Search

Ulster University, on behalf of the Irish Manufacturing Council, hosted the 39th International Manufacturing Conference (IMC39) at the Derry~Londonderry campus on the 24th and 25th of August 2023 [...]

This paper proposes a gesture-based control system for industrial robots. To achieve that goal, the performance of an image classifier trained on three different American Sign Language (ASL) fingerspelling image datasets is considered. Then, the three are combined into a single larger dataset, and the classifier is trained on that. The result of this process is then compared with the original three.

This paper introduces a methodology for precise object orientation determination using principal component analysis, with robust performance under significant noise conditions. It validates the potential to mitigate the challenges associated with axis-aligned bounding boxes in smart manufacturing environments. The proposed approach paves the way for improved alignment in robotic grasping tasks, positioning it as a computationally efficient alternative to ML methods employing oriented bounding boxes. the methodology demonstrated a maximum angle deviation of 3.5 degrees under severe noise conditions through testing with bolts in orientations of 0 to 180 degrees.

Additive manufacturing powders require a well-defined particle size distribution (PSD) and spherical morphology to ensure good flowability. To simplify characterisation, powders can be prepared using standard metallurgical techniques followed by optical imaging of the cross-sectioned particles. Measured PSDs of particle sections are typically underestimates of the true PSD; hence, stereological corrections are required. Variations arise in the histogram binning methods (central binning versus upper limit binning) of commonly used stereological corrections. Although the results show some sensitivity to the binning method used, the GCO method seemed reasonably robust to changes in the binning method. However, authors are encouraged to follow the method as it is intended within the literature, which was found to be especially true when using Saltykov’s method.

Defects formed during the spreading of powder, known as powder bed defects, are a major issue in additive manufacturing processes. Deep learning (DL)-based image classification models can be utilised to detect defects caused by the powder spreading process. The aim of this research was to review and compare the performance of the EfficientNet_v2 deep learning image classification model against the commonly used VGG-16 model on a selective laser sintering powder bed defects (SLS PBDs) dataset. It was observed that the EfficientNet_v2 model achieved higher performance than the commonly used VGG-16 model, with a defect prediction accuracy of 97.54% and model sensitivity of 96.3%.

Traditional manufacturing units face a lot of challenges in coping with growing demand. With the more demand, maintaining the quality of the product is another challenge. Industry 4.0 can be utilized to meet the growing demand without sacrificing the quality of the product. The aim of this research is to analyse the improvement in production capabilities when implementing smart manufacturing. Woodford Timber Products has streamlined its production processes and increased its operational efficiency by embracing innovative technology and techniques. The digitisation of manufacturing processes through sophisticated software is capable of designing, rendering, analysing, and converting to machine language for multi axis CNC machines. The organisation derives significant insights from enormous volumes of data using product data management optimising manufacturing schedules, minimising waste, and improving product quality.

Open-topped woven polypropylene cellular containers filled with dense granular ballasts are often used as emergency flood defence structures. The effectiveness of these systems is highly dependent on the interaction with their bedding surface. The characteristics of the foundation will often govern the system’s overall resistance to applied loading imposed by retained floodwater. However, the frictional relationship between polypropylene bulk bag flood defences and common bedding surfaces has not been extensively investigated. This study aims to reduce the reliance on arbitrary static friction coefficients by measuring and presenting actual data obtained through quantitative testing. This study presents the results of full-scale field testing to quantify the frictional resistance generated between filled polypropylene bulk bags and seven common bedding surfaces. Findings resulting from testing each interface scenario are expressed as coefficients of static friction. Test interfaces affording high frictional resistance comprised an unmade gravel road (µ = 0.74) and grass (µ = 0.64). Contrastingly, interfaces generating significantly lower frictional resistance were steel floated concrete (µ = 0.40) and polypropylene plastic (µ = 0.40). Test interfaces involving asphalt (µ = 0.54) and tamped concrete (µ = 0.56–0.58) were also investigated. This study recommends new friction coefficients necessary to characterise the structural stability analysis of bulk bag flood defences with greater accuracy. Practical advice based on experimental observation and field design experience is also given.

Ver resumen pdf.

For this experiment it was hypothesized that endothelin 1 (END1) acting through its receptors, endothelin receptor type A (ENDRA) and/or endothelin receptor type B (ENDRB), on granulosal cells (GC) of ovine preovulatory follicles would inhibit steroid production, and therefore, prevent the premature luteinization of granulosal cells of those follicles. The ovaries of 17 Suffolk ewes were removed by mid-ventral laparotomy after two doses of PGF2a (5 mg; three h apart). Granulosal cell were harvested under sterile conditions from the two or three largest follicles. Aliquots of 15 µL of GC suspension from each presumed preovulatory follicle were cultured in 96-well plates containing a solution of 185 µL of Ham´s F12 supplemented medium and treatments (single and combined effects of LH, END1, ENDRA and ENDRB). According to estradiol (E2) concentration in follicular fluid (FF), presumed preovulatory follicles were categorized in estrogen-active and estrogen-inactive follicles. Data from 20 estrogen-active follicles (collected from 13 ewes) challenged to 10 different treatments were examined by analysis of variance. Accumulation of E2 and progesterone (P4) by cultured GC was not affected by either LH or END1 or the END receptor blockers. Granulosal cells from follicles with the greatest concentration of E2 in FF produced on the average significantly more E2 and P4 than follicles with lower concentrations of E2 in FF (P < 0,0001). In conclusion, under the experimental conditions of this study, END1 did not affect steroid production by GC from preovulatory follicles in sheep, and therefore the working hypothesis was not supported.