DEEP LEARNING FRAMEWORK FOR WOVEN COMPOSITE ANALYSIS

paper, we focus on exploring the relationship between weave patterns and their mechanical properties in woven fiber composites through Machine Learning. Specifically, we explore the interactions between woven architectures and in-plane stiffness properties through Deep Convolutional Neural Network (DCNN) and Generative Adversarial Network (GAN). Our research is important for exploring how woven composite’s pattern is related to its mechanical properties and accelerating woven composite design as well as optimization. We focus on two tasks: (1) Stiffness prediction: Predicting in-plane stiffness properties for given weave patterns. Our DCNN extracts high-level features through several convolutional and fully connected layers to determine the final predictions. (2) Weave pattern prediction: Predicting weave patterns for target stiffness properties, which can be treated as the reverse task of the first one. Due to many-to-one mapping between weave patterns and the composite properties, we utilize a Decoder Neural Network as our baseline model and compare its performance with GAN and Genetic Algorithm. We represent the weave patterns as 2D checkerboard models and use finite element analysis (FEA) to determine in-plane stiffness properties, which serve as input data for our ML framework. We show that: (1) for stiffness prediction, DCNN can predict stiffness values for a given weave pattern with relatively high accuracy (above 93%); (2) for weave pattern prediction, the GAN model gives the best prediction accuracy (above 92%) while Decoder Neural Network has the best time efficiency. HAOTIAN FENG

Experimental Heat Transfer and Boundary Layer Measurements on a Ceramic Matrix Composite Surface

Ceramic matrix composites (CMCs) are quickly becoming more prevalent in the design of gas turbines due to their advantageous weight and thermal properties. While there are many advantages, the CMC surface morphology differs from that of conventional cast airfoil components. Despite a great deal of research focused on the material properties of CMCs, little public work has been done to investigate the impact that the CMC surface morphology has on the boundary layer development and resulting heat transfer. In this study, a scaled-up CMC weave pattern was developed and tested in a low-speed wind tunnel to evaluate both heat transfer and boundary layer characteristics. Results from these experiments indicate that the CMC weave pattern results in augmented heat transfer and flow field properties that significantly vary locally when compared with a smooth surface.

Lithuanian National Costume in the 19th Century and in the 2nd Half of the 20th Century: Cultural Pollution and Remains of Authenticity

Lithuanian authors, authors abroad, and artists have presented Lithuanian folk clothes in their works. However, the oldest examples of these representations are not very reliable, because the authors painted them according to the descriptions of other people or copied works among each other. In the 20th century, the national costume of Lithuania changed considerably. Attention was not given to ethnographic regional peculiarities; instead, similar materials were chosen without any analysis. This article performs a comparative analysis of folk (the 19th century to the first half of the 20th century) and national (the second half of the 20th century) Lithuanian costumes to establish signs of cultural pollution and remaining authenticity. Over 500 articles of clothing with different purposes are collected from Lithuanian museums. Fabric parameters, such as raw materials, weaving technique, weave, pattern, decoration elements, etc., are established. The research results show that authentic folk clothes of the 19th century differ from the national costume of the second half of the 20th century in their cut, decoration, and patterns. No differences between ethnographic regions survived in the national costumes. Thus, at present, we must preserve our tangible heritage and re-create, as authentically as possible, national costume for folk songs and dance ensembles, folk restaurants, and rural tourism homesteads.

Detection of Remote Sensing Warp Tension during Weaving on Plain Twill and Satin Fabric

Warp tensions were measured while a machine was operating on a woven cotton fabric with three different woven patterns. This study was carried out with image analysis methods using a high speed camera. Three weave pattern types: plain, twill and satin were woven on the same weaving machine, and thus it could be understood how weave pattern differences affect warp tension. Each of these three weaves was woven in three weft densities: 20, 28 and 45 wefts per cm. These fabrics were able to be made on a weaving machine with an automatic dobby. It was aimed to investigate warp tension differences for three basic weave patterns while keeping all machine settings constant. The weave settings of the dobby were changed for plain, twill and satin weaves. Warp tension calculation was based on the warp elasticity theory. Warp elasticises were measured by image processing methods in MATLAB using a high-speed camera. It was aimed to improve upon the new method of warp extension measurement of fabric when the loom is in operation. It was observed that the warp tension in plain fabric was higher than for twill and satin under the same conditions.

