Deterioration Of Accuracy Research Articles

Computation of singular solutions to the Helmholtz equation with high order accuracy

Solutions to elliptic PDEs, in particular to the Helmholtz equation, become singular near the boundary if the boundary data do not possess sufficient regularity. In that case, the convergence of standard numerical approximations may slow down or cease altogether. We propose a method that maintains a high order of grid convergence even in the presence of singularities. This is accomplished by an asymptotic expansion that removes the singularities up to several leading orders, and the remaining regularized part of the solution can then be computed on the grid with the expected accuracy.The computation on the grid is rendered by a compact finite difference scheme combined with the method of difference potentials. The scheme enables fourth order accuracy on a narrow 3×3 stencil: it uses only one unknown variable per grid node and requires only as many boundary conditions as needed for the underlying differential equation itself. The method of difference potentials enables treatment of non-conforming boundaries on regular structured grids with no deterioration of accuracy, while the computational complexity remains comparable to that of a conventional finite difference scheme on the same grid. The method of difference potentials can be considered a generalization of the method of Calderon's operators in PDE theory.In the paper, we provide a theoretical analysis of our combined methodology and demonstrate its numerical performance on a series of tests that involve Dirichlet and Neumann boundary data with various degrees of “non-regularity”: an actual jump discontinuity, a discontinuity in the first derivative, a discontinuity in the second derivative, etc. All computations are performed on a Cartesian grid, whereas the boundary of the domain is a circle, chosen as a simple but non-conforming shape. In all cases, the proposed methodology restores the design rate of grid convergence, which is fourth order, in spite of the singularities and regardless of the fact that the boundary is not aligned with the discretization grid. Moreover, as long as the location of the singularities is known and remains fixed, a broad spectrum of problems involving different boundary conditions and/or data on “smooth” segments of the boundary can be solved economically since the discrete counterparts of Calderon's projections need to be calculated only once and then can be applied to each individual formulation at very little additional cost.

Accuracy-preserving boundary flux quadrature for finite-volume discretization on unstructured grids

In this paper, we derive a general accuracy-preserving boundary flux quadrature formula for the second- and third-order node-centered edge-based finite-volume discretizations on triangular and tetrahedral grids. It is demonstrated that the general formula reduces to some well-known formulas for the second-order scheme in the case of linear fluxes. Some special boundary grids are also examined. In particular, a simple one-point quadrature formula, which is typically used for quadrilateral grids, is shown to be exact for quadratic fluxes on triangular grids with uniformly spaced straight boundaries. Numerical results are presented to demonstrate the accuracy of the general formula, and accuracy deterioration caused by incompatible boundary flux quadrature formulas. In general, the third-order accuracy is lost everywhere in the domain unless the third-order accuracy is maintained at boundary nodes. It is also demonstrated that the third-order scheme does not require high-order curved elements for curved boundaries but requires accurate surface normal vectors defined, which can be estimated by a quadratic interpolation from a given grid, to deliver the designed third-order accuracy.

A Varying-Coefficient Expectile Model for Estimating Value at Risk

This article develops a nonparametric varying-coefficient approach for modeling the expectile-based value at risk (EVaR). EVaR has an advantage over the conventional quantile-based VaR (QVaR) of being more sensitive to the magnitude of extreme losses. EVaR can also be used for calculating QVaR and expected shortfall (ES) by exploiting the one-to-one mapping from expectiles to quantiles, and the relationship between VaR and ES. Previous studies on conditional EVaR estimation only considered parametric autoregressive model set-ups, which account for the stochastic dynamics of asset returns but ignore other exogenous economic and investment related factors. Our approach overcomes this drawback and allows expectiles to be modeled directly using covariates that may be exogenous or lagged dependent in a flexible way. Risk factors associated with profits and losses can then be identified via the expectile regression at different levels of prudentiality. We develop a local linear smoothing technique for estimating the coefficient functions within an asymmetric least squares minimization set-up, and establish the consistency and asymptotic normality of the resultant estimator. To save computing time, we propose to use a one-step weighted local least squares procedure to compute the estimates. Our simulation results show that the computing advantage afforded by this one-step procedure over full iteration is not compromised by a deterioration in estimation accuracy. Real data examples are used to illustrate our method. Supplementary materials for this article are available online.

