Optimized Ensemble Algorithm for Predicting Metamaterial Antenna Parameters

Metamaterial Antenna is a subclass of antennas that makes use of metamaterial to improve performance. Metamaterial antennas can overcome the bandwidth constraint associated with tiny antennas. Machine learning is receiving a lot of interest in optimizing solutions in a variety of areas. Machine learning methods are already a significant component of ongoing research and are anticipated to play a critical role in today's technology. The accuracy of the forecast is mostly determined by the model used. The purpose of this article is to provide an optimal ensemble model for predicting the bandwidth and gain of the Metamaterial Antenna. Support Vector Machines (SVM), Random Forest, K-Neighbors Regressor, and Decision Tree Regressor were utilized as the basic models. The Adaptive Dynamic Polar Rose Guided Whale Optimization method, named AD-PRS-Guided WOA, was used to pick the optimal features from the datasets. The suggested model is compared to models based on five variables and to the average ensemble model. The findings indicate that the presented model using Random Forest results in a Root Mean Squared Error (RMSE) of (0.0102) for bandwidth and RMSE of (0.0891) for gain. This is superior to other models and can accurately predict antenna bandwidth and gain.     

Coupled equilibrium model of hybridization error for the DNA microarray and tag-antitag systems

In this work, a detailed coupled equilibrium model is presented for predicting the ensemble average probability of hybridization error per chip-hybridized input strand, providing the first ensemble average method for estimating postannealing microarray/TAT system error rates. Following a detailed presentation of the model and implementation via the software package NucleicPark, under a mismatched statistical zipper model of duplex formation, error response is simulated for both mean-energy and randomly encoded TAT systems versus temperature and input concentration. Limiting expressions and simulated model behavior indicate the occurrence of a transition in hybridization error response, from a logarithmically convex function of temperature for excess inputs (high-error behavior), to a monotonic, log-linear function of temperature for dilute inputs (low-error behavior), a novel result unpredicted by uncoupled equilibrium models. Model scaling behavior for random encodings is investigated versus system size and strand-length. Application of the model to TAT system design is also undertaken, via the in silico evolution of a high-fidelity 100-strand TAT system, with an error response improved by nine standard deviations over the performance of the mean random encoding     

Hemoglobin correction for near-infrared pH determination in lysed blood solutions

The near-infrared (NIR) measurement of blood pH relies on the spectral signature of histidine residing on the hemoglobin molecule. If the amount of hemoglobin in solution varies, the size of the histidine signal can vary depending on changes in either the pH or hemoglobin concentration. Multivariate calibration models developed using the NIR spectra collected from blood at a single hemoglobin concentration are shown to predict data from different hemoglobin levels with a bias and slope. A simple, scalar path length correction of the spectral data does not correct this problem. However, global partial least-square (PLS) models built with data encompassing a range of hemoglobin concentration have a cross-validated standard error of prediction (CVSEP) similar to the CVSEP of data obtained from a single hemoglobin level. It will be shown that the prediction of pH of an unknown sample using a global PLS model requires that the unknown have a hemoglobin concentration falling within the range encompassed by the global model. An alternative method for correcting the predicted pH for hemoglobin levels is also presented. The alternative method updates the single-hemoglobin-level models with slope and intercept estimates from the pH predictions of data collected at alternate hemoglobin levels. The slope and intercept correction method gave SEP values averaging to 0.034 pH units. Since both methods require some knowledge of the hemoglobin concentration in order for a pH prediction to be made, a model for hemoglobin concentration is developed using spectral data and is used for pH correction.     

An Optimized Ensemble Model for Prediction the Bandwidth of Metamaterial Antenna

Metamaterial Antenna is a special class of antennas that uses metamaterial to enhance their performance. Antenna size affects the quality factor and the radiation loss of the antenna. Metamaterial antennas can overcome the limitation of bandwidth for small antennas. Machine learning (ML) model is recently applied to predict antenna parameters. ML can be used as an alternative approach to the trial-and-error process of finding proper parameters of the simulated antenna. The accuracy of the prediction depends mainly on the selected model. Ensemble models combine two or more base models to produce a better-enhanced model. In this paper, a weighted average ensemble model is proposed to predict the bandwidth of the Metamaterial Antenna. Two base models are used namely: Multilayer Perceptron (MLP) and Support Vector Machines (SVM). To calculate the weights for each model, an optimization algorithm is used to find the optimal weights of the ensemble. Dynamic Group-Based Cooperative Optimizer (DGCO) is employed to search for optimal weight for the base models. The proposed model is compared with three based models and the average ensemble model. The results show that the proposed model is better than other models and can predict antenna bandwidth efficiently.     

