Abstract: For heap leaching of uranium ore, the seepage of the leaching solution in the uranium ore heap has significant effect on the leaching behavior. Therefore, it is important to study the permeability characteristics of the uranium ore heap. In the present work, ten groups of samples with different particle size distribution fractal dimensions were prepared using the fragmented uranium ore for heap leaching taken from a uranium mine in South China; experiments were conducted to obtain the permeability and the flow state index of each sample using the self manufactured apparatus for liquid saturated seepage experiment; analyses were made for the effect of the particle size distribution fractal dimension on permeability of the samples; and support vector machine (SVM) was used to establish the SVM models for predicting the permeability and the flow state index, respectively. The results show that, under laboratory conditions, the liquid saturated seepage in the uranium ore heap for heap leaching follows non-Darcy exponential law with flow state index ranging from 1.1 to 1.5; the permeability decreases gradually with the increase of the particle size distribution fractal dimension; and the SVM models for predicting the permeability and the flow state index can give the predicted results with relative errors of less than 8% and 7%, respectively; so as to satisfy the requirements for engineering application.

Key words: heap leaching, uranium ore heap, liquid saturated seepage, support vector machine (SVM), fractal dimension