Suboptimal Parameter Research Articles

A deep learning-based model for defect detection in laser-powder bed fusion using in-situ thermographic monitoring

Additive manufacturing of metal components with laser-powder bed fusion is a very complex process, since powder has to be melted and cooled in each layer to produce a part. Many parameters influence the printing process; however, defects resulting from suboptimal parameter settings are usually detected after the process. To detect these defects during the printing, different process monitoring techniques such as melt pool monitoring or off-axis infrared monitoring have been proposed. In this work, we used a combination of thermographic off-axis imaging as data source and deep learning-based neural network architectures, to detect printing defects. For the network training, a k-fold cross validation and a hold-out cross validation were used. With these techniques, defects such as delamination and splatter can be recognized with an accuracy of 96.80%. In addition, the model was evaluated with computing class activation heatmaps. The architecture is very small and has low computing costs, which means that it is suitable to operate in real time even on less powerful hardware.

Recursive Approach of Sub-Optimal Excitation Signal Generation and Optimal Parameter Estimation

Optimal Input Design (OID) methodologies are developed to find a signal that could best estimate a set of parameters of a given model. Their application in constrained nonlinear systems, especially when the search space limits or the initial conditions are unknown, may present several difficulties due to the numerical instability related to the optimization processes. A good choice over the parameters possible ranges is a trade-off among numerical stability, search space size, and effectiveness, and it is hardly found. To deal with this problem, this paper proposes a series of changes in the Sub-Optimal Excitation Signal Generation and Optimal Parameter Estimation (SOESGOPE) methodology. First, the limits over the parameters are tightly adjusted according to their confidence. A recursive approach runs the optimization methodology, analyzes the solution’s feasibility and marginal costs given by the Lagrange Multipliers, and selects a direction that could improve the system’s response. This approach improves the convergence and the assertiveness of the estimation process. To validate this approach, some cases, including a parameters estimation of a mobile robot nonlinear system, are tested.

A Financial Market Model for the Netherlands: A Methodological Refinement

The Committee Parameters (Langejan et al. (2014)) advises to use the KNW-model (after Koijen et al. (2010)) to generate a representative scenario set for feasibility studies of pension funds. The scenario set enables a stochastic analysis of such feasibility studies. The underlying KNW-model is based on an affine factor model for the term structure. Stock returns, bond returns, interest rates, and inflation depend on observed factors and two latent factors. As such, the model contains relations between key financial risk factors of pension funds. CPB’s task is to estimate the model on Dutch data and, if appropriate, to calibrate some parameters in order to fit the recommendations of the Committee Parameters. Draper (2014) describes the current methods for this estimation and calibration.The calibration aims to adjust the Ultimate Forward Rate (UFR) and certain long-term expectations and covariances of the variables in the model. However, this calibration process introduces some arbitrariness. More specifically, the resulting parameter set may deviate substantially from the maximum likelihood set, even when taking the restrictions of the calibration into account. Instead of calibrating the model, we show how to impose restrictions in a continuous-time affine term structure model. In this way, the parameters correspond to the optimum of a constrained maximum likelihood estimation. The results suggest that the method in Draper (2014) provides suboptimal parameter estimates.The main result of this paper is the derivation of closed-form expressions for the long-term (unconditional) expectations, covariances, and the term structure. The expressions are required for the constrained likelihood optimization, and replace simulations for a long-run analysis of parameter sets. Our results apply to a wide range of continuous-time affine term strucure models with the Markov property, including the models in Dai and Singleton (2002) and Koijen et al. (2010).The model is outlined in Section 2. Section 3 provides expressions for the mean and covariance of possibly transformed variables in a VAR(1)-model. Section 4 presents closed-form expressions for some characteristics of the term structure in terms of the parameters. The estimation results are in Section 5. We draw conclusions in Section 6.

