Article
Engineering, Multidisciplinary
Jianmin Wang, Jianjun Zhao, Zhenghe Liu, Zhijun Kang
Summary: A method is proposed to eliminate the influence of multiple outliers on parameters by constructing a correction model that considers outliers, developing a full search algorithm to estimate parameters and outliers simultaneously, and successfully eliminating the influence of outliers.
Article
Computer Science, Artificial Intelligence
Qing-Yan Chen, Da-Zheng Feng, Wei -Xing Zheng, Xiang -Wei Feng
Summary: Point set registration (PSR) is a competitive technique that captures the overall structure between two point-set patterns. It is divided into two sub-problems: searching for point set correspondence (PSC) and estimating spatial transformation matrix (STM). To address errors and outliers, an efficient PSR algorithm with dual terms based on total least squares (DT-TLS) is proposed, along with a soft decision-making framework and TLS-based criterion. Experimental results demonstrate that the proposed algorithm outperforms state-of-the-art algorithms in multi-view computer vision tasks.
PATTERN RECOGNITION
(2023)
Article
Engineering, Electrical & Electronic
Shaohui Lv, Haiquan Zhao, Lijun Zhou
Summary: A more flexible algorithm called Maximum Mixture Total Correntropy (MMTC) is proposed to address the limitation of the Maximum Total Correntropy (MTC) algorithm by combining Mixture Correntropy (MC) and Total Least Square (TLS) methods. Numerical simulations demonstrate that MMTC outperforms MTC in parameter estimation of the error-in-variable (EIV) model under impulse noise background. The theoretical analysis is also validated by the simulation results.
Article
Biochemical Research Methods
Ruoyu Tang, Xinyu He, Ruiqi Wang
Summary: The study presents a general computational method for constructing maps between different cell fates and parametric conditions by systematic perturbations. The method does not require accurate parameter measurements or bifurcations. The maps obtained can help in understanding how systematic perturbations drive cell fate decisions and transitions, providing valuable information for predicting and controlling cell states.
Article
Automation & Control Systems
Kaio D. T. Rocha, Marco H. Terra
Summary: This paper proposes a robust Kalman filter for uncertain linear discrete-time systems, formulating a robust regularized least-squares estimation problem using the penalty function method. It is shown that under reasonable conditions, the robust filter is stable and guarantees a bounded error variance for quadratically stable systems.
SYSTEMS & CONTROL LETTERS
(2021)
Article
Automation & Control Systems
Shujun Fan, Ling Xu, Feng Ding, Ahmed Alsaedi, Tasawar Hayat
Summary: This paper addresses the identification problem of discrete-time linear time-invariant errors-in-variables systems with colored output noise. It introduces the multi-innovation theory and proposes correlation analysis-based algorithms to improve parameter accuracy. Simulation results validate the effectiveness of these algorithms.
INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
(2021)
Article
Mathematics, Applied
Andrea Carracedo Rodriguez, Linus Balicki, Serkan Gugercin
Summary: The AAA algorithm is a popular tool for data-driven rational approximation of multivariate functions. It is a data-driven method that only requires function evaluations and does not need access to the full state-space model. The method can also be extended to multi-input/multi-output dynamical systems and has a connection to tangential interpolation theory.