Blood Flow,Negative Pressure,and Hemolysis During Hemodialysis

Blood flow, measured by an ultrasound flowmeter, and the extent of hemolysis were assessed during a single, routine dialysis in 100 patients. Before and after the hemodialysis session, blood was drawn for measurements of haptoglobin (HPT), hemoglobin (HGB), albumin (ALB), and lactate dehydrogenase (LDH). The average values were as follows: pump speed 510 mL/ min, real blood flow 422 mL/min, arterial chamber pressure -350 mm Hg, and venous chamber pressure 279 mm Hg. Haptoglobin concentrations were higher in patients with central vein catheters compared to patients with arteriovenous access. The meaning of this finding is unclear. Mean HPT concentrations increased significantly less during hemodialysis (2.37%) than concentrations of ALB (11.3%), HGB (9.17%), and LDH (18.2%), indicating that some hemolysis is present in all dialyses. In dialyses with arterial chamber pressures more negative than -350 the median concentration of ALB (8.70%) increased significantly more than the median concentration of HGB (7.99%). This indicates significantly more hemolysis in dialyses with more negative pressures compared to those with less negative arterial chamber pressures. Median LDH increased more in dialyses with more negative prepump chamber pressures (16.19% vs 13.78%), but not significantly; however, LDH increases were significantly higher than either HGB or ALB, thus indicating significantly more hemolysis in dialyses with more negative pressures compared to those with less negative arterial chamber pressures. Erythropoietin dose was not significantly different in patients dialyzed with more or less negative arterial chamber pressures (17 645 ± 1226 U/week vs 16 308 ± 1506 U/week). We conclude that dialyses with negative arterial chamber pressures greater than -350 mm Hg cause slightly higher hemolysis than dialyses with less negative arterial chamber pressures, but this increased hemolysis is not associated with an increased requirement of erythropoietin dose. Whether this increased hemolysis is of clinical significance is uncertain.     

大跨拱支预应力网壳结构的模糊优化设计研究

针对武汉长江防洪模型展示大厅采用的大跨拱支预应力网壳结构,考虑优化目标函数与约束条件的模糊性,建立了该类结构体系的模糊优化设计数学模型;求解时首先通过约束水平截集法,将模糊优化模型转化为一系列确定性优化模型;然后基于拱支预应力网壳的结构特点,将预应力构件(拉索和吊杆)截面尺寸、预应力作用取值(构件初始应变)以及非预应力构件截面尺寸等优化设计变量分别归为不同的优化级别,采用分级优化思想对确定性优化模型进行求解;最后依据结构经济性与安全性平衡的目标求出结构的最优约束水平,从而得到最优的结构设计方案,并同时确定结构的合理预应力分布,由此形成了拱支预应力网壳结构的两阶段三级模糊优化设计方法。     

Assessment of environmental radon hazard using human respiratory tract models

Radon is a natural radioactive gas derived from geological materials. It has been estimated that about half of the total effective dose received by human beings from all sources of ionizing radiation is attributed to 222Rn and its short-lived progeny. In this paper, the use of human respiratory tract models to assess the health hazard from environmental radon is reviewed. A short history of dosimetric models for the human respiratory tract from the International Commission on Radiological Protection (ICRP) is first presented. The most important features of the newest model published by ICRP in 1994 (as ICRP Publication 66) are then described, including the morphometric model, physiological parameters, radiation biology, deposition of aerosols, clearance model and dose weighting. Comparison between different morphometric models and comparison between different deposition models are then given. Finally, the significance of various parameters in the lung model is discussed, including aerosol parameters, subject related parameters, target and cell related parameters, and parameters that define the absorption of radon from the lungs to blood. Dosimetric calculations gave a dose conversion coefficient of 15 mSv/WLM, which is higher than the value 5 mSv/WLM derived from epidemiological studies. ICRP stated that dosimetric models should only be used for comparison of doses in the human lungs resulted from different exposure conditions.     

Effect of rework,rejects and inspection on lot size with work-in-process inventory

The economic order quantity and economic production quantity models are the most commonly used inventory models in production environments for the calculation of optimum lot size. However, these models are based on the unrealistic assumption that every process produces good quality products every time. Moreover, the impact of inspection is neglected in all extended inventory control models involving work in process inventory. By taking both imperfect production and lot size inspection into consideration, this paper presents a more realistic approach for the modelling of optimum lot size and total cost with a focus on the work in process inventory. A mathematical model is derived for optimum lot size based on the minimisation of the average cost. Our approach incorporates the effect of rework, rejects and inspection on work in process inventory. The significant effect of imperfect production and inspection on optimum lot size is evaluated via numerical examples. In comparison to existing models, the proposed model is a more generalised and flexible form of inventory model for independent demands.     

