Method for controlling a combustion device

Abstract

The invention relates to a method for controlling a combustion process of a gas turbine. A combustion chamber, a control device which stores a calculation model of the combustion process and an exhaust gas measuring device are required. First, permissible limit values for nitrogen oxides and carbon monoxide as harmful substances are determined. The actual values of at least one of the two harmful substances are continuously measured in the exhaust gas. If a signal is given for reducing the power of the gas turbine to the lowest possible value, the minimum fuel delivery is calculated at which the limit values are adhered to. The fuel delivery is then reduced until either the calculated minimum fuel delivery is reached or until the proportion of the continuously measured harmful substance reaches the permissible limit value.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method for controlling a combustion device, wherein it is important to achieve a minimum power. BACKGROUND

[0002] Various methods for controlling a combustion device are known from the prior art. The methods are usually matched to the respective combustion device and differ depending on the type of combustion device and its purpose. However, the present consideration is made with regard to a state in which the combustion device can be operated with as low a power as possible within the permissible range. In connection therewith, it is known from the prior art that the relative content of harmful substances, in particular the carbon monoxide load, increases as one moves further away from the rated power in the partial load range. In this regard, the operation of the combustion device is limited to a minimum value at which the limit values are reliably adhered to, with the proviso that the content of harmful substances is taken into account.

[0003] It is common from the prior art to determine by means of calculation and from laboratory experiments what share of the respective harmful substance has at a reduced power. Thus, in creating the rules for controlling the combustion process, the previously determined minimum power is specified as a minimum value.

[0004] From WO201361301A1 a gas turbine plant is known which has a method in which the share of carbon monoxide is detected. Here, the gas turbine plant comprises a gas turbine with a combustion chamber in which fuel is burned in the operation of the gas turbine. The delivery of fuel is adjusted here by a control device in accordance with a power reference value. In addition, there is a detection module by means of which the carbon monoxide concentration can be determined. Here, the power reference value is adjusted in the control device in accordance with the carbon monoxide concentration.

[0005] However, it is sometimes desirable to operate the combustion device with as low a power as possible, in particular in order to avoid shutdowns. SUMMARY

[0006] It is therefore an object of the present invention to be able to achieve a lower power than is common to date, at which limit values given for harmful substances are adhered to.

[0007] The object proposed is solved by the method according to the invention according to the teaching of embodiments of the invention and alternatively according to the teaching of embodiments of the invention. The method according to the invention is explained in embodiments of the invention as a further improvement of the combination of embodiments of the invention. Advantageous method steps are explained in embodiments according to the invention.

[0008] Such a method serves to control the combustion process of a combustion device. It is initially irrelevant what type of combustion device is involved here. The method can advantageously be used at least to control the combustion process in the combustion chamber of a gas turbine. It is necessary in any case that the combustion device comprises at least one combustion chamber and that at least one burner is arranged at the combustion chamber. By means of the burner, fuel and the required intake air for the combustion of the fuel can be delivered into the combustion chamber for combustion. Furthermore, a control device for carrying out the method is present. For this purpose, a calculation model of the combustion process is stored in the control device. On the basis of the stored calculation model, it can be ascertained here whether the harmful substances are theoretically within the permissible range or not in the case of a preset power.

[0009] For this purpose, it is necessary to store permissible limit values for the proportion of nitrogen oxides in the calculation model of the control device. Furthermore, it is similarly necessary to determine permissible limit values for the proportion of carbon monoxide. The two values can be defined as invariable quantities or it can be proposed that they can be matched to local conditions. The permissible limit values can relate, for example, to legal provisions.

[0010] Since an exact, completely error-free calculation is not feasible in practice due to a plurality of possible influencing factors, such as environmental conditions and completely constant combustion in the combustion device, for example, due to operating fluctuations, it is necessary to take into account tolerances so that the respective target values are reached on the basis of the permissible limit values for the respective proportions of nitrogen oxides and carbon monoxide. If it is determined that the state of the target values of the harmful substances is present, it can be reliably assumed that the permissible limit values are not exceeded.

