中文摘要：

以盐酸环丙沙星、水和乙醇三元物系所得的结晶动力学数据为基础，建立了一个预测结晶动力学数据的神经网络模型。结果表明。用此技术建立的模型对结晶动力学的预测比用传统方法（半经验的回归模型）要精确可靠。

英文摘要：

Neuron network model is constructed to predict crystallization kinetics data, based on the crystallization kinetics data of the system with ciprofloxacin hydrochloride, H2O and ethanol. And the results confirm that this model is better to predict the crystallization kinetics data than other traditional methods.