Optimization of HV electrode systems by neural networks using a new learning method

Mukherjee, P. K.; Trinitis, Carsten; Steinbigler, H.

doi:10.1109/94.556552

Cited by 21 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Uniform and complex electric stress distributions along the surface of axi-symmetric insulators in multi-dielectric arrangements by a neural network were presented in Bhattacharya, Chakravorti, and Mukherjee (2001). A new learning method was introduced in Mukherjee, Trinitis, and Steinbigler (1996) for the optimization of HV electrode systems by neural networks. Application of artificial neural networks for optimization of electrode contour was reported in Chakravorti and Mukherjee (1994).…”

Section: Introductionmentioning

confidence: 99%

A neural network-based estimation of electric fields along high voltage insulators

Aydoğmuş

2009

Expert Systems with Applications

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A neural network-based estimation of electric fields along high voltage insulators

Aydoğmuş

2009

Expert Systems with Applications

View full text Add to dashboard Cite

“…Finite element method (FEM) is employed in this work for numerical electric field calculation. The geometrical parameters of the insulator are changed by assuming a general base or initial structure for bushing geometry to find the appropriate electric field distribution and capacitance by means of GA. ANN is used for electrode optimization in Chakravorti and Mukherjee (1994) and later, the proposed method in this work was improved in optimization of electrode contours by modifying the proposed optimization algorithm (Mukherjee et al , 1996). These works were followed in Lahiri and Chakravorti (2005) in which the same general problem was solved to optimize outer geometrical parameters of electrode by a hybrid algorithm including ANN and SA.…”

Section: Introductionmentioning

confidence: 99%

Optimum design of high voltage bushings by rational Bézier curves

Jemsi

Vahidi

Naghizadeh

et al. 2012

COMPEL - The International Journal for Computation and Mathematics in Electrical and Electronic Engineering

View full text Add to dashboard Cite

Purpose -The purpose of this paper is to propose a new approach for designing different parts of a high voltage bushing. It also aims to consider technical and economical criteria for the optimum solution of the design problem. Design/methodology/approach -A novel method for finding the optimal contours of different elements of high voltage bushings, including ceramic insulator, electrode, and flange angle is presented. The rational Bézier curves are used for defining the surface of the insulators and conductors of the equipment. Then, these curves are optimally adjusted to obtain an appropriate techno-economical solution. The utilized optimization method is the improved bacterial foraging algorithm (BFA) with variable step sizes. In the design procedure, two-dimensional finite element method (2D FEM) is used to calculate the performance parameters in each step of the design procedure. In order to evaluate the performance of the proposed algorithm, optimal design of different elements of a 110 kV bushing using BFA and genetic algorithm is presented, compared, and discussed as well.Findings -The results of this research show that the technical design criteria and economical costs are satisfied by the proposed method. It is concluded that the rational Bézier curves can be implemented for other similar applications and optimal design of other equipment in the electrical engineering field combined with heuristic optimization techniques. Originality/value -Bezier curves are used for the first time for bushing design purpose. Two heuristic techniques are also implemented in order to facilitate the comparison and avoid local solutions.

show abstract

“…Niching methods extend Genetic Algorithms (GAs) by promoting the formation of stable subpopulations in the neighborhood of local and global optima. In this paper, we present a process for automatic design of electrodes using the Charge Simulation Method (CSM) [2] [3] coupled with niching and constrained optimization techniques. Our approach is radically different from those developed elsewhere.…”

Section: Introductionmentioning

confidence: 99%

“…Our approach is radically different from those developed elsewhere. In effect, shape optimal design is generally carried out by directly parametrizing the shape of devices [3] [4]. The proposed method consists in identifying an optimal shape from an optimal equipotential line resulting from a system of point charges.…”

Section: Introductionmentioning

confidence: 99%

Niching genetic algorithms for optimization in electromagnetics. II. Shape optimization of electrodes using the CSM

1998

View full text Add to dashboard Cite

In this paper, we present a new approach for automatic design of electrodes. The investigated method consists in identifying an optimal shape from an optimal equipotential resulting from a system of point charges. The electric field and potential are computed using the point charge simulation method. Niching genetic algorithms and constrained optimization techniques are applied to the electrode benchmark in order to find multiple optimal profiles.

show abstract

Optimization of HV electrode systems by neural networks using a new learning method

Cited by 21 publications

References 7 publications

A neural network-based estimation of electric fields along high voltage insulators

A neural network-based estimation of electric fields along high voltage insulators

Optimum design of high voltage bushings by rational Bézier curves

Niching genetic algorithms for optimization in electromagnetics. II. Shape optimization of electrodes using the CSM

Contact Info

Product

Resources

About