[0011] In order to be able to optimally control the combustion process, it is furthermore particularly advantageous if the type and / or the quality of the fuel used is known. For this purpose, it is feasible on the one hand that the parameters are input as presets in the control device for this purpose. Alternatively, it can be proposed that the type or the quality of the fuel is measured or determined before the fuel is delivered to the combustion device and the result is transmitted to the control device.

[0012] Furthermore, for the present method, a signal for setting a minimum power is required as part of the control of the combustion device, so that the fuel delivery to the combustion device is subsequently reduced.

[0013] In order to achieve as low a power as possible, it is now proposed according to the invention that the combustion device has an exhaust gas measuring device by means of which at least the actual proportion of the relevant harmful substances in the exhaust gas can be detected.

[0014] By means of the now known actual proportion of the harmful substances, the power of the combustion device can be further reduced until the target values are reached, as long as it is ensured that the other limit values are reliably adhered to.

[0015] In the prior art, due to the unknown level of the actual occurring harmful substance load, a sufficient safety margin for actually adhering to the limit values has to be taken into account when calculating the lowest possible fuel delivery or generated power and thus a large tolerance between the permissible limit value and the target value is necessary, whereas by means of the process control according to the application, the power can be further reduced by means of a reduction in the tolerance.

[0016] If now the relevant harmful substances, carbon monoxide and nitrogen oxides, are considered, three different methods according to the application for achieving the lowest possible power result.

[0017] In the first method according to the application, the actual proportion of nitrogen oxides in the exhaust gas is continuously measured. From the known actual proportion of nitrogen oxides, the calculation model in the control device now calculates whether a reduction in fuel and thus in power is possible if the proportion of nitrogen oxides does not exceed the target value. In parallel thereto, it is determined in the control device from the calculation model to what extent the amount of fuel can be reduced until the proportion of carbon monoxide theoretically reaches the target value. Due to the unknown actual content in the exhaust gas, a larger tolerance for the permissible limit value has to be adhered to here.

[0018] Similarly, in the second method according to the application, the actual proportion of carbon monoxide in the exhaust gas is continuously measured. From the known actual proportion of carbon monoxide, the calculation model in the control device now calculates a possible reduction in fuel or power until the target value for the proportion of carbon monoxide is reached. In contrast to the first method, in parallel it is determined whether a reduction in fuel is possible if the calculated proportion of nitrogen oxides does not exceed the target value.

[0019] Subsequently, the fuel delivery is now reduced with further continuous monitoring of the actual proportion of nitrogen oxides (in the first method) or carbon monoxide (in the second method) until the calculated minimum fuel quantity is reached. If the measured actual proportion of nitrogen oxides or carbon monoxide has previously reached the target value, the reduction in fuel delivery has already been stopped. In both methods, a hypothetical minimum fuel delivery and thus a minimum power is thus obtained, at which the one or both harmful substances, nitrogen oxides or carbon monoxide, have reached the target value. It can be reliably assumed here that the two permissible limit values are adhered to.

[0020] The third method according to the application combines the first method with the second method, in which the exhaust gas measuring device monitors not only the actual proportion of nitrogen oxides, but also the actual proportion of carbon monoxide. As in the other methods, the extent to which the fuel or the power can be reduced by the control device according to the calculation model in the case of a signal approaching the minimum power until one of the two target values or both target values is reached. Since both values are continuously detected, the tolerance for the permissible limit values for the two harmful substances can be chosen relatively small.

[0021] As before, the fuel delivery or the power is reduced until the previously calculated minimum fuel delivery is reached. If the state occurs in which one of the two measured values of the actual proportion of nitrogen oxides or carbon monoxide reaches the target value, the fuel reduction is stopped.

[0022] In the simplest case, the calculation of the minimum fuel delivery possible is carried out once after the signal has been given for the slow-down of the combustion device to the minimum power. However, it is particularly advantageous to carry out the calculation repeatedly in further progress, as long as the minimum power is desired, so that new possibilities for further reducing the fuel delivery or the power, as long as given, can be utilized. Correspondingly, the minimum fuel delivery is recalculated on the basis of the target values of the harmful substances given and the proportion of nitrogen oxides or carbon monoxide measured at the time of the recalculation, without exceeding the target values at the minimum fuel delivery.