Terrain Analysis Enhancements to the Height Above Nearest Drainage Flood Inundation Mapping Method

AbstractFlood inundation remains challenging to map, model, and forecast because it requires detailed representations of hydrologic and hydraulic processes. Recently, Continental‐Scale Flood Inundation Mapping (CFIM), an empirical approach with fewer data demands, has been suggested. This approach uses National Water Model forecast discharge with Height Above Nearest Drainage (HAND) calculated from a digital elevation model to approximate reach‐averaged hydraulic properties, estimate a synthetic rating curve, and map near real‐time flood inundation from stage. In 2017, rapid snowmelt resulted in a record flood on the Bear River in Utah, USA. In this study, we evaluated the CFIM method over the river section where this flooding occurred. We compared modeled flood inundation with the flood inundation observed in high‐resolution Planet RapidEye satellite imagery. Differences were attributed to discrepancies between observed and forecast discharges but also notably due to shortcomings in the derivation of HAND from National Elevation Dataset as implemented in CFIM, and possibly due to suboptimal hydraulic roughness parameter. Examining these differences highlights limitations in the HAND terrain analysis methodology. We present a set of improvements developed to overcome some limitations and advance CFIM outcomes. These include conditioning the topography using high‐resolution hydrography, dispersing nodes used to subdivide the river into reaches and catchments, and using a high‐resolution digital elevation model. We also suggest an approach to obtain a reach specific Manning's n from observed inundation and validated improvements for the flood of March 2019 in the Ocheyedan River, Iowa. The methods developed have the potential to improve CFIM.

A simulation-based approach to improve decoded neurofeedback performance

The neural correlates of specific brain functions such as visual orientation tuning and individual finger movements can be revealed using multivoxel pattern analysis (MVPA) of fMRI data. Neurofeedback based on these distributed patterns of brain activity presents a unique ability for precise neuromodulation. Recent applications of this technique, known as decoded neurofeedback, have manipulated fear conditioning, visual perception, confidence judgements and facial preference. However, there has yet to be an empirical justification of the timing and data processing parameters of these experiments. Suboptimal parameter settings could impact the efficacy of neurofeedback learning and contribute to the ‘non-responder’ effect. The goal of this study was to investigate how design parameters of decoded neurofeedback experiments affect decoding accuracy and neurofeedback performance. Subjects participated in three fMRI sessions: two ‘finger localizer’ sessions to identify the fMRI patterns associated with each of the four fingers of the right hand, and one ‘finger finding’ neurofeedback session to assess neurofeedback performance. Using only the localizer data, we show that real-time decoding can be degraded by poor experiment timing or ROI selection. To set key parameters for the neurofeedback session, we used offline simulations of decoded neurofeedback using data from the localizer sessions to predict neurofeedback performance. We show that these predictions align with real neurofeedback performance at the group level and can also explain individual differences in neurofeedback success. Overall, this work demonstrates the usefulness of offline simulation to improve the success of real-time decoded neurofeedback experiments.

Nonlinear spatial-time-varying filtering and identification algorithms for correlation-extremum dynamic systems with random structure and their linearization

The correlation-extremum systems theory is extended to stochastic systems with random structure or with switching parameters and the new suboptimal (due to the nonlinear system state and measurement equations) filtering and parameter identification algorithms and their linearized form are derived, which provide adaptive features and reliable operation for the proposed combined correlation-extremum dynamic systems with random structure under environment influences, and represent new solutions of the linearization problem for the case of great estimation errors. The obtained linearized solution allows the simplification of the filter a priori performance investigation at the signal processing system design stage.

Queue management for two-user cognitive radio with delay-constrained primary user

In this paper, two novel Queue Management Policies (QMP) are proposed for Quality of Service (QoS) enhancement of a two-user Cognitive Radio Network (CRN) comprising a Primary User (PU) and Secondary User (SU), the latter having non-causal information on PU’s messages (or packets). Specifically, we aim to maximize the throughput of the SU while satisfying the delay criterion of the Primary User (PU). The first proposed QMP is a hybrid interweave/overlay scheme where all the SU’s resources are devoted to the transmission of PU’s packets. The second proposed QMP adaptively uses all or some of the SU’s resources towards the transmission of the PU’s packet, this decision being based on the packet’s delay experienced in the PU queue. For this adaptive QMP, a novel multi-regime Markov fluid queue model is proposed via which closed-form expressions are derived and validated for the exact delay distribution for Poisson PU traffic and exponentially distributed packet lengths. Using this analytical tool, we optimally tune the parameters of the adaptive QMP and we show through numerical examples that it consistently outperforms the hybrid interweave/overlay model as well as two other conventional schemes in terms of SU throughput. We also show that the performance improvement attainable by the proposed QMP depends on the intensity of PU traffic as well as the channel conditions. A heuristic suboptimal parameter tuning scheme is also proposed with lesser computational complexity.