Combined State of Charge and State of Health estimation over lithium-ion battery cell cycle lifespan for electric vehicles

A combined SOC (State Of Charge) and SOH (State Of Health) estimation method over the lifespan of a lithium-ion battery is proposed. First, the SOC dependency of the nominal parameters of a first-order RC (resistor-capacitor) model is determined, and the performance degradation of the nominal model over the battery lifetime is quantified. Second, two Extended Kalman Filters with different time scales are used for combined SOC/SOH monitoring: the SOC is estimated in real-time, and the SOH (the capacity and internal ohmic resistance) is updated offline. The time scale of the SOH estimator is determined based on model accuracy deterioration. The SOC and SOH estimation results are demonstrated by using large amounts of testing data over the battery lifetime.

Dielectric spectroscopy‐based estimation of VCD in CHO cell culture

AbstractBACKGROUNDIn mammalian cell culture processes, viable cell density (VCD) is one of the most crucial parameters for monitoring of the current status of cell growth and protein production. Among the various real‐time monitoring tools available for VCD in Chinese hamster ovary (CHO) cell culture, dielectric spectroscopy is one of the more reliable due to its simplicity, robustness and accuracy. These attributes enable efficient monitoring and control of the corresponding processes. However, one of the major drawbacks of dielectric spectroscopy is frequent deterioration of accuracy during the late‐growth phase of CHO cell culture. This can be caused by abrupt changes in the cell physiology, as well as environmental conditions.RESULTSIn this study, various kinds of chemometric tools were employed to understand the spectral dynamics of dielectric spectroscopy during extended CHO cell culture, including different cell growth phases. Then, the performance of different multivariate regression models were compared with conventional Cole‐Cole equation‐based approaches. The models were compared in their ability to estimate VCD in CHO cell culture from dielectric spectra from several batches of bioreactor operations. Multivariate analysis of dielectric spectral datasets clearly revealed the time‐varying evolution of spectral dynamics during the transition between cell growth phases. The best prediction performance from the dielectric spectroscopy was realized using a regression model based on locally‐weighted partial least squares (LWPLS). This estimation model best compensated for spectral dynamics observed during the different cell growth phases.CONCLUSIONDielectric spectroscopy combined with LWPLS allows achievement of optimal and stable estimation of VCD during CHO cell culture, regardless of inherent differences in cell growth phases. © 2014 Society of Chemical Industry

Graph ensemble boosting for imbalanced noisy graph stream classification.

Many applications involve stream data with structural dependency, graph representations, and continuously increasing volumes. For these applications, it is very common that their class distributions are imbalanced with minority (or positive) samples being only a small portion of the population, which imposes significant challenges for learning models to accurately identify minority samples. This problem is further complicated with the presence of noise, because they are similar to minority samples and any treatment for the class imbalance may falsely focus on the noise and result in deterioration of accuracy. In this paper, we propose a classification model to tackle imbalanced graph streams with noise. Our method, graph ensemble boosting, employs an ensemble-based framework to partition graph stream into chunks each containing a number of noisy graphs with imbalanced class distributions. For each individual chunk, we propose a boosting algorithm to combine discriminative subgraph pattern selection and model learning as a unified framework for graph classification. To tackle concept drifting in graph streams, an instance level weighting mechanism is used to dynamically adjust the instance weight, through which the boosting framework can emphasize on difficult graph samples. The classifiers built from different graph chunks form an ensemble for graph stream classification. Experiments on real-life imbalanced graph streams demonstrate clear benefits of our boosting design for handling imbalanced noisy graph stream.

Determination of In Situ Trap Properties in CCDs Using a “Single-Trap Pumping” Technique

The science goals of space missions from the Hubble Space Telescope through to Gaia and Euclid require ultraprecise positional, photometric, and shape measurement information. However, in the radiation environment of the space telescopes, damage to the focal plane detectors through high-energy protons leads to the creation of traps, a loss of charge transfer efficiency, and a consequent deterioration in measurement accuracy. An understanding of the traps produced and their properties in the CCD during operation is essential to allow optimization of the devices and suitable modeling to correct the effect of the damage through the postprocessing of images. The technique of “pumping single traps” has allowed the study of individual traps in high detail that cannot be achieved with other techniques, such as deep level transient spectroscopy, whilst also locating each trap to the subpixel level in the device. Outlining the principles used, we have demonstrated the technique for the A-center, the most influential trap in serial readout, giving results consistent with the more general theoretical values, but here showing new results indicating the spread in the emission times achieved and the variation in capture probability of individual traps with increasing signal levels. This technique can now be applied to other time and temperature regimes in the CCD to characterize individual traps in situ under standard operating conditions such that dramatic improvements can be made to optimization processes and modeling techniques.

