Search Results (61)

A novel and innovative method was developed to fabricate defect-free composite hollow fiber (HF) membranes using drop-casting under continuous flow. The synthesized Pebax-1657—based membranes were examined for gas separation processes, focusing on the separation of CO₂ from CH₄ and CO gases. The separation performance of the membranes was rigorously assessed under realistic binary gas mixture conditions to evaluate their selectivity and performance. The effect of pressure on separation performance was systematically investigated, with transmembrane pressures up to 10 bar being applied at a temperature of 298 K. Remarkable CO₂/CH₄ selectivities of up to 110 and CO₂/CO selectivities of up to 48 were achieved, demonstrating the robustness and effectiveness of these composite HF membranes, suggesting their suitability for high-performance gas separation processes under varying operational conditions. Overall, this study introduces a novel approach for scaling up the fabrication of HF membranes and provides valuable insights into their application in CO₂ separation technologies, offering the potential for advancements in areas such as natural gas processing and carbon capture from CO-containing streams. Full article

Directed Energy Deposition Laser Beam (DED-LB) is a promising additive manufacturing technique that uses a laser source and a powder stream to build or repair metal components. Repair applications offer significant economic and environmental benefits but are more challenging to develop, especially for components that are difficult to process due to their intricate geometries and materials. Process conditions can change precipitously, and it is essential to implement monitoring systems that ensure high process stability and, consequently, superior end-product quality. In the present work, a mid-wave infrared coaxial camera was used to monitor the melt pool geometry. To simulate the challenging repair process conditions of the DED-LB process, experimental tests were carried out on substrates with different thicknesses. The stability of the deposition process on nickel-based superalloys was analyzed by means of MATLAB algorithms. Thus, the effect of open-loop and closed-loop monitoring with back control on laser power on the process conditions was assessed and quantified. Metallographic analysis of the produced samples was carried out to validate the analyses performed by the monitoring system. The occurrence of production defects (lack of fusion and porosity) related to parameters not directly controllable by monitoring systems, such as penetration depth and dilution, was determined. Full article

In tasks such as wood defect repair and the production of high-end wooden furniture, ensuring the consistency of the texture in repaired or jointed areas is crucial. This paper proposes the WTSM-SiameseNet model for wood-texture-similarity matching and introduces several improvements to address the issues present in traditional methods. First, to address the issue that fixed receptive fields cannot adapt to textures of different sizes, a multi-receptive field fusion feature extraction network was designed. This allows the model to autonomously select the optimal receptive field, enhancing its flexibility and accuracy when handling wood textures at different scales. Secondly, the interdependencies between layers in traditional serial attention mechanisms limit performance. To address this, a concurrent attention mechanism was designed, which reduces interlayer interference by using a dual-stream parallel structure that enhances the ability to capture features. Furthermore, to overcome the issues of existing feature fusion methods that disrupt spatial structure and lack interpretability, this study proposes a feature fusion method based on feature correlation. This approach not only preserves the spatial structure of texture features but also improves the interpretability and stability of the fused features and the model. Finally, by introducing depthwise separable convolutions, the issue of a large number of model parameters is addressed, significantly improving training efficiency while maintaining model performance. Experiments were conducted using a wood texture similarity dataset consisting of 7588 image pairs. The results show that WTSM-SiameseNet achieved an accuracy of 96.67% on the test set, representing a 12.91% improvement in accuracy and a 14.21% improvement in precision compared to the pre-improved SiameseNet. Compared to CS-SiameseNet, accuracy increased by 2.86%, and precision improved by 6.58%. Full article

Laser cladding has unique technical advantages, such as precise heat input control, excellent coating properties, and local selective cladding for complex shape parts, which is a vital branch of surface engineering. During the laser cladding process, the parts are subjected to extreme thermal gradients, leading to the formation of micro-defects such as cracks, pores, and segregation. These defects compromise the serviceability of the components. Ultrasonic vibration can produce thermal, mechanical, cavitation, and acoustic flow effects in the melt pool, which can comprehensively affect the formation and evolution for the microstructure of the melt pool and reduce the microscopic defects of the cladding layer. In this paper, the coupling model of temperature and flow field for the laser cladding of 45 steel 316L was established. The transient evolution laws of temperature and flow field under ultrasonic vibration were revealed from a macroscopic point of view. Based on the phase field method, a numerical model of dendrite growth during laser cladding solidification under ultrasonic vibration was established. The mechanism of the effect of ultrasonic vibration on the solidification dendrite growth during laser cladding was revealed on a mesoscopic scale. Based on the microstructure evolution model of the paste region in the scanning direction of the cladding pool, the effects of a static flow field and acoustic flow on dendrite growth were investigated. The results show that the melt flow changes the heat and mass transfer behaviors at the solidification interface, concurrently changing the dendrites’ growth morphology. The acoustic streaming effect increases the flow velocity of the melt pool, which increases the tilt angle of the dendrites to the flow-on side and promotes the growth of secondary dendrite arms on the flow-on side. It improves the solute distribution in the melt pool and reduces elemental segregation. Full article

