Post Processing Research Articles

Prediction Enhancement of Machine Learning Using Time Series Modeling in Gas Turbines

Abstract This study is based on time-series data taken from the combined cycle heavy-duty utility gas turbines. For analysis, first a multistage vector autoregressive model is constructed for the nominal operation of the powerplant assuming sparsity in the association among variables and this is used as a basis for anomaly detection and prediction. This prediction is compared with the time-series data of the plant-operation containing anomalies. The comparative advantage based on prediction accuracy and applicability of the algorithms is discussed for the postprocessing. Next, the long-memory behavior of residuals is modeled, and heterogeneous variances are observed from the residuals of the generalized additive model. Autoregressive fractionally integrated moving average (ARFIMA) and generalized autoregressive conditional heteroskedasticity (GARCH) models are employed to fit the residual process, which significantly improve the prediction. Rolling one-step-ahead forecast is studied. Numerical experiments of abrupt changes and trend in the blade-path temperature are performed to evaluate the specificity and sensitivity of the prediction. The prediction is sensitive given reasonable signal-to-noise ratio and has lower false positive rate. The control chart is able to detect the simulated abrupt jump quickly.

100 Gbps PAM4 ultra-thin photodetectors integrated on SOI platform by micro transfer printing.

The integration of compact high-bandwidth III-V active devices in a scalable manner is highly significant for Silicon-on-insulator (SOI) photonic integrated circuits. To address this, we demonstrate the integration of pre-fabricated 21 × 57 µm2 InGaAs photodetector (PD) coupons with a thickness of 675 nm to a 500 nm SOI platform using a direct bonding micro-transfer printing process. The common devices are coupled to the Si waveguides via butt, grating and evanescent coupling schemes with responsivities of 0.13, 0.3 and 0.6 A/W respectively, in line with simulations. The thin device facilitates simplified high-speed connections without the need for an interlayer dielectric. A back-to-back data communication rate of 50 Gb/s is achieved with on-off keying and with post processing of four-level pulse-amplitude modulation (PAM4) 100 Gb/s is realized. Potentially, around 1 million devices per 75 mm InP wafer can be attained. The integration of compact PDs exhibited in this work can be extended to modulators and lasers in the future.

Analysis of singularity in advection-diffusion-reaction equation with semi-analytical boundary elements

In order to solve the advection-diffusion-reaction equation with singularity by the boundary element method, the domain integral is handled by dual reciprocity method and the semi-analytical elements are implemented to approximate the unknown physical fields. By collocating internal points in the domain and approximating the primer field with radial basis function, the integral of domain can be settled and boundary-only integral equation can be derived. In this paper, we propose two new elements containing asymptotic expressions to model the unknown temperature and singular flux fields, while the known fields are simulated by linear interpolation. The unknowns in the new elements are amplitude coefficients which can be used to evaluate the intensity of flux. Compared with conventional boundary element, the proposed method allows to avoid direct evaluation of singular values. Thus, the accuracy can be guaranteed since all the unknowns have finite values. Benefiting from the semi-analytical elements, the mesh refinement for the region with singular physical field can be avoided. Moreover, the first three amplitude coefficients as well as the whole physical fields are obtained at the same time without any post process. We demonstrate the excellent numerical results through several examples in plates with singularities.

Pile Capacity Checking Tool based on Distributed Fibre Optic Sensing for Instrumented Pile Load Test

