Improving the Efficiency of Fuel Combustion with the Use of Various Designs of Embrasures

Abstract

Currently, $N{O}_X$ emission requirements for thermal power plants and power equipment are being tightened. Regime and technical measures are being developed to improve the efficiency of fuel combustion in boilers. Due to the high cost of field studies, and in some cases the impossibility of conducting them, mathematical modeling tools allow one to work out technical and tactical measures. In this paper, the multidisciplinary STAR-CCM+ platform with GMU-45 type burners is used to simulate the combustion of gaseous fuel in a digital model of an energy boiler of the type TGME-464. By conducting numerical experiments, the possibility of reducing $N{O}_X$ emissions by using flue gas recirculation is considered, and the efficiency of burner devices is compared when using different embrasure configurations.

1. Introduction

When analyzing the global volume of emissions of harmful substances following the signing of the Kyoto Protocol, which was ratified by 192 countries, it is possible to observe a particular inertia in reducing emissions in the energy industry. In the global energy sector, the maximum effect of reducing the volume of emissions of harmful substances is attributed to increased efficiency of thermal power plants [1], increased share of electricity production without burning organic fuel at nuclear power plants and hydroelectric power plants, and replacing burned coal with gas and fuel oil [2].

The technology of optimizing the fuel combustion process is especially relevant for thermal power plants that run on solid and liquid fuels. Thermal power plants are the main sources of $N{O}_X$ emissions into the atmosphere. To gradually reduce the negative impact of thermal power plants on the atmosphere, European and Russian Federation legislation establishes standards for specific emissions of pollutants into the atmosphere from boiler plants. In all countries, low values of maximum permissible concentrations (hereinafter MPCs) are established and periodically revised downwards. There are monthly average, daily average, hourly, and maximum single concentrations of nitrogen oxides.

In the furnaces of power and hot water boilers, the formation of all types of nitrogen oxides is observed in the zone with a temperature above 1800 K, called the zone of active combustion. The final output of $NO$ depends entirely on its main characteristics: excess air ratio, maximum combustion temperature, and residence time in the high-temperature region. It is known that the temperature factor has the most significant effect on the formation of nitrogen oxides during the combustion of both natural gas and fuel oil. Thus, the formation of fast nitrogen oxides begins at a temperature of 1170–1350 K and is completed in the range from 50 to 100 K; fuel oxides of nitrogen are formed at temperatures of 1100 to 1400 K, reaching a maximum at 1400 K and 1800 K [3]. Due to the high average integral temperature during the combustion of natural gas, thermal nitrogen oxides primarily contribute to the formation of $N{O}_X$ emissions. It is impossible to carry out instrumental measurements of the flame temperature and excess air in the flame zone directly during the operation of the boiler. Additionally, for thermal calculations, it is necessary to consider the aerodynamic parameters of the furnace devices, as the combustion of fuel occurs at high speed, and the combustion zone can move along the height of the furnace at a speed of up to 8 m/s. The influence of the excess air coefficient on $N{O}_X$ emission with increasing load appears insignificant. The flue gas sampling point for the completeness of fuel combustion and the amount of emissions of harmful substances in steam boilers is usually located behind the convective superheater in the downstream flue of the boiler. This is a section of the water economizer (regime section), in which the oxygen content is measured and the coefficient of excess air is determined. Simultaneously, small values of air twist angles at the level of the first row of burners contribute to an increase in the amount of heat transferred to the superheater.

According to their typology, measures to reduce $N{O}_X$ emissions are divided into two main groups: regime and technological. All technological measures aim to reduce the temperature in the active combustion zone and reduce the oxygen concentration in this zone.

The main technological methods for reducing $N{O}_X$ emissions from gas and fuel oil combustion include:

-

Reduction of the maximum temperature (recirculating of flue gases, injecting moisture into the combustion zone, lowering the temperature of hot air, sectioning the furnace with two-light screens, dispersing the flame along the height of the furnace);

-

Reduction of oxidant concentration (non-stoichiometric fuel combustion, staged combustion, burners with prolonged mixture formation, burners with an adjustable proportion of primary air, and combustion in the pre-furnace with a lack of oxidizer are all combustion methods);

-

Special methods (combustion in a fluidized bed, preliminary heat treatment of fuel, special heater designs).

