BP神经网络优化肉桂醛加氢制肉桂醇反应条件

Optimization Reaction Condition of Selective Hydrogenation of Cinnamaldehyde by BP Neural Network

利用均匀设计的实验结果,采用寻优后的具有泛化能力的3层BP神经网络,优化肉桂醛选择加氢制备肉桂醇反应条件。结果表明,催化剂用量对转化率影响最大,当催化剂用量为0.25～0.3g/mL,反应温度95～100℃,反应压力2.5～3MPa时肉桂醛的转化率最高,达100%;温度对选择性影响最大,当溶剂用量为70mL,反应温度80℃,反应压力1～1.5MPa时肉桂醇的选择性最高,为72%。

A BP (back propagation) neural network was trained by the experimental results of homogeneous design using optimal algorithm. Optimization of cinnamaldehyde is carried out through selective hydrogenation. Trained BP neural network for the reaction has shown that the catalyst amount influences the conversion mostly, the highest conversion would reach 100% under the following reaction conditions: the catalyst amount is 0.25~0.3 g/mL, the reaction temperature is 95～100 ℃, and the pressure is 2.5~3 MPa; the reaction temperature influences the selectivity mostly, the highest selectivity would reach 72% under the following reaction conditions: when solvent amount is 70 mL and reaction temperature is 80 ℃, pressure is 1~1.5 MPa.