[0023] If an exceeding of the permissible limit values is determined according to the continuous measurement or according to the recalculation, the fuel delivery is increased. In contrast thereto, the fuel delivery can be reduced again if a falling below the two target values is determined according to the recalculation.

[0024] The recalculation can be set at regular intervals on the one hand. For example, the time period can be chosen so that after a change in the fuel delivery and thus in the power, the proportion of the harmful substances changing in consequence thereof levels out to a substantially constant value.

[0025] On the other hand, the continuous measurement of nitrogen oxides and / or carbon monoxide can lead to a recalculation. For example, a comparison between the measured values and the permissible limit values and / or the target values can be carried out continuously, wherein a recalculation for adjusting the fuel quantity is triggered when a pre-set absolute or relative difference is reached. It can be proposed here that the difference is chosen to be small when the target value is exceeded and, in contrast thereto, the difference is chosen to be greater when the target value is undershot.

[0026] A greater potential for reducing the fuel quantity can be achieved when the combustion device comprises at least one main burner and at least one secondary burner. It is initially irrelevant what type of combustion device is involved in this, wherein it is proposed that the burners have different combustion characteristics. Analogous to what is necessary when using a single burner or a single burner type, the main burner as well as the secondary burner can deliver fuel and / or intake air into the combustion chamber.

[0027] What is possible in the presence of a main burner as well as a secondary burner is to extend the calculation model as follows: The distribution of the fuel onto the main burner as well as onto the secondary burner is calculated. Thus, given the fuel quantity and the distribution of the fuel, the expected share of carbon monoxide and the expected share of nitrogen oxides can be calculated. In comparison with the target values for nitrogen oxides or carbon monoxide, the minimum fuel delivery can then be determined as before, at which at least the actual share of nitrogen oxides or carbon monoxide reaches the target value given the distribution of the fuel. Correspondingly, the fuel delivery can also be reduced as before until the calculated minimum fuel delivery is reached.

[0028] However, it is particularly advantageous to calculate the optimal distribution of the fuel when using a main burner as well as a secondary burner. To this end, the comparison between the values of the calculated shares of nitrogen oxides and carbon monoxide and the target values can be carried out iteratively when changing the fuel distribution and reducing the fuel quantity until the smallest possible difference between the calculated shares of the harmful substances and the target values is reached.

[0029] The method applies in a particularly advantageous manner when the secondary burner is a so-called pilot burner.

[0030] In this, it is advantageous in the method for determining a favorable distribution of the fuel to take into account in which harmful substance the difference between the actually measured value or the theoretically calculated share and the target value is greater.

[0031] If the difference between the share of carbon monoxide and the target value thereof is greater than the difference between the share of nitrogen oxides and the target value thereof in the first case, it is advantageous to change the distribution of the fuel as follows: increase the share of the main burner and reduce the share of the secondary burner.

[0032] In contrast, it is advantageous in the second case if the difference between the share of nitrogen oxides and the target value thereof is greater than the difference between the share of carbon monoxide and the target value thereof, it is advantageous to change the distribution of the fuel as follows: increase the share of the secondary burner and reduce the share of the main burner.

[0033] If, after the calculation of the optimum distribution of the fuel quantities to the main burner and to the pilot burner, it is calculated that the two target values are undershot, or if, after setting the corresponding fuel quantities, it is determined from the measurement of the harmful substances that the two target values are undershot, the fuel delivery can then be further reduced.

[0034] If there is an intake air measurement device by means of which at least one property of the intake air can be determined, a further improvement of the method is achieved, in particular for reducing the necessary tolerances. It is particularly advantageous here if the temperature of the intake air and the air humidity are known in the calculation model. Correspondingly, the values can be taken into account when calculating the minimum fuel quantity and the optimum distribution of the fuel.

[0035] It is in principle sufficient at the outset to create a calculation model by means of known calculation bases (e.g. combustion properties, properties of the combustion device, type of fuel), in which the actually measured proportion of the harmful substances represents a variable for the calculation.