Pile load test instrumented using Distributed Optical Fibre Sensing (DOFS) offers a superior ability to assess pile capacity through continuous strain profiles as opposed to the discrete measurement method such as Vibrating Wire Strain gauges. Due to the vast amount of spatial data generated by the sensor, processing the raw data from DOFS is not always straightforward, often done manually with careful interpretation of various factors affecting the measurements. Experts are reconciling with different data processing algorithms to assess pile capacity (t-z curve) from the DOFS data and verify it, leading to a demand for standard practice. Currently, the verification is done by comparing the pile capacity assessed from DOFS data to the analysis from the commercially available numerical software. The software typically requires input parameters such as extensive soil data, structure geometry, and loading intensity. Some of these parameters are limited (due to bureaucracy or trade secrecy) for the pile load test contractor to verify the pile capacity computation. This paper presents a checking tool that has been developed to automate the processes including raw file parsing, initial processes (trimming, positioning, and averaging), and post processes (pile capacity computation and checking). The post processes utilised the simplified Finite Element Method (FEM) to perform the checking which only utilised the DOFS data. In this paper, an application of the developed tool to a DOFS instrumented pile load test in Kuala Lumpur is reported. This exercise provides the information that the tool can be helpful to verify the pile capacity computation from DOFS data with promising performance reliability at 95%.

Head movement during natural group conversation and inter-annotator agreement on manual annotation

During speech communication and conversational turn-taking, listeners direct their head and eyes to receive meaningful auditory and visual cues. Features of these behaviors may convey listener intent. This study designed a test environment, data collection protocol and procedures, and investigated head movement behaviors during self-driven conversations among multiple partners. Nine participants were tested in cohorts of three. Participants wore a headset with sensors tracked by an infrared camera system. Participants watched an audio-video clip, followed by a 5-min undirected discussion. The entire session was video recorded for annotation purposes. Two annotators independently coded the video files using the EUDICO Linguistic Annotator software application. Annotations were then co-registered with the head tracking data in post processing. Inter-annotator agreement demonstrated the desired reliability, thereby validating the procedures designed. Movement trajectories showed that there were individual differences on the head yaw distribution. The combination of objective measures of head movement and manual annotation of conversation behaviors provides a rich data set for characterizing natural conversations in ecologically valid settings. The measurement procedures and coding system developed here is a first step towards characterizing head movements during conversations needed to predict listening intent and to create actions based on those predictions.

Improving the reliability of power modules through layered diffusion solder interconnects – Comparative study based on experiments and FE-simulation

The demand for reliable, lead-free and high temperature stable solder interconnects for power electronic application has increased in recent years. The die attach solder materials are exposed to higher electrical currents and temperatures due to advancing designs and wide band-gap semiconductors. The diffusion soldering material Sn-Cu-Sn with its high post processing melting temperature is a promising option for high temperature die attach interconnects. The aim of this work is to predict potential failure mechanisms in the die attach of automotive power modules due to simulation based thermo-mechanical analysis of the diffusion solder Sn-Cu-Sn layer and SAC. The FE-model is validated with experimental results from active power cycling tests. The simulation indicates a potential for lifetime increase by usage of layered diffusion solder. The created FE-simulation model could be used for simulation-based design for reliability to reduce material costs and development time.

A post processing technique to achieve nanofinishing for functionality enhancement of Ti-6Al-4V femoral head fabricated by Laser Powder Bed Fusion

A post processing technique to achieve nanofinishing for functionality enhancement of Ti-6Al-4V femoral head fabricated by Laser Powder Bed Fusion

Reduction of ASZ size and enhancement of mechanical properties of TLP bonded IN718 joint using post bond heat treatment process

Reduction of ASZ size and enhancement of mechanical properties of TLP bonded IN718 joint using post bond heat treatment process

Machinability Characterization of 3D Printed PEEK

The 3D printing technology is being applied more and more every day, this is a consequence of its applicability and low waste generation, becoming one of the best options to obtain good quality pieces. Sometimes, post machining processes are necessary to fulfil tight tolerances or achieve complex geometries by means of the connection between different pieces printed using this technology. The field of knowledge and studies focused on 3D printing is in constant evolution. There are plenty of materials that can be used to apply 3D printing technology. Among them, PEEK is one of the best options when good mechanical properties are required. Being applied in aeronautic or automobile industry, is also used in biomedical applications, such as prosthesis or mechanical components among others. Within the machining processes, milling, turning, and drilling are the most widespread. Orthogonal cutting is a machining process in which the cutting edge of the tool is perpendicular to the cutting speed, and it is commonly used when a simple and pure study of the mechanism behind a material removal process is required. In this study, tests that analyze the orthogonal cutting on 3D printed PEEK samples using different orientations (0o and ±45o) have been conducted. The influence of cutting speed (30, 60 and 90 m/min) and depth of cut (50, 100 and 150 μm) is studied through the analysis of cutting forces and surface finish quality. As a general approximation, it can be seen that the fiber orientation affects significantly to the forces monitored but unexpectedly, lightly to the surface finish.

