综合采用正交试验与神经网络方法对甘蔗收获机剥叶元件的性能进行了分析优化.通过正交试验初步确定各因素水平的较优值,根据正交试验条件与结果对BP神经网络进行训练,在此基础上拓宽各因素水平的取值,利用神经网络的非线性映射能力对各因素进行仿真分析,从而确定最优组合为:2号材料、螺旋角10°、3排、交错深度6 mm、前角0°.
The orthogonal experiment method and BP neural network are integrated to analyze and optimize the cleaning performance. The orthogonal experiment method is used to determine the optimal combination of factors. And then the conditions and results of orthogonal experiment are employed to train the BP neural network. According to the above results, the levels of factor are broadened to some extent. And the BP neural network with nonlinear mapping ability is utilized to simulate and analyze the performance of cleaning element, in the optimal combination of factors, the materical is No. 2, the spiral angle is 10°, the number of elements is 3, the intersection depth is 6 mm, and the front angle is 0°.