[0036] Furthermore, it is also advantageous, due to the complexity of the combustion process and possible variations of the combustion properties, for example due to different local environmental conditions, that the calculation model can be adapted. To this end, it is advantageous to store the calculation parameters together with as many existing state data as possible continuously. Actual states of the combustion device or gas turbine (temperature data, vibration data, etc.), type and / or quality of the fuel, temperature and / or air humidity of the intake air, actual proportion of nitrogen oxides and / or carbon monoxide in the exhaust gas belong to the state data. The adaptation of the calculation model can be carried out regularly or continuously with consideration of the stored data. In this case, the so-called self-learning method can be used in a particularly advantageous manner.

[0037] The method is generally not limited to the fuel type. It can also be proposed that different fuels are used in the presence of a main burner and a pilot burner. In principle, the method can be used advantageously if the fuel is gaseous. BRIEF DESCRIPTION OF DRAWINGS

[0038] In the following figures, a combustion device and time curves are schematically sketched. The figures show:

[0039] Figure 1 a schematic diagram of a combustion device according to the application;

[0040] Figure 2 a schematic diagram showing possible time curves of parameters when applying the method according to the application. DETAILED DESCRIPTION

[0041] A combustion device 01 according to the application is schematically sketched in the drawing. The combustion device 01 comprises a combustion chamber 02 together with a main burner 03 and a secondary burner 04 arranged thereat. Fuel 23 and intake air 21 can be fed to the burners 03, 04. Exhaust gas 25, i.e. flue gas, flows out of the combustion chamber 02.

[0042] A control device 11 is present for controlling the method, in which control device 11 a calculation model 12 is saved and which, in the embodiment, comprises a data memory 13. Different characteristic variables are transmitted to the control device 11. On the one hand, a maximum proportion of nitrogen oxides 16 and a maximum proportion of carbon monoxide 17 are fixedly preset. Here, respectively allowed limit values or target values can be involved. In the first case, the target values can be calculated by the control device. It is likewise possible for not only the allowed limit values but also the respective target values to be transmitted as a preset to the control device 11.

[0043] Furthermore, the type or quality 24 of the fuel 23 needs to be known in the calculation model. For this purpose, it is exemplarily proposed that the quality 24 is continuously detected and transmitted to the control device 11. Furthermore, it is proposed in the embodiment that the temperature and the air humidity 22 of the intake air 21 are measured and transmitted to the control device 11.

[0044] Furthermore, it is important for the method according to the application that the actual proportion of nitrogen oxides 26 and / or the actual proportion of carbon monoxide 27 in the exhaust gas 25 is continuously measured and transmitted to the control device 11.

[0045] The method according to the application is triggered by a signal for approaching the minimum power, for which the respectively required desired power 15 is transmitted to the control device 11.

[0046] When the method is executed in the control device 11 on the basis of the calculation model 12 saved there, the minimum fuel feed and, in this case, the optimum distribution to the main burner 03 and the secondary burner 04 are calculated. On the basis of the calculation result, the corresponding main valve 05 for controlling the fuel flow to the main burner 03 and the corresponding secondary valve 06 for controlling the fuel flow to the secondary burner 04 are actuated by the control device 11.

[0047] In Figure 2 a possible method sequence in time with different characteristic variables is exemplarily shown. Starting from the normal power of the combustion device, a signal approaches the minimum power P sollIn the control device 11, a minimum power or minimum fuel supply is now calculated based on the calculation model 12, at which the predetermined limit values ​​for the proportion of nitrogen oxides and the proportion of carbon monoxide are observed (i.e. at least one target value is reached). In this case, the target value NOx is preset in the control device. max According to the calculation, the fuel supply and thus the power P are now reduced. ist The increase in the proportion of pollutants, in this case the proportion of nitrogen oxides NOx ist The carbon monoxide content (not shown here) is usually accompanied by a reduction in power—see time T2 .

