摘要

In order to increase grey prediction precision for non-equigap series, this paper proposes an extended GM(1,1) power model for non-equidistant series (non-equidistant GM (1,1) power model) and constructed a non-linear optimization model to obtain the whitened parameters in the model. The results show that the form of non-equidistant GM (1,1) power model is more flexible than the traditional ones. Finally, an engineering example of non-equidistant sequence is cited, which is significantly improved by non-equidistant GM(1,1) power model.