Controlling the Furnace Process in Coal-Fired Boilers
The unstable trends that exist in the market of fuel supplied to thermal power plants and the situations in which the parameters of their operation need to be changed (or preserved), as well as the tendency toward the economical and environmental requirements placed on them becoming more stringent, are factors that make the problem of controlling the combustion and heat transfer processes in furnace devices very urgent. The solution to this problem has two aspects. The first involves development of a combustion technology and,accordingly, the design of a furnace device when new installations are designed. The second involves modernization of already existing equipment. In both cases,the technical solutions being adopted must be properly substantiated with the use of both experimental and calculation studies.
The experience Central Boiler-Turbine Institute Research and Production Association (Ts KTI) and Zi O specialists gained from operation of boilers and experimental investigations they carried out on models allowed them to propose several new designs of multicell and maneuverable—in other words,
controllable—furnace devices that had been put in operation at power stations for several years. Along with this, an approximate zero-one-dimensional, zone wise calculation model of the furnace process in boilers had been developed at the Tsk Ti, which allowed Tsk Ti specialists to carry out engineering calculations of the main parameters of this process and calculate studies of furnaces employing different technologies of firing and combustion modes .
Naturally, furnace process adjustment methods like changing the air excess factor, stack gas recirculation fraction, and distribution of fuel and air among the tiers of burners, as well as other operations written in the boiler operational chart, are used during boiler operation.However, the effect they have on the process is limited in nature. On the other hand, control of the furnace process in a boiler implies the possibility of making substantial changes in the conditions under which the combustion and heat transfer proceed in order to considerably expand the range of
1
loads, minimize heat losses, reduce the extent to which the furnace is contaminated with slag, decrease the emissions of harmful substances, and shift to another fuel. Such a control can be obtained by making use of the following three main factors:
(i) the flows of oxidizer and gases being set to move in the flame in a desired aerodynamic manner;
(ii) the method used to supply fuel into the furnace and the place at which it is admitted thereto;
(iii) the fineness to which the fuel is milled.
The latter case implies that a flame-bed method is used along with the flame method for combusting fuel.The bed combustion method can be implemented in three design versions: mechanical grates with a dense bed, fluidized-bed furnaces, and spouted-bed furnaces.
As will be shown below, the first factor can be made to work by setting up bulky vorticisms transferring large volumes of air and combustion products across and along the furnace device. If fuel is fired in a flame, the optimal method of feeding it to the furnace is to admit it to the zones near the centers of circulating vorticisms, a situation especially typical of highly intense furnace devices. The combustion process in these zones features a low air excess factor (α< 1) and a long local time for which the components dwell in them, factors that help make the combustion process more stable and reduce the emission of nitrogen oxides .
Also important for the control of a furnace process when solid fuel is fired is the fineness to which it is milled; if we wish to minimize incomplete combustion, the degree to which fuel is milled should be harmonized with the location at which the fuel is admitted into the furnace and the method for supplying it there, for the occurrence of unburned carbon may be due not only to incomplete combustion of large-size fuel fractions, but also due to fine ones failing to ignite (especially when the content of volatiles Daff < 20%).
Owing to the possibility of pictorially demonstrating the motion of flows, furnace aerodynamics is attracting a great deal of attention of researchers and designers who develop and improve furnace devices. At the same time, furnace aerodynamics lies at the heart of mixing (mass transfer), a process the quantitative
2
parameters of which can be estimated only indirectly or by special measurements. The quality with which components are mixed in the furnace chamber proper depends on the number, layout, and momenta of the jets flowing out from individual burners or nozzles, as well as on their interaction with the flow of flue gases, with one another, or with the wall.
It was suggested that the gas-jet throw distance be used as a parameter
determining the degree to which fuel is mixed with air in the gas burner channel. Such an approach to estimating how efficient the mixing is may to a certain degree be used in analyzing the furnace as a mixing apparatus. Obviously, the greater the jet length (and its momentum), the longer the time during which the velocity gradient it creates in the furnace will persist there, a parameter that determines how completely the flows are mixed in it. Note that the higher the degree to which a jet is turbulence at the outlet from a nozzle or burner, the shorter the distance which it covers, and, accordingly, the less completely the components are mixed in the furnace volume. Once through burners have advantages over swirl ones in this respect.
