In this paper, Artificial Neural Networks (ANN) have been applied in thermal processes related to liquid immersed distribution transformers. This technique allows to monitor and to estimate the heating of transformers...
详细信息
In this paper, Artificial Neural Networks (ANN) have been applied in thermal processes related to liquid immersed distribution transformers. This technique allows to monitor and to estimate the heating of transformers in order to improve its performance and to reduce the insulation degradation. The insulation degradation in transformers is usually computed taking into account the hot spot temperature. However, the characteristics and properties related with this temperature are not very well known and its identification has been a difficult task. On the other hand, the ability of neural networks in to solve complex and diversified problems make them attractive for estimation of thermal processes associated with transformers.
Some classes of nonlinear controlled systems are considered. This systems have a linear or bilinear approximation. For different classes of the allowed controls the sufficient conditions of boundedness of the integral...
详细信息
Some classes of nonlinear controlled systems are considered. This systems have a linear or bilinear approximation. For different classes of the allowed controls the sufficient conditions of boundedness of the integral vortex are obtained. The dependence of this characteristic on degree of perturbations is discussed. External estimation of the integral vortex by means of the Lyapunov functions is constracted.
Linearizations of nonlinear functions that are based on Jacobian matrices often cannot be applied in practical applications of nonlinear estimation techniques. An alternative linearization method is presented in this ...
详细信息
Linearizations of nonlinear functions that are based on Jacobian matrices often cannot be applied in practical applications of nonlinear estimation techniques. An alternative linearization method is presented in this paper. The method assumes that covariance matrices are determined on a square root factored form. A factorization of the output covariance from a nonlinear vector function is directly determined by "perturbing" the nonlinear function with the columns of the factored input covariance, without explicitly calculating the linearization and with no differentiations involved. The output covariance is more accurate than that obtained with the ordinary Jacobian linearization method. It also has an advantage that Jacobian matrices do not have to be derived symbolically. (C) 1997 Elsevier Science Ltd.
This paper considers state estimation for a discrete-time hidden Markov model (HMM) when the observations are delayed by a random time, The delay process is itself modeled as a finite state Markov chain that allows an...
详细信息
This paper considers state estimation for a discrete-time hidden Markov model (HMM) when the observations are delayed by a random time, The delay process is itself modeled as a finite state Markov chain that allows an augmented state HMM to model the overall system. State estimation algorithms for the resulting HMM are then presented, and their performance is studied in simulations. The motivation for the model stems from the situation when distributed sensors transmit measurements over a connectionless packet switched communications network.
In this contribution, Genetic Programming (GP) is used to develop inferential estimation models using experimental data. GP performs symbolic optimisation, automatically determining both the structure and the complexi...
详细信息
In this contribution, Genetic Programming (GP) is used to develop inferential estimation models using experimental data. GP performs symbolic optimisation, automatically determining both the structure and the complexity of an empirical model. After a tutorial example, the usefulness of the technique is demonstrated by the development of an inferential estimation model of a plasticating extruder. A statistical analysis procedure is used as a guide in the selection of the final model structure. For the industrial case study, the inferential models obtained using the GP algorithm are compared to those obtained using a linear, finite impulse response model and a feedforward artificial neural network (FANN). For this application, the GP technique produces models with a significantly lower Root Mean Square (RMS) error than the other techniques.
The square-root information filter algorithm and the related covariance smoother algorithm have been generalized to handle singular state-transition matrices and perfect measurements. This allows the use of SRIF techn...
详细信息
The square-root information filter algorithm and the related covariance smoother algorithm have been generalized to handle singular state-transition matrices and perfect measurements. This allows the use of SRIF techniques for problems with delays and state constraints. The generalized algorithms use complete QR factorization to isolate deterministically known parts of the state and nonsingular parts of the state transition and disturbance influence matrices. These factorizations and the corresponding changes of coordinates are used to solve the recursive least-squares problems that are basic to the SRIF technique. Numerical stability, computation time, and storage requirements are comparable to the traditional SRIF algorithms. (C) 1999 Elsevier Science Ltd. All rights reserved.
Artificial neural networks are dynamic systems consisting of highly interconnected and parallel nonlinear processing elements. Systems based on artificial neural networks have high computational rates due to the use o...
详细信息
Artificial neural networks are dynamic systems consisting of highly interconnected and parallel nonlinear processing elements. Systems based on artificial neural networks have high computational rates due to the use of a massive number of simple processing elements and the high degree of connectivity between these elements. The ability of neural networks to realize some complex nonlinear function makes them attractive for system identification. This paper presents a novel approach to solve robust parameter estimation problems for nonlinear model with unknown-but-bounded errors and uncertainties. More specifically, a modified Hopfield network is developed and its internal parameters are computed using the valid-subspace technique. These parameters guarantee the network convergence to the equilibrium points. A solution for the robust estimation problem with unknown-but-bounded error corresponds to an equilibrium point of the network. Simulation results are presented as an illustration of the proposed approach.
There has been recent interest in using orthonormalised forms of fixed denominator model structures for system identification. A key motivating factor in the employment of these forms is that of improved numerical pro...
详细信息
There has been recent interest in using orthonormalised forms of fixed denominator model structures for system identification. A key motivating factor in the employment of these forms is that of improved numerical properties. Namely, for white input perfect conditioning of the least-squares normal equations is achieved by design. However, for the more usual case of coloured input spectrum, it is not clear what the numerical conditioning properties should be in relation to simpler and perhaps more natural model structures. This paper provides theoretical and empirical evidence to argue that in fact, even though the orthonormal structures are only designed to provide perfect numerical conditioning for white input, they still provide improved conditioning for a wide variety of coloured inputs.
In this work a new algorithm to tune a linear controller for a nonlinear system is presented. The algorithm iteratively minimizes a criterion of the control performance and it can be seen as an extension of the recent...
详细信息
In this work a new algorithm to tune a linear controller for a nonlinear system is presented. The algorithm iteratively minimizes a criterion of the control performance and it can be seen as an extension of the recently introduced model-free tuning algorithm to the nonlinear control problem. Indeed, the model-free tuning is used for the linear part and a model of the linearization of the closed-loop system around its current trajectory is used for the nonlinear part of the estimation. The algorithm requires an initial feedback controller that stabilizes the closed-loop for the desired reference signal and reference signals in its vicinity. It is also assumed that the closed-loop outputs are similar for this set of reference signals.
This paper establishes that when using a least squares criterion to estimate an output error type model structure, then the measurement noise induced variability of the frequency response estimate depends on the estim...
详细信息
This paper establishes that when using a least squares criterion to estimate an output error type model structure, then the measurement noise induced variability of the frequency response estimate depends on the estimated (and hence also on the true) pole positions. This dependence on pole position is perhaps counter to prevailing wisdom that for any 'shift invariant' model structure, the variability depends only on model order, data length, and input and noise spectral densities. That is, it is counter to the belief that variance error is model-structure independent.
暂无评论