The Soft Sensor Method For Double-elbow Solid Mass Flow Measurement

Gas-solid two phase flow is one of the most important branches of multiphase flow,for the complexity of its flow state and the diversity of its flow pattern directly influence detection precision of the solid mass flow rate, it’s difficult to meet the requirements of industrial field, which severely restricted the application of gas-solid two phase flow in industrial production.Many detection methods of gas-solid two phase flow mass flow rate are studied,especially the double elbow method. Its measuring principle is one elbow flow meter which is installed before gas-solid mixing point to detect gas phase differential pressure,the other is installed after mixing points to measure the differential pressure when the gas-solid two phase flow through the elbow meter. Two soft measurement models are established based on BP neural network and RBF network with the input of the pressure difference1?p between inside and outside of elbow wall before gas solid mixed, and the pressure difference2?p after gas solid mixed. The measuring relative error is within10% and 7%, respectively.In order to improve the double elbow method measuring principle, the solid phase particles space into the principle should be considered Two soft measurement models are established based on BP neural network and RBF network with the input of the pressure difference1?p between inside and outside of elbow wall before gas solid mixed, and the pressure difference2?p after gas solid mixed and12???pp. By comparison the RBF soft measurement model with three input variables is better with fast learn speed and strong approximation ability, and its relative error is within 3%, which can meet the requirement of industrial field and provide an effective method to measure the solid mass flow rate of gas solid two phase.