Physical phantom of typical Korean male for radiation protection purpose

Dose distribution within a human body can be measured using physical anthropomorphic phantoms. In an effort to establish reference Korean physical model, the first Korean physical phantom of average Korean adult male was constructed using computed tomography (CT) images of a healthy volunteer. The body dimension of the subject was close to that of average Korean male. The source images were obtained using fusion positron emission tomography machine at Radiation Health Research Institute in Korea, and ported into rapid prototyping process. The physical phantom was composed of three tissue-equivalent materials: epoxy resin, urethane foam and polyurethane representing bone, lungs and soft tissues, respectively. The densities of the tissue-equivalent materials were close to those recommended by the International Commission on Radiation Units and measurements. To facilitate dose mapping, the phantom was sliced into 2 cm sections. Hole grids for thermoluminescence (TL) dosemeter chips were drilled. To verify the appropriateness of the physical phantom, organ doses of selected organs were measured for reference photon beam, and compared with those computed by tomographic model constructed from the same CT images. Absorbed doses converted from TL relative response showed good agreement within 7% with those calculated.     

Perihospitalization hemoglobin-epoetin associations in U.S. hemodialysis patients, 1998 to 2003

While hospitalization is common for hemodialysis patients, perihospitalization associations between hemoglobin levels and epoetin doses are not well characterized. U.S. Medicare claims were used to identify 71,360 hemodialysis patients hospitalized from 1998 to 2003. Hemoglobin levels, epoetin doses, and epoetin responsiveness index (ERI) were compared by calendar year. In the prehospitalization month, the mean hemoglobin levels increased from 10.96 g/dL in 1998 to 11.76 in 2003 and the mean epoetin doses from 63,715 to 75,012 U; corresponding values in the hospitalization month were 10.53 and 11.19 g/dL, and 66,623 and 80,569 U. In each year, prehospitalization hemoglobin levels were achieved within 2 months, but ERI declined to prehospitalization levels within 12 months only in 2000. With mixed models, hemoglobin declines in the 3 prehospitalization months grew between 1998 (-0.1362 g/dL/month) and 2003 (-0.2003 g/dL/month). Epoetin responsiveness index slopes were J-shaped, with values of 287.9, 221.1, and 356.5U/month per g/dL in 1998, 2000, and 2003. In the 3 postadmission months, a modest increase in the rapidity of hemoglobin recovery was seen (+0.2538 g/dL/month in 1998, +0.2743 in 2003), with increasing rates of ERI change (+8.7 U/month/g/dL in 1998, +146.8 in 2003). While time to recovery of prehospitalization hemoglobin levels remained constant year to year, epoetin doses and ERI did not, suggesting that optimum perihospitalization anemia management practices have yet to be determined.     

Flood warning by ensemble of multiple hydrological models: a case study for Typhoon Morakot

After the catastrophic disaster brought by Typhoon Morakot in 2009, the enhancement of flood warning technology cannot wait in Taiwan. In recent years, ensemble flood warning has exhibited advantages in extending lead time, quantifying uncertainty and raising confidence in issuing warnings. Unlike most ensembles aimed at integrating meteorological variations, this study generates the ensemble through the combination of multiple conceptually different hydrological models in order to avoid possible bias by applying a single model for a flood forecast. Taking Typhoon Morakot as the study case, the townships in Chiayi City/County are selected as the study areas to compare the performance of ensemble warning with that given by individual models. The results indicate that the ensemble warning shows better accuracy than individual models by giving higher overall correctness, revealing the fact that hydrological ensemble is no less important than meteorological ensemble in acquiring better flood warning performance.     

Comparison of observed lung retention and urinary excretion of thorium workers and members of the public in India with the values predicted by the ICRP biokinetic model

The daily intake of natural Th and its contents in lungs, skeleton and liver of an Indian adult population group were estimated using radiochemical neutron activation analysis (RNAA) technique. These data on daily intake (through inhalation and ingestion) were used to compute Th contents in lungs and other systemic organs such as skeleton and liver using the new human respiratory tract model (HRTM) and the new biokinetic model of Th. The theoretically computed Th contents in lungs, skeleton and liver of an average Indian adult are 2.56, 4.00 and 0.17 microg, respectively which are comparable with the corresponding experimentally measured values of 4.31, 3.45 and 0.14 microg in an urban population group living in Mumbai. The measured lung contents of Th in a group of five occupational workers were used to compute their total body Th contents and the corresponding daily urinary excretions. The computed total body contents and daily urinary excretions of Th in the five subjects compared favourably with their measured values. These studies, thus, validate the new biokinetic model of Th in natural as well as in occupational exposures in Indian conditions.     

