Patent Application
20170314778
This Application claims priority under 35 USC 119 to German Application No. DE 10 2016 005 155.8, filed Apr. 28, 2016 the entire disclosure of which is incorporated herein by reference.
The invention relates to a pulsed combustion system for material treatment or material synthesis (pulse dryer, pulse combustor, pulsation reactor), having at least one burner, at least one pulsating flame, a combustion chamber and at least one gas column that is capable of resonating (e.g. in the combustion chamber or in a resonance tube), wherein a raw material that is to be treated can be introduced into and discharged from this system, and also relates to a corresponding method for operating the system.
The vast majority of all technical or industrial firing installations and combustion systems are configured and operated such that the combustion process proceeds on average in time-constant fashion with the exception of minor turbulent fluctuations, the magnitude of which is at least one order of magnitude smaller than the average values of the combustion process (such as the average flow velocity, the average temperature of the flame or of the exhaust gas stream, the average static pressure in the combustion chamber, etc.). This means that the conversion of the fuel used is continuous over time and—as a consequence—the release of heat from the combustion process and the mass flow of resultant exhaust gas (combustion products) also have time-constant values for a fixed burner setting.
Departing herefrom, phenomena or “abnormalities” occasionally arise which, in the literature, are termed combustion chamber pulsations, self-induced combustion instabilities or thermoacoustic pulsations. These are characterized in that, on reaching a stability limit, the initially stationary (i.e. time-constant) combustion process suddenly automatically tips over into a time-periodic, pulsating combustion process which is self-induced and whose time function can be closely approximated as sinusoidal. Attendant with this change, the heat release rate(s) of the flame(s) and thus the thermal combustion power of the combustion installation and also the exhaust gas stream in and from the combustion chamber and also the static pressure in the combustion chamber themselves also become periodic-transient, i.e. pulsating.
The apparition of these combustion instabilities often results in a different pollutant emission regime compared to stationary operation of the combustion system, and also causes, in addition to increased noise load to the installation surroundings, markedly increased mechanical and/or thermal loads on the installation structure (e.g. combustion chamber walls, combustion chamber cladding, etc.), which can even destroy the combustion system or individual components.
It is therefore easy to see that it is imperative to avoid the undesired apparition of the above-described phenomena in combustion systems configured for a time-constant combustion process in which the static pressure in the combustion chamber or in upstream or downstream installation components should also have constant values (constant-pressure combustion).
However, the situation is quite different in a small number of very specific combustion installations in which the above phenomenon of self-induced, periodic combustion instabilities is brought about on purpose and is used to generate a periodic combustion process with periodic heat release rate of the flame and periodic, pulsating exhaust gas stream (pulsating hot gas stream) in the combustion chamber and in downstream installation components (e.g. heat exchanger, chemical reactors, etc.).
These pulsed combustion installations can be split into those in which the heat from the pulsating combustion process is transferred e.g. for heating or for generating service water or steam (purely thermal use), or into pulsating reactors which primarily serve for physical and/or chemical treatment of a raw material (e.g. drying, calcination, thermally controlled material synthesis, etc.) and which are often termed pulse dryer or pulse combustor or pulsation reactor.
The raw material can also be a raw material mixture. The raw material or the raw material mixture can exist in solid form but also in liquid form, gas form or vapor form.
The advantage of these installations compared to conventional combustion systems using stationary operation lies in the periodic-transient and turbulent exhaust gas stream, averaged over time, in the combustion chamber or in downstream components (e.g. heat exchanger, reaction spaces, resonance tube, etc.).
Both with respect to solid walls (combustion chamber wall, wall of a heat exchanger, steam generator, etc.) and with respect to material introduced for treatment into the hot gas stream with defined treatment temperature, the heat transfer from the hot gas to the walls or the material is markedly higher, by 2 to 5 times, than a stationary, turbulent flow, on average, of identical average flow speed and identical temperature.
