Kl. Lo et al., HYBRID APPROACH USING COUNTERPROPAGATION NEURAL-NETWORK FOR POWER-SYSTEM NETWORK REDUCTION, IEE proceedings. Generation, transmission and distribution, 144(2), 1997, pp. 169-174
A hybrid counterpropagation neural network and Ward-type equivalent ap
proach for power system network reduction is proposed for improving th
e conventional external system equivalent technique. The proposed Ward
-type equivalent technique not only possesses the good properties of t
he extended Ward equivalent, but can also update the parameters of the
equivalent model for representing real-time topology changes of the e
xternal system. Another improvement is that a counterpropagation neura
l network is used to match the boundary equivalent power injections, T
he new hybrid approach combines the simplicity of Ward-type equivalent
techniques with the speed of artificial neural networks. Test results
demonstrate that the hybrid approach and highly accurate is very effi
cient and compared to the external system equivalent.