Optimal Linear Biased Estimation Based on Generalized Contraction Mapping

Estimation methods are generalized in this paper by the idea of “scalar-vector-matrix”. A generalized contraction mapping (GCM) framework is proposed for searching the optimal linear biased estimation. First, based on the latent model and the mean square error criterion, four different biased estimation methods are analyzed. They are the improved principal component estimation (PCE), the improved principal component estimation (IPCE), the ridge estimation (RE), and the generalized ridge estimation (GRE). A suboptimal ridge parameter for the RE is given. Four estimation performance theorems for the four methods are obtained using the traditional contraction mapping (CM) framework. The theoretical results can ease the difficulty of choosing methods for application. Second, we generalize the CM framework into the generalized contraction mapping (GCM) framework, and the optimal linear biased estimation method based GCM is given theoretically by the geometric tools of rotation, contraction, and reflection. Therefore, the GCM framework further improves the estimation performance. Finally, a numerical experiment is designed to validate the correctness of the theorems in the paper.

Quantitative Analysis of Intravoxel Incoherent Motion (IVIM) Diffusion MRI using Total Variation and Huber Penalty Function.

Quantitative analysis in intravoxel incoherent motion (IVIM) imaging commonly uses voxel-wise estimation of the bi-exponential model, which might not be reliable for clinical interpretation. Improving model fitting performance and qualitative and quantitative parametric estimation, two novel methodologies are proposed here. Five IVIM analyses methodologies: (a) Bi-exponential (BE) model, (b) Segmented BE method with two-parameter fitting (BEseg-2), (c) Segmented BE method with one-parameter fitting (BEseg-1), (d) BE with adaptive Total Variation penalty function (BE+TV) and (e) BE with adaptive Huber penalty function (BE+HPF) were evaluated. Relative root-mean-square error (RRMSE), relative bias (RB) and relative parameters (Drel,Drel∗,&frel) were calculated to estimate the accuracy of methods in simulations. Empirical datasets from 14 patients with bone tumor were analyzed using these methodologies. Coefficient of variation (CV) were estimated for each IVIM parameter in tumor volume to measure the precision of the estimation methods invivo. Both BE+TV and BE+HPF showed consistently lower RRMSE (~10-42%) and lower RB (-4 to 8%) at all noise levels, compared to BE, BEseg-2 and BEseg-1 (RRMSE: ~15-120% and RB: -20 to 62%). Estimated Drel,Drel∗&frel for both BE+TV and BE+HPF methods were ~1 (0.96-1.08), whereas BE, BEseg-2 and BEseg-1 showed sub-optimal parameter estimation (0.80-1.62). For clinical data BE+TV and BE+HPF showed 30-50% improved CV in estimating D, D*, and f than BE and improved CV in estimating D* (7-23%) and f (26-30%) than BEseg-2 and BEseg-1. Bi-exponential model with penalty function showed quantitatively and qualitatively improved IVIM parameter estimation for both simulated and clinical dataset of bone tumors, thus potentially making this approach suitable for clinical applications in future.

An Improved Parameter Value Optimization Technique for the Reflectionless Transmission-Line Model of the Cochlea

The passive reflectionless transmission-line model can reproduce physiological cochlear properties. We have applied an optimization technique to obtain good parameter values that give a desired cochlear property. However, we found that, in some case, it is difficult to find the sub-optimal parameter values because of complex dependencies among parameters. In this paper, we explicitly formulate some of the cochlear properties in the reflectionless transmission-line model, and improve the optimization technique using these formulae.

