Comparison of Estimation Sonic Shear Wave Time Using Empirical Correlations and Artificial Neural Network

Wellbore instability and sand production onset modeling are very affected by Sonic Shear Wave Time (SSW). In any field, SSW is not available for all wells due to the high cost of measuring. Many authors developed empirical correlations using information from selected worldwide fields prediction. Recently, researchers have used different Artificial Intelligence methods estimating SSW. Three existing Carroll, Freund, Brocher estimate in this paper, while a fourth new correlation established. For comparing with results, another study's Neural Network (ANN) was used. The same data that adopted ANN study here where it comprised 1922 measured points other nine parameters Gamma Ray, Compressional Sonic, Caliper, Neutron Log, Density Deep Resistivity, Azimuth Angle, Inclination True Vertical Depth one Iraqi directional well. based only on (CSW) predicting way developing previous correlations, CSW. A comparison demonstrated utilizing better higher R2 equal 0.966 lower statistical coefficients than four Brocher, had 0.7826, 0.7636, 0.6764, 0.8016, respectively, show best three existing. use or future calculations relevant decision makers number limitations target accuracy.