Nitrogen-Doped Single-Walled Carbon Nanotubes by Floating-Catalyst CVD Process

Due to the properties of single-walled carbon nanotube (SWCNT), thin film SWCNTs have been used for many versatile applications. Direct synthesis of the nanotube films with desired electronic structures without prerequisite of post treatment process then becomes one important step. Here, we present the synthesis of nitrogen-doped SWCNTs by floating-catalyst chemical vapor deposition process. With low nitrogen level incorporated into the sp2 carbon network, substitutional and pyridinic nitrogens become predominance. The electronic structure of N-doped SWCNT film could be changed into n-type doping. The localized structures of carbon and nitrogen bonding environments were also revealed by x-ray absorption spectroscopy. The formation of p-n junction was also obtained from the I-V characteristic of N-doped SWCNT heterojunction diode revealing n-type behavior of the sample.

A-182 The journey From Conceptualizing, Designing, and Implementing Our Integrated Smart Core Lab at Kokilaben Dhirubhai Ambani Hospital (KDAH) & Medical Research Institute, Mumbai, India

Abstract Background KDAH established in 2009, is a 550 bedded tertiary care, multispecialty hospital in Mumbai, India. Annually the hospital manages about 300 000 patients, the clinical lab delivers >1 750 000 results with Compound Annual Growth Rate of 18 %. In the last 14 years, the clinical lab has expanded; increasing the test services, adding equipment, recruiting staff, and requesting for more space. There is increasing pressure to improve speed of service to impact early patient outcomes keeping an awareness on cost, non-availability of space for expansion. There is also demand to bring in new clinical tests and increase outreach beyond the hospital. In October 2018, the laboratory and hospital management decided to apply principles of Lean and design thinking to laboratory processes. The Smart Core Laboratory conceptualized was an outcome of this work and is focused on delivering results in the most efficient way, in terms of quality, cost, and speed. Methods The team of experts from various lab disciplines and lean consultants, started evaluating the end-to-end process from sample received at the central station to results review, reporting and sample disposal 1. Process Map: The process steps are classified and evaluated by the core team as value added (VA) and non-value added (NVA) & business-value added or necessary waste. Process Maps charts were generated for the following sample flow. a. Central Sample Receipt to Section. Verification, Registration & Distribution Process.b.In Lab preanalytical, analytical, post analytical processes: Clinical Chemistry, Immunoassay, Infectious Serology. (Serum, Plasma Fluoride / Heparin)Hematology (EDTA Whole Blood)Hemostasis (Citrate Plasma) 1. Value Stream Map A detailed Value Stream Map of all the actions for the above processes was generated to determine VA and NVA process steps 3. Spaghetti Mapping: Motion Study Observation and analysis of the Spaghetti mapping diagrams exposed inefficient layouts and identified areas with a lot of unnecessary movement. This helped to decide optimization of the flow and re-organization of the entire lab section Results & Discussion Based on the analysis of the data generated from the lean tools and identifying process waste and non-value added steps the following immediate improvements were implemented a. Combining steps extra-processing at centrifugation at each lab sectionb.Decreasing sample vacutainers by reducing number of sample drawsc.Eliminating Overproduction by removing extra barcode printing and registersd.Using Staff Creatively: Reorganize IQC, Calibration and maintenance to off peak hourse.Space Optimization. Conclusion This consolidation approach has enabled us to free up at least 341 sq ft prime lab space (≈14 % reduction), decrease staff motion & sample transportation. Approximately 50% simplification in overall processes is achieved. It also allowed reduction in additional patient sample draws by at least 40 %. The team could use this lean analysis to conceptualize and design a multidisciplinary Automation solution that would consolidate 5 sections namely Clinical Chemistry, Immunoassay, Infectious Serology, Hematology and Hemostasis in the Smart Core Laboratory.

