVL, and LPIN1) involved in each stage of milk fat synthesis and demonstrated the interactions between genes and pathways. We also examined the effects of other substances (melanin, nicotinic acid, SA, etc.). Furthermore, we evaluated the influence of β-sitosterol, sodium butyrate, Met arginine, and Camellia oleifera Abel on milk fat synthesis to improve the mechanism of milk fat synthesis in cows and provide a mechanistic reference for the use of CLA in inhibiting milk fat biosynthesis.

The cart–inverted pendulum system (CIPS) is a typical example of underactuated mechanical systems. For the CIPS with friction and disturbances, a gain-scheduled model predictive control method is proposed to achieve the upright stabilization objective of the single inverted pendulum (SIP) while controlling the cart to reach a desired new position. To this end, first, a dynamic equation of the CIPS with friction and disturbances is formulated based on the Newton–Euler equation. On the basis of the dynamic equation of the CIPS, its motion characteristics and control process are analyzed. Next, the given dynamic equation of the CIPS is linearized to obtain a series of linearized models at seven different pendulum angles. Then, seven model predictive controllers (MPCs) are designed based on the above-linearized models, respectively. Introducing the idea of the gain-schedule, a gain-scheduled MPC (GSMPC) is designed to switch one of these seven MPCs to realize the control objective of the CIPS, according to the actual pendulum angle of the SIP during the control process. Finally, multi-group simulations that consider the friction and disturbances of the CIPS are implemented to demonstrate the effectiveness of the proposed gain-scheduled model predictive control method.

Pulsed thermography is an established nondestructive evaluation technology that excels at detecting and characterizing subsurface defects within specimens. A critical challenge in this domain is the accurate estimation of defect depth. In this paper, a new publicly accessible pulsed infrared dataset for PVC specimens is introduced. It was enriched with 3D positional information to advance research in this area. To ensure the labeling quality, a comparative analysis of two distinct data labeling methods was conducted. The first method is based on human domain expertise, while the second method relies on 3D CAD images. The analysis showed that the CAD-based labeling method noticeably enhanced the precision of defect dimension quantification. Additionally, a sophisticated deep learning model was employed on the data, which were preprocessed by different methods to predict both the two-dimensional coordinates and the depth of the identified defects.

Industrial radiography is a pivotal non-destructive testing (NDT) method that ensures quality and safety in a wide range of industrial sectors. Conventional human-based approaches, however, are prone to challenges in defect detection accuracy and efficiency, primarily due to the high inspection demand from manufacturing industries with high production throughput. To solve this challenge, numerous computer-based alternatives have been developed, including Automated Defect Recognition (ADR) using deep learning algorithms. At the core of training, these algorithms demand large volumes of data that should be representative of real-world cases. However, the availability of digital X-ray radiography data for open research is limited by non-disclosure contractual terms in the industry. This study presents a pipeline that is capable of modeling synthetic images based on statistical information acquired from X-ray intensity distribution from real digital X-ray radiography images. Through meticulous analysis of the intensity distribution in digital X-ray images, the unique statistical patterns associated with the exposure conditions used during image acquisition, type of component, thickness variations, beam divergence, anode heel effect, etc., are extracted. The realized synthetic images were utilized to train deep learning models, yielding an impressive model performance with a mean intersection over union (IoU) of 0.93 and a mean dice coefficient of 0.96 on real unseen digital X-ray radiography images. This methodology is scalable and adaptable, making it suitable for diverse industrial applications.

In this paper, we propose a new optimal trajectory planning method for the manipulator to optimize its operating efficiency and ensure the smoothness of the motion process. The position sequences in joint space corresponding to a specified trajectory in task space are obtained by using the inverse kinematic algorithm, and the seventh-degree B-spline curve interpolation method is used to construct the joint trajectory with controllable start–stop kinematic parameters, and continuous velocity, acceleration and jerk. The kinematic constraints of the manipulator are transformed into the control point constraints of the B-spline curve, the sequential quadratic programming method is used to solve the optimal motion time node, and then the time-optimal continuous jerk trajectory satisfying the nonlinear kinematic constraints is planned. Simulation results show that the proposed trajectory planning method provides the ideal trajectory for the joint controller, ensuring the manipulator can smoothly track any specified trajectory in the task space in the shortest time.

In response to the growing inspection demand exerted by process automation in component manufacturing, non-destructive testing (NDT) continues to explore automated approaches that utilize deep-learning algorithms for defect identification, including within digital X-ray radiography images. This necessitates a thorough understanding of the implication of image quality parameters on the performance of these deep-learning models. This study investigated the influence of two image-quality parameters, namely signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), on the performance of a U-net deep-learning semantic segmentation model. Input images were acquired with varying combinations of exposure factors, such as kilovoltage, milli-ampere, and exposure time, which altered the resultant radiographic image quality. The data were sorted into five different datasets according to their measured SNR and CNR values. The deep-learning model was trained five distinct times, utilizing a unique dataset for each training session. Training the model with high CNR values yielded an intersection-over-union (IoU) metric of 0.9594 on test data of the same category but dropped to 0.5875 when tested on lower CNR test data. The result of this study emphasizes the importance of achieving a balance in training dataset according to the investigated quality parameters in order to enhance the performance of deep-learning segmentation models for NDT digital X-ray radiography applications.

Pulsed thermography is an indispensable tool in the field of non-destructive evaluation. However, the data generated by this technique can be challenging to analyze and require expertise to interpret. With the rapid progress in deep learning, image segmentation has become a well-established area of research. This has motivated efforts to apply deep learning methods to non-destructive evaluation data processing, including pulsed thermography. Despite this trend, there has been a lack of public pulsed thermography datasets available for the evaluation of various spatial-temporal deep learning models for segmentation tasks. This paper aims to address this gap by presenting the PVC-Infrared dataset for deep learning. In addition, we evaluated the performance of popular deep-learning-based instance segmentation models on this dataset. Furthermore, we examined the effect of the number of frames and data transformations on the performance of these models. The results of this study suggest that appropriate preprocessing techniques can significantly reduce the size of the data while maintaining the performance of deep learning models, thereby speeding up the data processing process. This highlights the potential for using deep learning methods to make non-destructive evaluation data analysis more efficient and accessible to a wider range of practitioners.

Advanced materials such as continuous carbon fiber-reinforced thermoplastic (CFRP) laminates are commonly used in many industries, mainly because of their strength, stiffness to weight ratio, toughness, weldability, and repairability. Structural components working in harsh environments such as satellites are permanently exposed to some sort of damage during their lifetimes. To detect and characterize these damages, non-destructive testing and evaluation techniques are essential tools, especially for composite materials. In this study, artificial intelligence was applied in combination with infrared thermography to detected and segment impact damage on curved laminates that were previously submitted to a severe thermal stress cycles and subsequent ballistic impacts. Segmentation was performed on both mid-wave and long-wave infrared sequences obtained simultaneously during pulsed thermography experiments by means of a deep neural network. A deep neural network was trained for each wavelength. Both networks generated satisfactory results. The model trained with mid-wave images achieved an F1-score of 92.74% and the model trained with long-wave images achieved an F1-score of 87.39%.