Small object detection, which is frequently applied in defect detection, medical imaging, and security surveillance, often suffers from low accuracy due to limited feature information and blurred details. This paper proposes a small object detection method named YOLO-DHGC, which employs a two-stream structure with dense connections. Firstly, a novel backbone network, DenseHRNet, is introduced. It innovatively combines a dense connection mechanism with high-resolution feature map branches, effectively enhancing feature reuse and cross-layer fusion, thereby obtaining high-level semantic information from the image. Secondly, a two-stream structure based on an edge-gated branch is designed. It uses higher-level information from the regular detection stream to eliminate irrelevant interference remaining in the early processing stages of the edge-gated stream, allowing it to focus on processing information related to shape boundaries and accurately capture the morphological features of small objects. To assess the effectiveness of the proposed YOLO-DHGC method, we conducted experiments on several public datasets and a self-constructed dataset. Exceptionally, a defect detection accuracy of 96.3% was achieved on the Market-PCB public dataset, demonstrating the effectiveness of our method in detecting small object defects for industrial applications. Full article

Large Multimodal Models are emerging general AI models capable of processing and analyzing diverse data streams, including text, imagery, and sequential data. This paper explores the possibility of exploiting multimodality to develop more interpretable AI-based predictive tools for the water sector, with a first application for sewer defect detection from CCTV imagery. To this aim, we test the zero-shot generalization performance of three generalist large language-vision models for binary sewer defect detection on a subset of the SewerML dataset. We compared the LMMs against a state-of-the-art unimodal Deep Learning approach which has been trained and validated on >1 million SewerML images. Unsurprisingly, the chosen benchmark showcases the best performances, with an overall F1 Score of 0.80. Nonetheless, OpenAI GPT4-V demonstrates relatively good performances with an overall F1 Score of 0.61, displaying equal or better results than the benchmark for some defect classes. Furthermore, GPT4-V often provides text descriptions aligned with the provided prediction, accurately describing the rationale behind a certain decision. Similarly, GPT4-V displays interesting emerging behaviors for trustworthiness, such as refusing to classify images that are too blurred or unclear. Despite the significantly lower performance from the open-source models CogVLM and LLaVA, some preliminary successes suggest good potential for enhancement through fine-tuning, agentic workflows, or retrieval-augmented generation. Full article

As semiconductor chip manufacturing technology advances, chip structures are becoming more complex, leading to an increased likelihood of void defects in the solder layer during packaging. However, identifying void defects in packaged chips remains a significant challenge due to the complex chip background, varying defect sizes and shapes, and blurred boundaries between voids and their surroundings. To address these challenges, we present a deep-learning-based framework for void defect segmentation in chip packaging. The framework consists of two main components: a solder region extraction method and a void defect segmentation network. The solder region extraction method includes a lightweight segmentation network and a rotation correction algorithm that eliminates background noise and accurately captures the solder region of the chip. The void defect segmentation network is designed for efficient and accurate defect segmentation. To cope with the variability of void defect shapes and sizes, we propose a Mamba model-based encoder that uses a visual state space module for multi-scale information extraction. In addition, we propose an interactive dual-stream decoder that uses a feature correlation cross gate module to fuse the streams’ features to improve their correlation and produce more accurate void defect segmentation maps. The effectiveness of the framework is evaluated through quantitative and qualitative experiments on our custom X-ray chip dataset. Furthermore, the proposed void defect segmentation framework for chip packaging has been applied to a real factory inspection line, achieving an accuracy of 93.3% in chip qualification. Full article