Due to the similarity between convective heat transfer and the material transfer, the above statement also holds for the material transfer: in the case of the periodic-transient, pulsating flow, the transfer rate of gaseous/vaporous materials from the hot gas to the material to be treated, or from the material into the hot gas stream, increases by similar values due to the almost complete lack of boundary layers that exist, as is known, in the case of stationary flows and present barriers to diffusion and transfer.
Owing to these relationships, material that is to be treated experiences steeped heating gradients in the pulsating hot gas streams (“thermo-shock treatment”).
In order to now ensure reliable operation of pulsed combustion installations or pulsation reactors, it is important to keep the two main parameters of the combustion pulse—pulse frequency and pulse amplitude—as constant as is possible in the context of an industrial-scale process. If these change significantly during the process, it is to be expected that the advantages (e.g. in heat transfer or in material treatment) over the conventional, stationary process will be lost and/or that the homogeneity of the resulting product will suffer.
With respect to the pulse frequency which arises in the case of the formation of self-induced combustion instabilities by harmonic feedback in the burner-flame-combustion chamber-resonator system, the literature is unanimous in that these are dependent essentially on the geometry of volumes that are acoustically active or capable of resonating (such as the combustion chamber, the resonance tube, etc.) and on the gas temperature.
Therefore, if these two main variables influencing the pulse frequency remain unchanged during the process, e.g. the heat transfer or the material treatment or the material synthesis, then the frequency of the combustion/pressure/flow pulses also remains approximately constant.
In the context of the amplitude of the pulses, however, other influencing variables must also be taken into account.
This will become clear from the following example: if, in a pulsating reactor with fixed, predefined frequency and an amplitude of the combustion pulse that thus results (automatically in idle operation without addition of material), raw material is added to the pulsating hot gas with different mass flows (e.g. 50 kg/h and 100 kg/h) for thermal treatment, then different raw material addition rates cause different degrees of damping of the pulse amplitude, since the pulse energy that originally acted only on the gas pulse of the hot gas must then be spread over hot gas with different degrees of particle or droplet loading.
In extreme cases, it is even possible, with a sufficiently large addition per unit time of material to be treated, to completely suppress the pulsation in the reactor, thus of course eliminating the advantages of the periodic-transient process for the production of specific material qualities (e.g. particle sizes, specific surface areas, reactivity, etc.) in the product. In this context, it should once again be remembered that a reduction in the pulse amplitude (characterized for example as the amplitude of the static pressure pulse in the combustion chamber or of the speed of the hot gas flow in the combustion chamber or in the resonance tube) reduces the convective heat and material transfer between the hot gas and the walls or the material to be thermally treated, and as a result the advantages of the transient process, e.g. with regard to specific material properties that can be achieved, disappear.
In other words, any change in either the mass flows of the added raw materials or the specific material properties of various reactants (raw material density, humidity content, particle size distribution, solids content in the case of suspensions, etc.) change the pulse amplitude of the pulsating hot gas flow when material is added, and thus the result of the material treatment.
With a view to a possible material throughput that is as large as possible when required, it is fundamentally of interest to run an installation with the largest possible pulse amplitude during idle operation.
However, when operating a pulsed combustion system, it is also necessary to ensure on one hand that the pressure or velocity amplitudes occurring in the combustion process have a safe upper limit in order to reliably avoid mechanical or thermal overloading of the installation structure.
On the other hand, however, it is also necessary to reliably avoid flashbacks or separation or extinction of the flame in order to ensure stably pulsating long-term operation.
The term flashback is to be understood as follows: in pulsed combustion driven by the above-described process of, in particular, self-induced combustion instabilities (flame pulses), the mass flow of fuel/air mixture (e.g. in the case of premixed combustion) issuing from the burner changes in time-periodic fashion, and therefore the burner outlet velocity (axial velocity component) also changes in time-periodic fashion, while the flame speed (combustion speed) of the mixture issuing from the burner has a constant value when the composition (and thus the air ratio of the premix) is constant.
If, for a time during the pulsation period, the value of the flow velocity of the issuing mixture drops below the value of the flame speed, the flame moves upstream, counter to the flow direction, and almost into the burner. The moments in which the flow velocity of the issuing mixture is necessarily below the value of the flame speed increases with increasing amplitude of the pressure/combustion pulses.
