1887

Abstract

Backpropagation neural network is trained to recognize and classify constituents of carbonate rock and give appropriate acoustic velocity value to each constituents from their color difference. Carbonate minerals;Calcite and Dolomite are known can be distinguished visually using acid solution alizarin red s where calcite turns red while dolomite remain colorless. pore space which is impregnated using blue dyed epoxy will seen blue. Lavenberg marquadt training method is used to train the network. it reaches convergence after 25 iterations and yields nearly perfect classification. The result is 2D matrix (classified image) where the pixel value represents each constituents. The network is then trained to give the acoustic velocity value to each constituents derived from table of minerals constants. We assume that pore spaces is fully filled by brine water so their acoustic value is brine's. This results a acoustic velocity map or velocity model which is the used in virtual experiment of wave propagation simulation. Acoustic wave is propagated through the velocity model at various frequency from 10-35 KHz. The travel time and velocity are measured at each frequency of wave. we find the positive dispersion where velocity of wave increases as frequency increases.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.20148931
2012-06-04
2021-10-22
Loading full text...

Full text loading...

http://instance.metastore.ingenta.com/content/papers/10.3997/2214-4609.20148931
Loading
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error