Considering that the formation of nitrogen oxides decreases when excess oxygen is reduced, a natural means of suppressing their formation is to reduce the excess air supplied through the burners. Reduction of excess air is only possible as long as it does not lead to a sharp increase in losses with chemical underburning.

With a multi-tiered arrangement of burners, non-stoichiometric combustion can be an effective method—this involves organizing separate reduction (less than 1) and oxidizing (greater than 1.25) combustion zones in the furnace while keeping the total amount of excess air constant. In this case, $N{O}_X$ production decreases in the reducing combustion zone due to the lack of oxygen, and in the oxidizing zone, $N{O}_X$ decreases due to a reduction in combustion temperature. In practice, this is implemented through an imbalance of the coefficient $\alpha$ in the burner devices or in the tiers of burners. Introducing non-stoichiometric combustion can reduce $N{O}_X$ emissions by up to 35–50% without negatively impacting its main technical and economic indicators.

Recirculation of flue gases into the combustion chamber is the simplest means of reducing the temperature level and oxygen concentration in the combustion zone and is successfully implemented at enterprises and thermal power plants in both Russia and Europe [4]. Flue gas recirculation with the potential to reduce emissions by up to 60% has a specific capital cost 2 times lower than two-stage combustion and 7 times lower than installation of low-emission burners. Of course, these indicators for each type of boiler are determined based on the type of unit and are calculated. The recirculation of combustion products into the combustion chamber leads to a decrease in the temperature of the torch and the alignment of the temperature fields in the furnace, which leads to an intensive suppression of the formation of thermal $N{O}_X$. The reduction effect is greater when burning natural gas and is determined by such factors as the place of sampling of recirculation gases, the degree of recirculation r (%), the conditions for their introduction into the combustion chamber, the distribution of recirculation gases over the volume of the combustion chamber, and the state of the boiler plant.

When burning natural gas, the greatest effect is achieved when flue gases are fed directly into the fuel (up to 3.5–4.5% for each percentage of recirculation). The schemes where recirculation gases are introduced into the primary air with a decrease in the output of nitrogen oxides to 60% have found the greatest distribution. The optimal amount of recirculation gases is up to r ≤ (15–22%). A further increase in the degree of recirculation leads to a decrease in the efficiency of the boiler due to a decrease in the combustion temperature and chemical underburning.

N. Orfanoudakis et al. [5] compared the environmental and economic parameters of measures aimed at suppressing $N{O}_X$ emissions. The work was conducted primarily to study the best available technologies under the IPPC Directive. According to the findings, flue gas recirculation, with a maximum availability and suppression efficiency of up to 75% (for gaseous fuel), has an installation cost that is 2.5 times lower than the implementation of staged combustion and the installation of specially designed burners.

The efficiency of reducing $N{O}_X$ from the supply of recirculation gases directly to the burner device is affected by the place of the entry and the configuration of the burner porthole, which affect the intensity of the mixing processes of the media flowing from the burner channels. This is indicated in [6] by the form of the fields of the composite jet at the cut of the burner embrasure. The lowest concentration of nitrogen oxides in this case, for swirl burners, is observed in the burner with a separate pipe for supplying recirculation gases. Deng Pan et al. [7] described an experimental study with a laboratory setup, including using mathematical modeling tools, of the effect of flue gas recirculation (the recirculation gas is mixed with the formation of an oxidizer before being fed into the burner) on natural gas combustion. They found that as the recycled gas feed rate was increased to 20%, $NO$ emissions decreased by about 85%, and combustion instability can be related to the level of $N{O}_X$ emissions in premix burners. Yanyan Ji et al. [8] conducted a simulation of the operation of a diffusion gas boiler with flue gas recirculation. In addition to gas-fired boilers, flue gas recirculation justifies its implementation:

-

On pulverized coal boilers: Jun Li et al. [9] studied the effect of internal recirculation (0–30%) of flue gases from a 55 MW pulverized coal-fired boiler with coal and/or biomass co-firing. According to the results obtained in the study, due to the organization of internal recirculation at this boiler, it was possible to achieve an overall reduction in the amount of $N{O}_X$ by 18%;

