A novel dynamic neural network structure is developed for simulating and controlling dynamic systems. Using this kind of neural network, a controller for a hovering platform was investigated. The network description a...
详细信息
A novel dynamic neural network structure is developed for simulating and controlling dynamic systems. Using this kind of neural network, a controller for a hovering platform was investigated. The network description and attempts made to simulate the hovering platform behavior are presented.< >
The natural mathematical notation when developing models of physical systems is often that of differential-algebraic equations. It is well-known that many mathematical models of interest have high index, and that ther...
详细信息
This paper presents a new method for computing the discriminant vectors of the Foley–Sammon optimal set. First, an equivalent criterion is presented to replace the Fisher criterion; then, the problem of computing the...
详细信息
This paper presents a new method for computing the discriminant vectors of the Foley–Sammon optimal set. First, an equivalent criterion is presented to replace the Fisher criterion; then, the problem of computing the discriminant vectors in R n is transformed into the maximum problem in a subspace. Several theorems relating to the method are also presented. Experimental results show that the present method is superior to the positive pseudoinverse method, and the perturbation method in terms of correct classification rate.
This paper presents optimum pairing and ordering (P/O) conditions for simultaneous reduction in total capacitance, sensitivity and output noise of cascade SC filters. First, investigating relations among conditions pr...
详细信息
In this paper we present applications of three types of fiizzy neural networks: (1) crisp signals used to evaluate fûzzy weights;(2) fuzzy signals combined with fuzzy weights;and (3) fuzzy signals transformed by ...
详细信息
Signal parameter estimation from sensor array data is of great interest in a variety of applications, including radar, sonar, and radio communication. A large number of high-resolution (i.e., model-based) techniques h...
详细信息
Signal parameter estimation from sensor array data is of great interest in a variety of applications, including radar, sonar, and radio communication. A large number of high-resolution (i.e., model-based) techniques have been suggested in the literature. The vast majority of these require knowledge of the spatial noise correlation matrix, which constitutes a significant drawback. A novel instrumental variable (IV) approach to the sensor array problem is proposed. By exploiting temporal correlatedness of the source signals, knowledge of the spatial noise covariance is not required. The asymptotic properties of the IV estimator are examined, and an optimal IV method is derived. Simulations are presented examining the properties of the IV estimators in data segments of realistic lengths.< >
It is demonstrated how a neural network may be used to generate a time-suboptimal control policy in continuous time. Approximation of optimal feedback by neural networks is proposed. Training data are taken from a set...
详细信息
It is demonstrated how a neural network may be used to generate a time-suboptimal control policy in continuous time. Approximation of optimal feedback by neural networks is proposed. Training data are taken from a set of precalculated open-loop trajectories from which a neural network extracts the information about the mapping in question. Simulation results are provided for a simple example.< >
Artificial feedforward networks are studied as nonlinear function approximators used to identify forward and inverse mappings of discrete time dynamic systems. They are found to provide significant advantages over oth...
详细信息
Artificial feedforward networks are studied as nonlinear function approximators used to identify forward and inverse mappings of discrete time dynamic systems. They are found to provide significant advantages over other modelling techniques such as polynomial approximations, especially if the extrapolation beyond the region covered by the learning data is involved. We apply the neural network methodology to a simple second order approximation of a single-machine infinite-bus power system controlled by means of modifying the reactance of the line. Accurate off-line identification of forward and inverse dynamics of the system is performed by means of single hidden layer neural networks, and both models are then used in a direct inverse control configuration. The controller simulations show very good quality of transients for severe short-circuit fault.
The authors consider the analysis of stochastic Petri net models with generally distributed transition firing times. A so-called hybrid state analysis method is developed. The basic idea is to make the extended state,...
详细信息
The authors consider the analysis of stochastic Petri net models with generally distributed transition firing times. A so-called hybrid state analysis method is developed. The basic idea is to make the extended state, i.e., the hybrid state, which is composed of a marking and the enabling times of enabled transitions under the marking, propagate as a Markov process by the inclusion of supplementary variables. The recursive equation of hybrid state density functions is given.< >
The authors consider a fuzzy controller that processes fuzzy information. They discuss the model of the fuzzy controller, with fuzzy inputs for error and change in error, using a max-min neural network. A new learning...
详细信息
The authors consider a fuzzy controller that processes fuzzy information. They discuss the model of the fuzzy controller, with fuzzy inputs for error and change in error, using a max-min neural network. A new learning algorithm, a modified delta rule, is derived. The generalization property of the neural net can be used to find a controller output for new fuzzy values of error and change in error. An example is presented showing the applicability of the fuzzy neural controller.< >
暂无评论