Iterative Parameter Research Articles

Hybrid spectrum conjugate gradient algorithm in electromagnetic tomography

Electromagnetic tomography is a process detection technology based upon the principles of electromagnetic induction. The forward problem model and sensitivity distribution matrix of electromagnetic tomography are introduced as the basis of the inverse problem. The search direction and iterative parameters of the conjugate gradient algorithm are modified to improve the quality and convergence of image reconstruction. A new spectral parameter conjugate gradient algorithm is described to modify the search direction, which is used to control the angle between the old and new search directions. The search direction is determined according to the iteration of each step in order to find the optimal solution. Combining the advantages of the Fletcher-Reeves and Polak-Ribiere-Polyak algorithms in the nonlinear conjugate gradient algorithm, they are mixed in a specific proportion to obtain a new hybrid conjugate gradient algorithm. In order to verify the effectiveness of the modified conjugate gradient algorithm, three physical models of electromagnetic tomography system are constructed, and the modified conjugate gradient algorithm is compared with the traditional algorithm. The experimental results show that the reconstructed image quality of the modified spectral conjugate gradient algorithm is higher and has better numerical performance. The hybrid conjugate gradient algorithm highlights the advantages of the Fletcher-Reeves and Polak-Ribiere-Polyaks algorithms. The convergence speed is faster than the Polak-Ribiere-Polyak method, and the imaging quality is higher than the other algorithms.

Fast and Stable Iterative Algorithm for Searching Wheel–Rail Contact Point Based on Geometry Constraint Equations

Abstract This paper proposes a fast and stable iterative algorithm for wheel–rail contact geometry based on constraint equations, which can be implemented in dynamic wear simulations that real-time profile updating is needed. Further, critical factors that determine convergence and iteration stability are analyzed. A B-spline is adopted for wheel–rail profile modeling because it does not contribute to changes in the global shape of curves. It is found that the smoothness of the first and second derivative curves significantly affects the numerical stability of the Jacobian matrix, which determines the increments in iterations. Moreover, a damped Newton's iteration formula with a scaling factor of 0.5 is proposed considering the convergence rate and out-of-bound issues for the updated step. The influence of the initial iteration parameters on the convergence is studied using Newton fractals. The range within ±3 mm, centered on the target contact point, is found to be an unconditionally stable domain. The proposed method could achieve convergence within 10 and 30 steps under thread and flange contact conditions, respectively.

Predictive compensation based quantization iterative learning control for nonlinear nonaffine discrete‐time systems

AbstractIn this work, the problems of predictive compensation, unknown nonlinearity, and nonaffine structure are considered simultaneously for a quantized iterative learning control (QILC) design and analysis under a data‐driven framework. The compensation strategy can avoid deteriorated data transmission quality owing to limited channel capacities. First, a dynamic linearization methodology is employed to transform the nonlinear plant into a virtual iterative linear data model (iLDM) which includes all the input signals over a time‐window from the initial time instant to the current one. The iLDM is also used as a predictive model to estimate the unavailable information caused by the encoding–decoding mechanism. Then, a predictive compensation‐based QILC is proposed by optimizing quadratic functions, which includes an output prediction mechanism, a quantized iterative learning updating law, a quantized iterative parameter estimation law, and a resetting algorithm. The result is also extended to a class of MIMO nonlinear nonaffine discrete‐time systems. The developed control laws are data‐driven and independent of any system information. The theoretical results are proved by the use of contraction mapping principle and induction method. Examples are provided to verify the effectiveness of the proposed methods.

Speech Signal Analysis With a Refined Iterative Adaptive Method

A new speech signal analysis method referred to as a refined iterative adaptive method (RIAM) is introduced in this paper. Based on time-varying adaptive sinusoidal modeling, the RIAM method tries to determine in an iterative adaptive manner the instantaneous components of time-varying quasi-periodic multi-component signals such as voiced speech. The proposed method can adjust the current analysis parameters to the time-varying characteristics of the speech signal. This is done using a refined iterative sinusoidal parameter estimation algorithm based on a frequency correction mechanism combined with an adaptive scheme. The experiments on voiced speech demonstrate that the proposed RIAM algorithm outperforms some well-known state-of-the-art approaches. The RIAM algorithm provides a higher signal to reconstruction error ratio (SRER) of 42.267 dB with an improvement of 19.752 dB, 5.054 dB, and 2.552 dB compared to the conventional sinusoidal model (SM), adaptive harmonic model (aHM), and extended adaptive quasi harmonic model (eaQHM), respectively.