Continuous analogue to iterative optimization for PDE-constrained inverse problems

The parameters of many physical processes are unknown and have to be inferred from experimental data. The corresponding parameter estimation problem is often solved using iterative methods such as steepest descent methods combined with trust regions. For a few problem classes also continuous analogues of iterative methods are available. In this work, we expand the application of continuous analogues to function spaces and consider PDE (partial differential equation)-constrained optimization problems. We derive a class of continuous analogues, here coupled ODE (ordinary differential equation)–PDE models, and prove their convergence to the optimum under mild assumptions. We establish sufficient bounds for local stability and convergence for the tuning parameter of this class of continuous analogues, the retraction parameter. To evaluate the continuous analogues, we study the parameter estimation for a model of gradient formation in biological tissues. We observe good convergence properties, indicating that the continuous analogues are an interesting alternative to state-of-the-art iterative optimization methods.     

Distribution of natural thorium in the tissues of a whole body

The distribution of thorium in the tissues of a whole body donor to the United States Transuranium and Uranium Registries is described. This case, identified by the USTUR as Case 0212, had two documented intakes of plutonium and americium from occupational accidents while employed at Hanford but no known occupational exposure to thorium. Concentrations of 239+240Pu, 241Am, and 232Th in the tissues are compared and the distribution of these isotopes in this case is evaluated. The distribution data for 232Th are compared to those from previous studies of thorium in human tissues resulting from environmental exposure and to an individual exposed to Thorotrast (colloidal ThO2) in a medical diagnostic procedure. The 232Th distribution data from this work are also compared against ICRP 30 and ICRP 69 models for the behaviour of thorium in the human body.     

A Stochastic Simplex Approximate Gradient (StoSAG) for optimization under uncertainty

We consider a technique to estimate an approximate gradient using an ensemble of randomly chosen control vectors, known as Ensemble Optimization (EnOpt) in the oil and gas reservoir simulation community. In particular, we address how to obtain accurate approximate gradients when the underlying numerical models contain uncertain parameters because of geological uncertainties. In that case, ‘robust optimization’ is performed by optimizing the expected value of the objective function over an ensemble of geological models. In earlier publications, based on the pioneering work of Chen et al. (2009), it has been suggested that a straightforward one‐to‐one combination of random control vectors and random geological models is capable of generating sufficiently accurate approximate gradients. However, this form of EnOpt does not always yield satisfactory results. In a recent article, Fonseca et al. (2015) formulate a modified EnOpt algorithm, referred to here as a Stochastic Simplex Approximate Gradient (StoSAG; in earlier publications referred to as ‘modified robust EnOpt’) and show, via computational experiments, that StoSAG generally yields significantly better gradient approximations than the standard EnOpt algorithm. Here, we provide theoretical arguments to show why StoSAG is superior to EnOpt. © 2016 The Authors. International Journal for Numerical Methods in Engineering Published by John Wiley & Sons, Ltd.     

High-Movement Human Segmentation in Video Using Adaptive N-Frames Ensemble

A wide range of camera apps and online video conferencing services support the feature of changing the background in real-time for aesthetic, privacy, and security reasons. Numerous studies show that the Deep-Learning (DL) is a suitable option for human segmentation, and the ensemble of multiple DL-based segmentation models can improve the segmentation result. However, these approaches are not as effective when directly applied to the image segmentation in a video. This paper proposes an Adaptive N-Frames Ensemble (AFE) approach for high-movement human segmentation in a video using an ensemble of multiple DL models. In contrast to an ensemble, which executes multiple DL models simultaneously for every single video frame, the proposed AFE approach executes only a single DL model upon a current video frame. It combines the segmentation outputs of previous frames for the final segmentation output when the frame difference is less than a particular threshold. Our method employs the idea of the N-Frames Ensemble (NFE) method, which uses the ensemble of the image segmentation of a current video frame and previous video frames. However, NFE is not suitable for the segmentation of fast-moving objects in a video nor a video with low frame rates. The proposed AFE approach addresses the limitations of the NFE method. Our experiment uses three human segmentation models, namely Fully Convolutional Network (FCN), DeepLabv3, and Mediapipe. We evaluated our approach using 1711 videos of the TikTok50f dataset with a single-person view. The TikTok50f dataset is a reconstructed version of the publicly available TikTok dataset by cropping, resizing and dividing it into videos having 50 frames each. This paper compares the proposed AFE with single models and the Two-Models Ensemble, as well as the NFE models. The experiment results show that the proposed AFE is suitable for low-movement as well as high-movement human segmentation in a video.     

