Abstract: The non-uniqueness principle of grey system theory means that the solution of grey model is not unique under the condition of incomplete information. GM(1,1) is a grey prediction model with incomplete structural information, but its simulation and prediction results obviously do not meet that principle. To this end, in view of the network model of GM(1,1), the connotation of grey action quantity of GM(1,1) is systematically analyzed, and the interval grey number form of grey action quantity under uncertainty of influencing factors is restored. On this basis, a new mean difference GM(1,1) model is constructed. The new model has a better model structure and is fully compatible with the traditional mean difference GM(1,1). Then the new GM(1,1) model is employed to forecast the electricity energy consumption in China, showing that the reasonableness of the new model is superior. Hence, the grey attribute of the model is restored in this research, which is of positive significance to enrich the theoretical framework and improve the structure system of grey prediction model.

Key words: grey theory; prediction model; non-uniqueness principle; mean difference GM(1,1); grey action quantity; interval grey number