Recognition of Steam Jet Condensation Regime in Water Pipe Flow Based on Condensation Pressure Oscillation Features

A series of experiments on steam injected into owing water in a vertical pipe were carried out. The inlet steam flow rate and the owing water temperature were in the range of 0~800 kg•m^-2•s^-1and 30~70℃, respectively. The features of condensation pressure signals at different scales over a time-frequency plane were revealed by the method of wavelet multiresolution analysis. Absolute mean and relative energy of condensation pressure signals at levels 2~5 were selected as the characteristic parameters. Principal component analysis was applied to extract four principal component parameters, and the accumulative contribution rate exceeds 85%. A regime intelligent recognition system is constructed based on the neural network model. The recognition rates of Chugging, Ocil-I, Ocil-II and stable condensation in the pipe flow system were all over 91.7%.