Abstract
Artificial neural network are closely modeled on biological processes for data processing, specifically the nervous system and its basic unit the neuron. The neuron receives multiple signals from other neuron through its dendrites each signal multiplied by a weighting function (Coefficient). These signals are added in the cell body or soma and when each component signal reaches a threshold value, a signal known as action potential is sent out through the axon which is the neurons output channel. Data obtained from a case study well were modeled and analyzed for predicting the pressure profile. The result showed that the pressure profile increased from 0.00 psia/cp and had an asymptotic limit at 220000 psia/cp at 400 psia-pressure and for temperature 100°F respectively. Then it increased from 0.01 psia/cp to 0.18 psia/cp (specific gravity 1.2) at a temperature of 100°F, the result was compared with those of Earlougher [3] and found to be valid for (1) radial flow (2) isotropic and homogeneous medium (3) general gas law applies. The case study well-opukusi 8 (escravos swamp field) south east Nigeria was monitored for 36 days period when it was shut-in for the pressure to build up and attain steady state to enable the prediction of productivity index which has the value of 0.60 which is indicative of a declining productivity.