The authors propose a new methodology for the combined solution of the topological identification observability analysis and bad data processing problems in power systems. The solution is based on a pattern analysis approach. An efficient framework is suggested for solving data acquisition and processing problems, as well as joining pattern analysis and analytical procedures. Two different techniques of pattern analysis are combined to produce a classifier and an estimator with unique characteristics to deal with noisy environments. Unobservable network areas, multiple interacting bad data and bad critical measurements can be efficiently treated. The patterns required for the training process can be acquired from the SCADA system and/or from load-flow simulations. Test results are presented for the IEEE 24-busbar reliability test system. >

Solving data acquisition and processing problems in power systems using a pattern analysis approach

The authors describe a parallel distributed topology classifier. The idea is to determine the system configuration in a very fast way, even in the presence of incorrect or unavailable switch/breaker status and analog measurements. A new supervised learning algorithm suitable for very large training sets is introduced. >

Neural networks for topology determination of power systems

In this paper, a new approach for the determination of power system network topology based on Artificial Neural Networks (ANN) has been suggested. For the determination of power system network topology, three models of ANN based on Multilayer perceptron using Backpropagation Algorithm (BPA), Functional Link Network (FLN) and Counterpropagation Network (CPN) have been utilized and tested for both noisy as well as noise free data sets. ANN models based on BPA, FLN and CPN have been tested on IEEE 14-bus, IEEE 57-bus and a 75-bus practical Indian system. It has been established that the CPN based model predicts network topology more accurately as compared to the FLN and BPA based models in all test cases. Further, the CPN model is able to determine the network topology even if the network is unobservable for which the conventional network topology algorithm [8] fail to determine the topology.

Dušica Ćalović

Papers

An indirect method for the detection of changes in power network configuration