Dingding Chen

Neural network training data selection using memory reduced cluster analysis for field model development

A system and method for selecting a training data set from a set of multidimensional geophysical input data samples for training a model to predict target data. The input data may be data sets produced by a pulsed neutron logging tool at multiple depth points in a cases well. Target data may be responses of an open hole logging tool. The input data is divided into clusters. Actual target data from the training well is linked to the clusters. The linked clusters are analyzed for variance, etc. and fuzzy inference is used to select a portion of each cluster to include in a training set. The reduced set is used to train a model, such as an artificial neural network. The trained model may then be used to produce synthetic open hole logs in response to inputs of cased hole log data.

Variable Input Neural Network Ensembles in Generating Synthetic Well Logs

This paper discusses a hybrid method for construction of neural network ensembles (NNE) in generating synthetic well logs that is often driven by the needs of simulating unobtainable actual logs, reducing the operational cost, reconstruction of missing and/or bad log data, and minimizing the hazards associated with using radioactive sources. In this method, several computer-driven routines are developed to rank the candidate neural network inputs as a function of data partition, network complexity and initialization. Then a network pool is automatically formed having the selected candidate networks characterized with multi-set inputs and different hidden nodes. The ensemble optimization is performed using a multi-objective genetic algorithm by aggregating the ensemble validation error, complexity, and negative correlation into a single quantity of merit. The simulations applied to actual field examples demonstrate that using multi-set-input NNE is more robust than using single-set-input NNE with significantly reduced uncertainty and improved prediction accuracy on the new data for some applications.