The application of support vector machines to forecasting problems is becoming popular, lately. Several comparisons between neural networks trained with error backpropagation and support vector machines have shown adv...The application of support vector machines to forecasting problems is becoming popular, lately. Several comparisons between neural networks trained with error backpropagation and support vector machines have shown advantage for the latter in different domains of application. However, some difficulties still deteriorate the performance of the support vector machines. The main one is related to the setting of the hyperparameters involved in their training. Techniques based on meta-heuristics have been employed to determine appropriate values for those hyperparameters. However, because of the high noneonvexity of this estimation problem, which makes the search for a good solution very hard, an approach based on Bayesian inference, called relevance vector machine, has been proposed more recently. The present paper aims at investigating the suitability of this new approach to the short-term load forecasting problem.展开更多
Abstract Mehrotra-type predictor-corrector algorithm is one of the most effective primal-dual interior- point methods. This paper presents an extension of the recent variant of second order Mehrotra-type predictor-cor...Abstract Mehrotra-type predictor-corrector algorithm is one of the most effective primal-dual interior- point methods. This paper presents an extension of the recent variant of second order Mehrotra-type predictor-corrector algorithm that was proposed by Salahi, et a1.（2006） for linear optimization. Basedon the NT direction as Newton search direction, it is shown that the iteration-complexity bound of thealgorithm for semidefinite optimization is which is similar to that of the correspondingalgorithm for linear optimization.展开更多
It is very necessary for investigation on mechanism of windblown sand movement to understand and find out effective measures of preventing and reducing windblown sand, This also deals with some general features and ho...It is very necessary for investigation on mechanism of windblown sand movement to understand and find out effective measures of preventing and reducing windblown sand, This also deals with some general features and hot spots in the scientific forelands. such as multi-scale problems. interactions among. multi-physical-fields, randomness and nonlinearity as well as complex systems. In recent years, a series of experiments in wind tunnels and theoretical modeling as well as computer simulation have been taken by our research group （the Laboratory of Environmental Mechanics on Windblown Sand Movement in Lanzhou University ）in a cross-disciplinary （mechanics and geography） viewpoint. Several original and essential studies were explored such as the main regularities of charges on sand particles, the mechanisms of electric field in windblown sand flux, the effects induced by the electric field on the flux, the microwave propagations, the evolution process of windblown and flux under mutual couple interations among several physical fields, and the simulation of the main features of Aeolian sand ripples.展开更多
文摘The application of support vector machines to forecasting problems is becoming popular, lately. Several comparisons between neural networks trained with error backpropagation and support vector machines have shown advantage for the latter in different domains of application. However, some difficulties still deteriorate the performance of the support vector machines. The main one is related to the setting of the hyperparameters involved in their training. Techniques based on meta-heuristics have been employed to determine appropriate values for those hyperparameters. However, because of the high noneonvexity of this estimation problem, which makes the search for a good solution very hard, an approach based on Bayesian inference, called relevance vector machine, has been proposed more recently. The present paper aims at investigating the suitability of this new approach to the short-term load forecasting problem.
基金supported by Natural Science Foundation of Hubei Province under Grant No.2008CDZ047
文摘Abstract Mehrotra-type predictor-corrector algorithm is one of the most effective primal-dual interior- point methods. This paper presents an extension of the recent variant of second order Mehrotra-type predictor-corrector algorithm that was proposed by Salahi, et a1.（2006） for linear optimization. Basedon the NT direction as Newton search direction, it is shown that the iteration-complexity bound of thealgorithm for semidefinite optimization is which is similar to that of the correspondingalgorithm for linear optimization.
文摘It is very necessary for investigation on mechanism of windblown sand movement to understand and find out effective measures of preventing and reducing windblown sand, This also deals with some general features and hot spots in the scientific forelands. such as multi-scale problems. interactions among. multi-physical-fields, randomness and nonlinearity as well as complex systems. In recent years, a series of experiments in wind tunnels and theoretical modeling as well as computer simulation have been taken by our research group （the Laboratory of Environmental Mechanics on Windblown Sand Movement in Lanzhou University ）in a cross-disciplinary （mechanics and geography） viewpoint. Several original and essential studies were explored such as the main regularities of charges on sand particles, the mechanisms of electric field in windblown sand flux, the effects induced by the electric field on the flux, the microwave propagations, the evolution process of windblown and flux under mutual couple interations among several physical fields, and the simulation of the main features of Aeolian sand ripples.
