The Experts below are selected from a list of 270 Experts worldwide ranked by ideXlab platform
Xinjiang Lu - One of the best experts on this subject based on the ideXlab platform.
-
Probabilistic Inference-Based Least Squares Support Vector Machine for Modeling under Noisy Environment
IEEE Transactions on Systems Man and Cybernetics: Systems, 2016Co-Authors: Bi Fan, Xinjiang Lu, Han-xiong LiAbstract:—The least squares support vector machine (LS-SVM) has emerged as a popular data-driven modeling method and been extensively studied in the machine learning community. However, the LS-SVM is sensitive to noisy data and may not be effective when the level of noise is high. In this paper, a probabilistic LS-SVM is proposed to have a more reliable performance. First, a distributed LS-SVM is constructed with parameters estimated from data samples. Due to distributed nature of multiple LS-SVM, the Stochastic Property of parame-ters can be easily obtained and processed. Using the distribution characteristics of these parameters, the final outcome is derived through the probabilistic inference and thus be evaluated statisti-cally. Its parallel structure is also suitable for parallel computing to reduce computing time. Both simulations and experiments demonstrate the effectiveness of the proposed probabilistic LS-SVM. Index Terms—Distributed LS-SVM, noisy data, proba-bilistic inference, probabilistic least squares support vector machine (LS-SVM).
-
Probabilistic Inference-Based Least Squares Support Vector Machine for Modeling Under Noisy Environment
IEEE Transactions on Systems Man and Cybernetics: Systems, 2016Co-Authors: Xinjiang Lu, Han-xiong LiAbstract:The least squares support vector machine (LS-SVM) has emerged as a popular data-driven modeling method and been extensively studied in the machine learning community. However, the LS-SVM is sensitive to noisy data and may not be effective when the level of noise is high. In this paper, a probabilistic LS-SVM is proposed to have a more reliable performance. First, a distributed LS-SVM is constructed with parameters estimated from data samples. Due to distributed nature of multiple LS-SVM, the Stochastic Property of parameters can be easily obtained and processed. Using the distribution characteristics of these parameters, the final outcome is derived through the probabilistic inference and thus be evaluated statistically. Its parallel structure is also suitable for parallel computing to reduce computing time. Both simulations and experiments demonstrate the effectiveness of the proposed probabilistic LS-SVM.
-
Online Probabilistic Extreme Learning Machine for Distribution Modeling of Complex Batch Forging Processes
IEEE Transactions on Industrial Informatics, 2015Co-Authors: Xinjiang Lu, Minghui HuangAbstract:An effective model of batch forging processes is crucial to ensure the quality conformance control of batch productions. However, obtaining this model has proven difficult due to a variety of the raw forgings produced by manufacturing error, material variation, geometric defects, etc. In this paper, a novel online probabilistic extreme learning machine (ELM) is proposed to model batch forging processes. A probabilistic ELM is first developed to extract the distribution information of the batch forging processes from the data. Due to the highly linear structure of the ELM, the Stochastic Property of the forging process is easily derived and processed. By using the characteristics of the online ELM, a strategy is then developed to update the distribution model as new forging process data are collected. Finally, case studies on complex batch forging processes demonstrate the effectiveness of the proposed online probabilistic ELM.
Tielong Shen - One of the best experts on this subject based on the ideXlab platform.
-
a Stochastic logical system approach to model and optimal control of cyclic variation of residual gas fraction in combustion engines
Applied Thermal Engineering, 2016Co-Authors: Yuhu Wu, Madan Kumar, Tielong ShenAbstract:Abstract In four stroke internal combustion engines, residual gas from the previous cycle is an important factor influencing the combustion quality of the current cycle, and the residual gas fraction (RGF) is a popular index to monitor the influence of residual gas. This paper investigates the cycle-to-cycle transient behavior of the RGF in the view of systems theory and proposes a multi-valued logic-based control strategy for attenuation of RGF fluctuation. First, an in-cylinder pressure sensor-based method for measuring the RGF is provided by following the physics of the in-cylinder transient state of four-stroke internal combustion engines. Then, the Stochastic Property of the RGF is examined based on statistical data obtained by conducting experiments on a full-scale gasoline engine test bench. Based on the observation of the examination, a Stochastic logical transient model is proposed to represent the cycle-to-cycle transient behavior of the RGF, and with the model an optimal feedback control law, which targets on rejection of the RGF fluctuation, is derived in the framework of Stochastic logical system theory. Finally, experimental results are demonstrated to show the effectiveness of the proposed model and the control strategy.
