The invention relates to controlling combustion dynamics in a gas turbine engine. More particularly, this invention relates to controlling combustion dynamics by biasing airflow to a combustion flame in the gas turbine engine.
Gas turbine engines are known to include a compressor for compressing air, a combustor for producing a hot gas by burning fuel in the presence of the compressed air produced by the compressor, and a turbine for expanding the hot gas to extract shaft power. Gas turbine engines using annular combustion systems typically include a plurality of individual burners disposed in a ring about an axial centerline for providing a mixture of fuel and air to an annular combustion chamber disposed upstream of the annular turbine inlet vanes. Other gas turbines use can-annular combustors wherein individual burner cans feed hot combustion gas into respective individual portions of the arc of the turbine inlet vanes. Each can includes a plurality of main burners disposed in a ring around a central pilot burner.
During operation, the combustion flame can generate combustion oscillations, also known as combustion dynamics. Combustion oscillations in general are acoustic oscillations which are excited by the combustion itself. The frequency of the combustion oscillations is influenced by an interaction of the combustion flame with the structure surrounding the combustion flame. Since the structure of the combustor surrounding the combustion flame is often complicated, and varies from one combustor to another, and because the combustion flame itself may vary over time, it is difficult to predict the frequency at which combustion oscillations occur. As a result, combustion oscillations may be monitored during operation and parameters may be adjusted in order to influence the interaction of the combustion flame with its environment.
A combustion flame emits sound energy during combustion. A more uniform flame will generate more uniform acoustics, but perhaps with higher peak amplitude at a particular frequency than a less uniform flame. When an emitted frequency of combustion coincides with a resonant frequency of the combustion chamber the system may operate in resonance, and the resulting combustion dynamics may damage the gas turbine components, or at least reduce their lifespan.
One known way to reduce the interaction of the combustion flame with the combustion acoustics is to reduce the coherence of the flame, i.e. reduce the spatio-temporal uniformity of the flame. A flame with less uniform combustion throughout its volume is likely to perturb the gas turbine less than a uniform flame because the energy released is spatially distributed and therefore decreases its coupling to the system resonant frequencies or acoustic modes. This is the well known Rayleigh criterion. As a result, combustion dynamics of flames with less uniform combustion throughout its volume are less likely to be exacerbated than by a more uniform flame.
One way that has been utilized to reduce flame coherence has been to vary the fuel/air ratio throughout the flame. Main premix burners often have a swirler that swirls an airflow flowing through the burner. Fuel outlets in the burner introduce a flow of fuel into the airflow to produce a fuel/air mixture of a certain ratio. The fuel/air ratio from main burners may be varied. For example, some of the main burners of a combustor may be controlled by one fuel stage, and the remaining burners of the combustor by another stage. Since the structure of the main burners and swirlers in them are uniform throughout the burners in the combustor, varying the fuel from burner to burner varies the fuel/air ratio. Since each fuel/airflow has a different amount of fuel when it reaches the combustion flame, the combustion/temperature of the combustion flame varies throughout its volume and the flame is less coherent.
Such a fuel biasing of the combustion flame has drawbacks. Separate fuel stages are very expensive to manufacture and complicated to operate. Further, localized regions of leaner and richer combustion within the combustion flame produce less than optimal emissions.
Another way that has been utilized to reduce flame coherence has been to vary portions of the combustion flame axially with respect to other portions of the combustion flame which results in a less uniform combustion flame, thereby reducing combustion dynamics. This has been accomplished, in one example, by increasing the volume of fuel/air flow through one burner with respect to another burner. This has also been accomplished by positioning burners in different locations axially with respect to other burners in a combustor. However, these configurations may not work under all situations, so there remains room in the art for combustor configurations to reduce flame coherence and associated combustion instabilities.
The invention is explained in the following description in view of the drawings that show:
The inventors have devised an innovative way to configure a combustor utilizing premix main burners (i.e. burners) so that different burners will deliver fuel/air flows having a differing parameter which will, in turn, reduce flame coherence and associated combustion dynamics. The differing parameter need not be the fuel/air ratio, so that combustion dynamics may be controlled without sacrificing optimized emissions.
