关键词: MATLAB, 人工神经网络（ANN）, BP网络

Abstract: Artificial nerve network（ANN） is an interdisciplinary subject rising in recent years. In forecast, ANN will establish a non-linear reflection relation between the input and output , then automatically simulate the natural relation between the various mineralization factors and carry out the whole optimal searching, thus reducing the human intervention and improving the accuracy of the resource forecast. The back propagation network-BP network is mostly used. As MATLAB can provide many tool boxes for tracing the foreign advanced computing methods and mathematical models, the learning, training, fitting and forecast (simulation) processes of BP network model can be conveniently realized by using the nerve network tool boxes in MATLAB. Based on this thought, taking as example the forecast for the lead-zinc ore prospect area in Xunbei region, Shaanxi Province, BP artificial nerve network system for mineral resource forecast is set up by transferring the internal functions on MATLAB platform, based on which the forecast for the perspective area is made.

Key words: MATLAB, Artificial nerve network（ANN） BP network