Full text loading...
-
Fuzzy Rules Based System in order to Characterization of Cavities from Gravity
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 80th EAGE Conference and Exhibition 2018, Jun 2018, Volume 2018, p.1 - 5
Abstract
In this paper a new method is presented for shape factor estimation of microgravity anomalies. The method is based on training of a designed neural network (NN) with a vast training data of several objects with different shapes as: Vertical Cylinder, Sphere and Horizontal Cylinder. The input of the NN is the residual anomaly of the selected principle profile and the output is the depth and shape factor of the related object. To extract the most probable shape of the object, three main If-then fuzzy rules are used and the membership degree of the shape to the fuzzy set of {near to: Vertical Cylinder, Sphere and Horizontal Cylinder} is achieved. The method is tested for real data measured on a mining shaft in Kalgoorlie Gold Mine in Australia. The most advantage of the method is its ability to specify how much the gravity target is near to the estimated shape and depth with no need to pre-assumption about its shape.