It is now possible for the flame, in that phase of the pulsation period in which the flow velocity of the issuing mixture rises back up to its maximum value, to be driven back out of the burner and thus once again, at least for another time interval, burns in the desired axial position outside the burner, that is to say downstream of the burner outlet.
However, it is particularly problematic if the flame, as it moves into the burner, then adopts an anchoring position inside the burner that is so stable that even during the highest flow velocity of the issuing mixture, arising within the pulsation period of the pulse of the burner outlet velocity, the flame cannot be driven out of the burner and thus burns permanently within the burner or within the burner casing, and thermally damages or even destroys the burner.
In summary, the term flashback is thus understood as the flame momentarily or permanently burning within the burner and not, as actually desired, in a separated, stable position axially downstream of the burner outlet, outside the burner.
On the basis of the above explanations, constructors and operators of pulsed combustion installations for thermal material treatment or material synthesis are faced with the following problem:
On one hand, the strength (that is to say the amplitude) of the pulse of the hot gas flow in a pulsed combustion installation is a very important variable which determines both the properties of the material that is treated or synthesized and also the homogeneity of the thermally treated material. The amplitude is therefore an essential adjustment parameter in the context of thermal material treatment/material synthesis in pulsation reactors.
In that context, excessive values of the pulse amplitude of the material-carrying hot gas pulse can lead to a flashback and thus to damage to or destruction of the reactor, which must then be immediately counteracted by switching off the flame, that is to say the burner.
It must be understood that switching off the installation due to undesired flashback also means that the material (product) that was last thermally treated and/or synthesized under undefined thermal conditions has to be considered as waste.
As a consequence, a more or less onerous cleaning process of the material-carrying reactor sections is necessary before re-starting the reactor and resumption of thermal material treatment or synthesis, in order to reliably remove, from the reactor, material which was treated or synthesized in a thermally undefined manner.
The present invention is therefore based on the object of specifying a device with which, during operation of a pulsation reactor, the amplitudes of the hot gas flow pulse in the combustion chamber can adopt large values, in order to thus also for example be able to increase the product throughput rates and thus the reactor throughput that can be achieved, and with which at the same time the risk of undesired flashback of the pulsating flame into the burner is effectively avoided. The object is also to propose a method for operating this device.
This object is achieved, according to the invention, in that the burner for generating a pulsating flame for generating a pulsating hot gas flow is designed as a swirl burner and has a diffuser as element at its outlet.
A method according to the invention is thus characterized in that the pulsating fuel/air mixture flowing to the pulsating flame is guided through a swirl burner and a diffusor adjoining this burner.
The diffuser is in particular a conical, preferably metallic diffuser whose free cross-sectional area increases in the axial flow direction.
An opening half-angle of the inventive element, termed diffuser, can be chosen in the range between 3 degrees and 45 degrees.
An axial extent of the diffuser (diffuser length) can, depending on the embodiment, be between 0.5 times and 10 times the free diameter of the burner outlet.
The diffuser can be positioned directly downstream of a swirler on the burner. It is however also possible to provide, immediately at the outlet of the swirler, another cylindrical pipe element to which the diffuser then adjoins in the axial flow direction.
In that context, the invention is based on the surprising finding that, when using a pulsating premix swirl flame or a pulsating, quick-mixing diffusion swirl flame as a “driver” for a pulsed combustion reactor or pulsation reactor for generating the pulsating hot gas stream necessary for material treatment or material synthesis therein, a burner outlet designed as a diffuser and having the correct choice of diffuser length and diffuser opening angle automatically prevents return of the pulsating flame into the burner—that is to say flashback—even in the case of very large amplitudes of the combustion pulse in the combustion chamber.
One possible explanatory model of the action of the invention as self-adapting flashback prevention is based on the fact that, when the pulsating flame migrates upstream in those moments of the pulsation period when the burner outflow velocity is low, the flame itself blocks part of the free flow cross section area of the diffuser. This is based on the knowledge that, in the case of a swirl burner, that area (perpendicular to the axial direction) taken up by the flame or the inner recirculation region of the swirl flow, which underpins the flame, is then essentially independent of the outflow velocity, in the axial direction of the burner, of the fuel/air mixture.