A Hybrid Neural Network Model for Marine Dissolved Oxygen Concentrations Time-Series Forecasting Based on Multi-Factor Analysis and a Multi-Model Ensemble

Dissolved oxygen (DO) is an important indicator of aquaculture, and its accurate forecasting can effectively improve the quality of aquatic products. In this paper, a new DO hybrid forecasting model is proposed that includes three stages: multi-factor analysis, adaptive decomposition, and an optimization-based ensemble. First, considering the complex factors affecting DO, the grey relational (GR) degree method is used to screen out the environmental factors most closely related to DO. The consideration of multiple factors makes model fusion more effective. Second, the series of DO, water temperature, salinity, and oxygen saturation are decomposed adaptively into sub-series by means of the empirical wavelet transform (EWT) method. Then, five benchmark models are utilized to forecast the sub-series of EWT decomposition. The ensemble weights of these five sub-forecasting models are calculated by particle swarm optimization and gravitational search algorithm (PSOGSA). Finally, a multi-factor ensemble model for DO is obtained by weighted allocation. The performance of the proposed model is verified by time-series data collected by the pacific islands ocean observing system (PacIOOS) from the WQB04 station at Hilo. The evaluation indicators involved in the experiment include the Nash–Sutcliffe efficiency (NSE), Kling–Gupta efficiency (KGE), mean absolute percent error (MAPE), standard deviation of error (SDE), and coefficient of determination (R²). Example analysis demonstrates that: ① The proposed model can obtain excellent DO forecasting results; ② the proposed model is superior to other comparison models; and ③ the forecasting model can be used to analyze the trend of DO and enable managers to make better management decisions.     

ADAPTIVE RESPONSE SURFACE METHOD - A GLOBAL OPTIMIZATION SCHEME FOR APPROXIMATION-BASED DESIGN PROBLEMS

For design problems involving computation-intensive analysis or simulation processes, approximation models are usually introduced to reduce computation lime. Most approximation-based optimization methods make step-by-step improvements to the approximation model by adjusting the limits of the design variables. In this work, a new approximation-based optimization method for computation-intensive design problems - the adaptive response surface method(ARSM), is presented. The ARSM creates quadratic approximation models for the computation-intensive design objective function in a gradually reduced design space. The ARSM was designed to avoid being trapped by local optima and to identify the global design optimum with a modest number of objective function evaluations. Extensive tests on the ARSM as a global optimization scheme using benchmark problems, as well as an industrial design application of the method, are presented. Advantages and limitations of the approach are also discussed     

OPTIMUM DESIGN OF A SINGLE-ENDED FLYBACK CONVERTER

In this paper, an optimum design method is employed for a single ended flyback converter with four outputs. The mathematical model contains 16 variables and 16 constraints to minimize the weighted sum of the power loss and weight of the converter. The Augmented Lagrangian Multiplier Penalty Function Technique is adopted in forming the CAD program with FORTRAN IV. By treating the switching frequency as a constant in each computer run, a set of suboptimum design solutions were obtained by varying the frequency from 20 KHz to 60 KHz in 10 KHz steps. The U-shaped curve is observed by plotting the total loss characteristic against frequency. The following optimum design results are also obtained:

1) The optimum de duty ratio is about 0.3 to 0.34.

2) The converter should work at the critical mode between the discontinuous and current mode.

3) The air gap of the transformer core should be large enough (e.g.to 1.3 mm).     

Optimization of adaptive-optics systems closed-loop bandwidth settings to maximize imaging-system performance

We present the results of research aimed at optimizing adaptive-optics closed-loop bandwidth settings to maximize imaging-system performance. The optimum closed-loop bandwidth settings are determined as a function of target-object light levels and atmospheric seeing conditions. Our work shows that, for bright objects, the optimum closed-loop bandwidth is near the Greenwood frequency. However, for dim objects without the use of a laser beacon the preferred closed-loop bandwidth settings are a small fraction of the Greenwood frequency. In addition, under low light levels selection of the proper closed-loop bandwidth is more critical for achieving maximum performance than it is under high light levels. We also present a strategy for selecting the closed-loop bandwidth to provide robust system performance for different target-object light levels.