Each fuel/air flow may be characterized by the same swirl number but a different mass flow rate. The swirl number (S) is defined as the ratio of the axial flux of the angular momentum (Gϕ) to the axial thrust (Gx) times the exit radius (R),
In an embodiment the fuel/air flows emanating from each burner may have the same fuel/air ratio. As a result of a uniform fuel/air ratio from burner to burner, localized areas of varying temperature within the combustion flame may be reduced or eliminated. By eliminating these localized areas, the less than optimal emissions associated with them are also eliminated.
A different flow from one burner to the next may result from directing differing flows to respective burners, or by varying the geometry within a burner to influence the airflow there through, or both. Maintaining the same fuel/air ratio may be accomplished by mechanically configuring each fuel outlet to produce this result, or by fuel control via staging, or a combination of both.
As can be seen in
When the flows into the main burners 14, 15 are conditioned in this manner the swirlers (not shown) within the main burners 14, 15 may be the same throughout all the main burners 14, 15. In this manner the respective flow of air that does make it to a particular burner will be subject to the same swirl as other flows. The only thing that will change is the mass flow rate of air flowing through the particular burner with respect to other burners. As a result this configuration for conditioning respective flows lends itself well to a retrofit application, where a flow conditioning plate 24 may be installed on existing combustors 10. Adding a flow conditioning plate 24 to existing combustors 10 is a simple and relatively inexpensive way to condition the supply flow 20 into flows tailored for respective burners. Since most combustors 10 that could be retrofitted in this manner already have fuel staging, the fuel staging may be adjusted as necessary to produce the same fuel/air ratio from each burner, which would reduce or eliminate varying temperature within the combustion flame, thereby reducing emissions. It is also envisioned where the fuel/air ratio may still be varied in fuel/air flows from burner to burner. This provides an added degree of control and/or fine tuning. Similarly, the fuel/air ratio may be adjusted during operation such that at times the fuel/air ratios of all the respective flows are the same, and at other times, the fuel/air ratio of all the respective flows are different. This may be necessary when other factors are considered, such as transient operating conditions etc. It is also envisioned that the flow conditioning plate 24 may be used in conjunction with the teachings below.
Further, for sake of simplicity it has been assumed that the supply air 20 may have an essentially uniform pressure throughout its volume before being conditioned when a flow conditioner 24 is used. The same assumption is made about the region into which the airflows leaving the burners flow. This simplification contributes to a more ready understanding of the invention because the pressure drop from before the conditioning plate 24 to the region downstream of the burners would be the same regardless of what path the supply air takes between the conditioning plate 24 and the region downstream of the burners. Thus it is easier to envision how different burner/swirler geometries may influence the flow through the respective burner. Similarly, in embodiments where no conditioning plate 24 is used, it is assumed that the supply air 20 may have an essentially uniform pressure throughout its volume before entering respective burners, and after leaving the burners. Here again it is easier to envision how different burner/swirler geometries may influence the flow through the respective burner. However, the inventors understand that pressure variations may occur throughout the volumes of each of these areas of assumed uniform pressure, and these pressures and locations of pressure variations may change during operation. In embodiments where all main burner fuel outlets are controlled by a single stage and uniform fuel/air ratios among all flows are desired, it is understood that perfect uniformity for fuel/air ratios may not always be achieved. Such operating variations are envisioned and may be tolerable, depending on the design. Such variations are likely to be less than variations present in existing fuel biasing combustors, and so combustors as disclosed herein are still likely to have improved emissions when compared to fuel biasing combustors. Minor lack of uniformity may be tolerable if, for instance, the cost saving associated with a single stage controlling the fuel to all the main burners 14, 15 is preferred. When more uniformity is desired then staging the control the fuel among the main burners may be preferred, despite the added cost.
