Evaluation of Bacteriological Parameters in Water Using Artificial Neural Network

This paper deals with the application of artificial neural network (ANN) for the evaluation of bacteriological parameters in water. It dependents on temperature, conductivity, dissolved oxygen, total dissolved solids, depth of water, chlorides, phosphates, nitrates, biochemical oxygen demand, total Kjeldahl nitrogen, fecal coliform, total coliform and fecal steptococci before and after the domestic waste mixing zone of River Kabini, tributary of Cuavery at Nanjanagud, Mandya district, Karnataka. The ANN predicted values are close to the actual laboratory tested values. In this paper 150 actual measured values and laboratory tested values have been taken. For predictions of values using ANN, input and outputs parameters, learning rate parameters, error tolerance, number of cycles to reduce the randomly assigned weights are required, for processing this, the back propagation algorithm and delta rule are required, to input these values to ANN the actual measured and laboratory tested values are used as input and output parameters. The learning rate parameter is 0.55, error tolerance is 0.001 and 5600 number of cycles have been chosen. The first ANN pattern chosen is 10-11-11-3 (ten neuron in input layer, two hidden layers of eleven neuron each and three neuron in output layer) and second parameter is 0.55, error tolerance is 0.001 and 4500 number of cycles, have been chosen. The ANN pattern chosen is 10-12-12-13 (ten neuron in input layer, two hidden layers of eleven neuron each and three neuron in output layer). Back propagation algorithm has been used to train the network, and delta rule is used to adjust the weights and to reduce the errors. The network predicted values, measured and laboratory tested values have been shown in figures and graphs.