Power Systems Application of Artificial Neural Networks. (ANN)

appliancepartΤεχνίτη Νοημοσύνη και Ρομποτική

19 Οκτ 2013 (πριν από 3 χρόνια και 11 μήνες)

99 εμφανίσεις

Power Systems Application of Artificial Neural
Networks
. (ANN)



Introduction


Brief history.


Structure



How they work


Sample Simulations. (EasyNN)



Why use them (Merits and Demerits)?



Current Applications



NN & Power Systems


Conclusion


Introduction


Derived from biological neuron.


Connection of processing nodes that
transfer activity to the next.


Likened to the Brain.


Mostly one
-
way connection.


Architecture


Feedforward




Feedback


Network Layers


Perceptron


Architecture


Feedforward





Perceptron
















Network









How They Work.


Inputs to NN compared with pre
-
processed data.


Outputs compared to desired response.


NN “learn” system patterns.


NN are “trained” to respond.



Learning


Supervised


Unsupervised


Reinforced



EasyNN software
.

EasyNN


Not the best model for NN








Output Graph

Why Use NN (Merits)?


Adaptive Learning


Self Organization


Real
-
Time Organization.


Fault Tolerance.

Current Applications.


Sales Forecasting


Industrial Process Control


Data Validation


Risk Management (Insurance)


Banking (loan processing).


Medical

Power Systems Application.



Integrated Power Systems.


Load Forecasting (Blackout)


Power Plants.


Many other areas.


Conclusion.


NN


great potential for future systems.


Need for much more research.


Could prove vital in power management
and distribution.