Thus, an outflow of the fresh mixture from the burner is then possible only through the free annular gap remaining between the flame, or the inner recirculation zone of the swirl flow that forms the flame, and the diffuser wall. Since the area of this annular gap perpendicular to the burner axis decreases upstream within the diffuser due to the conical geometry of the latter while the area taken up by the flame remains constant, for continuity reasons the axial flow velocity of the fresh mixture flow issuing from the burner increases monotonically until it reaches a value at which the flame, due to its limited flame speed, can no longer migrate further upstream but rather slows down and remains stationary.
Conversely, the same mechanism also prevents excessive separation (with subsequent extinction) of a pulsating flame from the burner outlet at those time points in the pulsation period in which particularly high, instantaneous axial velocities of the pulsating fresh mixture flow exist:
First, the flame is displaced downstream in that phase of the pulsation period in which the flow velocity of the issuing mixture rises to its maximum, indeed due to the rising flow velocities of the fresh mixture flow within the diffuser. However, this simultaneously increases the size of the free annular gap area between the diffuser wall and the central flame which, in the case of a swirl burner, occupies the same area irrespective of its axial position. For continuity reasons, this causes the axial flow velocity to drop again, thus ultimately also effectively counteracting excessive separation of the flame from the burner outlet (which would have the consequence that the flame could even be extinguished entirely).
It is expressly stated at this point that the pulsating flame burning within the diffuser is not considered to be an inadmissible flashback.
In particular, a high-temperature-resistant design of the diffuser, for example using a ceramic internal cladding or by forced cooling using a twin-wall embodiment of the diffuser with air or water as the coolant, makes it simple for a person skilled in the art to ensure durably safe operation of the pulsating flame at/in the diffuser.
It is also to be noted that the problem of undesired flashback as a consequence of the pulsating flow in a pulsed combustion reactor or pulsation reactor is significant only if the flame is fully premixed (fuel and combustion air are molecularly mixed with one another spatially upstream of the burner) or if it burns as a quick-mixing diffusion flame. In the latter case, there is e.g. a nozzle-mixing flame in which the fuel and combustion air are brought together only within the burner, and preferably at the burner outlet.
Other advantages and features of the invention emerge from the following description of an exemplary embodiment, in which:
Fuel is to be understood for example as fuel gases such as natural gas, methane, hydrogen, or liquid fuels such as alcohol, etc. Within the context of the present invention, combustion air is to be understood in general terms as an oxidant which supplies the oxygen necessary for combustion. In addition to air, this for example also includes pure oxygen or oxygen-enriched air etc.
This combustion air stream or this fuel/air mixture is guided via a swirler 3 within the swirl burner 1 such that the mass flow 5 issuing from the burner outlet 4 has, in addition to movement in the axial direction, a rotational movement 6 in the circumferential direction (tangential velocity component or “swirl”).
With this rotational movement 6, the mass flow 5 flows into a diffuser 7. The walls of the diffuser have an essentially conical profile with an opening half-angle 8. This opening half-angle 8 is in a range between 3° and 45° and is measured with respect to the axial direction.
The burner outlet 4 has an essentially circular cross section and can have an axial extent 9 which is preferably, for example, in the range between 0 and 0.5 m.
The axial length 10 of the conical or frustoconical diffuser 7 can for example be between 0.1 and 1 m. It is therefore, relative to the dimensions of the burner outlet 4, between approximately 0.5 times and 10 times the free diameter of the burner outlet.
Toward the end 11 of the diffuser 7, a swirl flame 12 forms in the mass flow 5 of the issuing fuel/air mixture.
This swirl flame 12 is in particular also characterized by a central recirculation region 13, which is an important characteristic of a swirl-stabilized flame.
