CAJ | 학술논문

缓释制剂的处方优化属于多因素、多水平的复杂优化问题，人工神经网络很适于处理这类复杂的多变量非线性关系。本文在对缓释制剂特性及影响其处方设计质量的重要因素进行细致分析的基础上，应用BP人工神经网络建立缓释制剂处方质量预测模型。研究结果表明，BP神经网络可以有效地进行缓释制剂处方质量预测，是缓释制剂处方优化的有力工具。
완석제제적처방우화속우다인소、다수평적복잡우화문제，인공신경망락흔괄우처리저류복잡적다변량비선성관계。본문재대완석제제특성급영향기처방설계질량적중요인소진행세치분석적기출상，응용BP인공신경망락건립완석제제처방질량예측모형。연구결과표명，BP신경망락가이유효지진행완석제제처방질량예측，시완석제제처방우화적유력공구。
Optimization on sustained release preparation formulation is a multi-factor, multi-level complex optimiza-tion problem. Artificial neural network is very suitable for dealing with such complex multivariable nonlinear system. Based on analyzing the characteristics of sustained release preparation and the main influential factors of its quality, this paper focused on building a quality forecast model for sustained release preparation formulation by using BP neural network. The results showed that the BP neural network can effectively forecast the quality of sustained re-lease preparation formulation. It is a powerful optimization tool of sustained release preparation formulation.