Research on Prediction of Hydrogen Consumption of Fuel Cell Vehicles Based on Improved GBRT Algorithm

With the increasingly prominent problems of environmental pollution and energy crisis, the use of new energy has gradually become an important research direction. Among new energy sources, hydrogen energy has broad application prospects due to its high efficiency and environmental protection. Hydrogen energy vehicles are one of the important application forms of hydrogen energy. With the advancement of fuel cell technology and hydrogen storage technology, hydrogen energy vehicles have now reached a commercial stage of a certain scale. The fuel cell is the core of the hydrogen energy vehicle. Studying the hydrogen consumption of the fuel cell when the hydrogen energy vehicle is driving under different working conditions is helpful to study the actual performance of the fuel cell. In this paper, the characteristic parameters related to hydrogen consumption are extracted by analyzing the driving data of fuel cell vehicles. Use the extracted feature parameters to train the improved GBRT algorithm model, and use the trained algorithm model to predict the hydrogen consumption status of fuel cell vehicles in the future time period.