基于BP神经网络和AHP的软胶囊滴丸产品质量软测量建模

为了解决软胶囊滴丸产品质量控制系统的建模问题，文中提出了一种基于BP神经网络和层次分析法（AHP）的二级软测量建模方法．首先详细分析了软胶囊滴丸生产过程质量控制的性能要求和影响产品生产过程质量的因素，给出软胶囊滴丸产品的二级层次结构质量指标体系和操作变量；在此基础上，利用BP神经网络进行建模，获得以工艺参数为输入，以各二级质量指标为输出的软胶囊滴丸生产过程质量子模型；然后，用AHP确定各二级指标的权重，最终建立软胶囊滴九生产过程质量控制的二级软测量模型．仿真实验结果表明了该方法的有效性．

A Capsule Drop Quality Modeling Based on BP Neural Network and AHP

A novel method aiming at the modeling of capsule drop quality control system was presented based on BP neural network and AHP（ analysis hierarchy process）. The process parameters and a two-level hierarchy index system of soft capsule drops quality was proposed based on the analysis to the production process. On this basis, the model was established based on BP neural network, whose inputs are the process parameters and outputs were the secondary quality indexes. AHP was then used to determine the weights of the secondary quality indexes and a two-level hierarchy index system model of capsule drop quality was established. The simulation resuits show the effectiveness of the method.