Numerical simulations for the optimization of online wash systems for gas turbines

Abstract

During gas turbine operation, large amounts of small particles are ingested along with the air inflow. As a result, the blade shape changes over time, hence affecting the air flow pattern and reducing efficiency. Field tests and operator experience have shown that online water washing systems help to regain power lost due to contaminant build up on the compressor blades. This technology involves injecting fine atomized sprays of water through nozzles into the air stream. Poorly designed online washing systems may result in inefficient operation that does not lead to the expected improvement in gas turbine efficiency. For this reason, the efficient design of compressor online washing systems is crucial if the desired output of performance recovery is to be realized. An optimum design requires controlling the water spray characteristics at the inlet to the compressor first stage. The current work is based on a numerical approach to propose a generalized and structured methodology for optimizing gas turbine online washing systems. The approach is applied to optimize the design of the online washing system for the General Electric gas turbine model MS5002. Comparisons between the optimized design and the current design are also presented. Numerical optimization results show that an optimum design of the online washing system for MS5002 gas turbine is feasible for the following nozzles system: 11 axial nozzles and 12 radial nozzles, equally spaced. The optimized nozzle diameter is 1.5 mm, mass flow rate is 0.1672 kg/s, and half-cone angle is 15 degrees for axial nozzles. For the radial nozzles, the optimized nozzle diameter is 1.1 mm, mass flow rate is 0.1533 kg/s, and half-cone angle is 54 degrees. With this system droplet size in the range of 55-150 um, impact velocity of 110-113 m/s, and water covers the entire outlet boundary of bell mouth uniformly.