The focused beam X-ray fluorescence elemental quantification software package PIQUANT

Adoption of a robust X-ray fluorescence spectroscopy quantification and spectrum fitting routine calls for careful consideration on the inner workings and databases incorporated in its architecture. For analysis of micro-XRF data returned from the Planetary Instrument for X-ray Lithochemistry (PIXL), integrated on the Mars 2020 Perseverance rover, the University of Washington and the National Aeronautics and Space Administration (NASA) have invested in the production of an in-house micro-XRF software package, PIQUANT, capable of supporting quantitative elemental analysis of whole rock and geological materials. The PIQUANT software uses an iterative fundamental parameters physics-based model to convert X-ray peak intensity into elemental concentration, and has minimal reliance on calibration using standards. It also incorporates polycapillary optic transmission correction to account for photon passage in the X-ray optic of micro-XRF systems. This work introduces the key features of PIQUANT's architecture, its databases, models, assumptions and summarizes features available as part of the analysis products it generates. A working example of a quantification process available with this software is presented within.

Aircraft conceptual design using metaheuristic-based reliability optimisation

The reliability optimisation methodology is developed to solve a conceptual design problem of a fixed-wing Unmanned Aerial Vehicle (UAV). The reliability quantification is based on the most probable point (MPP) concept, leading to a double-loop optimisation problem. The design problem is formulated as an outer-loop multiobjective optimisation for the main design problem and an inner-loop optimisation for estimating a reliability index (β) of a design solution. The goal of the outer-loop optimisation is to minimise the aircraft take-off weight, and simultaneously maximise β. The aerodynamic and stability properties of an aircraft solution are calculated by a vortex lattice method (VLM), while various types of empirical weight methods are obtained. The design problem is set up to have uncertainties in calculating aircraft empty weight and aerodynamic coefficients and derivatives. For the inner loop optimisation, the MPP is used for approximating the reliability index and probability of failure (pf). Multiobjective Meta-heuristic with Iterative Parameter Distribution Estimation (MMIPDE) and Success-History based Adaptive Differential Evolution (SHADE) are used for solving the outer- and inner-loop optimisations respectively. Four parameters setting strategies for running metaheuristics are proposed for use with the proposed metaheuristic-based reliability optimisation. The comparative results reveal that the best dynamic parameter setting from this study can reduce runtime by 22.5% compared to the traditional metaheuristic run while maintaining competitive results.

Quantitative SPECT/CT imaging of lead-212: a phantom study

BackgroundLead-212 (212Pb) is a promising radionuclide for targeted therapy, as it decays to α-particle emitter bismuth-212 (212Bi) via β-particle emission. This extends the problematic short half-life of 212Bi. In preparation for upcoming clinical trials with 212Pb, the feasibility of quantitative single photon-emission computed tomography/computed tomography (SPECT/CT) imaging of 212Pb was studied, with the purpose to explore the possibility of individualised patient dosimetric estimation.ResultsBoth acquisition parameters (combining two different energy windows and two different collimators) and iterative reconstruction parameters (varying the iterations x subsets between 10 × 1, 15 × 1, 30 × 1, 30 × 2, 30 × 3, 30 × 4, and 30 × 30) were investigated to evaluate visual quality and quantitative uncertainties based on phantom images. Calibration factors were determined using a homogeneous phantom and were stable when the total activity imaged exceeded 1 MBq for all the imaging protocols studied, but they increased sharply as the activity decayed below 1 MBq. Both a 20% window centred on 239 keV and a 40% window on 79 keV, with dual scatter windows of 5% and 20%, respectively, could be used. Visual quality at the lowest activity concentrations was improved with the High Energy collimator and the 79 keV energy window. Fractional uncertainty in the activity quantitation, including uncertainties from calibration factors and small volume effects, in spheres of 2.6 ml in the NEMA phantom was 16–21% for all protocols with the 30 × 4 filtered reconstruction except the High Energy collimator with the 239 keV energy window. Quantitative analysis was possible both with and without filters, but the visual quality of the images improved with a filter.ConclusionsOnly minor differences were observed between the imaging protocols which were all determined suitable for quantitative imaging of 212Pb. As uncertainties generally decreased with increasing iterative updates in the reconstruction and recovery curves did not converge with few iterations, a high number of reconstruction updates are recommended for quantitative imaging.

