Exhaust gas state calculation device, combustion control device, exhaust gas state calculation method, and exhaust gas state calculation program
Patent Information
- Application Number
- JP2025058790
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-03-31
- Publication Date
- 2026-10-14
AI Technical Summary
【0010】 本開示によれば、アンモニアと、アンモニアとは異なる燃料とを混合した混合燃料を燃焼するエンジンから排出された排ガスの状態を好適に検出することができる。
Smart Images

Figure 2026166065000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to an exhaust gas state calculation device, a combustion control device, an exhaust gas state calculation method, and an exhaust gas state calculation program.
Background Art
[0002] Dual fuel engines that generate power by burning a mixed fuel obtained by mixing two types of fuels are known.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of Invention
Problems to be Solved by the Invention
[0004] In dual fuel engines that burn the mixed fuel as described above, unburned components may be contained in the exhaust gas after combustion. It is necessary to appropriately switch the operating state of the dual fuel engine according to the state of the unburned components in the exhaust gas. For this reason, it is required to properly grasp the state of the unburned components in the exhaust gas.
[0005] The present disclosure has been made to solve the above-described problems, and an object thereof is to provide an exhaust gas state calculation device, a combustion control device, an exhaust gas state calculation method, and an exhaust gas state calculation program that can appropriately calculate the state of exhaust gas discharged from an engine that burns a mixed fuel obtained by mixing two types of fuels.
Means for Solving the Problems
[0006] To achieve the above objective, the exhaust gas state calculation device of the present disclosure is an exhaust gas state calculation device for calculating the state of unburned components in exhaust gas discharged from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, and comprises: a data acquisition unit that acquires one or more operating data from the operating data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; a storage unit that stores a calculation rule calculated based on the relationship between the operating data and the state of unburned components in the exhaust gas, which takes the operating data as input and outputs the state of unburned components in the exhaust gas; and a calculation unit that processes the operating data acquired by the data acquisition unit with the calculation rule stored in the storage unit to calculate the state of unburned components in the exhaust gas.
[0007] The combustion control device of this disclosure comprises the exhaust gas state calculation device described above and a control device for controlling the combustion of the engine, wherein the control device controls the operation of the engine based on the state of unburned material in the exhaust gas calculated by the exhaust gas state calculation device.
[0008] Furthermore, the exhaust gas state calculation method of this disclosure is an exhaust gas state calculation method for calculating the state of unburned components in exhaust gas discharged from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, and includes the steps of: acquiring one or more operating data from the operating data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; acquiring a calculation rule that takes the operating data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operating data and the state of unburned components in the exhaust gas; and processing the acquired operating data with the acquired calculation rule to calculate the state of unburned components in the exhaust gas.
[0009] Furthermore, the exhaust gas state calculation program of this disclosure is an exhaust gas state calculation program that calculates the state of unburned components in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, and causes a computer to execute the following: a process of acquiring one or more operating data from the operating data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; a process of acquiring a calculation rule that takes the operating data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operating data and the state of unburned components in the exhaust gas; and a process of processing the acquired operating data with the acquired calculation rule to calculate the state of unburned components in the exhaust gas. [Effects of the Invention]
[0010] According to this disclosure, the state of exhaust gas emitted from an engine that burns a mixed fuel, which is a mixture of ammonia and a fuel other than ammonia, can be suitably detected. [Brief explanation of the drawing]
[0011] [Figure 1] Figure 1 is a block diagram showing an example of a combustion system according to the first embodiment. [Figure 2] Figure 2 is a flowchart showing an example of the processing performed by the exhaust gas condition calculation device. [Figure 3] Figure 3 is a flowchart showing an example of the processing performed by a combustion control device. [Figure 4] Figure 4 is a block diagram showing an example of a program creation system. [Figure 5] Figure 5 is a block diagram showing an example of a calculation program creation device. [Figure 6] Figure 6 is an explanatory diagram showing an example of training data. [Figure 7] Figure 7 is a flowchart showing an example of the processing performed by the calculation program creation device. [Figure 8] Figure 8 is a block diagram showing an example of a combustion system according to the second embodiment. [Modes for carrying out the invention]
[0012] Preferred embodiments of the present disclosure will be described in detail below with reference to the drawings. However, these embodiments do not limit the present disclosure, and where there are multiple embodiments, they may be combinations of these embodiments. Furthermore, the components in the embodiments include those readily conceivable by those skilled in the art, those that are substantially identical, and those that are equivalent.