Parametric Study on the Evaporation Characteristics of Biodiesel-Diesel-Methanol Microemulsion Droplets

ABSTRACT In order to investigate the evaporation characteristics of biodiesel-diesel-methanol microemulsions, an experimental method of droplet evaporation with post processing techniques is established. The droplet normalized square diameter (NSD), droplet temperature and images during evaporation process are used to analyze the evaporation characteristics at different ambient temperatures. The results indicate that the methanol volume percentage, initial volume of droplet and ambient temperature have evident influence on the evaporation characteristics of droplets. When the initial volume of droplet is 0.5 μL, BD95M5 (biodiesel-diesel blend (95%)+methanol (5%)) and BD85M15 (biodiesel-diesel blend (85%)+methanol (15%)) droplets do not show the puffing and micro-explosion phenomena during evaporation process, but BD75M25 (biodiesel-diesel blend (75%)+methanol (25%)) droplet exhibits micro-explosion evidently at ambient temperature of 573K. Puffing is observed for BD85M15 droplet, and micro-explosion is for BD75M25 droplet at ambient temperature of 773K. The droplet lifetime is decreased and the mean evaporation rate (MER) is increased with the increase of ambient temperature and methanol volume percentage in microemulsion. The evaporation process ends at 5.938 s/mm2, 5.417 s/mm2 and 4.792 s/mm2 for BD95M5, BD85M15 and BD75M25, respectively, and MERs of them are 0.087 mm3/s, 0.095 mm3/s and 0.108 mm3/s at ambient temperature of 573K. The MERs of BD95M5, BD85M15 and BD75M25 are 0.225 mm3/s, 0.244 mm3/s and 0.309 mm3/s, respectively, at ambient temperature of 773K. With the increase of initial volume of droplet, the occurrence of puffing and micro-explosion is strengthened. The MER of droplets with initial volume of 0.6 μL is the largest, which increases by 13.1% and 12.8% compared with 0.4 μL droplets and 0.5 μL droplets, respectively.

Engineering of heat sources in laser precision machining

Requirement towards the enhancement of the results of technological operations of laser processing, their transformation into precise technological operations is solved, in general, by the choice of optimal technological schemes of laser processing and characteristics of a tool that is formed from laser beam characteristics. The main goal of these attempts is to minimize portions of energy that destroys and removes the material from the cavity of element being formed utilizing ultra-short pulses of electro-magnetic energy with high intensity that can bring balance to the instability of technological process, manage it outcome and maintain the quality of processing. On the other hand, it leads to a decrease in the efficiency of technological operations of laser processing. For example, linear speed of laser cutting of stent with nanosecond pulses are used is within the range of 25 to 30 millimeters per minute, whereas in the case of microsecond pulses - the processing speed can reach up to 250 mm/min and slight striation of side surface of the cutting kerf could be eliminated by means of post processing. Since the process of material destruction at laser beam processing is of a thermal nature in terms of heat dissipation and its propagation into the body of the work piece (especially the metal one), it is possible to control and modify the absorption of radiation by the surface of the work piece, the spatial shape of the heat source and thermal an physical properties of the work piece. This paper discusses perspective solutions for the successful engineering of laser beam irradiation process.

Confined self‐assembly of block copolymers in emulsion droplets: Preparation, shape and internal structure, and applications

AbstractConfined assembly of block copolymers (BCPs) in emulsion droplets receives considerable attention since self‐assembly in flexible geometry pioneers an efficient yet promising route to prepare unconventional polymer materials that cannot be acquired in the bulk state or solution. The deformability and interfacial property regulation of the soft emulsion droplets offer tremendous potential for the design and construction of BCP nanoparticles with desired shape and internal nanostructures. This review summarizes the recent developments of BCP assembly behavior in three‐dimensional (3D) soft oil emulsion droplets, including preparation methods, shape and internal structure control, as well as advanced applications of the generated BCP materials. Typical experimental parameters and post processes, including intrinsic property of BCPs, surfactant property, confinement degree, interfacial instability, additives, evaporation rate, post‐annealing, and disassembly, that have significant influence on the precise control and regulation of emulsion interface properties and the prepared assembly structures are well systematically summarized. Finally, the challenges and further perspective in this area are presented.

