Parameter optimization of a pure electric sweeper dust port by a backpropagation neural network combined with a whale algorithm

Abstract. Optimizing the structure of suction port is key to effectively improving performance sweeping vehicle. The CFD (computational fluid dynamics) method and gas–solid two-phase flow model are used analyse influence rule structural parameters height above ground on cleaning effect, which verified by real vehicle tests. data set was established an orthogonal test method, a BP (backpropagation) neural network fit evaluation indexes. fitting errors were all within 5 %, indicating that results this good. According relation output, whale algorithm should be further solve optimal parameters. show parameter combination β=63∘, d=168 mm h=12 mm. energy consumption optimized reduced, internal airflow loss reduced. particle residence time becomes shorter, can out from outlet faster, thus dust absorption effect. research provide theoretical reference for optimization matching sweepers.