Firstly, the mineral contents are c

Firstly, the mineral contents are calculated applying the evolutionary complex method. Because the objective function is a fourdimensional optimization problem, the number of vertices in each subpopulation is set to 8 in the program. To avoid the result falling into the local extreme value due to the small initial population, it is set 9 subpopulations. The local evolution is set to 15 times and global iteration is 20 times. The variation of the minimum objective function value of each subpopulation with the global iteration is shown in Fig. 10. In Fig. 11, the four columns on the left are the comparisons between the actual logging curves and the calculated ones. Obviously, they have a good coincidence. The five columns on the right are the mineral contents and the corresponding lithology, respectively. It can be seen that the tuff content is higher in the tuffaceous breccia and tuff, which is reasonable in reality. It indicates that the proposed volume model is suitable for both volcanic lavas and pyroclastic rocks. The specific mineral parameters used in the program are displayed in Table 1. Table 2 demonstrates the feasibility of the volume interpretation model and the effectiveness of the estimated mineral contents with the mean relative errors less than 3%.