SIAM JOURNAL ON SCIENTIFIC COMPUTING
(2023)
Article
Statistics & Probability
Jun Zhang, Leyi Cui
Summary: This paper considers nonlinear regression models with exponential parametric distortion measurement errors when neither the response variable nor the covariates can be directly observed. Nonlinear least squares and weighted nonlinear least squares estimation methods are proposed to estimate the parameters in the distortion functions under two identifiability conditions. The asymptotic results of estimators are studied, especially the difference between parametric calibrations and nonparametric calibrations. Simulation studies are conducted to demonstrate the performance of the proposed estimators.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2022)
Article
Automation & Control Systems
Lu Lu, Kai Zhou, Guangya Zhu, Xiaomin Yang, Badong Chen
Summary: Partial discharge (PD) location techniques are useful for monitoring the condition of electrical apparatus in power systems. This article proposes an adaptive filtering technique called the total least-squares (TLS)-Matern kernel (TLS-MK) algorithm to address the PD location problem. The TLS-MK algorithm effectively suppresses noise from direct and reflected waves of the PD source, and estimates the time difference for PD location. Simulations and experiments demonstrate that the proposed algorithm improves location accuracy compared to state-of-the-art methods.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Geochemistry & Geophysics
Yu Hu, Xing Fang, Wenxian Zeng, Hansjoerg Kutterer
Summary: The modern GNSS technique is an effective tool for observing crustal motions and plate tectonics dynamics. The multiframe transformation method is proposed to connect the time-varying GNSS coordinates, ensuring unique and consistent results. The introduction of variance component (VC) allows flexible indicator for land movement analysis.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Mathematics, Interdisciplinary Applications
Dmitriy Ivanov
Summary: This study proposes a new method for identifying the parameters of an induction motor with a skin effect. Parameter estimates were determined based on generalized total least squares. Simulation results showed high accuracy of the obtained estimates. The results of this research can be applied in the development of predictive diagnostic systems.
FRACTAL AND FRACTIONAL
(2023)
Article
Engineering, Electrical & Electronic
Gang Wang, Jingci Qiao, Rui Xue, Bei Peng
Summary: This paper investigates the kernel recursive least squares (KRLS) algorithm in the quaternion domain using the generalized Hamilton-real calculus method. The study shows the feasibility and performance of the proposed algorithm by first examining the quaternion recursive least squares (QRLS) algorithm and then generalizing it to the quaternion KRLS algorithm, with theoretical analysis and simulations demonstrating convergence and accuracy.
Article
Automation & Control Systems
Saurabh Pandey, Tao Liu, Qing-Guo Wang
Summary: In this paper, a bias-eliminated output error model identification method is proposed for single-input-single-output sampled systems with integer-type time delay, aiming at effectively estimating rational model parameters and load disturbance response, considering the impact of stochastic noise, accelerating the convergence rate of model parameter estimation, and proposing a method for evaluating the delay parameter.
IET CONTROL THEORY AND APPLICATIONS
(2021)
Article
Chemistry, Analytical
Xingxing Ai, Jiayi Zhao, Hongtao Zhang, Yong Sun
Summary: The paper proposes a new sparse sliding-window KRLS algorithm for online prediction of nonlinear time-varying channel state information (CSI) in MIMO systems. The algorithm maintains a small kernel dictionary size by selecting and discarding samples, achieving better predictive accuracy than the traditional algorithm.
Article
Automation & Control Systems
Jie Hou, Hao Su, Chengpu Yu, Fengwei Chen, Penghua Li, Haofei Xie, Taifu Li
Summary: In this article, a consistent subspace identification method (SIM) is proposed for block-oriented errors-in-variables Hammerstein systems. The existing SIMs using parity subspace based on noisy measurements may result in biased parameter estimates. We propose a scheme for consistent system parameter estimation, which estimates the noise-free Hankel matrix using available noisy measurements and noise variances. The effectiveness and merits of the proposed method are supported by two simulation examples.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Fen Zhao, Yinguo Li, Jie Hou, Ling Bai
Summary: The proposed entity importance estimation network using attention-based graph embedding addresses the issue of relation prediction in large-scale knowledge graphs, achieving a high F1 score in cases of missing relations. This approach effectively resolves the problems of missing relations and low precision in knowledge graphs question answering.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Thermodynamics
Jie Hou, Jiawei Liu, Fengwei Chen, Penghua Li, Tao Zhang, Jincheng Jiang, Xiaolei Chen
Summary: This paper proposes a model and data uncertainties-robust method based on an enhanced adaptive unscented Kalman filter (AUKF) for simultaneous estimation of model parameters and battery state of charge (SOC). The method integrates all unknown variables using an extended state observer and obtains uncertain model and data statistics through a covariance matching technique with adaptive forgetting factor. The proposed method is more robust to model uncertainties and data uncertainties compared to conventional SOC estimation methods.
