Objective The controllability changes of structural brain network were explored based on the control and brain network theory in young smokers, this may reveal that the controllability indicators can serve as a powerful factor to predict the sleep status in young smokers.Methods Fifty young smokers and 51 healthy controls from Inner Mongolia University of Science and Technology were enrolled. Diffusion tensor imaging (DTI) was used to construct structural brain network based on fractional anisotropy (FA) weight matrix. According to the control and brain network theory, the average controllability and the modal controllability were calculated. Two-sample t-test was used to compare the differences between the groups and Pearson correlation analysis to examine the correlation between significant average controllability and modal controllability with Fagerstr?m Test of Nicotine Dependence (FTND) in young smokers. The nodes with the controllability score in the top 10% were selected as the super-controllers. Finally, we used BP neural network to predict the Pittsburgh Sleep Quality Index (PSQI) in young smokers.Results The average controllability of the dorsolateral superior frontal gyrus, supplementary motor area, lenticular nucleus putamen, and lenticular nucleus pallidum in the young smokers’ group, and the modal controllability of the orbital inferior frontal gyrus, supplementary motor area, gyrus rectus, and posterior cingulate gyrus, were all significantly different from those of the healthy controls group (P<0.05). The average controllability of the right supplementary motor area (SMA. R) in the young smokers group was positively correlated with FTND (r=0.393 0, P=0.004 8), while modal controllability was negatively correlated with FTND (r=-0.330 1, P=0.019 2).Conclusion The controllability of the structural brain network in young smokers is abnormal. which may serve as an indicator to predict sleep condition. It may provide the imaging evidence for evaluating the cognitive function impairment in young smokers.