首页>
外国专利>
Method of estimating flow rate and of detecting leak of wide area water using recurrent analysis recurrent neural network and deep neural network
Method of estimating flow rate and of detecting leak of wide area water using recurrent analysis recurrent neural network and deep neural network
展开▼
机译:递归分析递归神经网络和深度神经网络估计流量和检测广域水泄漏的方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
Regression analysis, regression neural network and deep neural network flow rate prediction and leak detection method according to the present invention, the sensor data pre-processing step for pre-processing the sensor data for the pressure and flow rate over time in the pipeline; A flow estimating process of selecting a flow rate prediction model by evaluating a regression analysis model and a trained deep neural network (DNN) model, and predicting the flow rate using the flow rate prediction model; And a leak detection process for determining a leak in the pipeline and a leak occurrence point using a leak detection model and an abnormality detection model according to a trained recurrent neural network (RNN) model. Differences in pressure, flow rate, and flow rate calculated by pressure and pressure difference can be used to predict pipeline flow rates with high accuracy.
展开▼