Artificial neural-net based dynamic security assessment for electric power systems

Abstract

In the post-fault dynamic analysis of interconnected power systems, the critical fault clearing time (CCT) is one of the parameters of paramount importance. It constitutes a complex function of the pre-fault system condition, fault type and location, and protective relaying strategy. The evaluation of CCT involves elaborate computations that often include time-consuming solutions of nonlinear on-fault system equations. This paper describes an adaptive pattern recognition approach based on highly parallel information processing using artificial neural networks (ANN). High adaptation capabilities of these networks make them able to synthesize the complex mappings that carry the input attributesfeatures into the single valued space of the CCT's. Appropriate input feature selection makes this approach a candidate for successfully handling toDologicallv indeoendent dynamic security assessment process.

A feed-forward artificial neural network with enhanced feature selection for power system transient stability assessment

“…They pass the information to the neurons in the hidden layer in a weighted form. The neurons in the hidden layer process this information with a nonlinear function [12].…”

Application of Evolutionary Technique to Power System Vulnerability Assessment

“…In order to overcome such weaknesses, computational intelligence (CI) techniques, such as neural networks, particle swarm optimization, and evolutionary computation, have been proposed for VA [18][19][20][21][22][23][24][25][26][27]; especially neural networks for its pattern classification capability. As a pattern classifier, neural networks need extensive off-line simulation so as to acquire a large-enough set of training data to represent the different operating conditions of typical power systems.…”

Transient stability assessment of power systems described with detailed models using neural networks

“…Therefore, timedomain simulation is still the most accurate method for transient stability analysis and it can be applied to any level of power system models, but as mentioned before, the main problem of this method is that it is very time consuming. As a result, in recent years there have been several attempts in using computational intelligence based techniques like neural networks (NNs) for transient stability assessment (TSA) of power systems, e.g., see [14][15][16][17][18][19][20][21][22][23][24][25]. Neural networks are widely used for function approximation and/or classification problems, because no rigorous mathematical system modeling is required in order to train an NN to form the underlying mapping.…”

“…Sobajic and Pao [14] used NNs for prediction of the CCT for a small test power system. Djukanovic et al [15] used individual energy function normalized by the critical value of global energy function evaluated at fault clearing time (FCT) to predict energy margin and stability.…”