[0048] For example, it is now possible that the target value for carbon monoxide is already reached in the calculation, whereas the target value for nitrogen oxides is not equal to the measured value NOx. ist There is still a large difference between the values ​​of NOx and CO. This leads to an advantageous method of changing the fuel distribution so that a difference also results between the target value and the calculated value for carbon monoxide, with the result that a difference also occurs between the target value and the measured value for nitrogen oxides. ist The difference between NOx and NH is reduced - see time T3. Here, the fuel quantity can be reduced again until the target value NOx is substantially reached according to the calculation or corresponding measurement. max ——See time T4.

[0049] Now it is possible that the process stabilizes, with the proportion of pollutants decreasing over time—see time T5. Due to the continuous monitoring of at least one pollutant, it is possible to trigger a new calculation in the event of a given difference, so that the fuel supply and thus the power P ist A further reduction of is possible - see time T6.

Claims

1. A method for controlling a combustion process in a combustion device (01), the combustion device (01) comprising: a combustion chamber (02) in which fuel can be combusted together with intake air (21); and at least one burner capable of delivering the fuel and / or the intake air (21) into the combustion chamber (02); and a control device (11) in which a calculation model (12) of the combustion process is stored; and an exhaust gas measuring device capable of detecting the actual proportion of nitrogen oxides; the method comprises the following steps: - determining a permissible limit value (16) for the nitrogen oxide fraction and determining a target value as the limit value after deducting tolerances; - determining a permissible limit value (17) for the carbon monoxide content and determining a target value as the limit value after deducting tolerances; - continuously detecting the actual proportion of the nitrogen oxides in the exhaust gas (25); - Detection of signals for setting minimum power; - calculating a minimum fuel supply with the aid of the calculation model (12) at which the expected proportion of carbon monoxide reaches the target value, wherein the calculation model is created with the aid of a known calculation basis; - while continuously monitoring the actual proportion of nitrogen oxides in the exhaust gas (25), reducing the fuel delivery to the calculated minimum fuel delivery or until the target value for nitrogen oxides is reached.

2. The method according to claim 1, The calculation is performed repeatedly, wherein the fuel supply is increased if an exceeding of one of the limit values ​​is detected, and the fuel supply is further reduced if an undershooting of both limit values, after deducting the corresponding process tolerances, is detected.

3. The method according to claim 2, wherein said calculations are performed at regular intervals; or The calculation is performed as soon as a predefined difference between the measured actual proportion of pollutants in the exhaust gas (25) and the target value specified therefor is exceeded.

4. The method according to any one of claims 1 to 3, The combustion device (01) comprises at least one main burner (03) and at least one auxiliary burner (04), wherein the main burner (03) and the auxiliary burner (04) are capable of respectively delivering fuel and / or intake air (21) into the combustion chamber (02); and the method comprises the following steps: - determining the distribution of the fuel to the main burner (03) and the secondary burner (04) when calculating the minimum fuel supply, at which the expected proportion of carbon monoxide or the expected proportion of nitrogen oxides reaches the target value; - reducing the fuel delivery taking into account the previously calculated distribution of the fuel to the main burner (03) and the secondary burner (04).

5. The method according to claim 4, The auxiliary burner (04) is a pilot burner.

6. The method according to claim 4, wherein the fuel distribution is changed with a higher portion for the main burner (03) and a lower portion for the secondary burner (04) if the difference between the target value for carbon monoxide and the calculated or measured actual portion (27) is greater and if the difference between the target value for nitrogen oxides and the calculated or measured actual portion is smaller; In this case, the fuel supply is subsequently reduced further if undershooting of the two aforementioned target values ​​is detected.

7. The method according to claim 4, wherein, in the event of a greater difference between the target value for nitrogen oxides and the calculated or measured actual portion and a smaller difference between the target value for carbon monoxide and the calculated or measured actual portion (27), the fuel distribution is changed with a higher portion for the secondary burner (04) and a lower portion for the main burner (03); In this case, the fuel supply is subsequently reduced further if undershooting of two target values ​​is detected.

8. The method according to any one of claims 1 to 3, The intake air measuring device is capable of determining at least one property of the intake air (21), wherein the property is taken into account in the control device (11) when calculating the fuel supply and / or the distribution of the fuel.

9. The method according to any one of claims 1 to 3, Calculation parameters and existing status data are stored continuously, and the calculation model is adjusted regularly or continuously based on the stored data.