-

When burning kerosene: an experimental study by Mohsen Abdelaal et al. [10] with a laboratory installation is described (recycled gas is mixed with an oxidizer before being fed into the burner). In the course of the study, it was found that the flame remains stable up to a recirculation level of 30–40% and turns blue at higher recirculation rates. By implementing flue gas recirculation, a significant reduction in $N{O}_X$ emissions of more than 90% can be achieved through lower temperatures (estimated up to 25% reduction) and better mixing with the oxidizer;

-

In internal combustion engines (when burning natural gas). A special method of exhaust gas recirculation involves the recirculation of exhaust gases from one or more engine cylinders back into the engine intake manifold under any operating conditions [11];

-

When burning biofuels: in [12], an experimental study on the implementation of primary measures to reduce emissions of harmful substances in flue gases was carried out on a prototype stand of a biofuel boiler with a capacity of 20 kW. Flue gas recirculation in the experiment was organized by mixing primary air with flue gases after the boiler. The flue gas recirculation coefficient increased from 0% to 52% during the experiment. According to the study, it was concluded that the measured combustion temperatures are insufficient for the formation of thermal $N{O}_X$, which suggests that flue gas recirculation reduces mainly thermal $N{O}_X$;

-

When burning fuel mixtures based on landfill gases and products of thermal decomposition of industrial and municipal waste: in the work by K. Kikuchi et al. [13], they experimentally evaluated the combustion characteristics of self-recirculating combustion of exhaust gases and the effect of combustion air nozzle parameters using a recirculating flow burner. They confirmed the possibility of burning hydrogen with a low $N{O}_X$ content due to exhaust gas recirculation.

The burner device must provide a decrease in temperature and slow mixing of air with fuel, avoiding, in turn, chemically incomplete combustion of the fuel. During the reconstruction of the burner devices of steam boilers in operation at TPPs during technical re-equipment and reconstruction, they can not only achieve a reduction in $N{O}_X$ emissions, but also simultaneously improve the technical and economic indicators of the legal boiler, as well as increase the safety of its operation [14]. Furthermore, an important condition is to ensure high turbulence of combustion processes [15]. When modeling furnace processes, it is required to know the temperature of adiabatic combustion of fuel, the attenuation coefficient of radiation by the furnace medium, the emissivity of the furnace, considering the size of the radiating layer, the coefficient of thermal efficiency of the furnace screens, the resulting Boltzmann number, and the final calculated value of the gas temperature at the outlet of the furnace.

It has been found that burners with low thermal conductivity have higher flame temperatures. With an increase in the thermal conductivity of the burner, the length of the flame increases. Thermal radiation has little effect on flame temperature and height, but it has a significant effect on the structure of the flame for burner materials with low electrical conductivity. Flame stabilization on a quartz burner occurs at fuel jet velocities lower than on steel or aluminum burners. The thermal conductivity of the burner wall and the recirculation of heat through it play an essential role in flame stabilization. The length of the flame increases with the increasing thermal conductivity of the burner wall. From a practical point of view, a quartz burner may be the optimal solution for micro-diffusion flame applications with lower fuel flow rates [16].

Currently, in order to minimize pollutant emissions, many studies have been carried out to study the mechanism of the chemical reaction of fuel. This leads to the need in development for a new class of equipment and production technology. For the energy sector, decarbonization is one of the appropriate measures to achieve the goal of carbon neutrality. In the world, boilers are partially being switched from coal to natural gas, which makes it possible to reduce the emission of harmful substances. Replacing bituminous coal with methane can reduce CO₂ emissions by about 43%, while almost completely eliminating emissions of particulate matter, sulfur oxides, and mercury. This approach can also reduce nitrogen oxide emissions [17]. In reference [18], the uneven supply of secondary air and the inclination of the burner on the efficiency of fuel combustion in a boiler with a capacity of 500 MW were studied. By tilting the burner, $N{O}_X$ emissions can be reduced. To improve combustion stability, the authors of [19] proposed a conical swirler based on the distribution of two parts of the fuel operating at low operating conditions, which makes it possible to increase combustion stability, but at the same time it can lead to an increase in $N{O}_X$ emission.