A semiparametric Bayesian method with birth-death Markov Chain Monte Carlo algorithm for extreme mooring tension analysis

A series of 1: 50 model tests have been conducted to study the dynamics of a taut mooring system and two hybrid mooring systems attached to a deep-water semi-submersible in the 100-year wave condition of the South China Sea. Three typical environmental conditions were tested, including head sea, beam sea and quartering sea. The mixture Gamma- Generalized Pareto distribution (GPD) model is applied to study short term extreme dynamic mooring tensions based on the measured data, the results are compared with the estimations from mixture Gamma and two-parameter Weibull models. A birth-death Markov chain Monte Carlo (MCMC) sampling procedure is developed to fit the parameters of a mixture Gamma- GPD via Bayesian inference, and an extensive discussion of the de-clustering procedure, parameters of the prior distribution for threshold and MCMC iteration steps on Bayesian inference are carried out. It is found that the Weibull model underestimates the most probable largest dynamic mooring tensions. On the contrary, the mixture Gamma model overestimates the extreme dynamic mooring tension significantly. The mixture Gamma- GPD can present good estimations of extreme dynamic mooring tension, and its performance is robust to the sample definition, MCMC iteration steps and parameters of threshold prior distribution.

Lithium-ion cell equivalent circuit model identification by galvanostatic intermittent titration technique

When modelling lithium-ion batteries thorough identification of parameters across the entire operation domain is necessary to capture non-linear variations of properties caused by temperature or state of charge. This work presents a parameter identification method using galvanostatic intermittent titration technique (GITT) to create high resolution look-up tables and response surfaces for equivalent circuit models (ECM). Significant improvements are proposed over other parameter estimation method, such as HPPC. These improvements are open-circuit voltage change compensation, and ohmic resistance correction, which yield to better overall accuracy of the model. An iterative parameter identification algorithm (IPIA) is introduced to increase the robustness of the non-linear least square curve fitting for higher orders ECM. The method is applied to identify the parameters of 1RC and a 2RC ECM. The use of GITT in conjunction with IPIA has allowed greater fidelity of model response surfaces than previously published pulse identification methods. Experiments were conducted on Nickel-Manganese-Cobalt cathode lithium-ion cells. The methodology presented in this paper is intended to be applicable to any lithium-ion battery format or chemistry with minor adjustments.

Success history based adaptive multi-objective differential evolution variants with an interval scheme for solving simultaneous topology, shape and sizing truss reliability optimisation

A general approach based on the most probable point (MPP) method for solving reliability truss optimisation with simultaneous topology, shape and sizing (TSS) design variables is developed. The design problems are solved using double loop optimisation where the inner loop is for reliability index and the probability of failure approximation is solved by Harris Hawk Optimisation (HHO). The outer loop, the main TSS truss optimisation loop, is solved by two newly developed algorithms, namely interval success history based adaptive multi-objective differential evolution (iSHAMODE) and its hybrid variant with the whale optimisation algorithm (iSHAMODE-WO). Six TSS truss optimisation problems are evaluated. The results from the proposed method for reliability approximation and First Order Second Moment (FOSM) are compared and validated with Monte Carlo Simulation (MCS). The proposed method shows more consistent and accurate results compared to FOSM. Furthermore, the efficiency of the proposed optimisation algorithms (iSHAMODE and iSHAMODE-WO) is proved; they can outperform their predecessors and several state-of-the-art algorithms including multi-objective multi-verse optimisation algorithm (MOMVO), multi-objective grasshopper optimisation algorithm (MOGOA), multi-objective dragonfly optimisation (MODA), multi-objective salp swarm algorithm (MSSA), hybridisation of real-code population-based incremental learning and differential evolution (RPBILDE), and multi-objective meta-heuristic with iterative parameter distribution estimation (MMIPDE).