Identification of effective implementations of simulated annealing for optimizing thinning schedules for single forest stands

We explored the most effective implementation of simulated annealing (SA) to optimize thinning schedules for single even-aged stands, with respect to the reliability of optimality. To enable foresters to use this implementation in a variety of situations, four yield models were set as benchmarks. We defined 32 implementations, each a different combination of variable control methods, proposal densities, cooling functions, and controlling the width of the proposal density. We tested these implementations using 625 SA parameter sets, resulting from the combination of five values each for initial and final temperature, the number of temperature levels, and proposal density width. We defined an energy function that enables us to predict potentially appropriate ranges of temperature parameters that relate to minimum acceptance rate. We ran each implementation 40 times for each yield model with each parameter set. To evaluate the reliability of SA under optimal parameters, we compared the objective values for the 40 runs using the best parameter set between implementations. To evaluate robustness under suboptimal parameter settings, we compared the variation in the mean objective values for all parameter sets between implementations. The variable control method significantly affected the reliability of SA under optimal parameters. Use of the Cauchy distribution for proposal density improved robustness under suboptimal parameter settings. The widths of proposal densities were better controlled when not taking temperature into account. The effect of cooling-function choice was relatively non-significant. We identified the implementation that appears best for both evaluations, and discussed the adjustment of its parameters.

Joint reconstruction of dynamic PET activity and kinetic parametric images using total variation constrained dictionary sparse coding

Dynamic positron emission tomography (PET) is capable of providing both spatial and temporal information of radio tracers in vivo. In this paper, we present a novel joint estimation framework to reconstruct temporal sequences of dynamic PET images and the coefficients characterizing the system impulse response function, from which the associated parametric images of the system macro parameters for tracer kinetics can be estimated. The proposed algorithm, which combines statistical data measurement and tracer kinetic models, integrates a dictionary sparse coding (DSC) into a total variational minimization based algorithm for simultaneous reconstruction of the activity distribution and parametric map from measured emission sinograms. DSC, based on the compartmental theory, provides biologically meaningful regularization, and total variation regularization is incorporated to provide edge-preserving guidance. We rely on techniques from minimization algorithms (the alternating direction method of multipliers) to first generate the estimated activity distributions with sub-optimal kinetic parameter estimates, and then recover the parametric maps given these activity estimates. These coupled iterative steps are repeated as necessary until convergence. Experiments with synthetic, Monte Carlo generated data, and real patient data have been conducted, and the results are very promising.

Cognitive Radio Algorithms Coexisting in a Network: Performance and Parameter Sensitivity

This paper studies the performance of cognitive radios in a scenario where different pairs of radios adopt different cognition/decision making approaches. We want to assess: 1) if there is a category of cognitive radio algorithms that consistently outperforms the others and 2) how sensitive different algorithms are to suboptimal parameter setting. Our approach is to take a representative set of well-known classes of cognitive radio algorithms, mix and match them throughout thousands of simulations, and determine which seem to perform better. We find that choosing a cognitive radio algorithm means finding a balance between the best-case performance obtained by optimally setting all parameters, and the behavior in uncontrolled, unknown environments, where suboptimal decisions are likely to be made. The approaches we consider, namely reinforcement learning, optimization metaheuristics, multi-armed bandit solutions, and supervised learning, greatly differ in their performance. For example, schemes that are able to achieve a high throughput in our simulation study are more sensitive to suboptimally set parameters.

Data assimilation in integrated hydrological modelling in the presence of observation bias

Abstract. The use of bias-aware Kalman filters for estimating and correcting observation bias in groundwater head observations is evaluated using both synthetic and real observations. In the synthetic test, groundwater head observations with a constant bias and unbiased stream discharge observations are assimilated in a catchment-scale integrated hydrological model with the aim of updating stream discharge and groundwater head, as well as several model parameters relating to both streamflow and groundwater modelling. The coloured noise Kalman filter (ColKF) and the separate-bias Kalman filter (SepKF) are tested and evaluated for correcting the observation biases. The study found that both methods were able to estimate most of the biases and that using any of the two bias estimation methods resulted in significant improvements over using a bias-unaware Kalman filter. While the convergence of the ColKF was significantly faster than the convergence of the SepKF, a much larger ensemble size was required as the estimation of biases would otherwise fail. Real observations of groundwater head and stream discharge were also assimilated, resulting in improved streamflow modelling in terms of an increased Nash–Sutcliffe coefficient while no clear improvement in groundwater head modelling was observed. Both the ColKF and the SepKF tended to underestimate the biases, which resulted in drifting model behaviour and sub-optimal parameter estimation, but both methods provided better state updating and parameter estimation than using a bias-unaware filter.