Unlike conventional burners, porous ceramic radiant burners (PCRBs) have a high combustion rate, high thermal efficiency, and low pollutant emissions [20]. Innovative solutions such as pre-chamber turbulent jet ignition are currently being introduced, which provide longer burning time and increased lean burn limits when applied to homogeneous super lean premixes (HULPs) evaluated flame-based models used in RANS 3D Computational Fluid Dynamics (CFD) simulations for turbulent premixed combustion to assess their applicability limits for HULP mixtures [21]. Two different combined approaches are chosen: the flame zone model (FAM) and the coherent flame model (CFM). Both the FAM and CFM approaches can only predict the development of combustion with satisfactory accuracy when “classical” flame configurations are detected.

A significant body of work in numerical simulation is Reynolds-averaged simulation (RANS). With such modeling, it is difficult to achieve detailed flow and flame structures. Since the first computational experiment, a wide range of applications has appeared for modeling the flow of liquids and gases, interfacial interactions, flows with chemical reactions, and heat transfer using CFD. These include: Ansys (Fluid/CFX) [22,23,24], Star CCM+ [25,26], SigmaFlame [27,28], CEDRE [29,30], and Fire 3D [31].

In reference [32], the results of modeling the combustion process in burners with various designs and heat transfer on the walls of the chimney are presented in order to optimize the combustion efficiency and reduce the amount of harmful emissions. The results showed that the size of the air intake has only a minor effect on combustion efficiency. At the same time, the airflow pumped by the fan significantly increases this indicator. In particular, it has been shown that increasing the height of the chimney from 2 m to 4 m significantly increases the combustion efficiency from 63% to 94% with a 50% increase in incoming airflow due to natural convection. The paper notes that further optimization of combustion efficiency requires detailed parametric studies for various burner configurations.

A number of works are devoted to the optimization of the design of diffusion burners operating on natural gas. Burner thermal efficiency, and emission of nitrogen oxide, carbon monoxide, and unburned methane were used as targets for the optimization problem. In particular, numerical simulation was carried out for various given values of design variables (dimensions of channels for air and gas intake, etc.), after which the optimal design was selected using the “Pareto front concept” [33]. The paper noticed that optimization of the burner design increases the efficiency by 29.4%, and reduces emissions of carbon monoxide, nitrogen oxide, and unburned methane by 81.2%, 98.6%, and 83.9%, respectively.

Lehigh University’s Energy Research Center (ERC) has conducted research on optimizing combustion in boilers of various sizes and designs, operating on different fuels and with various burners [34]. Based on the research results, a specialized intelligent software package has been developed to optimize fuel combustion. For several years, the software package has been used to determine the optimal control settings to improve boiler performance while reducing $N{O}_X$ emissions. Depending on the application, the software package was also used to determine the parameters that have the greatest impact on performance, such as unburned carbon content in the ash and carbon monoxide (CO) emissions, as well as combustion conditions: flame front location and temperature in the combustion zone. The goals of optimizing the combustion process using this software package varied from project to project and included, among others, the definition of initial characteristics, the definition of minimum levels of $N{O}_X$, minimum specific thermal power, etc.

The works [35,36] present the results of a study of the influence of the design of the burner embrasure on the nature of combustion, performed by numerical simulation using CFD methods. There is a great potential for reducing harmful emissions by optimizing the geometry of the burner embrasure. The burner embrasure is a constructive component of the walls of the boiler furnace in the area where the working medium exits from the mouth of the burner. The opening of the burner torch, the degree of mixing of fuel and oxidizer, and the range of the torch depend on the design of the embrasure (type and size). The following types of embrasures are distinguished: diffuser–cylindrical, confuser, diffuser–confuser, confuser–cylindrical, cylindrical. Theoretically, confuser and diffuser–confuser embrasures should significantly improve the mixing of fuel with air, and make the division into combustion zones with a lack and excess of air less pronounced, as well as somewhat reduce flow rates. Due to this fact, the range of the flame should be reduced, which, accordingly, should cause a reduction in the formation of $N{O}_X$.