Building information modelling (BIM) is an appreciated technology in the field of architecture and construction management. Collaboration of information in BIM has not been fully utilized in the structural engineering stream as many engineers keep on working with previous prevailing design approaches. Failure to adequately facilitate automation in design could lead to structural defects, construction rework, or even structural clashes, with major financial implications. Given the inherent complexity of large-scale construction projects, the ‘manual design and detailing’ of structure is a challenging task and prone to human errors. Against this backdrop, this study developed a 4D building information management approach to facilitate automated structural models for professionals designing all the elements required in reinforced concrete (RC) structures like slabs, beams, and columns. The main contribution of this study is to obtain structural models directly from architecture models automatically, which reduces effort and possible errors in the previous prevailing approaches. The framework enables execution of all the model design works automatically through coding. This is achieved by executing a script which is beneficial for integrated project delivery (IPD). The 3D structural model in BIM software presented in this study extracts and transfers the geometrical data and links these data in Industry Foundation Classes (IFC) files using integration facilitated by Python 3.6 and IFCopenshell. The developed automated programme framework offers a cost-effective and accurate methodology to address the limitations and inefficiencies of traditional methods of structural modelling, which had been carried out manually. The authors have developed a novel tool to extract structural models from architectural models without proprietary software, greatly benefiting BIM managers by enhancing 3D BIM models. This advancement toward Open BIM, crucial for the architecture, engineering, and construction (AEC) industry’s future, is accessible to educators and beginners and highlights BIM’s effectiveness in improving structural analysis and productivity. The core finding of this study is to generate a structural model from an architecture model by automating the script with Python integration of IFC and IFCopenshell. The merits of the developed framework are reduced clashes, more economical structural modelling, and fully automated smart work as functions of the IPD. Full article

Operational amplifiers (op amps) are fundamental blocks that find wide application both as stand-alone devices and as crucial blocks embedded in various Systems on Chips (SoCs). Achieving high defect coverage, as well as performing defect localization in these circuits, has proven to be a difficult/expensive task, even with sophisticated testing circuitry. The ISO 26262 standard for functional safety (FuSa) includes the stringent requirement that an automotive IC must have a very high defect coverage. This reinforces the need to ensure the functionality of analog and mixed (AMS) circuits, especially in mission critical applications. This paper presents an all-digital op amp defect detection, diagnosis, and localization method that can be used both for production and in-field tests and discusses various implementation of the proposed method. We validated our results using extensive transistor-level simulations of multiple op amp architectures using TSMC 180 nm technology. Across op amp architectures and multiple implementation approaches, we achieved a worst-case and best-case defect coverage of 94.5% and 99%, respectively. Furthermore, in this work, we also propose a defect diagnosis and localization strategy using recorded bit streams from states of digital injectors and detectors. Full article

Weld line defects, commonly occurring during the plastic product manufacturing process, are caused by the merging of two opposing streams of molten plastic. The presence of weld lines harms the product’s aesthetic appeal and durability. This study uses artificial neural networks to forecast the ultimate tensile strength of a PA6 composite incorporating 30% glass fibers (GFs). Data were collected from tensile strength tests and the technical parameters of injection molding. The packing pressure factor is the one that significantly affects the tensile strength value. The melt temperature has a significant impact on the product’s strength as well. In contrast, the filling time factor has less impact than other factors. According to the scanning electron microscope result, the smooth fracture surface indicates the weld line area’s high brittleness. Fiber bridging across the weld line area is evident in numerous fractured GF pieces on the fracture surface, which enhances this area. Tensile strength values vary based on the injection parameters, from 65.51 MPa to 73.19 MPa. In addition, the experimental data comprise the outcomes of the artificial neural networks (ANNs), with the maximum relative variation being only 4.63%. The results could improve the PA6 reinforced with 30% GF injection molding procedure with weld lines. In further research, mold temperature improvement should be considered an exemplary method for enhancing the weld line strength. Full article

Recently, salt-assisted pyrolyzation has been deemed an emerging and efficient method for the preparation of biochars due to its facile operation as well as its good structural and chemical properties. In this work, biochars (MBCx) are prepared by heating bamboo powders in eutectic salts (Li₂CO₃ + K₂CO₃) at 500–600 °C in the air. Multiple technologies are employed to examine the physiochemical properties of bamboo biochars. Correlations between heating temperature and structural features and carbon dioxide uptakes of bamboo biochars have been investigated. The results show that heating temperature has a significant influence on the physicochemical properties of bamboo biochars. With the elevation of the heating temperature, the defect structures of bamboo biochars gradually ascend, especially when the heating temperature reaches 600 °C. MBCx biochars visibly exceed conventional bamboo biochar prepared via pyrolyzation in a nitrogen stream free of salt addition. Pyrolysis of bamboo in eutectic salts endows biochars with higher oxygen content and more carbon defects, which likely accounts for their better CO₂ capture activities. Full article

