Fouling in the compressor and turbine, from particles (such as dust, pollen, seed, insect and unburned hydrocarbon) deposited on the blade results in the change of blade profile, thus flow area is reduced. Many researches have demonstrated that fouling can result in the reduction of power and efficiency, and can result in the increase running costs. At present, the most effective control method for gas turbine fouling is off-line washing, but off-line washing may seriously affect the unit’s benefit. So, it is very important to determine the optimal washing cycle. In previous studies, many researchers determine the washing time based on the reduction level of thermodynamic parameters, such as the change of compressor pressure ratio. But, research demonstrated that other failure modes (such as erosion, corrosion and tip clearance increase) can result in the similar change. So, it is not reasonable based on the change of single thermodynamic parameter. However, the fouling in the compressor and turbine cannot directly be observed. In this paper, washing strategy model is proposed. At present, off-line washing criteria about gas turbines are mostly established according to compressors of clean gas turbines and the washing cost is neglected. This model pay attention to economical off-line washing interval based on the principle of minimum mean generalized crank washing cost. For the performance deterioration rate of gas turbine due to fouling, it can be obtained by combining reverse method with optical measuring technology. The washing cost includes running additional cost and shutdown cost. For running additional cost, it includes two aspects. The first running additional cost is the cost of power reduction due to the decrease of mass flow rate. The second running additional cost is the cost of the heat rate increase due to the decrease of gas turbine efficiency. For downtime cost, it is the cost of power loss due to the unit downtime. The calculation result indicates the optimal off-line interval is approximately 80 hours. Compared with traditional washing criteria, the model improved the availability of equipment. Compared with the actual interval, the model improved operation security. Finally, the influence factors such as washing number, shutdown time, and fuel price and degradation rate are analyzed. The result shows the model is greatly influenced by these factors.

This content is only available via PDF.
You do not currently have access to this content.