Article Information
Determination of Turbulence Coefficient using Genetic Algorithms

Keywords: Turbulence Loss Coefficient, Well Hydraulic Parameters, Optimization, Genetic Algorithm.

Mehran University Research Journal of Engineering & Technology

Volume 31 ,  Issue 4

Ashfaque  Ahmed  Memon,Naeem  Aziz  Memon,Kamran   Ansari

Abstract

An effort has been made to optimally determine the status of turbulence loss coefficient in the present study. The parameter is identified in the framework of GA (Genetic Algorithm), which consequently resulted in the development of a computer model. In accordance with principles of the GA, the objective function, sse (sum of square of errors), is minimized between 0.009785 and 0.017565; as a result the well hydraulic parameters are optimally identified. To check validity of the model, simulated draw downs are compared against the observed ones, which indicate mean difference between them varying from 0.0049m to 0.0124m. Furthermore, validity of the model is also endorsed through statistical analysis with model efficiency varying between 99.97 and 100.00%. The model is applied to 5 data sets of step drawdown pumping test, yielding 5 values of turbulence coefficient varying between 1.01 and 2.08. Out of these 5 optimized values of turbulence coefficient none of the values is equal to 2. This scenario of variation of turbulence coefficient substantiate that turbulence coefficient is a variable and not a constant (i.e. equal to 2.0, as suggested by Jacob and Singh) while considering turbulence loss coefficient as a constant is discarded.