Evaluation of cosine radial basis function neural networks on electric power load forecasting

Karayiannis, N.B.; Balasubramanian, M.; Malki, Heidar

doi:10.1109/ijcnn.2003.1223732

Cited by 17 publications

(10 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For linear generator functions of the form , and condition (13) holds as an equality if (14) For exponential generator functions of the form , and condition (13) holds as an equality if (15) The shapes of the radial basis functions can be determined in practice by fixing to an integer value between 2 and 4 and computing the values of their free parameters or according to (14) or (15) …”

Section: Estimating the Free Parameters Of Radial Basis Functionsmentioning

confidence: 99%

On the construction and training of reformulated radial basis function neural networks

Karayiannis

Randolph-Gips

2003

IEEE Trans. Neural Netw.

Self Cite

113

View full text Add to dashboard Cite

Presents a systematic approach for constructing reformulated radial basis function (RBF) neural networks, which was developed to facilitate their training by supervised learning algorithms based on gradient descent. This approach reduces the construction of radial basis function models to the selection of admissible generator functions. The selection of generator functions relies on the concept of the blind spot, which is introduced in the paper. The paper also introduces a new family of reformulated radial basis function neural networks, which are referred to as cosine radial basis functions. Cosine radial basis functions are constructed by linear generator functions of a special form and their use as similarity measures in radial basis function models is justified by their geometric interpretation. A set of experiments on a variety of datasets indicate that cosine radial basis functions outperform considerably conventional radial basis function neural networks with Gaussian radial basis functions. Cosine radial basis functions are also strong competitors to existing reformulated radial basis function models trained by gradient descent and feedforward neural networks with sigmoid hidden units.

show abstract

Section: Estimating the Free Parameters Of Radial Basis Functionsmentioning

confidence: 99%

On the construction and training of reformulated radial basis function neural networks

Karayiannis

Randolph-Gips

2003

IEEE Trans. Neural Netw.

Self Cite

113

View full text Add to dashboard Cite

show abstract

“…Zhao et al [4] also raise an improved RBF NN algorithm that uses PCA to implement reconstruction on input space and then creates network architecture based on the contribution rate of each principal component. In [5], a kind of cosine RBF neural network is proposed and successfully applied in power system load forecasting. In the issue of traffic flow forecasting, Zhang et al [6] propose an improved RBF NN method based on chaos theory and verifies its practicability by both simulation and actual using.…”

Section: Introductionmentioning

confidence: 99%

A degredation interval prediction method based on RBF neural network

Zhang

Sun

2014

2014 10th International Conference on Reliability, Maintainability and Safety (ICRMS)

View full text Add to dashboard Cite

In the area of reliability, remaining useful lifetime (RUL) prediction can help people establish reasonable maintenance strategies and then implement maintenance activities at a right time. In this paper, RBF neural network approach is applied in the degradation prediction process of a certain microwave component. A degradation model that describes how a certain degradation parameter changes over time is established and then the performance degradation trend can be obtained based on this model. And then a confidence interval prediction can be obtained based on traditional probability theory, which proves that the results have reached a high confidence level. Finally, the BP neural network approach is introduced as a comparison, and results indicate that the proposed method has higher precision and stability.

show abstract

“…In [7] Broomhead and Lowe introduced the RBFNN that, as it has been demonstrated in the literature [8], [27], [30], have the capability of approximating any given function. This kind of ANN have been successfully applied to many problems related with This work has been partially supported by the Spanish CICYT Project TIN2004-01419 nonlinear regression and function approximation [21], [35], [36]. An RBFNN is a two-layer, fully connected network in which each neuron implements a gaussian function as follows:…”

Section: Introductionmentioning

confidence: 99%

Supervised RBFNN Centers and Radii Initialization for Function Approximation Problems

Guillén¹,

Rojas²,

González³

et al. 2006

The 2006 IEEE International Joint Conference on Neural Network Proceedings

View full text Add to dashboard Cite

Radial Basis Function Neural Networks (RBFNNs) have been applied to solve problems of classification, function approximation and time series prediction. In the design of an RBFNN it is necessary to set the values for the positions of the centers and the radii for each RBF. In the literature it is usually performed an initialization step to set the positions of the centers and, once they are placed, the radii are calculated using a heuristic. In this paper, a new algorithm to set the value of those two parameters is presented. This new algorithm uses a supervised learning in such a way that the position of the centers will be constrained by the output of the function to be approximated. Since each center represents a neuron that is activated by the input vectors, the radii are initialized using the center's positions and their activation grades. In this way, the calculation of the radii is also influenced by the output of the target function, not like in other heuristics where only the positions of the centers or the input vectors are considered. As the experiments show, the new algorithm outperforms other algorithms previously used for this problem.

show abstract

Evaluation of cosine radial basis function neural networks on electric power load forecasting

Cited by 17 publications

References 23 publications

On the construction and training of reformulated radial basis function neural networks

On the construction and training of reformulated radial basis function neural networks

A degredation interval prediction method based on RBF neural network

Supervised RBFNN Centers and Radii Initialization for Function Approximation Problems

Contact Info

Product

Resources

About