10. The method according to any one of claims 1 to 3, The actual state of the combustion device (01) and / or the type and / or mass (24) of the fuel and / or the temperature of the intake air (21) and / or the air humidity (22) and / or the actual proportion of the nitrogen oxides and / or the carbon monoxide in the exhaust gas (25) are continuously stored.

11. A method for controlling a combustion process in a combustion device (01), the combustion device (01) comprising: a combustion chamber (02) in which fuel can be combusted together with intake air (21); and at least one burner capable of delivering the fuel and / or the intake air (21) into the combustion chamber (02); and a control device (11) in which a calculation model (12) of the combustion process is stored; and an exhaust gas measuring device capable of detecting the actual proportion of carbon monoxide (27); the method comprises the following steps: - determining a permissible limit value (16) for the nitrogen oxide content and determining a target value as the limit value after deducting a tolerance; - determining a permissible limit value (17) for the carbon monoxide fraction and determining a target value as the limit value after deducting tolerances; - continuously detecting the actual proportion (27) of carbon monoxide in the exhaust gas (25); - Detection of signals for setting minimum power; - calculating a minimum fuel supply with the aid of the calculation model (12) at which the expected proportion of nitrogen oxides reaches the target value, wherein the calculation model is created with the aid of a known calculation basis; - while continuously monitoring the actual proportion (27) of carbon monoxide in the exhaust gas (25), reducing the fuel supply until the calculated minimum fuel supply or until the target value for carbon monoxide is reached.

12. The method according to claim 11, The calculation is performed repeatedly, wherein the fuel supply is increased if an exceeding of one of the limit values ​​is detected, and the fuel supply is further reduced if an undershooting of both limit values, after deducting the corresponding process tolerances, is detected.

13. The method according to claim 12, wherein said calculations are performed at regular intervals; or The calculation is performed as soon as a predefined difference between the measured actual proportion of pollutants in the exhaust gas (25) and the target value specified therefor is exceeded.

14. The method according to any one of claims 11 to 13, The combustion device (01) comprises at least one main burner (03) and at least one auxiliary burner (04), wherein the main burner (03) and the auxiliary burner (04) are capable of respectively delivering fuel and / or intake air (21) into the combustion chamber (02); and the method comprises the following steps: - determining the distribution of the fuel to the main burner (03) and the secondary burner (04) when calculating the minimum fuel supply, at which the expected proportion of carbon monoxide or the expected proportion of nitrogen oxides reaches the target value; - reducing the fuel delivery taking into account the previously calculated distribution of the fuel to the main burner (03) and the secondary burner (04).

15. The method according to claim 14, The auxiliary burner (04) is a pilot burner.

16. The method according to claim 14, wherein the fuel distribution is changed with a higher portion for the main burner (03) and a lower portion for the secondary burner (04) if the difference between the target value for carbon monoxide and the calculated or measured actual portion (27) is greater and if the difference between the target value for nitrogen oxides and the calculated or measured actual portion is smaller; In this case, the fuel supply is subsequently reduced further if undershooting of the two aforementioned target values ​​is detected.

17. The method according to claim 14, wherein, in the event of a greater difference between the target value for nitrogen oxides and the calculated or measured actual portion and a smaller difference between the target value for carbon monoxide and the calculated or measured actual portion (27), the fuel distribution is changed with a higher portion for the secondary burner (04) and a lower portion for the main burner (03); In this case, the fuel supply is subsequently reduced further if undershooting of two target values ​​is detected.

18. The method according to any one of claims 11 to 13, The intake air measuring device is capable of determining at least one property of the intake air (21), wherein the property is taken into account in the control device (11) when calculating the fuel supply and / or the distribution of the fuel.

19. The method according to any one of claims 11 to 13, Calculation parameters and existing status data are stored continuously, and the calculation model is adjusted regularly or continuously based on the stored data.

20. The method according to any one of claims 11 to 13, The actual state of the combustion device (01) and / or the type and / or mass (24) of the fuel and / or the temperature of the intake air (21) and / or the air humidity (22) and / or the actual proportion of the nitrogen oxides and / or the carbon monoxide in the exhaust gas (25) are continuously stored.

