应用BP神经网络进行了管道缺陷参数的定量识别，识别结果的误差小于10%，完全满足实际检测的要求。
Use BP neural network to quantitatively recognize pipeline defect parameter, the error of recognition result is below 10%, and requirement of practical inspection can be completely fulfilled.
分别采用人工神经网络法和逐步回归分析法对原油管道蜡沉积实验数据进行分析处理，建立蜡沉积速率模型。
As the artificial neural network and stepwise regression analysis are respectively applied to the data of crude wax deposition experiment, the model of wax deposition velocity was established.
为了解决目前管道流量泄漏监测和定位问题，根据管道泄漏应力波和正常信号的区别，建立神经网络模型。
To solve the leak detection and location problem in pipeline flux, we built neural network model based on the difference of leak wave and normal signal.
应用推荐