TuneR: a framework for tuning software engineering tools with hands-on instructions in R

Numerous tools automating various aspects of software engineering have been developed, and many of the tools are highly configurable through parameters. Understanding the parameters of advanced tools often requires deep understanding of complex algorithms. Unfortunately, suboptimal parameter settings limit the performance of tools and hinder industrial adaptation, but still few studies address the challenge of tuning software engineering tools. We present TuneR, an experiment framework that supports finding feasible parameter settings using empirical methods. The framework is accompanied by practical guidelines of how to use R to analyze the experimental outcome. As a proof-of-concept, we apply TuneR to tune ImpRec, a recommendation system for change impact analysis in a software system that has evolved for more than two decades. Compared with the output from the default setting, we report a 20.9% improvement in the response variable reflecting recommendation accuracy. Moreover, TuneR reveals insights into the interaction among parameters, as well as nonlinear effects. TuneR is easy to use, thus the framework has potential to support tuning of software engineering tools in both academia and industry. Copyright © 2016 John Wiley & Sons, Ltd.

Approaches for Monitoring the Energy Consumption with Machine Learning Methods

In times of rising energy costs and increasing customer awareness of sustainable production methods, many manufacturers take measures to reduce their energy consumption. However, after the realization of such activities the energy demand often tends to increase again due to e.g. leaks, clogged filters, defect valves or suboptimal parameter settings. In order to prevent this, it is necessary to quickly identify such increases by continuously monitoring the energy consumption and counteracting accordingly. Currently, the monitoring is either performed manually or by setting static threshold values. The manual control can be time consuming for large amounts of sensor data. By setting static threshold values only a fraction of the inefficiencies are disclosed. Another option is to use anomaly detection methods from the area of machine learning, which compare the actual sensor values with the expected ones. In this paper an overview about existing anomaly detection methods, which can be applied for this purpose, is presented.

Two security transmission realization schemes for MIMO Gaussian wiretap channel with one helper

This paper presents noncooperative and cooperative security transmission schemes for multiple-input multiple-output (MIMO) Gaussian wiretap channel with one helper for antenna configuration with arbitrary number. In these two schemes, the transmitter performs beamforming based on generalized singular value decomposition (GSVD), where an appropriate power allocation algorithm is utilized. Meanwhile, the helper sends artificial noise for higher secrecy rate. However, the transmission strategies for the artificial noise are different in the two schemes. In the first scheme, the helper adopts GSVD-based beamforming. Nevertheless, in this scheme, the impact of the artificial noise on the information signal at the receiver is not considered. To solve the problem, the helper performs space projection (SP)-based beamforming in the second scheme. In this scheme, suboptimal weighting factors are introduced to reduce the computational complexity, which can be adapted to the change of the channel quality. Theoretical analysis for the performance of the two proposed schemes is then given. Furthermore, simulation results indicate that the two presented schemes perform better than the existed schemes without helper. They also show that in the second scheme the suboptimal parameter setting is better than equal parameter setting and quite close to optimal parameter setting.

IPO: a tool for automated optimization of XCMS parameters.

BackgroundUntargeted metabolomics generates a huge amount of data. Software packages for automated data processing are crucial to successfully process these data. A variety of such software packages exist, but the outcome of data processing strongly depends on algorithm parameter settings. If they are not carefully chosen, suboptimal parameter settings can easily lead to biased results. Therefore, parameter settings also require optimization. Several parameter optimization approaches have already been proposed, but a software package for parameter optimization which is free of intricate experimental labeling steps, fast and widely applicable is still missing.ResultsWe implemented the software package IPO (‘Isotopologue Parameter Optimization’) which is fast and free of labeling steps, and applicable to data from different kinds of samples and data from different methods of liquid chromatography - high resolution mass spectrometry and data from different instruments.IPO optimizes XCMS peak picking parameters by using natural, stable 13C isotopic peaks to calculate a peak picking score. Retention time correction is optimized by minimizing relative retention time differences within peak groups. Grouping parameters are optimized by maximizing the number of peak groups that show one peak from each injection of a pooled sample. The different parameter settings are achieved by design of experiments, and the resulting scores are evaluated using response surface models. IPO was tested on three different data sets, each consisting of a training set and test set. IPO resulted in an increase of reliable groups (146% - 361%), a decrease of non-reliable groups (3% - 8%) and a decrease of the retention time deviation to one third.ConclusionsIPO was successfully applied to data derived from liquid chromatography coupled to high resolution mass spectrometry from three studies with different sample types and different chromatographic methods and devices. We were also able to show the potential of IPO to increase the reliability of metabolomics data.The source code is implemented in R, tested on Linux and Windows and it is freely available for download at https://github.com/glibiseller/IPO. The training sets and test sets can be downloaded from https://health.joanneum.at/IPO.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0562-8) contains supplementary material, which is available to authorized users.

