Artificial process element. Sum function is known to

Artificial neural
networks (ANNs) are computational networks that endeavour to re-enact the
systems of nerve cells (neurons) in creature focal sensory systems 7 and to give
a mapping between input space and ideal space by understanding the intrinsic
connections between information with the assistance of preparing methods and
utilizing processors called neurons 18. Artificial Neural Network is utilized
to diminish the computational effort required for reliability analysis and
damage discovery 9. 

Since Artificial Neural
Network has been broadly used to demonstrate natural procedures because of the
capacity of ANN to demonstrate an astounding number of human’s qualities in
example; winning as a matter of fact and summing up from past example to solve
new issues 12. The general basic units of an Artificial Neural Network are
artificial neurons. Each neuron receives input data then processes it.  From that point forward, the neurons will
convey a single output. No particular architecture ought to be expected despite
the fact that Artificial Neural Network is made out of a gathering of
interconnected neurons that are frequently assembled in layers 13.

An Artificial neuron is
made out of five primary parts specifically: input, weights, sum function,
activation function and output 12. Information that enters the neuron from
other neuron is called inputs. Weights are known as the values that express the
impact of an input set or another process element in the previous layer on this
process element. Sum function is known to be the function that computes the
impact of inputs and weights absolutely on this process element. An artificial
neural network learns by altering the weights between the neurons in response
to the errors between the genuine output values and the target output values 16.

Artificial Neural Network (ANN) has become a new
tool and an efficient model for the prediction and forecasting of different
water quality variables in river systems. This is caused by the natural
vulnerabilities of contaminant source and water quality data 14. In addition,
ANN is likewise skilled in solving the non-linear functional relationship
between a few parameters and variables involved in the procedure under the
study. ANN is known as a potential and powerful modelling tool due to its
ability to learn and capture the behaviour of any complex and non-linear
process 15.