-
Notice of Removal: Control design for residual gas fraction in engine based on Stochastic logical dynamics
2015 54th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE), 2015Co-Authors: Yuhu Wu, Tielong ShenAbstract:In this paper, the optimal control scheme for residual gas fraction is proposed considering the Stochastic Property in engine. Initially, the receding horizon optimal control problem for the Stochastic logical dynamical systems with finite states is considered. Based on transition probability and semi-tensor product of matrix, a succinct algebraic expression of dynamic programming algorithm is derived to solve the receding horizon control problem. Then, an optimal controller designed to reduce the variation of residual gas fraction in the framework of Stochastic logician dynamical model, in which variable valve timing is taken as the control actuator. Validation results are demonstrated which conducted on a full-scaled gasoline engine test bench.
-
Model-Based Stochastic Optimal Air–Fuel Ratio Control With Residual Gas Fraction of Spark Ignition Engines
IEEE Transactions on Control Systems Technology, 2014Co-Authors: Jun Yang, Tielong Shen, Xiaohong JiaoAbstract:In this paper, a Stochastic optimal control scheme for the air-fuel ratio is proposed, which considers the cyclic variations of the residual gas fraction (RGF). Initially, a cylinder pressure-based measurement of the RGF is derived by following the physics of inlet-exhaust process. Then, a dynamical model is presented to describe the cyclic variation of the air charge, fuel charge, and combustion products under a cyclically varied RGF, where the RGF is modeled as a Markovian Stochastic process. Using this model, a feedback control law is derived, which optimizes the quadratic cost function in the Stochastic sense with respect to the Stochastic Property of the residual gas. The cost function reflects the tradeoff between the accuracy of the regulation of the air-fuel ratio with the fluctuation in the fuel injection. Finally, a sampling process-based statistical analysis for the RGF is presented based on the experiments conducted on a full-scaled gasoline engine test bench, and the proposed control law is validated based on a numerical simulation and experiments.
Yuhu Wu - One of the best experts on this subject based on the ideXlab platform.
-
a Stochastic logical system approach to model and optimal control of cyclic variation of residual gas fraction in combustion engines
Applied Thermal Engineering, 2016Co-Authors: Yuhu Wu, Madan Kumar, Tielong ShenAbstract:Abstract In four stroke internal combustion engines, residual gas from the previous cycle is an important factor influencing the combustion quality of the current cycle, and the residual gas fraction (RGF) is a popular index to monitor the influence of residual gas. This paper investigates the cycle-to-cycle transient behavior of the RGF in the view of systems theory and proposes a multi-valued logic-based control strategy for attenuation of RGF fluctuation. First, an in-cylinder pressure sensor-based method for measuring the RGF is provided by following the physics of the in-cylinder transient state of four-stroke internal combustion engines. Then, the Stochastic Property of the RGF is examined based on statistical data obtained by conducting experiments on a full-scale gasoline engine test bench. Based on the observation of the examination, a Stochastic logical transient model is proposed to represent the cycle-to-cycle transient behavior of the RGF, and with the model an optimal feedback control law, which targets on rejection of the RGF fluctuation, is derived in the framework of Stochastic logical system theory. Finally, experimental results are demonstrated to show the effectiveness of the proposed model and the control strategy.
-
Notice of Removal: Control design for residual gas fraction in engine based on Stochastic logical dynamics
2015 54th Annual Conference of the Society of Instrument and Control Engineers of Japan (SICE), 2015Co-Authors: Yuhu Wu, Tielong ShenAbstract:In this paper, the optimal control scheme for residual gas fraction is proposed considering the Stochastic Property in engine. Initially, the receding horizon optimal control problem for the Stochastic logical dynamical systems with finite states is considered. Based on transition probability and semi-tensor product of matrix, a succinct algebraic expression of dynamic programming algorithm is derived to solve the receding horizon control problem. Then, an optimal controller designed to reduce the variation of residual gas fraction in the framework of Stochastic logician dynamical model, in which variable valve timing is taken as the control actuator. Validation results are demonstrated which conducted on a full-scaled gasoline engine test bench.
