A New Approach For Residual Gravity Anomaly Profile Interpretations: Forced Neural Network (FNN)

摘要: This paper presents a new approach for interpretation of residual gravity anomaly profiles, assuming horizontal cylinders as source. The method, called Forced Neural Network (FNN), is introduced to determine the un- derground structure parameters which cause anomalies. New technologies are improved detect borders geological bodies in reliable way. In first phase one neuron used model system and back prop- agation algorithm applied find density difference. second phase, differences quantified mean square error computed. process iterated until small enough. After obtaining results case synthetic data, simulate real Gulf Mexico map, has form anticline structure, examined. Gravity values from cross section this case, result be very close those obtained with proposed method.