Nonlinear time-series modeling is fundamental to a wide variety of control and prediction problems. This letter focuses on the joint parameter time-delay estimation for an extended version nonlinear exponential autoregressive (ExpAR) model. To address difficulties posed by unknown improve accuracy, we first employ redundant rule transform ExpAR model into augmented identification Then invoke multi-innovation theory enhance data utilization propose new algorithm that combines stochastic...

Summary By using the collected batch data and iterative search, based on filtering identification idea, this article investigates proposes a filtered multi‐innovation generalized projection‐based method, gradient‐based least squares‐based method for equation‐error autoregressive systems described by models. These methods can be extended to other linear nonlinear scalar multivariable stochastic with colored noises.

Photosensitized oxygenation has been recognised as a modern method of incorporating oxygen into substrate, it offers environmentally benign alternatives to several conventional synthetic procedures. A metal-free aerobic selective sulfoxidation photosensitized by perylene diimide photocatalyst developed. The reaction utilizes visible light the driving force and molecular oxidant. advantages developed include high efficiency selectivity, extremely simple operation work-up procedure, mild...

We consider filter design of a linear system with parameter uncertainty. In contrast to the robust Kalman which focuses on worst case analysis, we propose methodology based iteratively solving tradeoff problem between nominal performance and robustness Our proposed can be computed online efficiently, is steady-state stable, less conservative than filter.

Summary This paper studies the data filtering‐based identification algorithms for an exponential autoregressive time‐series model with moving average noise. By means of filtering technique and hierarchical principle, is transformed into three sub‐identification (Sub‐ID) models, a three‐stage extended stochastic gradient algorithm derived identifying these Sub‐ID models. In order to improve parameter estimation accuracy, multi‐innovation (F‐3S‐MIESG) developed by using theory. The simulation...

We develop a novel computationally efficient and general framework for robust hypothesis testing. The new features way to construct uncertainty sets under the null alternative distributions, which are centered around empirical distribution defined via Wasserstein metric, thus our approach is data-driven free of distributional assumptions. convex safe approximation minimax formulation show that such renders nearly-optimal detector among family all possible tests. By exploiting structure least...

Lasso, or $\ell^1$ regularized least squares, has been explored extensively for its remarkable sparsity properties. It is shown in this paper that the solution to addition sparsity, robustness properties: it a robust optimization problem. This two important consequences. First, provides connection of regularizer physical property, namely, protection from noise. allows principled selection regularizer, and particular, generalizations Lasso also yield convex problems are obtained by...

The conventional image segmentation algorithm of the colorimetric sensor array is inefficient and vulnerable to interferences environment. Therefore, in order improve algorithm, an based on fuzzy C-means clustering (FCM) propo sed this study. Through information gray-scale distribution histogram, proposed divides different wave-peak regions, where pixels are relatively concentrated, into clusters determine number clusters. In addition, gray values these calculated initial cluster center....

Summary Modeling an exponential autoregressive (ExpAR) time series is the basis of solving corresponding prediction and control problems. This paper investigates hierarchical parameter estimation methods for ExpAR model. By identification principle, original nonlinear optimization problem transformed into combination a linear problem, then, we derive least squares stochastic gradient (LS‐SG) algorithm. Given difficulty determining step‐size in LS‐SG algorithm, approach proposed to obtain...

This study focuses on the recursive parameter estimation problems for non-linear exponential autoregressive model with moving average noise (the ExpARMA short). By means of gradient search, an extended stochastic (ESG) algorithm is derived. Considering difficulty determining step-size in ESG algorithm, a numerical approach proposed to obtain optimal step-size. In order improve accuracy, authors employ multi-innovation identification theory develop (MI-ESG) model. Introducing forgetting...

Summary This article concentrates on the recursive identification algorithms for exponential autoregressive model with moving average noise. Using decomposition technique, we transform original into a linear and nonlinear subidentification derive two‐stage least squares (LS) extended stochastic gradient (ESG) algorithm. In order to improve parameter estimation accuracy, employ multi‐innovation theory develop LS ESG A simulation example is provided test effectiveness of proposed algorithms.

This study employs the data filtering technique to investigate recursive identification problems for a non‐linear exponential autoregressive model with moving average noise, i.e. ExpARMA model. Whitening by linear filter, original is divided into filtered and coloured noise model, then filtering‐based extended stochastic gradient algorithm derived. In order improve parameter estimation accuracy, multi‐innovation theory used develop A simulation example given demonstrate superiority of...

This paper focuses on the recursive parameter estimation problem of exponential autoregressive (ExpAR) model. Applying Newton search and multi-innovation theory, a algorithm is presented for identifying ExpAR In order to improve computational efficiency, hierarchical identification principle employed decompose an model into two sub-models, derive algorithm. A simulation example provided demonstrate effectiveness proposed algorithms.