Open AccessAnalysis of Spray Uniformity of Sprayers Based on Deep Belief Network
Author(s) -
Shujiang Li,
Aijing Guo,
Maoyuan Li
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1626/1/012185
Subject(s) - nozzle , spray nozzle , artificial neural network , deep belief network , process (computing) , spray characteristics , inverse , nonlinear system , computer science , artificial intelligence , process engineering , environmental science , mathematics , engineering , mechanical engineering , physics , geometry , quantum mechanics , operating system
Improving spray quality has a important practical significance for controlling the drift of harmful substances, increasing the effective utilization rate of pesticides and reducing environmental pollution. The influence of factors on spray quality is analyzed, and the nozzle wear is regarded as one of the influencing factors for the first time in this paper. It also constructs the relationship between spray distribution uniformity and influencing factors Model by proposes a soft measurement method based on deep belief network. Besides, when the distribution coefficient of variation and another influencing factors is given, the pressure value corresponding to them can get by neural network inverse modeling method, And the pressure can be used as the controlled variable in the spraying process. Compared to BP, the DBN model has stronger nonlinear mapping capabilities, higher prediction accuracy, and high practical value.