Research on vehicle fuel consumption prediction model based on Cauchy mutation multiverse algorithm

Abstract

The establishment of a truck fuel consumption prediction model is helpful to improve fuel economy and reduce carbon emissions to protect the environment. In the plateau environment, complex geographical conditions have a dramatic impact on fuel consumption. Traditional fuel consumption models can not meet the requirements of plateau prediction accuracy and robustness. In this paper, a back propagation fuel consumption prediction model based on the Cauchy Multi-Verse optimizer (CMVO) considering plateau conditions is proposed. A Cauchy factor is introduced to improve the global search ability of MVO. Moreover, a tangent descent factor is introduced to reconstruct its travel distance rate (TDR), which significantly improves the convergence speed of algorithm. The experimental results show that the convergence time of CMVO-BP is 50% shorter than that of MVO-BP; Compared with logistic regression and RNN algorithm, the prediction accuracy is improved by 5.7%. Under the plateau high-speed condition, the accuracy of fuel consumption prediction can reach 97.5%, R2 coefficient score can reach 95.7.