神经网络专家系统在试井解释中的应用

论述了应用专家系统和人工神经网络自动识别试井解释模型的新方法．首先应用人工智能对试井解释模型的压力导数图进行识别，建立了用专家系统构成模块式识别系统；然后应用人工神经网络实现试井解释模型识别；该系统能根据新的试井资料识别试井解释模型；最后，神经网络和专家系统相结合用于试井解释．油田实际试井数据的识别结果说明该方法是一种非常有效的识别方法，对不完全的，有噪声的试井数据也能有效地识别，识别准确率较其它方法都有较大提高

Application of Neural Networks and Expert System in the Interpretation of Well Test

The paper expounds in detail a new method of identifying the well test interpretation model by using the expert system and artificial neural networks.First,artificial intelligence was used to identify the pressure derivative plots of well test model,and a moduler recognition system constructed with expert system was set up.Then the artificial neural networks was used to identify the well test interpretation model.This system can identify the well test interpretation model according to the new well test data.Finally the neural networks and expert system were used together for well test interpretation.Practical identification results show that the method is very effective. It can identify incomplete well test data even with noise.The identification accuracy is very high compared with other methods.