Identification of Flow Regimes Based on Adaptive Learning and Additional Momentum BP Neural Network

BP neural network is one of traditional methods to solve the inverse problems in Electrical Capacitance Tomography (ECT). To adopt this method, simple problems in industry can be solved well, but for the actual complicated industry environment it is limited. In this paper, based on the analysis of disadvantages in traditional BP neural network, adaptively adjustment learning rate is adopted and additional momentum factor are imposed. In the improved network, the capacitance values are input to train to obtain a mature network to identify the flow regimes. According to the experiment result, compared with the traditional BP neural network, the convergence speed is increased and the tending to local minimum is solved, which supplies a new method for flow regime identification in ECT system.