Han-xiong Li - One of the best experts on this subject based on the ideXlab platform.
-
Probabilistic Inference-Based Least Squares Support Vector Machine for Modeling under Noisy Environment
IEEE Transactions on Systems Man and Cybernetics: Systems, 2016Co-Authors: Bi Fan, Xinjiang Lu, Han-xiong LiAbstract:—The least squares support vector machine (LS-SVM) has emerged as a popular data-driven modeling method and been extensively studied in the machine learning community. However, the LS-SVM is sensitive to noisy data and may not be effective when the level of noise is high. In this paper, a probabilistic LS-SVM is proposed to have a more reliable performance. First, a distributed LS-SVM is constructed with parameters estimated from data samples. Due to distributed nature of multiple LS-SVM, the Stochastic Property of parame-ters can be easily obtained and processed. Using the distribution characteristics of these parameters, the final outcome is derived through the probabilistic inference and thus be evaluated statisti-cally. Its parallel structure is also suitable for parallel computing to reduce computing time. Both simulations and experiments demonstrate the effectiveness of the proposed probabilistic LS-SVM. Index Terms—Distributed LS-SVM, noisy data, proba-bilistic inference, probabilistic least squares support vector machine (LS-SVM).
-
Probabilistic Inference-Based Least Squares Support Vector Machine for Modeling Under Noisy Environment
IEEE Transactions on Systems Man and Cybernetics: Systems, 2016Co-Authors: Xinjiang Lu, Han-xiong LiAbstract:The least squares support vector machine (LS-SVM) has emerged as a popular data-driven modeling method and been extensively studied in the machine learning community. However, the LS-SVM is sensitive to noisy data and may not be effective when the level of noise is high. In this paper, a probabilistic LS-SVM is proposed to have a more reliable performance. First, a distributed LS-SVM is constructed with parameters estimated from data samples. Due to distributed nature of multiple LS-SVM, the Stochastic Property of parameters can be easily obtained and processed. Using the distribution characteristics of these parameters, the final outcome is derived through the probabilistic inference and thus be evaluated statistically. Its parallel structure is also suitable for parallel computing to reduce computing time. Both simulations and experiments demonstrate the effectiveness of the proposed probabilistic LS-SVM.
K. Yamada - One of the best experts on this subject based on the ideXlab platform.
-
an efficient robust adaptive filtering algorithm based on parallel subgradient projection techniques
IEEE Transactions on Signal Processing, 2002Co-Authors: I. Yamada, K. Slavakis, K. YamadaAbstract:This paper presents a novel robust adaptive filtering scheme based on the interactive use of statistical noise information and the ideas developed originally for efficient algorithmic solutions to the convex feasibility problems. The statistical noise information is quantitatively formulated as Stochastic Property closed convex sets by the simple design formulae developed in this paper. A simple set-theoretic inspection also leads to an important statistical reason for the sensitivity to noise of the affine projection algorithm (APA). The proposed adaptive algorithm is computationally efficient and robust to noise because it requires only an iterative parallel projection onto a series of closed half spaces that are highly expected to contain the unknown system to be identified and is free from the computational load of solving a system of linear equations. The numerical examples show that the proposed adaptive filtering scheme realizes dramatically fast and stable convergence for highly colored excited speech like input signals in severe noise situations.
-
An efficient robust adaptive filtering scheme based on parallel subgradient projection techniques
2001 IEEE International Conference on Acoustics Speech and Signal Processing. Proceedings (Cat. No.01CH37221), 2001Co-Authors: I. Yamada, K. Slavakis, K. YamadaAbstract:This paper presents a novel robust adaptive filtering scheme based on the interactive use of statistical noise information and an extension of the ideas developed originally for efficient algorithmic solutions to the convex feasibility problems. The statistical noise information is quantitatively formulated as Stochastic Property closed convex sets by the simple design formulae developed. The proposed adaptive algorithm is computationally efficient and robust to noise because it requires only an iterative parallel projection onto a series of closed half spaces highly expected to contain the unknown system to be identified. The numerical examples show that the proposed adaptive filtering scheme achieves low estimation error and realizes dramatically fast and stable convergence even for highly colored excited input signals in severely noisy situations.