[0013] [First Embodiment] <Combustion System> Figure 1 is a block diagram showing an example of a combustion system according to the first embodiment. The combustion system 10 shown in Figure 1 burns a mixed fuel obtained by mixing a first fuel with a second fuel different from the first fuel. Examples of the first fuel include ammonia, methanol, and hydrogen. In the following explanation, ammonia will be used as an example. Examples of the second fuel include diesel fuel. The combustion system 10 includes an engine 12, an ammonia supply device 14, a fuel supply device 16, an air supply device 18, an exhaust pipe 20, an SCR (Selective Catalytic Reduction) catalyst 22, an ASC (Ammonia Slip Catalyst) 24, and a combustion control device 29. The combustion control device 29 includes a control device 30 and an exhaust gas state calculation device 40.
[0014] The engine (engine body) 12 is a device that rotates a rotating shaft using energy generated by burning ammonia, and serves as a power source for a mobile body and a power source for rotating the rotating shaft of a generator. The engine 12 is a dual-fuel engine. A dual-fuel engine is a system that supplies two types of fuel to the combustion chamber and burns the supplied mixed fuel. The engine 12 is equipped with intake valves and exhaust valves that open and close the intake port and exhaust port, respectively. The engine 12 is also equipped with fuel injectors that inject fuel into each combustion chamber and spark plugs that ignite the fuel.
[0015] The ammonia supply device 14 supplies ammonia to the engine 12. Ammonia is a carbon-free fuel that emits less greenhouse gases including carbon dioxide compared to hydrocarbon fuels. The ammonia supply device 14 has a tank for storing ammonia, a pipe connecting the tank and the engine 12, a control valve for adjusting the pressure and supply amount of ammonia supplied to the engine 12, and the like. A part of the functions of the ammonia supply device 14 may be arranged inside the engine 12.
[0016] The fuel supply device 16 supplies fuel to the engine 12. The fuel is a fuel other than ammonia, a fuel that ignites more easily than ammonia, and is, for example, light oil. The fuel may also be fossil fuel other than light oil, biofuel, or the like. The fuel supply device 16 has a tank for storing fuel, a pipe connecting the tank and the engine 12, a control valve for adjusting the pressure and supply amount of fuel supplied to the engine 12, and the like. A part of the functions of the fuel supply device 16 may be arranged inside the engine 12.
[0017] The air supply device 18 supplies air to the engine 12. The air supply device 18 has a pipe for guiding outside air, and an adjustment unit for adjusting the supply of the inflowing air to the engine 12. A control valve or the like can be used as the adjustment unit.
[0018] The exhaust pipe 20 is a pipe connected to the engine 12, through which exhaust gas generated by combustion of ammonia and the like in the engine 12 flows.
[0019] The SCR catalyst 22 and the ASC 24 are exhaust gas treatment devices that convert ammonia into other substances when the exhaust gas flowing through the exhaust pipe contains an ammonia component, specifically, an unburned ammonia component. The SCR catalyst 22 is disposed on the downstream side of the engine 12 in the exhaust pipe 20. The SCR catalyst 22 is a selective reduction catalyst that promotes the reaction between ammonia and nitrogen oxides. The SCR catalyst 22 supports, for example, vanadium, molybdenum, tungsten, zeolite and the like as active components.
[0020] The ASC24 is positioned downstream of the SCR catalyst 22 in the exhaust pipe 20. The ASC24 oxidizes the ammonia contained in the exhaust gas. Note that the equipment placed in the exhaust pipe 20 is not limited to the SCR catalyst 22 and ASC24, but various other equipment for treating exhaust gas can be placed there.
[0021] The combustion control device 29 comprises a control device 30 and an exhaust gas state calculation device 40, and controls the operation of the engine 12. The control device 30 is an ECU (Electronic Control Unit). The control device 30 is connected to various parts of the engine 12. The control device 30 is connected to sensors that acquire the status of the engine 12 and the environment around the engine. The control device 30 is connected to the exhaust gas state calculation device 40. The control device 30 acquires input values from an operation unit, such as the accelerator, which inputs the operating conditions of the engine 12. Note that the operating conditions of the engine 12 may be acquired from a calculation unit instead of an operation unit. Based on the status of the engine 12, the environment around the engine, and the operation command, the control device 30 determines the operating conditions of the engine 12 and controls the operation of the engine 12 based on the determined operating conditions. The control device 30 controls the opening and closing timing of the fuel injection valve, the opening and closing timing of the intake and exhaust ports by the intake and exhaust valves, the opening and closing of valves in various parts, etc., and controls the supply amount and timing of ammonia and fuel, the supply amount and pressure of air, etc.