Knitting from Nature: Self-Sensing Soft Robotics Enabled by All-in-One Knit Architectures.

Self-sensing soft robotics that mimic the proprioception and exteroception abilities of natural biological systems have shown great potential in challenging applications. However, current add-on strategies that simply combine sensors with actuators by post processing generally suffer from poor compatibility in mechanical properties, interfacing problems, complex manufacturing, and high cost. Herein, we present knitted soft robotics with build-in textile-integrated multimodal sensors, where the knit structure is used not only as a physical actuating layer but also as a sensing functional component. Based on different knit-stitch arrangements, an all-in-one knitted electronic skin with functions of neurons, sensing, and actuation in a single knit-structured fabric layer is constructed. The knitted electronic skin is then integrated into knitted soft robotics, enabling a proprioceptive sense of actuation deformation and an exteroceptive perception of ambient stimuli with minimized interferences for actuation. In addition, the tuck stitches serve as an anisotropic strain-limiting layer to increase the actuating energy efficiency, which resolves the key conflict of softness and volumetric power density in soft actuators. This design strategy provides a convenient, low-cost, and customized method to bring about structural and functional integrability into soft actuators, greatly extending the adaptability of current soft robotics for real-world applications.

Post processing techniques used to improve the quality of 3D printed parts using FDM: State of art review and experimental work

Fused Deposition Modeling is very effective and important method of the additive manufacturing which is widely used in different sectors. Poor surface smoothness, weak mechanical characteristics, and insufficient dimensional precision frequently limit the ability of additive manufacturing to generate optimized and tailored polymeric parts. Post-processing techniques are typically used to remedy these problems. The goal of the current work is to give a thorough classification and explanation of post processing methods. A comprehensive analysis of post-processing is given, along with a discussion of the advantages and disadvantages of each technique. By taking into account a variety of factors, including the material, part geometry, processing time and treatment specificity, this work aims to enable the selection of the most appropriate post-processing. After a state of art review on post processing methods for 3D printing an experimental work carried out on one the post processing method i.e sanding is also presented here. The % improvement in surface roughness due to sanding ranges between 3 and 25%. The maximum.

In-situ Selective Laser Heat Treatment for microstructural control of additively manufactured Ti-6Al-4V

As-built Laser Powder Bed Fusion (LPBF) Ti-6Al-4V typically exhibits a fully acicular α′-martensite microstructure, and requires post-process heat treatment in order to decompose the martensite and achieve sufficient ductility. In the present study, we demonstrate a simple concept based on in-situ Selective Laser Heat Treatment (SLHT) that can effectively alter the microstructure and activate the decomposition of the α′-martensite into a lamellar α+β microstructure within a short time scale (~30s). SLHT consists of multiple rescanning of the printed part, with low energy density, triggering solid-state phase transformations. Operando X-ray diffraction has been performed on cuboid and thin wall geometries, and was augmented by thermal finite element simulations. Upon SLHT, a gradual formation of the β phase as well as an α/α′peak narrowing trend have been evidenced through X-ray diffraction, as an indication of the diffusional nature of α′-martensite decomposition. Moreover, through fine tuning of the process parameters at the final stage of SLHT, a controlled temperature evolution during cooling was achieved, leading to preservation of the β phase, a product of the decomposition, down to room temperature. Complementary microstructural characterizations via EBSD, SEM, and TEM confirm the presence of a lamellar α+β microstructure after SLHT. Our results evidence, for the first time, the fast kinetics of α′-martensite decomposition under in-situ SLHT. The approach is meant to be implemented at selected locations during the LPBF process, avoiding time-consuming post processing steps, and leading to composite-like, architected microstructures.