When the diameters of respective swirlers differ, but the swirlers are aerodynamically proportional, the fuel/air ratio of the flows from respective burners can be varied or can be the same. In an embodiment where the same fuel/air ratio is desired for all flows, this can be accomplished by mechanically configuring the respective fuel outlets without the need for staging among the main burners 14, 15, or by utilizing staging among the main burners 14, 15, or both. In an embodiment where the fuel/air ratio is to be the same from burner to burner, and the fuel outlets are mechanically configured to produce consistent fuel/air ratios throughout, multiple stages of fuel to control fuel to the main burners 14, 15 may not be needed. This is particularly advantageous because fuel staging is expensive to manufacture, operate and maintain. Eliminating a fuel stage for the main burners 14, 15 would result in a significant cost savings, without sacrificing the needed control over the combustion dynamics, and may even improve emissions over staged/fuel biasing schemes. Nonetheless, it is envisioned that staging among main burners 14, 15 may still be desired, and may afford a greater degree of control over combustion dynamics and emissions. The balance of cost versus desired control may determine which ultimate configuration is chosen, and this flexibility is the result of this innovative approach.
In another embodiment, the airfoils 36 of one swirler may be a different thickness than airfoils 36 of another swirler. If the remainder of the geometry is the same among swirlers, then the thicker blades of one swirler 36 will restrict the air flowing through that swirler. The mass flow rate of the air through the swirler is thus reduced, but the flow is characterized by the same swirl number as a flow emanating from a burner where the swirler airfoils 36 are relatively thinner. This can be seen in
This configuration may likewise be designed to produce the same fuel/air ratio in all fuel/air flows, or different fuel/air ratios. If the same fuel/air ratio is desired, the fuel outlets can be configured mechanically do produce the desired fuel/air ratios, without staging among the main burners 14, 15. The fuel may also be controlled with staging among the main burners 14, 15. Both techniques may also be used together to control fuel/air ratios.
Also shown schematically in
In another embodiment individual airfoils within one swirler may differ in geometry from other airfoils in the same swirler. Only one swirler may have airfoils of differing geometry, or as many as all of the swirlers may have airfoils of differing geometry. For example
In another embodiment the shape of the airfoil within the swirler differs from blade to blade within the swirler. For example, in the previous embodiments the discrete flow paths between adjacent airfoils in a swirler may have a rectangular cross section. As seen in
It can be seen that the inventors have devised an air biasing structure capable of reducing flame coherence, and associated combustion dynamics, in a manner not yet seen in the art. This structure provides greater design flexibility without sacrificing necessary control over combustion dynamics. Further, when the fuel/air ratio of all fuel/air flows flowing into the combustion flame are kept the same an entire stage of fuel controls for the main burners may be removed, saving substantial manufacturing and operating costs, while reducing emissions over fuel biasing schemes of the prior art.
While various embodiments of the present invention have been shown and described herein, it will be obvious that such embodiments are provided by way of example only. Numerous variations, changes and substitutions may be made without departing from the invention herein. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5197290 | Lee et al. | Mar 1993 | A |
5277023 | Bradley et al. | Jan 1994 | A |
5628182 | Mowill | May 1997 | A |
5735681 | Cheng | Apr 1998 | A |
6253538 | Sampath et al. | Jul 2001 | B1 |
6269646 | Lovett et al. | Aug 2001 | B1 |
6920758 | Matsuyama et al. | Jul 2005 | B2 |
6931853 | Dawson | Aug 2005 | B2 |
6966186 | Bachovchin et al. | Nov 2005 | B2 |
7181916 | Ziminsky et al. | Feb 2007 | B2 |
7631499 | Bland | Dec 2009 | B2 |
20040093851 | Dawson | May 2004 | A1 |
20060263731 | Hasegawa et al. | Nov 2006 | A1 |
20070119183 | Evolut | May 2007 | A1 |
20080072605 | Hagen et al. | Mar 2008 | A1 |
20080163627 | ELKady et al. | Jul 2008 | A1 |
20090255259 | Kastrup et al. | Oct 2009 | A1 |
Entry |
---|
Lefebvre, Arthur H. and Ballal, Dilip R; Gas turbine combustion: alternative fuels and emissions; 2010; Taylor & Francis Group, LLC; 3rd edition; p. 114-143. |
Number | Date | Country | |
---|---|---|---|
20120006029 A1 | Jan 2012 | US |