Application of an Artificial Neural Network Model to a Na-K Geothermometer

A new geothermometer model is proposed by applying data obtained from a known Na-K geothermometer to an artificial neural network. In this model, Na and K values were implemented as input signals and geothermometers as the output signal. Multi-layer perceptrons and back propagation were used as training algorithms for the artificial neural network. Reservoir temperatures of some geothermal fields in Turkey determined by this method are in accord with those determined from other methods.