Application of adaptive kinetic modelling for bias propagation reduction in direct 4D image reconstruction

Parametric imaging in thoracic and abdominal PET can provide additional parameters more relevant to the pathophysiology of the system under study. However, dynamic data in the body are noisy due to the limiting counting statistics leading to suboptimal kinetic parameter estimates. Direct 4D image reconstruction algorithms can potentially improve kinetic parameter precision and accuracy in dynamic PET body imaging. However, construction of a common kinetic model is not always feasible and in contrast to post-reconstruction kinetic analysis, errors in poorly modelled regions may spatially propagate to regions which are well modelled. To reduce error propagation from erroneous model fits, we implement and evaluate a new approach to direct parameter estimation by incorporating a recently proposed kinetic modelling strategy within a direct 4D image reconstruction framework. The algorithm uses a secondary more general model to allow a less constrained model fit in regions where the kinetic model does not accurately describe the underlying kinetics. A portion of the residuals then is adaptively included back into the image whilst preserving the primary model characteristics in other well modelled regions using a penalty term that trades off the models. Using fully 4D simulations based on dynamic [15O]H2O datasets, we demonstrate reduction in propagation-related bias for all kinetic parameters. Under noisy conditions, reductions in bias due to propagation are obtained at the cost of increased noise, which in turn results in increased bias and variance of the kinetic parameters. This trade-off reflects the challenge of separating the residuals arising from poor kinetic modelling fits from the residuals arising purely from noise. Nonetheless, the overall root mean square error is reduced in most regions and parameters. Using the adaptive 4D image reconstruction improved model fits can be obtained in poorly modelled regions, leading to reduced errors potentially propagating to regions of interest which the primary biologic model accurately describes. The proposed methodology, however, depends on the secondary model and choosing an optimal model on the residual space is critical in improving model fits.

Formulation of lyotropic liquid crystal containing mulberry stem extract: influences of formulation ingredients on the formation and the nanostructure

This study focused on the formulation of lamellar lyotropic liquid crystal (LLC) loaded with mulberry stem extract (MSE). The LLC formulation tested used two oils: n-dodecane or tridecyl salicylate, a co-solvent (propylene glycol) and a single (PEG-7 glyceryl cocoate) or mixed surfactant system. The mixed surfactant was PEG-7 glyceryl cocoate/PEG-40 hydrogenated castor oil/glyceryl monooleate. The LLC formation and phase behaviour were observed by polarized optical microscopy (POM) before and after MSE loading. Nanostructure determinations on these formulations following MSE loading used small angle X-ray scattering (SAXS) at 25-40°C. Lamellar LLCs are formed more easily with n-dodecane than tridecyl salicylate. Propylene glycol, in the aqueous phase (1 : 1), failed to form LLC due to suboptimal critical packing parameter (CPP) value. A single or mixed surfactant system also influenced the formation of lamellar LLC according to the chemical structure of both oils and especially the surfactants used. The four lamellar LLC formulations selected for MSE loading were PEG-7 glyceryl cocoate/tridecyl salicylate/water; mixed surfactant/tridecyl salicylate/water; PEG-7 glyceryl cocoate/n-dodecane/water and mixed surfactant/n-dodecane/water, named F1, F2, F3 and F4, respectively. MSE in F1 and F3 did not affect the lamellar structure, while MSE in F2 and F4 enlarged the lamellar structure. The SAXS data confirmed that the LLC formulations obtained were lamellar and the structure persisted with MSE. These lamellar formulations should find widespread application for MSE and perhaps other similar herbal cosmetics.