It is was proposed that the extent to which once through jets are mixed as they penetrate with velocity w2 and density ρ2 into a transverse (drift) flow moving with velocity w1 and having density ρ1 be correlated with the relative jet throw distance in the following way
Where ks is a proportionality factor that depends on the “pitch” between the jet axes (ks= 1.5–1.8).
The results of an experimental investigation in which the mixing of gas with air in a burner and then in a furnace was studied using the incompleteness of mixing as a parameter are reported in 5.
A round once through jet is intensively mixed with the surrounding medium in a furnace within its initial section, where the flow velocity at the jet axis is still equal to the velocity w2 at the nozzle orifice of radius r0.The velocity of the jet blown into the furnace drops very rapidly beyond the confines of the initial section, and the axis it has in the case of wall-mounted burners bends toward the outlet from the furnace.
One may consider that there are three theoretical models for analyzing the mixing of jets with flow rate G2 that enter into a stream with flow rate G1. The first
3
model is for the case when jets flow into a “free” space (G1= 0),the second model is for the case when jets flow into a transverse (drift) current with flow rate G1?G2，and the third model is for the case when jets flow into a drift stream with flow rate G1 At a = 0.07, the length of the round jet’s initial section is equal to 10 r0 and the radius the jet has at the transition section (at the end of the initial section) is equal to 3.3 r0. The mass flow rate in the jet is doubled in this case. The corresponding minimum furnace cross-sectional area Ff for a round once through burner with the outlet cross-sectional area Fb will then be equal to and the ratio Ff/Fb≈20. This value is close to the actual values found in furnaces equipped with once through burners. In furnaces equipped with swirl burners, a= 0.14 and Ff/Fb≈10. In both cases, the interval between the burners is equal to the jet diameter in the transition section d tr , which differs little from the value that has been established in practice and recommended in [7]. The method traditionally used to control the furnace process in large boilers consists of equipping them with a large number of burners arranged in several tiers. Obviously, if the distance between the tiers is relatively small, operations on disconnecting or connecting them affect the entire process only slightly. A furnace design employing large flat-flame burners equipped with means for controlling the flame core position using the aerodynamic principle is a step forward. Additional possibilities for controlling the process in TPE-214 and TPE-215 boilers with a steam output of 670 t/h were obtained through the use of flat-flame burners arranged in two tiers with a large distance between the tiers; this made it possible not only to raise or 4 lower the flame, but also to concentrate or disperse the release of heat in it [1]. A very tangible effect was obtained from installing multicell (operating on coal and open-hearth, coke, and natural gases) flat-flame burners in the boilers of cogeneration stations at metallurgical plants in Ukraine and Russia. Unfortunately, we have to state that, even at present, those in charge of selecting the type, quantity, and layout of burners in a furnace sometimes adopt technical solutions that are far from being optimal. This problem should therefore be considered in more detail. If we increase the number of burners nb in a furnace while retaining their total cross-sectional area (ΣFb=idem) and the total flow rate of air through them, their equivalent diameters deq will become smaller, as will the jet momentums GB, resulting in a corresponding decrease in the jet throw distance Hb and the mass they eject. The space with high velocity gradients also becomes smaller, resulting in poorer mixing in the furnace as a whole. This factor becomes especially important when the emissions of Box and CO are suppressed right inside the furnace using staged combustion (at αb < 1) under the conditions of a Fortinbras nonuniform distribution of fuel among the burners. In [1], a quantitative relationship was established between the parameters characterizing the quality with which once through jets mix with one another as they flow into a limited space with the geometrical parameter of concentration = with nb = idem and GB = idem. By decreasing this parameter we improve the mass transfer in the furnace; however, this entails an increase in the flow velocity and the expenditure of energy (pressure drop) in the burners with the same Fb. At the same time, we know from experience and calculations that good mixing in a furnace can be obtained without increasing the head loss if we resort to large long-range jets. This allows a much less stringent requirement to be placed on the degree of uniformity with which fuel must be distributed among the burners. Moreover, fuel may in this case be fed to the furnace location where it is required from process control considerations. For illustration purposes, we will estimate the effect the number of burners has on the mixing in a furnace at = = idem. schematically shows the plan views of two furnace chambers differing in the number of once through round nozzles (two and 5