Parameter optimization for nonlinear grey Bernoulli model on biomass energy consumption prediction

Abstract

Nonlinear Grey Bernoulli Model (NGBM(1,1)) and its derivative model utilize the specific power exponent function to manifest the nonlinear characteristics of the energy consumption data pattern. Because the modeling constraint conditions and the data processing mechanism are rarely considered in parameter optimization of NGBM(1,1) the aim of this paper is just to establish a novel NGBM(1,1) optimization model with constraints using Box–Cox transformation (BC-NGBM*), in which the constraint conditions of the power index in the power function transformation are discussed according to the principle of difference information and the data processing mechanism. Parameter optimization of BC-NGBM* would be solved collectively using Quantum Adiabatic Evolution (QAE) algorithm. 143 data sets from M4-competition are studied for confirming the effectiveness of BC-NGBM* with QAE algorithm Finally, using data from 2010 to 2018, BC-NGBM* is used to forecast biomass energy consumption in China, the United States, Brazil, and Germany. The proposed model demonstrates high accuracy in all cases and is efficient for short-term biomass energy consumption forecasting.