期刊名称：International Journal of Hybrid Information Technology
印刷版ISSN：1738-9968
出版年度：2016
卷号：9
期号：9
页码：57-66
出版社：SERSC
摘要：In sugarcane combine harvesting, not hulling rate and damage rate of sugarcane in the process of leaf stripping are rather high. Thus, multiple flexible leaf stripping system was proposed. Making using of general multiple flexible leaves stripping to sugarcane, leaf-stripping device with flexible plate teeth was developed. On the basis of the four-factor second order rotation combination experiment, a neural network prediction model between working performance influence factor and index was established. And according to using BP neural network and Genetic Algorithm, the facility parameter optimization was performed. The best parameter combination for the facility was the rotational speed of inlet roller623.2r/min, the rotational speed of first order leaf-stripping roller 951.6r/min, the rotational speed of second order leaf stripping roller 1129.4r/min, the rotational speed of deliver roller 846.7r/min. The results show that a new approach for performance prediction model building and parameter optimization of sugarcane leaf-stripping device with flexible plate teeth, combining BP neural network with Genetic Algorithm is reliable, accurate and feasible.
