Numerical Simulation And Intelligent Measurement Research Of Humidity Distribution In Wet Steam Condensation Flow

The steam turbine is a prime mover in the thermal power plant, and its work security and economy directly influence the economic benefit of power plant operation. In large-scale firepower aircrew, the admission of steam turbine is superheated steam, but wet steam in the last stage blade of low pressure cylinder. Therefore, considering the influence of the moisture steam on the security and economy of steam turbine operation, it is necessary and of great significance in theory and practice to collaborative simulate and intelligent measure the humidity distribution in the last stage of steam turbine.Supported by the Hunan Provincial Natural Science Foundation of project " Investigation of Exhaust Wetness and Diameter of Water Droplet Measurement Method in Steam Turbine Based on Mie Scattering of Laser ", combined with field synergy theory, grey relational analysis method and support vector machine method, numerical simulation and intelligent measurement research of humidity distribution in wet steam condensation flow are studied systematically. The main research contents and innovations are expressed as follows:(1)The model of wet steam condensation was set up based on the theory of droplet nucleation and growth. Then simulated based on computational fluid dynamics theory and field synergy principle, the distribution characteristic of humidity in condensation of wet steam was obtained.(2) The grey relational analysis method was used to deal with affecting factors of humidity measurement performance and further define the primary and secondary impacting factors. The results provide theory foundation for simplifying the operating conditions of humidity measurement in condensation of wet steam flow.(3)The prediction model of the humidity distribution in steam turbine was established according to support vector machine. After that, the parameters of support vector machine were optimized by genetic algorithm. The results indicate that the least relative error of intelligent measurement model of the humidity in wet steam is 2.2% by using nonlinear correction method.