21. A method for controlling a combustion process in a combustion device (01), the combustion device (01) comprising: a combustion chamber (02) in which fuel can be combusted together with intake air (21); and at least one burner capable of delivering the fuel and / or the intake air (21) into the combustion chamber (02); and a control device (11) in which a calculation model (12) of the combustion process is stored; and an exhaust gas measuring device capable of detecting an actual proportion of nitrogen oxides and an actual proportion of carbon monoxide (27); the method comprises the following steps: - determining a permissible limit value (16) for the nitrogen oxide fraction and determining a target value as the limit value after deducting tolerances; - determining a permissible limit value (17) for the carbon monoxide fraction and determining a target value as the limit value after deducting tolerances; - continuously detecting the actual proportion of nitrogen oxides and the actual proportion of carbon monoxide (27) in the exhaust gas (25); - Detection of signals for setting minimum power; - calculating a minimum total fuel delivery using the calculation model (12) at which the expected carbon monoxide fraction and the expected nitrogen oxide fraction each reach the target value, wherein the calculation model is created using a known calculation basis; - while continuously monitoring the actual proportion of nitrogen oxides and the actual proportion of carbon monoxide (27) in the exhaust gas (25), reducing the fuel supply until the calculated minimum fuel supply or until the respective target value for nitrogen oxides or carbon monoxide is reached.

22. The method according to claim 21, The calculation is performed repeatedly, wherein the fuel supply is increased if an exceeding of one of the limit values ​​is detected, and the fuel supply is further reduced if an undershooting of both limit values, after deducting the corresponding process tolerances, is detected.

23. The method according to claim 22, wherein said calculations are performed at regular intervals; or The calculation is performed as soon as a predefined difference between the measured actual proportion of pollutants in the exhaust gas (25) and the target value specified therefor is exceeded.

24. The method according to any one of claims 21 to 23, The combustion device (01) comprises at least one main burner (03) and at least one auxiliary burner (04), wherein the main burner (03) and the auxiliary burner (04) are capable of respectively delivering fuel and / or intake air (21) into the combustion chamber (02); and the method comprises the following steps: - determining the distribution of the fuel to the main burner (03) and the secondary burner (04) when calculating the minimum fuel supply, at which the expected proportion of carbon monoxide or the expected proportion of nitrogen oxides reaches the target value; - reducing the fuel delivery taking into account the previously calculated distribution of the fuel to the main burner (03) and the secondary burner (04).

25. The method according to claim 24, The auxiliary burner (04) is a pilot burner.

26. The method according to claim 24, wherein the fuel distribution is changed with a higher portion for the main burner (03) and a lower portion for the secondary burner (04) if the difference between the target value for carbon monoxide and the calculated or measured actual portion (27) is greater and if the difference between the target value for nitrogen oxides and the calculated or measured actual portion is smaller; In this case, the fuel supply is subsequently reduced further if undershooting of the two aforementioned target values ​​is detected.

27. The method according to claim 24, wherein, in the event of a greater difference between the target value for nitrogen oxides and the calculated or measured actual portion and a smaller difference between the target value for carbon monoxide and the calculated or measured actual portion (27), the fuel distribution is changed with a higher portion for the secondary burner (04) and a lower portion for the main burner (03); In this case, the fuel supply is subsequently reduced further if undershooting of two target values ​​is detected.

28. The method according to any one of claims 21 to 23, The intake air measuring device is capable of determining at least one property of the intake air (21), wherein the property is taken into account in the control device (11) when calculating the fuel supply and / or the distribution of the fuel.

29. The method according to any one of claims 21 to 23, Calculation parameters and existing status data are stored continuously, and the calculation model is adjusted regularly or continuously based on the stored data.

30. The method according to any one of claims 21 to 23, The actual state of the combustion device (01) and / or the type and / or mass (24) of the fuel and / or the temperature of the intake air (21) and / or the air humidity (22) and / or the actual proportion of the nitrogen oxides and / or the carbon monoxide in the exhaust gas (25) are continuously stored.

Images