Advances in Industrial and Manufacturing Engineering最新文献

Continuous production processes are often highly complex and involve machine failures as well as unscheduled process downtimes. Failures result in the production of waste and in high opportunity costs, but their causes are not always apparent to machine operators. As a result, identifying failure root causes and avoiding risky process states is of high interest for producers. This work presents an approach for a data-driven failure risk assessment that is validated on real-world process data of a nonwovens production line. In this approach, association rule mining is adapted to continuous processes for producing highly interpretable results in the form of association rules that represent the main causes for failures. The methodology includes data preparation, modelling of production states and the evaluation of root causes using an associative classification algorithm. The result of this paper is a method for an interpretable risk assessment in continuous production processes. By using the method in live production, causes of failures can be detected and interpreted. The universal structure of the developed method supports applications in many other continuous production processes.

For manually operated machine tools, the accuracy of the machine tool structure limits the accuracy of the parts produced. Such is not necessarily the case with computer numerically controlled (CNC) machine tools. This concept may not be immediately obvious to the engineering undergraduate. The method of error compensation is presented here in a manner that is accessible to the undergraduate engineering student. A homogeneous transformation matrix (HTM) model quantifies the geometric errors of a machine tool, which can be compensated for in software. The mathematical treatment is reduced to only essential elements to emphasize physical understanding. A key feature of this presentation is the application of the model to a three-axis milling machine. This illustration enables the undergraduate student to grasp the concept with ease. Another feature is that the entire model is developed from first principles, which does not require the student to invoke any empirical relationships. Three solved numerical problems illustrate the application of the model to practical situations. Information provided here may be used by the teacher as a template to introduce this subject at the undergraduate level.

Machine learning (ML) has become a key technology to leverage the potential of large data amounts that are generated in the context of digitized and connected production processes. In projects for developing ML solutions for production applications, the selection of hyperparameter optimization (HPO) techniques is a key task that significantly impacts the performance of the resulting ML solution. However, selecting the best suitable HPO technique for an ML use case is challenging, since HPO techniques have individual strengths and weaknesses and ML use cases in production are highly individual in terms of their application areas, objectives, and resources. This makes the selection of HPO techniques in production a very complex task that requires decision support. Thus, we present a structured approach for benchmarking HPO techniques and for integrating the empirical data generated within benchmarking experiments into decision support systems. Based on the data generated within a large-scale benchmarking study, the validation results prove that the usage of benchmarking data improves decision-making in HPO technique selection and thus helps to exploit the full potential of ML solutions in production applications.

The Wire and Arc Additive Manufacturing has called attention due to its potential in allowing the buildup of high integrity metallic parts using the commonly available welding robots in the industry. However, such a technology still presents some challenges, mainly related to obtaining optimal deposition parameters, which result in consistent layer geometry which leads to the robot and the welding power source programming. In this sense, the objective of this work is to analyze the influence of the parameters in bead and multi-layer wall geometries fabricated by the Cold Metal Transfer process to select the configurations that result in the best deposition control. The study was carried out in four steps: (i) deposition of single beads on plate, varying wire feed speed and travel speed that would result in stable and sound beads; (ii) deposition of five layer walls, considering both unidirectional and bidirectional path strategies, with and without stops between layers; (iii) deposition of ten and twenty layer walls, refining deposition parameters; and (iv) deposition of a one hundred layer wall, with the best parameter configuration among the previously tested, with bidirectional continuous strategy. The results showed that the geometry produced with a mean current of 62 A and torch travel speed of 8 mm/s along each layer and 24 mm/s on the transition between layers generated the best results, considering the natural cooling conditions. Also, the bidirectional path deposition presented the most regular geometries, when compared to the unidirectional strategy.

In the material extrusion (MEX) Additive Manufacturing (AM) technology, the layer-by-layer nature of the fabricated parts, induces specific features which affect their quality and may restrict their operating performance. Critical quality indicators with distinct technological and industrial impact are surface roughness, dimensional accuracy, and porosity, among others. Their achieving scores can be optimized by adjusting the 3D printing process parameters. The effect of six (6) 3D printing control parameters, i.e., raster deposition angle, infill density, nozzle temperature, bed temperature, printing speed, and layer thickness, on the aforementioned quality indicators is investigated herein. Optical Microscopy, Optical Profilometry, and Micro Χ-Ray Computed Tomography were employed to investigate and document these quality characteristics. Experimental data were processed with Robust Design Theory. An L25 Taguchi orthogonal array (twenty-five runs) was compiled, for the six control parameters with five levels for each one of them. The predictive quadratic regression models were then validated with two additional confirmation runs, with five replicas each. For the first time, the surface quality features, as well as the geometrical and structural characteristics were investigated in such depth (>500 GB of raw experimental data were produced and processed). A deep insight into the quality of the MEX 3D printed parts is provided allowing the control parameters’ ranking and optimization. Prediction equations for the quality features as functions of the control parameters are introduced herein, with merit in the market-driven practice.