Mobile Node Localization Focusing on Stop-and-Go Behavior of Indoor Pedestrians

Despite recent advances in localization technology for mobile devices, to provide real-time position information to people indoors is still a big challenge; usually there is a trade-off between localization accuracy and infrastructural costs (e.g., dense anchor deployment). A possible solution would be employing cooperative approaches which utilize estimated positions of surrounding mobile nodes to complement a small number of anchors. However, it often results in poor estimation accuracy since a temporary large position error due to node mobility easily propagates to neighbor nodes. This paper presents a novel cooperative localization algorithm that addresses this problem by focusing on “stop-and-go behavior” of indoor pedestrians. The key idea is to collaboratively find movement state (moving or static) of each node based on peer-to-peer distance measurement which is inherently necessary for cooperative localization, and use only static nodes as reference points for localization to avoid potential accuracy deterioration. Also, nodes in static state can reduce localization frequency to conserve battery power, keeping the tracking quality. Through extensive simulations, we have demonstrated the performance of our method in terms of accuracy and energy efficiency. The effectiveness in a real application scenario has been also confirmed using a measurement-based sensor model and real mobility traces.

Proper discretization of homogeneous differentiators

Homogeneous sliding-mode-based differentiators (HD) are known to provide for the high-accuracy robust estimation of derivatives in the presence of sampling noises and discrete measurements, provided that the differentiator dynamics evolve in continuous time. The popular one-step Euler discrete-time implementation is proved to cause differentiation accuracy deterioration, if the differentiation order exceeds 1. A novel discrete-time realization of the HD is proposed, which preserves the ultimate accuracy of the continuous-time HD also with discrete measurements.

Similarity Retrieval of Vector Images with Indirect Matching

Increase use of vector images have generated interests on effective ways of vector image retrieval. This paper presents a fast similarity retrieval of vector images. For reducing the computational cost in similarity evaluation, we proposed an indirect matching framework. This framework uses a small number of key images, which are previously randomly selected from a database. The basic idea of the framework is offline calculation of creating a similarity table with pre-selected key images. In the retrieval phase, by referring to the similarity table, the similarity of each database image to the given query is quickly estimated. Experimental results have shown that the retrieval time is greatly reduced by the proposed method without much deterioration of retrieval accuracy.

Finite-time sliding mode attitude control for a reentry vehicle with blended aerodynamic surfaces and a reaction control system

This paper proposes a finite-time robust flight controller, targeting for a reentry vehicle with blended aerodynamic surfaces and a reaction control system (RCS). Firstly, a novel finite-time attitude controller is pointed out with the introduction of a nonsingular finite-time sliding mode manifold. The attitude tracking errors are mathematically proved to converge to zero within finite time which can be estimated. In order to improve the performance, a second-order finite-time sliding mode controller is further developed to effectively alleviate chattering without any deterioration of robustness and accuracy. Moreover, an optimization control allocation algorithm, using linear programming and a pulse-width pulse-frequency (PWPF) modulator, is designed to allocate torque commands for all the aerodynamic surface deflections and on–off switching-states of RCS thrusters. Simulations are provided for the reentry vehicle considering uncertain parameters and external disturbances for practical purposes, and the results demonstrate the effectiveness and robustness of the attitude control system.

Comparison of Hydraulic Tomography with Traditional Methods at a Highly Heterogeneous Site

Over the past several decades, different groundwater modeling approaches of various complexities and data use have been developed. A recently developed approach for mapping hydraulic conductivity (K) and specific storage (Ss ) heterogeneity is hydraulic tomography, the performance of which has not been compared to other more "traditional" methods that have been utilized over the past several decades. In this study, we compare seven methods of modeling heterogeneity which are (1) kriging, (2) effective parameter models, (3) transition probability/Markov Chain geostatistics models, (4) geological models, (5) stochastic inverse models conditioned to local K data, (6) hydraulic tomography, and (7) hydraulic tomography conditioned to local K data using data collected in five boreholes at a field site on the University of Waterloo (UW) campus, in Waterloo, Ontario, Canada. The performance of each heterogeneity model is first assessed during model calibration. In particular, the correspondence between simulated and observed drawdowns is assessed using the mean absolute error norm, (L1 ), mean square error norm (L2 ), and correlation coefficient (R) as well as through scatterplots. We also assess the various models on their ability to predict drawdown data not used in the calibration effort from nine pumping tests. Results reveal that hydraulic tomography is best able to reproduce these tests in terms of the smallest discrepancy and highest correlation between simulated and observed drawdowns. However, conditioning of hydraulic tomography results with permeameter K data caused a slight deterioration in accuracy of drawdown predictions which suggests that data integration may need to be conducted carefully.

