Predictive Modeling and Correlated Response Optimization of Polymethylmethacrylate (PMMA)-Based Bio-Nano-Composite Material Using a Hybrid Module

Polymethylmethacrylate (PMMA) is commonly known as bone cement, having good biocompatibility, mechanical qualities. It is extensively used in the biomedical sector as a synthetic bone material, orthopedic surgery and dental applications. However, some primary machining is required to achieve the tailored shape, size and finish before application in the human body. This study focuses on the machining (drilling) behavior of the developed PMMA-based Hydroxyapatite (PMMA-HA) bio-nano- composites. The machining efficiency and parametric control were estimated using a combined principal component analysis (PCA) module and evaluation based on distance from average solution (EDAS). The Hydroxyapatite (HA) weight percentage (wt.%), spindle speed (SPEED) and tool material (TOOL) viz. HSS, Carbide and TiAlN are chosen according to the Taguchi-based experimental array. The objective is to get the best possible machining responses, such as the material removal rate (MRR), mean surface roughness (Ra) and circularity error ([Formula: see text] using the PCA-EDAS hybrid module. The optimal condition is found as the HSS drilling bit, 10%[Formula: see text]wt.%, SPEED-1428[Formula: see text]rpm with an improvement of 30.53%, 21.15% and 41.9% in MRR, Ra and [Formula: see text]-ERROR, respectively. The microstructural investigation scanning electron microscope (SEM) shows the excellent morphology and quality of the drilled hole in the proposed composites. Also, an X-ray diffraction (XRD) analysis of the prepared sample was done to ensure the proper reinforcement. The flexural test shows a significant expansion in the mechanical property due to the presence of HA in PMMA

Modeling Temporal Patterns with Dilated Convolutions for Time-Series Forecasting

Time-series forecasting is an important problem across a wide range of domains. Designing accurate and prompt forecasting algorithms is a non-trivial task, as temporal data that arise in real applications often involve both non-linear dynamics and linear dependencies, and always have some mixtures of sequential and periodic patterns, such as daily, weekly repetitions, and so on. At this point, however, most recent deep models often use Recurrent Neural Networks (RNNs) to capture these temporal patterns, which is hard to parallelize and not fast enough for real-world applications especially when a huge amount of user requests are coming. Recently, CNNs have demonstrated significant advantages for sequence modeling tasks over the de-facto RNNs, while providing high computational efficiency due to the inherent parallelism. In this work, we propose HyDCNN, a novel hybrid framework based on fully Dilated CNN for time-series forecasting tasks. The core component in HyDCNN is a proposed hybrid module, in which our proposed position-aware dilated CNNs are utilized to capture the sequential non-linear dynamics and an autoregressive model is leveraged to capture the sequential linear dependencies. To further capture the periodic temporal patterns, a novel hop scheme is introduced in the hybrid module. HyDCNN is then composed of multiple hybrid modules to capture the sequential and periodic patterns. Each of these hybrid modules targets on either the sequential pattern or one kind of periodic patterns. Extensive experiments on five real-world datasets have shown that the proposed HyDCNN is better compared with state-of-the-art baselines and is at least 200% better than RNN baselines. The datasets and source code will be published in Github to facilitate more future work.

A NEW CRITERION FOR DRILLING MACHINABILITY EVALUATION OF NANOCOMPOSITES MODIFIED BY GRAPHENE/CARBON FIBER EPOXY MATRIX AND OPTIMIZATION USING COMBINED COMPROMISE SOLUTION

This article describes new control criteria and robust optimization methodology to balance drilling parameters and machining characteristics. Experimentation was performed according to response surface methodology (RSM) using a TiAlN coated SiC tool. The full drilling force signal and cutting parameters tested are categorized into five stages, indicating the drilling tool-workpiece interactions’ different statuses. Principal component analysis (PCA) assigns real response priority weight during the aggregation of conflicting characteristics. The hybrid module of combined compromise solution and PCA (CoCoSo–PCA) is used to decide the optimal parametric setting. It efficiently undertakes a trade-off between minimal thrust ([Formula: see text][Formula: see text]N), torque ([Formula: see text][Formula: see text]Nm) surface roughness ([Formula: see text]m). A regression model between input parameters and output function was established using RSM quadratic model. The validation experiment shows significant improvement, and the proposed module can be recommended for quality-productivity characteristics control.

Robust deep convolutional neural network against image distortions

Many approaches have been proposed in the literature to enhance the robustness of Convolutional Neural Network (CNN)-based architectures against image distortions. Attempts to combat various types of distortions can be made by combining multiple expert networks, each trained by a certain type of distorted images, which however lead to a large model with high complexity. In this paper, we propose a CNN-based architecture with a pre-processing unit in which only undistorted data are used for training. The pre-processing unit employs discrete cosine transform (DCT) and discrete wavelets transform (DWT) to remove high-frequency components while capturing prominent high-frequency features in the undistorted data by means of random selection. We further utilize the singular value decomposition (SVD) to extract features before feeding the preprocessed data into the CNN for training. During testing, distorted images directly enter the CNN for classification without having to go through the hybrid module. Five different types of distortions are produced in the SVHN dataset and the CIFAR-10/100 datasets. Experimental results show that the proposed DCT-DWT-SVD module built upon the CNN architecture provides a classifier robust to input image distortions, outperforming the state-of-the-art approaches in terms of accuracy under different types of distortions.

Emergent Instructional Strategies for a Performance-Based Course That Is Taught Virtually

This chapter will focus on strategies for educators teaching courses relying on a performance-based set of practices but are forced to teach virtually. By reflecting on the author's personal experience of teaching a hybrid module course in Spring 2019 and teaching completely online towards the end of the semester of Fall 2020, Cornelius will touch on several observations from recent times to propel proposed solutions, tips, and advice. Highlighting lessons learned during the transition and useful resources, this chapter will focus on strategies to lessen the blow of teaching a course rooted in practice during a pandemic and how to dial it back towards a theory-based course while maintaining most or all intended learning outcomes. In light of the pandemic, there will be a large amount of weight on the theory side; how does one find the counterweight? While lending towards an innovative side while thinking of solutions, the chapter will describe ideas that educators may want to attempt to facilitate hands-on learning at home.

Synergistic 2D/3D Convolutional Neural Network for Hyperspectral Image Classification

Accurate hyperspectral image classification has been an important yet challenging task for years. With the recent success of deep learning in various tasks, 2-dimensional (2D)/3-dimensional (3D) convolutional neural networks (CNNs) have been exploited to capture spectral or spatial information in hyperspectral images. On the other hand, few approaches make use of both spectral and spatial information simultaneously, which is critical to accurate hyperspectral image classification. This paper presents a novel Synergistic Convolutional Neural Network (SyCNN) for accurate hyperspectral image classification. The SyCNN consists of a hybrid module that combines 2D and 3D CNNs in feature learning and a data interaction module that fuses spectral and spatial hyperspectral information. Additionally, it introduces a 3D attention mechanism before the fully-connected layer which helps filter out interfering features and information effectively. Extensive experiments over three public benchmarking datasets show that our proposed SyCNNs clearly outperform state-of-the-art techniques that use 2D/3D CNNs.