Investigation of the mechanical and forming behaviour of 3D warp interlock carbon woven fabrics for complex shape of composite material

Composite materials which are reinforced with 3D warp interlock fabrics have outstanding mechanical properties such as higher delamination resistance, ballistic damage resistance and impact damage tolerance by means of their improved structural properties. Textile reinforcements are exposed to large deformations in the production stage of composite materials which have complex shape. Although good formability properties of 3D warp interlock fabrics in forming process were already proven by recent studies, further information is needed to elucidate forming behaviours of multi-layer fabrics which is produced with high stiffness yarns like carbon. In this study, 3D warp interlock carbon fabrics were produced on a prototype weaving loom and the same carbon yarn was used in two fabric directions with equal number of yarn densities. Fabrics were differentiated with regard to the presence of stuffer warp yarn, weave pattern and parameters of binding warp yarn which are angle and depth. Therefore, the effect of fabric architecture on the mechanical and formability properties of 3D warp interlock carbon fabrics could be clarified. Three different breaking behaviours of fabrics were detected and they were correlated with crimp percentages of yarn groups. In addition, the bending and shear deformations were analysed in view of parameters of fabric architectures. Two distinct forming behaviours of fabrics were determined according to the distribution of deformation areas on fabrics. Moreover, the optimal structure was identified for forming process considering the fabric architecture.

Preliminary study of weave pattern influence on microplastics from fabric laundering

Nature, including the oceans, is polluted by the presence of plastics. Different products can be found, including plastic bottles, toys, toothpaste tubes, fruit meshes, etc. Small pieces of plastics, known as microplastics, have been found in the oceans and there is concern that their impact is increasing. Some of those microplastics are considered to come from the textile products. In this article, the authors will suggest how to minimize the environmental problem of the presence of microplastics in wastewaters from textile laundries. The aim of this study is to determine the influence of some parameters related to the design of fabrics. A relationship between microplastic release and parameters from weave design will be established. A fiberglass filter was used to analyze water from the laundry. Results demonstrated that the interlacing coefficient influences the number of particles in the wastewaters. Moreover, it was demonstrated that the higher the density of yarns/cm in the fabric, the lower the quantity of fibers could be found in the wastewater. Results demonstrated that the presence of weave patterns, such as a plain pattern, work better from the sustainability point of view than a twill. The interlacing coefficient and the weft density are important to prevent the microplastic release.

Influence of Pin Offset and Weave Pattern on the Performance of Al-Cu Joints Reinforced with Graphene Particles

The friction stir welding technique is currently employed for different alloy combinations like magnesium, titanium, copper, nickel-based composites. Dissimilar materials may be efficiently joined by the process and this has been used in several structural applications such as automotive components, aircraft structure, shipbuilding, and space shuttle external tank and train bodies. Even though improvement in the joint strength was achieved, problems such as voids, tunnels, cracks persisted in the weldments. The objective is to achieve defect-free joints. Tool pin offset for joining of aluminum alloy enables better stirring action and an easy flow of plasticised material in the weld nugget. Hence, to overcome the above-mentioned problem, a technique of combined effect of weaving, tool pin offset, and reinforcement of self-lubricated graphene nanoplatelets was attempted. The tensile strength obtained with the effect of tool rotational speed under weave weld with pin offset condition with reinforcement was 13.82 % higher than the weld obtained under the same condition without reinforcement. IMCs such as AlCu, Al2Cu, and Al4Cu9 large size band layers were observed with the linear welding condition, whereas weave welding conditions resulted in the formation of the thin uniform layer.

Experimental Heat Transfer and Boundary Layer Measurements on a Ceramic Matrix Composite Surface

Ceramic matrix composites (CMCs) are quickly becoming more prevalent in the design of gas turbines due to their advantageous weight and thermal properties. While there are many advantages, the CMC surface morphology differs from that of conventional cast airfoil components. Despite a great deal of research focused on the material properties of CMCs, little public work has been done to investigate the impact that the CMC surface morphology has on the boundary layer development and resulting heat transfer. In this study, a scaled-up CMC weave pattern was developed and tested in a low speed wind tunnel to evaluate both heat transfer and boundary layer characteristics. Results from these experiments indicate that the CMC weave pattern results in augmented heat transfer and flow field properties that significantly vary locally when compared to a smooth surface.

Microbial Barrier Properties of Cotton Fabric—Influence of Weave Architecture

The subject of the paper focuses on the effect of weave architecture on microbial barrier properties of woven fabrics or more precisely on identifying crucial elements of weave architecture that dominantly influence bacteria penetration in dry condition. For that purpose, 12 samples of cotton fabrics were woven and examined. In their structure, all samples had the same yarns (36 tex) in warp and weft, same densities of warp (24 yarns/cm), two weft densities (24 and 20 yarns/cm) and six different basic weave structures. Microbial barrier permeability was determined according to a previously developed test method in cooperation with University Hospital Center Zagreb. Bacterial endospores of apathogenic species of the genus Bacillus: Geobacillus stearothermophilus and Bacillus atrophaeus were used. The effect of weave pattern on microbial barrier properties was significant. Weave patterns, decisively determined the number of influencing pores and its sizes in woven fabrics, as well as the yarn floating which jointly almost perfectly correlated with bacteria penetration through the woven fabric. Multiple linear regression of pore numbers and floating threads produced equations which correspond in 99% to the measuring results for densities 24/24 and 24/20, and more than 98% considering both densities of the set. Among compared weave patterns, satin weave had significantly lower permeability of microorganisms (six–seven times) than basket weave (the highest), for both densities.