The flame 12 burns in a pulsating manner into a combustion chamber, not shown in greater detail here, fluidically connected to the diffuser 7, where it generates an oscillating hot gas flow. To this hot gas flow, a quantity of material for material treatment or material synthesis is to be added as required. Once treated or synthesized in the hot gas flow, this material is then separated out again, for example in a cyclone or a hot gas filter, which are also not represented.
The pulsation of the flame 12 is self-induced and, in the burner-flame-combustion chamber-reaction space-separation device system, feeds back on the inflowing mass flow 5.
Owing to the previously mentioned oscillation of the hot gas flow, the flame 12 pulsates and therefore oscillates the rotating mass flow 5 flowing into it, which in turn maintains the pulsation and so on and so forth (that is to say that the mass flow issuing from the burner has a time-dependent—generally approximately sinusoidal—time profile).
The oscillating mass flow 5 causes the velocity in the axial direction of the burner outlet flow (axial velocity component of the burner outflow) to change, while at the same time the flame speed 14 of the flame 12 that forms remains constant. In that context, the flame speed is the speed with which the flame 12 propagates in the fuel/air mixture flowing to it, counter to the outflow direction of this mixture.
This causes a shift in the axial position of the flame that forms: when the outflow velocity of the mass flow 5 drops, the flame 12 migrates into the diffuser 7, for example as far as “position 1” shown in
Now, however, the geometric extent of the flame 12 transversely to the outflow direction is essentially constant irrespective of its axial position. This is associated with the special properties of a swirl-stabilized flame, as already stated above.
The fact that the cross-sectional areas Al and A2 available for the flow in axial positions 1 and 2 are different in size in said positions, owing to the conical shape of the diffuser 7, causes a corresponding change, between positions 1 and 2, in the size of the annular areas 15 and 17 formed at those positions between the flame 12 (of constant geometric extent) and the wall 16 of the diffuser:
Since the cross section of the diffuser broadens conically in the flame propagation direction between the area Al and the area A2, the free annular area of the annular gap, that surrounds the flame 12 and lies between the flame 12 and the wall 16 of the diffuser 7, increases accordingly. In that context, the swirl-stabilized flame 12 acts like a solid body through which the mass flow 5 of issuing fuel/air mixture cannot flow, in part due to the central, inner recirculation zone 13 present therein.
In position 2, the free annular area 15 causes an axial velocity U2 with an axial impulse flow İ2. If the pulsating flame 12 now migrates as described within the pulsation period due to the time-dependent and initially decreasing burner outlet mass flow Mzu(t) and thus the decreasing outlet velocity U, upstream in the direction of the burner 1, it will reach position 1 shown in
In position 1, the free annular area 17 is, as stated, smaller than the free annular area 15 of position 2 owing to the conical shape of the diffuser 7. Thus, in position 1, when the area through which the flow cannot pass and which is taken up by the swirl-stabilized flame 12 characterized by the central recirculation zone is the same, the axial flow velocity U1 increases and thus the axial impulse flow İ1 also increases there. The axial impulse flow İ1 is therefore greater than the axial impulse flow İ2.
Owing to the axial flow velocity U1 rising as explained, the flame 12 is thus prevented from migrating further upstream into the diffuser 7 and thus into the burner 4. Rather, the flame 12 halts in approximately this position 1 until the burner outlet mass flow 5, rising again within the pulsation period and accordingly also having a burner outlet velocity which is increasing again, pushes the flame 12 back downstream into the original position 2, which is to be considered safe.
This ensures that the flame 12 cannot migrate as far as the burner 1 or as far as the swirler 3, and cannot establish itself therein in the event of a flashback.
It should also be noted, at this point, that the wall 16 of the diffuser 7 and the wall 18 of the burner outlet 4 can be either uncooled or, as shown on the left-hand side of
It is also possible to protect the walls 16 and 18 using a ceramic cladding.
The device described here, and the operation thereof, make it possible to set the pulse amplitude of a pulsed combustion reactor to high values even when the installation is idling (i.e. with no addition of material), without this implying the risk of a flashback into the burner, since such a flashback is reliably prevented by the diffuser provided, according to the invention, on the swirl burner used.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2016 005 155.8 | Apr 2016 | DE | national |