Article
Automation & Control Systems
Jie Hou, Hao Su, Chengpu Yu, Fengwei Chen, Penghua Li, Haofei Xie, Taifu Li
Summary: In this article, a consistent subspace identification method (SIM) is proposed for block-oriented errors-in-variables Hammerstein systems. The existing SIMs using parity subspace based on noisy measurements may result in biased parameter estimates. We propose a scheme for consistent system parameter estimation, which estimates the noise-free Hankel matrix using available noisy measurements and noise variances. The effectiveness and merits of the proposed method are supported by two simulation examples.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Zhifan Li, Qijun Deng, Fengwei Chen, Pan Sun, Jiangtao Liu, Wenshan Hu, Hong Zhou
Summary: This article proposes a method of input design for wireless power transfer (WPT) systems based on receding horizon D-optimization. The system's dynamic behavior is characterized by a discrete-time, single-input, single-output (SISO) model, and the model parameters are estimated using a recursive least-squares (RLS) method. The excitation signal of the system is designed based on receding horizon D-optimization to improve the quality of parameter estimates. The sequential quadratic programming (SQP) method is employed to solve the receding horizon D-optimization for improving convergence and efficiency of the input design process. Numerical and experimental results demonstrate the effectiveness of the proposed method.
IEEE JOURNAL OF EMERGING AND SELECTED TOPICS IN POWER ELECTRONICS
(2023)
Article
Engineering, Electrical & Electronic
Zhe Liu, Hongsheng Hu, Yu-Gang Su, Yue Sun, Fengwei Chen, Pengqi Deng
Summary: This article presents a double-receiver single capacitance coupled wireless power transfer system with a three-plate compact coupler to achieve free-position charging and superior system performance. The system consists of two independent receivers: one with a LCLC-S topology for constant-voltage output, and the other with a LCLC-M topology for constant-current output, without affecting each other. By circuit analysis and a simplified equivalent circuit of the three-plate coupler, the CV and CC characteristics are theoretically analyzed. The effectiveness of the proposed system is verified through prototype testing, showing stable output power for each receiver at any position within the transmitting plate.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2023)
Article
Automation & Control Systems
Jie Hou, Zhen Yang, Taifu Li, Huiming Wang, Jincheng Jiang, Xiaolei Chen
Summary: A full-parameter constrained parsimonious subspace identification method is proposed to model the DC-DC converters, incorporating the steady-state a priori information. Compared with traditional data-driven methods, the subspace-based method can estimate both model structure and parameters simultaneously with appropriate computational complexity. Moreover, the proposed algorithm can accurately estimate the system parameters with a smaller variance compared to the traditional full-parameter constrained subspace approach. Experimental results on a DC-DC synchronous buck converter verify the effectiveness and superiority of the proposed method.
CONTROL THEORY AND TECHNOLOGY
(2023)
Article
Engineering, Electrical & Electronic
Fengwei Chen, Hongsheng Hu, Lei Zhao, Arturo Padilla, Jie Hou
Summary: In this paper, a linear parameter-varying (LPV) Hammerstein model is proposed to better handle the nonlinearities in wireless power transfer systems. A method to identify the model structure and estimate the parameters is also presented. Compared with other data-driven models, the LPV Hammerstein model is able to accurately describe the system dynamics and has the advantages of simplicity and accuracy.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2023)
Article
Computer Science, Information Systems
Jingwen Hou, Weisi Lin, Guanghui Yue, Weide Liu, Baoquan Zhao
Summary: Personalized image aesthetics assessment aims to estimate aesthetic experiences based on individual preferences. This research proposes a method that directly estimates personalized aesthetic experiences from the interaction between image contents and user preferences, without the need for prior knowledge on generic aesthetics assessment. Extensive experiments show that the proposed method outperforms previous personalized methods and generic methods in terms of both personalized and generic aesthetics assessment.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Proceedings Paper
Automation & Control Systems
Hao Su, Jie Hou
Summary: This paper proposes a Bias-Correction Least Squares (BCLS) algorithm based on the least squares method and consistency for estimating parameters of Errors-In-Variables (EIV) Hammerstein-Wiener systems using available noisy measurements. The main contribution of this study is to derive the estimation bias of the least squares method and recursively represent the monomial of noiseless measurements as available measurements. The effectiveness of the algorithm is demonstrated through simulation examples.
2022 41ST CHINESE CONTROL CONFERENCE (CCC)
(2022)