P. Tahmasbi; A. Hezarkhani
Abstract
In the present research, comparative evaluation of various learning algorithms in neural network modeling was performed for ore grade estimation in Sonjail porphyry copper deposit. The main goal of the following investigation would be optimizing the network architecture and to present an architectural ...
Read More
In the present research, comparative evaluation of various learning algorithms in neural network modeling was performed for ore grade estimation in Sonjail porphyry copper deposit. The main goal of the following investigation would be optimizing the network architecture and to present an architectural optimization trend to better performing the copper grade estimation within the region. Therefore, 12 algorithms were investigated back propagation learning algorithms. Based on this research it is merged that by applying the LM and BFG algorithms, there would be the best performance. The reasons why the other algorithms have the same performance would be presented within the paper as well. The input parameters are coordinates and the outputs are the copper grades for each specified point. To obtain the optimal structure, a network with different layers has been applied, which it has acquired 12 neurons within one layer. To investigate the data normal shapes, various normal shape has been acquired in the [0 1], which could merged the best results. Finally to get the best network optimizations several transfer functions have been applied, and the sigmoid transfer function illustrated least error when the transfer function is selected. Considering the optimal conditions, the R2 value has merged 0.946 for network which could be the result showing that the optimal network architecture causes estimation improvement.
A. Hezarkhani; P. Tahmasbi; O. Asghari
Abstract
Separation of alteration zones is one of the important processes in evaluation and identification of mining activities that provide great help to have better view of the region and its mineralization. Most of the alteration separation is based on petrological investigations and the other methods are ...
Read More
Separation of alteration zones is one of the important processes in evaluation and identification of mining activities that provide great help to have better view of the region and its mineralization. Most of the alteration separation is based on petrological investigations and the other methods are less applied. Therefore, in this research, there is an attempt by applying RBPNN (Radial Basis Probabilistic Neural Network) to separate these alteration zones. Because of the special structure and easy designing of these networks, they are usually capable to solve the classification problem. The input data were 28 element analyses related to 45 geochemical samples and its outputs were classified alteration zones (potassic, transition, phyllic) that was coding for every inputs data. After selection the training and testing data, the network has been prepared for training and then the data were inputs and the results were outputs. According to the results, the network could distinguish the difficult spatial relation between the inputs, with 28 spatial variables and classify those correctly. The calculated MSE (Mean Square Error) is 0.0163, which shows the good performance of network in this field.