Based on the analysis of scientific and technical literature, it can be concluded that one of the most promising areas for reducing greenhouse gas emissions is the optimization of the design of burners and the use of flue gas recirculation. In this work, a study of various designs of embrasures for the GMU-45 burner of the TGME-464 boiler was carried out using flue gas recirculation.

The work is organized as follows. Section 2 describes the model. In Section 3, we summarize and discuss the results obtained. Finally, Section 4 collects some conclusions and outlines the prospects for future work.

2. Materials and Methods

2.1. The 3D Model of a Steam Boiler

In the multidisciplinary platform STAR-CCM+, a calculation model of the boiler unit model E-500-13.8-560GMN (another name TGME-464, manufacturer Taganrog Boiler-Making Works <<Krasny Kotelshchik>> (Taganrog, Russia)) with gas–oil burners of the GMU-45 type was constructed (Figure 1). These boilers are widely used at Russian power plants, particularly Ulyanovsk Thermal Power Plant No. 1, Cheboksarskaya Thermal Power Plant No. 2, Novocheboksarskaya Thermal Power Plant No. 3, Penza Thermal Power Plant No. 1, Saranskaya Thermal Power Plant No. 2, Saratovskaya Thermal Power Plant No. 5, Samarskaya Thermal Power Plant, Syzranskaya Thermal Power Plant, and Volzhsky CHPP Automobile Thermal Power Plant, etc. This steam boiler is a steam generator designed to carry out natural circulation of the working medium in the evaporative heating surface with its forced, direct-flow movement through a water economizer and a superheater. The rated capacity of the boiler is 500 t of steam per hour, with steam parameters of temperature at 833 K and pressure at 14 MPa. The volume of the combustion chamber is 1610 m³. The concentration of $N{O}_X$ according to the regime map increases with increasing load, reaching a maximum capacity of 500 t of steam per hour, the flue gas temperature at the chimney inlet is 393 K, and fuel consumption (natural gas) is 39,000–40,000 m³/h. All eight burners are operating at the same time.

2.2. Mathematical Model of the Boiler with Standard Burners

Modeling of this boiler unit with standard burners is considered in more detail in the work. Substantiation of the applied mathematical model, selection of the grid, and analysis of the convergence of the grid were carried out [26].

2.3. 3D Model of Modified Burners

The burner embrasure of the boiler unit is a hole in the wall of the furnace, into which the outlet part of the burner is inserted (Figure 2). Furthermore, the embrasure is a structural element of the furnace screen. The shape and type of the embrasure can affect the characteristics of the torch, including the rate of outflow of the air–fuel mixture, and the opening of the torch.

To assess the possibility of reducing $N{O}_X$ emissions in the multidisciplinary platform STAR-CCM+, calculation models of GMU-45 burners with various embrasures were built: standard (Figure 3a), diffuser–confuser (Figure 3b), confuser (with constriction) (Figure 3c), confuser–cylindrical (Figure 3d), diffuser–cylinder (Figure 3e), cylindrical (Figure 3f). For each type of embrasure, a model and computational grid of the TGME-464 boiler were built. With a “basic” size of 0.2 m in the computational grid, the number of elements in the simulated boiler with 8 burners was 23,626,694 units. The size of the elements in the combustion chamber area was 15% of the “base” size, and in the burner area it was 6%.

2.4. Checking the Adequacy of the Model

The adequacy of the model was verified by comparing the obtained experimental data with the simulation data for the TGME-464 boiler unit with a standard GMU-45 burner at different recirculation coefficients. The comparison is shown in Figure 4. The marked data in the figure were obtained as follows: (1) we have full-scale experiments on the operating equipment of the CHPP [37], they were obtained for various recirculations, and we modeled the same parameters; (2) in order to show the efficiency of using recirculation even with the use of a burner with a standard porthole, it was decided to divide the data to the corresponding data at recirculation r = 0%. As we can see, the error of the obtained data does not exceed 3%.