Joint parameter estimation and decoding in a distributed receiver

SummaryThis paper presents an algorithm for iterative joint channel parameter (carrier phase, Doppler shift, and Doppler rate) estimation and decoding of transmission over channels affected by Doppler shift and Doppler rate using a distributed receiver. This algorithm is derived by applying the sum‐product algorithm (SPA) to a factor graph representing the joint a posteriori distribution of the information symbols and channel parameters given the channel output. In this paper, we present two methods for dealing with intractable messages of the SPA. In the first approach, we use particle filtering with sequential importance sampling for the estimation of the unknown parameters. We also propose a method for fine‐tuning of particles for improved convergence. In the second approach, we approximate our model with a random walk phase model, followed by a phase tracking algorithm and polynomial regression algorithm to estimate the unknown parameters. We derive the Weighted Bayesian Cramer‐Rao Bounds for joint carrier phase, Doppler shift, and Doppler rate estimation, which take into account the prior distribution of the estimation parameters and are accurate lower bounds for all considered signal‐to‐noise ratio values. Numerical results (of bit error rate and the mean‐square error of parameter estimation) suggest that phase tracking with the random walk model slightly outperforms particle filtering. However, particle filtering has a lower computational cost than the random walk model‐based method.

Assessing the robustness and scalability of the accelerated pseudo-transient method

Abstract. The development of highly efficient, robust and scalable numerical algorithms lags behind the rapid increase in massive parallelism of modern hardware. We address this challenge with the accelerated pseudo-transient (PT) iterative method and present a physically motivated derivation. We analytically determine optimal iteration parameters for a variety of basic physical processes and confirm the validity of theoretical predictions with numerical experiments. We provide an efficient numerical implementation of PT solvers on graphical processing units (GPUs) using the Julia language. We achieve a parallel efficiency of more than 96 % on 2197 GPUs in distributed-memory parallelisation weak-scaling benchmarks. The 2197 GPUs allow for unprecedented tera-scale solutions of 3D variable viscosity Stokes flow on 49953 grid cells involving over 1.2 trillion degrees of freedom (DoFs). We verify the robustness of the method by handling contrasts up to 9 orders of magnitude in material parameters such as viscosity and arbitrary distribution of viscous inclusions for different flow configurations. Moreover, we show that this method is well suited to tackle strongly nonlinear problems such as shear-banding in a visco-elasto-plastic medium. A GPU-based implementation can outperform direct-iterative solvers based on central processing units (CPUs) in terms of wall time, even at relatively low spatial resolution. We additionally motivate the accessibility of the method by its conciseness, flexibility, physically motivated derivation and ease of implementation. This solution strategy thus has a great potential for future high-performance computing (HPC) applications, and for paving the road to exascale in the geosciences and beyond.

An unconditionally convergent RSCSCS iteration method for Riesz space fractional diffusion equations with variable coefficients

In this paper, a respectively scaled circulant and skew-circulant splitting (RSCSCS) iteration method is employed to solve the Toeplitz-like linear systems arising from time-dependent Riesz space fractional diffusion equations with variable coefficients. The RSCSCS iteration method is shown to be convergent unconditionally by a novel technique, and only requires computational costs of O(NlogN) with N denoting the number of interior mesh points in space. In theory, we obtain an upper bound for the convergence factor of the RSCSCS iteration method and discuss the optimal value of its iteration parameter that minimizes the corresponding upper bound. Meanwhile, we also design a fast induced RSCSCS preconditioner to accelerate the convergence rate of the Krylov subspace iteration method likes generalized minimal residual (GMRES) method. Numerical results are presented to show that the efficiencies of our proposed RSCSCS iteration method and the preconditioned GMRES method with the RSCSCS preconditioner.

