Comparison of Optimized Neural Network with Fuzzy Logic for Ore Grade Estimation

This paper presents a new approach to improve the performance of neural network method to estimate the grade values. The true value of ore body grade which is calculated based on the accurate data is a challenge of the mining industry. The main goal of the following investigation would be the performance comparison of various learning algorithms in neural network that could apply for ore grade estimation. Up to now, there is not presented procedure to determine the network structure for some complicated cases, therefore; design and production of neural network would be almost dependent on the userʹs experience. To prevent this problem, neural network based on genetic algorithm (ANN-GA) was applied to present the advantages, e.g. assay estimations. To show the performance of this method, three prevalent estimation methods such as artificial neural network (ANN) and fuzzy logic (FL). One of the most important problems in neural network designing is the topology and the value parameter accuracy that if those elements selection was correctly and optimally, the designer would achieve a better result. To test this new method, it was evaluated by a case study. First, the parameters and topology of neural network determined without applying the genetic algorithm (trial and error method) and in order to consider the effect of genetic algorithm on results, it was applied GA to obtain the parameters (the input number, number of neurons and layers in the hidden layers, the momentum and the learning rates) and then network performance. The results indicate that this method could improve the network performance rather than ANN and FL, also the mean square error (MSE) and R values decreased and increased respectively