3. Results and Discussion

3.1. Experiment Plan

Within the framework of this study, a calculation boiler model TGME-464 with modified oil–gas burners of the GMU-45 type was used to study the temperature distribution and $N{O}_X$ concentration along the height of the combustion chamber at a distance of 1 m from the steam boiler embrasure. The initial data for the experiment are presented in

Table 1. An experimental plan for studying the emission of harmful substances in a combustion chamber with one burner and a steam boiler.

Methane consumption	kg/s	0.7334
Speed of primary air at the inlet	kg/s	3.46
Speed of secondary air at the inlet	kg/s	4.59
Degree of recirculation	%	0	5	8	15
The resulting temperature of the mixture of air and recirculation gases	K	477	481	483	487
Mass fraction of CO2 component in secondary air	%	0	0.005	0.008	0.0157
Mass fraction of H2O component in secondary air	%	0	0.011	0.018	0.032

.

3.2. Discussion of the Results of a Computational Experiment

The studies were carried out for a steam load of the boiler of 400 t/h, which corresponds to a gas flow rate for the burner of 3936 m³/h. The air supply is divided into two streams: primary and secondary. Air flows before entering the burner embrasure pass swirling devices. Gas is supplied to the space between the central pipe with a diameter of 245 × 8 mm and a pipe with a diameter of 325 × 9 mm. On the side of the burner, the gas collector ends with a truncated cone, in which holes are drilled, from which methane exits at an angle to the axis of the air flow. The study of the influence on the characteristics of the torch of various types of embrasures was carried out at the degree of recirculation of flue gases r = 0, 5, 8, and 15%. The simulation results are presented below in Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9. All results are presented in a plane section along the central axis of the burner.

According to the presented results, it can be seen that the use of different types of embrasures affects the maximum temperature of the flame, its length, and shape. To evaluate the results obtained, we present the flame structure in terms of temperature and the content of the $N{O}_X$ fraction, as well as the dependence of the temperature distribution along the length of the combustion chamber and the mass content of $N{O}_X$ on the different content of recirculation gases. It can also be concluded that for shallow combustion chambers, the use of the diffuser–cylindrical and cylindrical type of embrasures is impractical, since there will be a possibility of a torch throw and a dynamic effect on the opposite wall of the furnace (the core of the torch is displaced to the level of 2.9 m and further at various degrees of recycling).

The results show that by changing the type of embrasure, not only the boundary and length of the torch change, but also its radiation characteristics. For all types of embrasures, with an increase in the degree of recirculation, a decrease in temperature and a decrease in the maximum concentration of $N{O}_X$ are observed. The lowest temperatures in the core of the torch are achieved when using the diffuser–confuser type of embrasure (the maximum reduction of 63% was achieved with a recirculation of 15%), and if you look at the structure of the torch, you can see that the torch has the largest opening angle and is shortened. The lowest concentrations of $N{O}_X$ are observed in the converging (with narrowing) type of embrasure. It is also significant that when implementing flue gas recirculation in this mode on this type of boiler, together with the confuser type of embrasure, it has a higher efficiency in terms of nitrogen oxide emissions, but a possible decrease in boiler efficiency should be taken into account. The presented results make it possible to determine the optimal operating mode of the burner, taking into account $N{O}_X$ emissions and boiler efficiency.

4. Conclusions

This paper presents the results of modeling the processes of natural gas combustion in a power boiler of the TGME-464 type and burners of the GMU-45 type using various types of embrasure with flue gas recirculation in the STAR-CCM+ multidisplinary platform. It has been established that by changing the type of embrasure, not only the boundaries and length of the torch change, but also its radiation and environmental characteristics. The influence of the type of burner porthole on the shape, length of the torch, temperature of the torch core, as well as on the formation of $N{O}_X$ is determined. It is shown that by changing the amount of recirculation gas supply, it is possible to reduce the formation of nitrogen oxides, and using the confuser (with a narrowing) shape of the embrasure, it is possible to achieve a reduction in $N{O}_X$ emissions (up to 63%) compared to standard types of embrasure. It was also noted that the area of generation of nitrogen oxides and the zone of maximum core temperature varies along the length of the torch and depends on the shape of the embrasure.

On the basis of the studies carried out and the conclusions formulated, we further plan to study the simultaneous combination of various types of embrasures to stabilize combustion.