【摘要】 针对所建立的具有仿人智能和机器学徒特征的自组织经验进化预测模型，运用软件工程方法学系统分析了预测软件的性能要求与模块组成，设计和开发了相应的软件。使用面向对象设计方法开发的预测软件结构合理、界面友好，具有联机帮助功能。以高炉铁水硅含量预测为案例对预测软件进行了测试。在线测试结果表明：设计的软件运行稳定，在具有一定预测经验的前提下，软件系统的预测效果优于对应条件下的人工预测结果。更多还原

【Abstract】 A self-organized prediction system based on experience evolution, which has characteristic of simulating human and machine apprentice, is investigated by developing prototype of the model with software engineering methodology and by analyzing the properties and function of the prediction system. The prediction system, which is with online help, is developed with object-oriented methodology. It is therefore function-satisfactory, user-friendly. The case of prediction for silicon content in hot metal is tested in a commercial blast furnace by using the prediction software. It shows that the software works well in the online trial and the prediction results are better than these made by human expert when the system has some prediction experience.更多还原