Sequential Classification by Exploring Levels of Abstraction

In the paper we describe a sequential classification scheme that iteratively explores levels of abstraction in the description of examples. These levels of abstraction represent attribute values of increasing precision. Specifically, we assume attribute values constitute an ontology (i.e., attribute value ontology) reflecting a domain-specific background knowledge, where more general values subsumes more precise ones. While there are approaches that consider levels of abstraction during learning, the novelty of our proposal consists in exploring levels of abstraction when classifying new examples. The described scheme is essential when tests that increase precision of example description are associated with costs – such a situation is often encountered in medical diagnosis. Experimental evaluation of the proposed classification scheme combined with ontological Bayes classifier (i.e., a näıve Bayes classifier expanded to handle attribute value ontologies) demonstrates that the classification accuracy obtained at higher levels of abstraction (i.e., more general description of classified examples) converges very quickly to the classification accuracy for classified examples represented precisely. This finding indicates we should be able to reduce the number of tests and thus limit their cost without deterioration of the prediction accuracy.

Hierarchical Minutiae Matching for Fingerprint and Palmprint Identification

Fingerprints and palmprints are the most common authentic biometrics for personal identification, especially for forensic security. Previous research have been proposed to speed up the searching process in fingerprint and palmprint identification systems, such as those based on classification or indexing, in which the deterioration of identification accuracy is hard to avert. In this paper, a novel hierarchical minutiae matching algorithm for fingerprint and palmprint identification systems is proposed. This method decomposes the matching step into several stages and rejects many false fingerprints or palmprints on different stages, thus it can save much time while preserving a high identification rate. Experimental results show that the proposed algorithm can save almost 50% searching time compared with traditional methods and illustrate its effectiveness.

A case of false hypoglycemia by SMBG due to improper storage of glucometer test strips

Many types of glucometer test strips deteriorate in the presence of humidity. Storing test strips outside their container or removing them from their individual wrappers could worsen their measurement accuracy. While the measurement accuracy can gradually deteriorate under such conditions, some patients are unaware of this aspect of test strip care and handling. Although such a deterioration in glucometer accuracy is often seen in clinical practice, there are few reports that warn about this issue. This is the first case report of pseudohypoglycemia caused by the inappropriate storage of test strips, which resulted in inadequate use of medications and deterioration of glycemic control. In clinical practice, patients sometimes handle glucometers in unexpected ways. This can arise due to inadequate instruction. Patients should be advised to read the manual associated with the device carefully and to store the test strips in their original container, closing the cap tightly, and to take each test strip out of its individual wrapper just before it is used. Careful patient instruction is needed to ensure safe and stable diabetic control.

A Quasi-Implicit Time-Advancing Scheme for Flow in a Three-Dimensional Curved Duct

A quasi-implicit time-marching scheme for solving unsteady incompressible three-dimensional flows on cell-centered unstructured meshes is developed. The finite-volume formulation is used for the spatial derivatives, and the flow variables at the cell face are obtained using the pressure correction. The nonlinear equations resulting from the fully implicit scheme are linearized without deterioration of the overall super-linear time accuracy. The system matrices are solved using the CG iterative method, known as the P-BiCGSTAB method for the momentum equation and the P-CG method for the pressure Poisson equation. The model is applied to simulate fully developed laminar flow in both a 90° curved 3-D circular duct and a 90° curved 3-D square duct. Steady solution is obtained in an unsteady time-marching manner. Computed results compare well with experimental data and other numerical results. It is demonstrated that the present method can be applied to unsteady incompressible laminar 3-D flow with a complex geometry on the unstructured grid system.

Forecasting Bank Leverage

Standard early warning models to predict bank failures cannot be estimated during periods of few or zero failures, precluding any updating of such models during times of good performance. Here we address this problem using an alternative approach, forecasting the simple leverage ratio (equity/assets) as a continuous variable that does not suffer from the small sample problem. Out-of-sample performance shows some promise as a supplement to the standard approach, despite measurable deterioration in prediction accuracy during the crisis years.

