Operation Optimization of Air Cooling System Based on Random Forest Algorithm

A direct air cooling unit has remarkable water-saving performance, which can effectively solve the contradiction of rich coal and poor water. However, the turbine outlet stream pressure fluctuates greatly at different environmental conditions. This study examines the optimization of the direct air cooling unit's turbine outlet stream pressure operation under variable working conditions. The high back pressure in summer can easily cause the unit to shut down. The condenser tube bundle can also easily freeze during the low-load operation in winter. Optimizing the operating back pressure of the unit is conducive to improving the efficiency and ensuring the safe of the unit. This study determines the influencing factors of the air cooling units. Based on the actual operating data, the calculation model of the cooling system is established. The optimum value of turbine outlet stream pressure at different conditions is obtained. Then, the benefit of under different load and ambient temperature conditions is compared. The operating back pressure is reduced after the optimization control is put into. Furthermore, the effect of optimal control is good.