In laser powder bed fusion processes, keyholes are the gaseous cavities formed where laser interacts with metal, and their morphologies play an important role in defect formation and the final product quality. The in-situ X-ray imaging technique can monitor the keyhole dynamics from the side and capture keyhole shapes in the X-ray image stream. Keyhole shapes in X-ray images are then often labeled by humans for analysis, which increasingly involves attempting to correlate keyhole shapes with defects using machine learning. However, such labeling is tedious, time-consuming, error-prone, and cannot be scaled to large data sets. To use keyhole shapes more readily as the input to machine learning methods, an automatic tool to identify keyhole regions is desirable. In this paper, a deep-learning-based computer vision tool that can automatically segment keyhole shapes out of X-ray images is presented. The pipeline contains a filtering method and an implementation of the BASNet deep learning model to semantically segment the keyhole morphologies out of X-ray images. The presented tool shows promising average accuracy of 91.24% for keyhole area, and 92.81% for boundary shape, for a range of test dataset conditions in Al6061 (and one AliSi10Mg) alloys, with 300 training images/labels and 100 testing images for each trial. Prospective users may apply the presently trained tool or a retrained version following the approach used here to automatically label keyhole shapes in large image sets. Full article

Deep learning based visual cognition has greatly improved the accuracy of defect detection, reducing processing times and increasing product throughput across a variety of manufacturing use cases. There is however a continuing need for rigorous procedures to dynamically update model-based detection methods that use sequential streaming during the training phase. This paper reviews how new process, training or validation information is rigorously incorporated in real time when detection exceptions arise during inspection. In particular, consideration is given to how new tasks, classes or decision pathways are added to existing models or datasets in a controlled fashion. An analysis of studies from the incremental learning literature is presented, where the emphasis is on the mitigation of process complexity challenges such as, catastrophic forgetting. Further, practical implementation issues that are known to affect the complexity of deep learning model architecture, including memory allocation for incoming sequential data or incremental learning accuracy, is considered. The paper highlights case study results and methods that have been used to successfully mitigate such real-time manufacturing challenges. Full article

The present work is focused on nickel catalysts supported on La₂O₃-CeO₂ binary oxides without and with the addition of Cu to the active component for the dry reforming of methane (DRM). The catalysts are characterized using XRD, XRF, TPD-CO₂, TPR-H₂, and low-temperature N₂ adsorption–desorption methods. This work shows the effect of different La:Ce ratios (1:1 and 9:1) and the Cu addition on the structural, acid base, and catalytic properties of Ni-containing systems. The binary LaCeO_x oxide at a ratio of La:Ce = 1:1 is characterized by the formation of a solid solution with a fluorite structure, which is preserved upon the introduction of mono- or bimetallic particles. At La:Ce = 9:1, La₂O₃ segregation from the solid solution structure is observed, and the La excess determines the nature of the precursor of the active component, i.e., lanthanum nickelate. The catalysts based on LaCeO_x (1:1) are prone to carbonization during 6 h spent on-stream with the formation of carbon nanotubes. The Cu addition facilitates the reduction of the Cu-Ni catalyst carbonization and increases the number of structural defects in the carbon deposition products. The lanthanum-enriched LaCeO_x (9:1) support prevents the accumulation of carbon deposition products on the surface of CuNi/La₂O₃-CeO₂ 9:1, providing high DRM activity and an H₂/CO ratio of 0.9. Full article

To avoid overloading a network, it is critical to continuously monitor the natural environment and disseminate data streams in synchronization. Based on self-maintaining technology, this study presents a technique called self-configuration management (SCM). The purpose is to ensure consistency in the performance, functionality, and physical attributes of a wireless sensor network (WSN) over its lifetime. During device communication, the SCM approach delivers an operational software package for the radio board of system problematic nodes. We offered two techniques to help cluster heads manage autonomous configuration. First, we created a separate capability to determine which defective devices require the operating system (OS) replica. The software package was then delivered from the head node to the network’s malfunctioning device via communication roles. Second, we built an autonomous capability to automatically install software packages and arrange the time. The simulations revealed that the suggested technique was quick in transfers and used less energy. It also provided better coverage of system fault peaks than competitors. We used the proposed SCM approach to distribute homogenous sensor networks, and it increased system fault tolerance to 93.2%. Full article