The Research Of The Optimal Back Pressure In Direct Air-cooled Condenser Units

In some coal-rich but short of water areas of China, air-cooled condenser units have been widely used. The direct air-cooled condenser units is important to energy saving.After analysis the development, composition and characteristics of direct air-cooled condenser units as well as the effect of direct air-cooled condenser system does in the whole thermal, and thermodynamics-related knowledge is applied here to analyze the cold end model in different working conditions by a mechanism based method, then obtain the dominant factors which affect unit back pressure, including the exhaust steam flew rate, face velocity upon axial flow fan and environmental temperature. The calculation of power dissipation of fan group under different wind velocities can be effectively resolved by similarity law.The establishment of data model is described in detail from some air-cooled condenser units. The model of unit back pressure is established by BP neural networks and least squares support vector machine by using the obtained steady data The evaluation results show that these two models are both characterized by better learning capacity and generalization capacity. Least squares support vector machine is better than BP neural networks.