Evaluation of localization errors for craniospinal axis irradiation delivery using volume modulated arc therapy and proposal of a technique to minimize such errors

PurposeTo dosimetrically evaluate the effects of improper patient positioning in the junction area of a VMAT cranio-spinal axis irradiation technique consisting of one superior and one inferior arc and propose a solution to minimize these patient setup errors. MethodsFive (n=5) cranio-spinal axis irradiation patients were planned with 2 arcs: one superior and one inferior. In order to mimic patient setup errors, the plans were recalculated with the inferior isocenter shifted by: 1, 2, 5, and 10mm superiorly, and 1, 2, 5, and 10mm inferiorly. The plans were then compared with the corresponding original, non-shifted arc plans on the grounds of target metrics such as conformity number and homogeneity index, as well as several normal tissue dose descriptors. “Gradient-optimized” plans were then created for each patient in an effort to reduce dose discrepancies due to setup errors. ResultsPercent differences were calculated in order to compare each of the eight shifted plans with the original non-shifted arc plan, which corresponds to the ideal patient setup. The conformity number was on average lower by 0.9%, 2.7%, 5.8%, and 9.1% for the 1, 2, 5, and 10mm inferiorly-shifted plans and 0.4%, 0.8%, 2.8%, and 6.0% for the respective superiorly-shifted plans. The homogeneity indices were, averaged among the five patients and they indicated less homogeneous dose distributions by 0.03%, 0.3%, 1.0%, and 2.8% for the inferior shifts and 0.2%, 1.2%, 6.3%, and 15.3% for the superior shifts. Overall, the mean doses to the organs at risk deviate by less than 2% for the 1, 2, and 5mm shifted plans. The 10mm shifted plans, however, showed average percent differences, over all studied organs, from the original plan of up to 5.6%. Using “gradient-optimized” plans, the average dose differences were reduced by 0.2%, 0.5%, 1.2%, and 2.1% for 1, 2, 5, and 10mm shifts, respectively compared to the originally optimized plans, and the maximum dose differences were reduced by 11.7%, 8.5%, 12.4%, and 13.9% on average for the 1, 2, 5, and 10mm shifted plans. ConclusionsSetup errors related to isocenter shifting should be minimized in order to provide the patient with the most dosimetrically accurate treatment possible. Errors of 1–2mm can negatively affect the quality of the delivered treatment, most notably in the arc junction area, but the deterioration of the treatment plan accuracy is not as problematic as in the cases of larger errors such as 5–10mm. By employing a new planning technique, the dose differences due to setup errors can be greatly reduced.

Comparison of Twelve-Lead Electrocardiogram Using a Glove-Based Recording System With Standard Methodology

We tested the clinically relevant diagnostic accuracy of a new electrocardiographic (ECG) recording system incorporating all 10 ECG electrodes in a single-size glove worn on the patient's left arm and placed on the chest. The PhysioGlove (PG) was designed to allow fast, reproducible, electrode placement with only minimal training. The American College of Cardiology/American Heart Association ECG recording guidelines and others have repeatedly highlighted the unacceptable progressive deterioration in ECG accuracy mainly resulting from a performer's lack of proficiency and diligence, leading to frequent electrode misplacement. We studied 428 consenting adult patients with a broad spectrum of anthropomorphic characteristics and ECG and cardiovascular pathologic entities. The chest girth was measured to ascertain the single-size PG clinical diagnostic accuracy in ≥90% of this patient population. For each patient, a PG and standard-cable electrocardiogram were consecutively recorded and interpreted by experienced electrocardiographers. The study included 3 phases: phase 1, run-in (n = 120); phase 2, comparative diagnostic accuracy (n = 208); and phase 3, randomized, blinded, diagnostic accuracy (n = 100). Of the entire study population (n = 428), 92% fit the chest girth range of 85 to 118 cm (34 to 47 in.), representing the reference standard clinical diagnostic PG chest girth range. The phase 2 PG diagnostic accuracy was 91.3% for entire chest girth range and 95.7% for the 89.4% of patients with a chest girth within the reference range. The mean PG diagnostic accuracy in phase 3 was 93% (95% confidence interval 89% to 95%). In conclusion, compared with standard-cable electrocardiograms, the PG demonstrated excellent diagnostic accuracy (93% to 95.7%) in ≥90% of a typical western adult patient population. The PG's ease of use and minimal training requirements offer a promising tool to markedly improve ECG clinical diagnostic accuracy in most adult western patients.

An alternate point-wise scheme for electric field integral equations

A point-wise approach that can be used efficiently in the numerical solution of Electric Field Integral Equations is introduced. The algorithm is based on the so-called magic distance concept, which defines exactly the point-to-point equivalent of a four-dimensional integral. Magic distance values are rigorously obtained in the electrostatic case and their usefulness is demonstrated. The concept is generalized to the electrodynamic case, resulting in a family of very simple magic-distance inspired algorithms that maintain the overall accuracy of Galerkin formulations without appreciable deterioration in the overall accuracy. (c) 2013 Wiley Periodicals, Inc. Int J RF and Microwave CAE, 2013.