A modified variant of HSS preconditioner for generalized saddle point problems

Recently, Zhang [Numerical Linear Algebra with Applications, 2018: e2166] constructed an efficient variant of Hermitian and skew-Hermitian splitting (HSS) preconditioner for generalized saddle point problems, and gave the corresponding theoretical analysis and numerical experiment. In this paper, based on the efficient variant of the HSS (EVHSS) preconditioner, we relax the strong condition on the iterative parameter [Formula: see text], generalize the algorithms and further present the modified variant of Hermitian and skew-Hermitian splitting (HSS) (MVHSS) preconditioner for solving the generalized saddle point problem. Moreover, by similar theoretical analysis, we obtain that the MVHSS iterative method is convergent under weaker conditions. Finally, an example is given to verify the effectiveness of the proposed method. MSC: 65F10; 65F15; 65F50

NIMBLE

Emerging Metaverse applications demand reliable, accurate, and photorealistic reproductions of human hands to perform sophisticated operations as if in the physical world. While real human hand represents one of the most intricate coordination between bones, muscle, tendon, and skin, state-of-the-art techniques unanimously focus on modeling only the skeleton of the hand. In this paper, we present NIMBLE, a novel parametric hand model that includes the missing key components, bringing 3D hand model to a new level of realism. We first annotate muscles, bones and skins on the recent Magnetic Resonance Imaging hand (MRI-Hand) dataset [Li et al. 2021] and then register a volumetric template hand onto individual poses and subjects within the dataset. NIMBLE consists of 20 bones as triangular meshes, 7 muscle groups as tetrahedral meshes, and a skin mesh. Via iterative shape registration and parameter learning, it further produces shape blend shapes, pose blend shapes, and a joint regressor. We demonstrate applying NIMBLE to modeling, rendering, and visual inference tasks. By enforcing the inner bones and muscles to match anatomic and kinematic rules, NIMBLE can animate 3D hands to new poses at unprecedented realism. To model the appearance of skin, we further construct a photometric HandStage to acquire high-quality textures and normal maps to model wrinkles and palm print. Finally, NIMBLE also benefits learning-based hand pose and shape estimation by either synthesizing rich data or acting directly as a differentiable layer in the inference network.

A generalized two-sweep shift splitting method for non-Hermitian positive definite linear systems

In this paper, based on the shift splitting of the coefficient matrix, a generalized two-sweep shift splitting (GTSS) method is introduced to solve the non-Hermitian positive definite linear systems. Theoretical analysis shows that the GTSS method is convergent to the unique solution of the linear systems under a loose restriction on the iteration parameter. Numerical experiments are reported to the efficiency of the GTSS method.

Correlation Analysis of Network Big Data and Film Time-Series Data Based on Machine Learning Algorithm

To expand the application of machine learning in movie data, in order to explore the correlation between network big data and film time-series data, based on the machine learning algorithm, the correlation and multifractal characteristics of happiness index (HI) and film box office (BO) were studied and described by introducing multifractal crossover method. On this basis, some indicators are introduced to optimize the neural network model so that the optimization model can describe and predict the box office and other related information well. The results show that the critical values of the happiness index and box office show a linear change trend with the increase of freedom, and the corresponding change curves of the happiness index and box office show obvious nonlinear characteristics, which can be divided into slow increase stage, steady increase stage, and approximately gentle stage. With the increase of iteration parameter q value, the change trend of the long-term and short-term curves of the generalized Hurst function is basically the same, and the difference between the two is getting smaller and smaller, while the difference between the two curves is getting bigger and bigger with the increase of q value of Renyi function. The changing trend of the dynamic Hurst index in the sliding window period all shows that it first rises rapidly to a certain value, then fluctuates rapidly with the increase of time, then drops rapidly to a constant value, and finally continues to show repeated small range fluctuation. Under the influence of time-series parameter α, the original sequence changes the most, the replacement sequence changes the medium, and the corresponding rearrangement sequence changes the least. The overall distribution of box office prediction data conforms to the characteristics of linear variation. The prediction index of the optimized HI-LSTM (Happiness Index-Long term short term memory neural network) model is higher in the box office, indicating that the model has better performance in describing and predicting the box office. This study can provide a theoretical basis for the correlation study of network big data and film data.

