Sei sulla pagina 1di 18

ARTIFICIAL NEURAL NETWORKS

What is a Neural Network?

An Artificial Neural Network (ANN) is an information processing paradigm that is inspired by the way biological nervous systems, such as the brain, processes the information. The neural network consists of a number of neurons which are the elementary processing units that are connected together according to some pattern of connectivity.

Neural network with their remarkable ability to drive meaning from complicated and imprecise data. The neural network system help where we cannot formulate an algorithmic solution. Neural network process the information in a similar way the human brain does. The disadvantage is that because the network finds out how to solve the problem by itself, its operation can be unpredictable

Information flow in a Neural cell:

Dendrites receive activation from other neurons. Soma process the incoming activations and convert them into output activations.

Axons act as transmission lines to send activation to other neurons.


Synapses the junctions allows signal transmission between the axons and dendrites.

An artificial neural network may be simulated by first deducing the essential features of neurons and their interconnections.

McCulloch and Pitts(1943) are generally recognized as the designers of the first neural network. Hebbs(1949)developed the first learning rule. In the 1950 and 60s many researchers (Block,Minsky,Papert and Rosenblatt) worked on multi layer ann.

A very simplified model of real neurons is known as a threshold logic unit (TLU)

Training of ANN is carried out by presenting the network with examples called training patterns. During training the synaptic weight get modified to model the given problem.

Single layer Feed-Forward network

Multi layer Feed-Forward Network


Recurrent Network

Supervised Learning
Unsupervised Learning Reinforced Learning

To train the neural network to perform some task, we must adjust the weights of each unit in such a way that the error between the desired output and the actual output is reduced. This process requires that the neural network compute the error derivative of the weights(EW). The back propagation is the most widely used method for determining the EW.

Design of the ANN Training Set Generation and Number of Training Set

Selection of Training Parameters


Weight Initialization Sufficiency of Knowledge

ANN is the functional imitation of human brain which simulates the human intuition in making decisions and drawing conclusion.
ANNs are being extensively applied in the different areas of power system problems , where large computational time is required e.g. contingency analysis ,load forecasting.

THANK YOU

Potrebbero piacerti anche