Flux assisted tungsten inert gas welding (FATIG) welding is a modified version of tungsten inert gas (TIG) welding to achieve a higher depth of penetration. In the present work, nanoparticles SiO₂, Al₂O_3, Fe₂O₃, and CuO mix with acetone and coated on the joint before welding. Bead on plate welding using different variants of FATIG welding performed on Hastelloy C-22. A comparative study of these variants called Activated tungsten inert gas (ATIG) and Flux bound tungsten inert gas (FBTIG) welding was conducted to find out their effects on depth of penetration, depth to width (D//W) ratio, surface appearance, and slag detachability. In addition, the influence of acidic and basic nature of flux on weld bead geometry and surface appearance are analyzed. Acidic fluxes produce a smoother weld surface than basic oxide fluxes; additionally, acidic flux slag is less sticky than basic flux slag. activated TIG (ATIG) welding with SiO₂ flux increases penetration and D/W ratio by 125% and 190%, respectively compared to normal TIG welding.

Additive Manufacturing (AM) offers new design and manufacturing opportunities of thin-wall microchannel heat exchangers for aerospace and industrial applications. Laser Powder Directed Energy Deposition (LP-DED) is an AM process providing large scale manufacturing of thin-wall microchannel heat exchangers. Successful industrialization of the LP-DED process requires critical quantification and understanding of the metallurgical, geometric, and process limitations. Specifically, understanding the as-built surface texture, inclusive of roughness and waviness, is significant due to its effects on the friction factor and pressure drop within a heat exchanger. This experimental study completed a design of experiments (DOE) to determine the critical build parameters that impact surface texture for enclosed thin-wall samples. This study summarizes the characterization work of the LP-DED process for 1 mm enclosed walls with an Fe–Ni–Cr (NASA HR-1) alloy. The LP-DED parameters including laser power, powder feedrate, travel speed, layer height, and rotary atomized powder feedstock were modified in the experiment. An evaluation of the DOE samples and resulting surface texture is provided along with conclusions from these experiments. Results indicate that 3D areal and 2D profile (directional) surface texture is estimated by 2x the powder diameter that becomes captured or partially melted on the trailing edge of the melt pool. The fine powder showed a higher sensitivity to parameter changes but resulted in a higher density material and 23% reduction in roughness. Surface texture was also shown to vary between closed channel shapes (internal) due to ricochets, recirculation, and higher volume of powder available to bond compared to external (outer) surfaces. The understanding of the LP-DED process as-built surface texture is essential to fluid flow applications such as heat exchanges and can modify performance for enhanced heat transfer or can be a detriment to pressure drop.

Polymer blending is one of the popular methods for producing tailor-made materials by combining the properties of individual polymers. Binary polymer blends have quite commonly been used over the past few decades. Recently, researchers have shifted their focus towards ternary polymer blends and this study aims to investigate a ternary polymer blend system consisting of low-density polyethylene (LDPE), polystyrene (PS) and polymethyl methacrylate (PMMA). The LDPE/PS/PMMA blend was processed by melt blending using a twin-screw extruder. The effects of the extrusion process parameters (i.e., screw speed and barrel set temperatures) and the blend composition on the mechanical, rheological and thermal properties of the polymer blend and the degree of crystallinity of the LDPE matrix were studied. Three different screw speeds (i.e., 50 rpm, 100 rpm and 150 rpm), two different barrel set temperatures (i.e., 200 °C and 220 °C), and two different component mass ratios (i.e., 70/10/20 and 70/20/10) were studied. The results showed that the tensile properties of the LDPE/PS/PMMA blend were significantly influenced by its microstructure. Yield strength and Young's modulus decreased at first and then increased with increasing screw speed. The blend processed at a barrel set temperature of 220 °C was found to have better tensile properties than the blend processed at 200 °C. Furthermore, the blend with a PS content of 10 wt% possessed better tensile properties than the blend with a PS content of 20 wt%. Regardless of the blend compositions and the process settings, the LDPE/PS/PMMA blends reported better mechanical properties than those of pure LDPE with a Young's Modulus of 240 MPa and a yield stress of 10.47 MPa. The rheology of the blend was also significantly affected by the process parameters and the blend composition. However, different process parameters and mass ratios did not indicate a significant influence on the melting temperature (around 109.5 °C) and the degradation initiation temperature (around 252.3 °C) of the LDPE/PS/PMMA blend, but both the melting temperature and the degradation initiation temperature of the ternary blend were found to be slightly lower than those of pure LDPE. The degree of crystallinity of the LDPE matrix was also affected by both the screw speed and the barrel set temperature. The results revealed that, better mechanical properties can be achieved by blending PS and PMMA with LDPE without significantly affecting the thermal properties compared to those of pure LDPE.

Additive manufacturing parameters of high-performance polymers greatly affect the thermal history and consequently quality of the end-part. For fused deposition modeling (FDM), this may include printing speed, filament size, nozzle, and chamber temperatures, as well as build plate temperature. In this study, the effect of thermal convection inside a commercial 3D printer on thermal history and crystalline morphology of polyetheretherketone (PEEK) was investigated using a combined experimental and numerical approach. Using digital scanning calorimetry (DSC) and polarized optical microscopy (POM), crystallinity of PEEK samples was studied as a function of thermal history. In addition, using finite element (FE) simulations of heat transfer, which were calibrated using thermocouple measurements, thermal history of parts during virtual 3D printing was evaluated. By correlating the experimental and numerical results, the effect of printing parameters and convection on thermal history and PEEK crystalline morphology was established. It was found that the high melting temperature of PEEK, results in fast melt cooling rates followed by short annealing times during printing, leading to relatively low degree of crystallinity (DOC) and small crystalline morphology.