Sparse Quadratic Approximation for Graph Learning.

Learning graphs represented by M-matrices via an l1-regularized Gaussian maximum-likelihood method is a popular approach, but also one that poses computational challenges for large scale datasets. Recently proposed methods cast this problem as a constrained optimization variant of precision matrix estimation. In this paper, we build on a state-of-the-art sparse precision matrix estimation method and introduce two algorithms that learn M-matrices, that can be subsequently used for the estimation of graph Laplacian matrices. In the first one, we propose an unconstrained method that follows a post processing approach in order to learn an M-matrix, and in the second one, we implement a constrained approach based on sequential quadratic programming. We also demonstrate the effectiveness, accuracy, and performance of both algorithms. Our numerical examples and comparative results with modern open-source packages reveal that the proposed methods can accelerate the learning of graphs by up to 3 orders of magnitude, while accurately retrieving the latent graphical structure of the data. Furthermore, we conduct large scale case studies for the clustering of COVID-19 daily cases and the classification of image datasets to highlight the applicability in real-world scenarios.

Strain-rate dependence of mechanical characteristics of PLLA with different MW

Abstract Dynamic mechanical properties of polymers for biomedical applications are crucial parameters for development and engineering of new medical devices. Here, the time-dependent material behavior is a key factor for durability. Varying the strain rate is a convenient implementation of time-dependency for uniaxial testing. This study investigates time-dependence of Poly(L-Lactide) (PLLA) through uniaxial testing with different strain rates and PLLA with different molecular weight. The results show strain dependence for elongation at break and yield stress, Young’s modulus however is not rate dependent. An increase in elongation at break is also seen with increasing molecular weight of PLLA. Plastic strain increases significantly only for PLLA with an intermediate inherent viscosity. Results show distinct time dependencies regarding strain rate for PLLA with slightly different inherent viscosities. For stent-related mechanical material characteristics, higher molecular weight PLLA seems to be advantageous. This study only considers base materials, although appropriate thermal, mechanical as well as chemical post processing approaches for further adjustment of different properties have already been shown. A combination of the best possible base material and a suitable post-processing should be targeted.

Comparison of clinically available dynamic susceptibility contrast post processing software to differentiate progression from pseudoprogression in post-treatment high grade glioma

IntroductionThe purpose of this retrospective study was to compare two, widely available software packages for calculation of Dynamic Susceptibility Contrast (DSC) perfusion MRI normalized relative Cerebral Blood Volume (rCBV) values to differentiate tumor progression from pseudoprogression in treated high-grade glioma patients. Material and MethodsrCBV maps processed by Siemens Syngo.via (Siemens Healthineers) and Olea Sphere (Olea Medical) software packages were co-registered to contrast-enhanced T1 (T1-CE). Regions of interest based on T1-CE were transferred to the rCBV maps. rCBV was calculated using mean values and normalized using contralateral normal- appearing white matter. The Wilcoxon test was performed to assess for significant differences, and software-specific optimal rCBV cutoff values were determined using the Youden index. Interrater reliability was evaluated for two raters using the intraclass correlation coefficient. Results41 patients (18 females; median age = 59 years; range 21–77 years) with 49 new or size-increasing post-treatment contrast-enhancing lesions were included (tumor progression = 40 lesions; pseudoprogression = 9 lesions). Optimal rCBV cutoffs of 1.31 (Syngo.via) and 2.40 (Olea) were significantly different, with an AUC of 0.74 and 0.78, respectively. Interrater reliability was 0.85. DiscussionWe demonstrate that different clinically available MRI DSC-perfusion software packages generate significantly different rCBV cutoff values for the differentiation of tumor progression from pseudoprogression in standard-of-care treated high grade gliomas. Physicians may want to determine the unique value of their perfusion software packages on an institutional level in order to maximize diagnostic accuracy when faced with this clinical challenge. Furthermore, combined with implementation of current DSC-perfusion recommendations, multi-center comparability will be improved.