An Argument in Favor of Deep Brain Stimulation for Uncommon Movement Disorders: The Case for N-of-1 Trials in Holmes Tremor.

Deep brain stimulation (DBS) is part of state-of-the-art treatment for medically refractory Parkinson’s disease, essential tremor or primary dystonia. However, there are multiple movement disorders that present after a static brain lesion and that are frequently refractory to medical treatment. Using Holmes tremor (HT) as an example, we discuss the effectiveness of currently available treatments and, performing simulations using a Markov Chain approach, propose that DBS with iterative parameter optimization is expected to be more effective than an approach based on sequential trials of pharmacological agents. Since, in DBS studies for HT, the thalamus is a frequently chosen target, using data from previous studies of lesion connectivity mapping in HT, we compared the connectivity of thalamic and non-thalamic targets with a proxy of the HT network, and found a significantly higher connectivity of thalamic DBS targets in HT. The understanding of brain networks provided by analysis of functional connectivity may thus provide an informed framework for proper surgical targeting of individual patients. Based on these findings, we argue that there is an ethical imperative to at least consider surgical options in patients with uncommon movement disorders, while simultaneously providing consistent information regarding the expected effectiveness and risks, even in a scenario of surgical-risk aversion. An approach based on n-of-1 DBS trials may ultimately significantly improve outcomes while informing on optimal therapeutic targets and parameter settings for HT and other disabling and rare movement disorders.

LPLB: An approach for the design of a lightweight convolutional neural network

SUMMARYConvolutional neural network (CNN) is one of the widely used deep neural network architecture for data analytics in the Internet of Things (IoT). However, due to its severe resource requirements, deploying CNN on resource‐constrained edge devices is quite challenging. Moreover, IoT services demand fast data analytics in order to be useful in their context. Hence, ensuring the deployment of CNN models on IoT edge devices is crucial. To this purpose, this article proposes a framework LPLB (less parameters and less bits) for the design of a lightweight CNN with reduced number of parameters and lesser storage requirements while preserving the model accuracy. LPLB consists of three steps: CNN training, iterative parameter drop and post‐training quantization. The performance of LPLB is evaluated on three datasets: German traffic sign recognition dataset, Kaggle hand gesture recognition dataset, and CIFAR‐10 dataset. Experiments conducted in the study reveal that the proposed framework is able to reduce the number of parameters by 11× in case of German traffic sign recognition dataset, 12.5× in case of Kaggle hand gesture recognition dataset, and 14× in case of CIFAR‐10 dataset. Moreover, the storage requirements are reduced by 44.42× in case of German traffic sign recognition dataset, 50× in case of Kaggle hand gesture recognition dataset, and 56.7× in case of CIFAR‐10 dataset. This is achieved without a significant drop in the accuracy. The accuracy drop in case of German traffic sign recognition dataset is 0.215%, 0.18% in case of Kaggle hand gesture recognition dataset, and 0.148% in case of CIFAR‐10 dataset. The proposed framework is generic and can be used for the design of lightweight CNN models corresponding to other use‐cases as well. This will make real‐time classification/recognition possible in delay‐sensitive applications like self‐driving cars, advanced driver assistance systems, health monitoring, elderly posture recognition and so forth.

Variational data assimilation for a sea dynamics model

Abstract The 4D variational data assimilation technique is presented for modelling the sea dynamics problems, developed at the Marchuk Institute of Numerical Mathematics of the Russian Academy of Sciences (INM RAS). The approach is based on the splitting method for the mathematical model of sea dynamics and the minimization of cost functionals related to the observation data by solving an optimality system that involves the adjoint equations and observation and background error covariances. Efficient algorithms for solving the variational data assimilation problems are presented based on iterative processes with a special choice of iterative parameters. The technique is illustrated for the Black Sea dynamics model with variational data assimilation to restore the sea surface heat fluxes.