[0022] <Exhaust gas condition calculation device> The exhaust gas state calculation device 40 calculates the state of unburned components in the exhaust gas by calculation based on the operating conditions of the engine 12 obtained from the control device 30. In other words, the exhaust gas state calculation device 40 is a soft sensor that calculates the state of unburned components in the exhaust gas without measuring the state of unburned components in the exhaust gas with a physical sensor. Examples of unburned components in the exhaust gas include ammonia, methanol, and hydrogen. The state of unburned components in the exhaust gas includes, for example, a first state in which the operating state of the engine 12 can be maintained, a second state in which the operating state of the engine 12 is switched, and a third state in which the state transitions between the first and second states. More specifically, the first state can be a state in which harmful unburned ammonia components are within the range of a predetermined safety management standard. The second state can be a state in which there are many harmful unburned ammonia components, and the amount of unburned ammonia components exceeds a predetermined safety management standard. The third state can be a state in which the safety management standard for unburned ammonia components is met, but the level is somewhat high and it is necessary to reduce the concentration. The above criteria for the first, second, and third states are merely examples, and the first, second, and third states may be defined by other criteria.
[0023] The exhaust gas state calculation device 40 includes an operation data acquisition unit 42, a calculation unit 44, and a storage unit 46. The operation data acquisition unit 42 acquires operation data such as control values of the engine 12. The operation data acquired by the operation data acquisition unit 42 are parameter values used to calculate the state of unburned matter in the exhaust gas. The data acquisition unit can acquire one or more operation data from the following operation data that have low correlation with each other. Here, the multiple operation data are one data selected from operation data with high correlation with each other, and it is preferable to use a combination of data where the correlation with the state of unburned matter in the exhaust gas is above a threshold, and the correlation with each other is below other thresholds, that is, a combination that has a high correlation with the state of unburned matter in the exhaust gas, and the selected data do not have a strong correlation with each other. By combining the above data, the state of unburned matter in the exhaust gas can be calculated with high accuracy. The correlation of the data can be obtained by performing a correlation analysis on the acquired data. The data to be acquired may be operation data of the engine 12, including the results of actually measuring the state of unburned matter in the exhaust gas, or the results of a simulation.
[0024] Examples of operating data include measured values of engine speed, intake manifold pressure, intake manifold temperature, exhaust pressure, exhaust temperature, intake air temperature, intake air pressure, fuel flow rate, ammonia flow rate, and in-cylinder pressure, as well as IMEP (indicated mean effective pressure) calculated from in-cylinder pressure, maximum in-cylinder pressure, and ECU setting values.
[0025] The operating data preferably combines four values: values related to the combustible material fed into the engine, values related to the air fed into the engine, values related to the engine's combustion pressure, and values related to the pressure or temperature of the gas flowing through the engine. Values related to ambient temperature may also be included. In addition, key parameters indicating the engine's operating state, such as engine speed, may also be included.
[0026] Examples of values related to the combustible materials fed into the engine include the mixing ratio of the first fuel (ammonia, etc.) and the second fuel (diesel, etc.), or the flow rate and supply amount of the first fuel (ammonia, etc.) and the second fuel (diesel, etc.). The mixing ratio of ammonia and diesel has a high correlation with the ammonia concentration. An example of a value related to the air fed into the engine is the coefficient of variation (cov) of the indicated mean effective pressure (IMEP). The coefficient of variation is obtained by dividing the measured value of the in-cylinder pressure or the standard deviation of the IMEP calculated from the in-cylinder pressure by the mean value. Examples of values related to the combustion pressure of the engine include exhaust pressure and intake pressure. An example of a value representing the pressure of the air flowing through the engine is the excess air ratio.
[0027] The calculation unit 44 is a calculation device and includes calculation circuits such as a CPU. The calculation unit 44 processes the data acquired by the operation data acquisition unit 42 to calculate the state of unburned material in the exhaust gas at a specific location in the exhaust piping 20.
[0028] The memory unit 46 is a memory that stores various information such as the calculation contents and programs of the arithmetic unit 44, and includes at least one of the following: RAM, main memory such as ROM, and external memory such as HDD. The memory unit 46 stores the learned program 48. The learned program 48 is a program that performs machine learning by inputting operating data of specified parameters of the engine 12 and outputting the state of unburned material in the exhaust gas at a specific position in the exhaust pipe 20.
[0029] The ammonia calculation process of the exhaust gas state calculation device 40 will be explained using Figure 2. Figure 2 is a flowchart showing an example of the process of the exhaust gas state calculation device. The process in Figure 2 can be executed by performing various processes in the calculation unit 44.
[0030] The arithmetic unit 44 obtains a learned program from the storage unit 46 (step S12). The arithmetic unit 44 obtains target operating data for the engine 12 (step S14). The arithmetic unit 44 inputs the engine operating data into the learned program and calculates the state of unburned gas in the exhaust (step S16). The arithmetic unit 44 outputs the calculated state of unburned gas in the exhaust to the control device 30 (step S18). The arithmetic unit 44 determines whether the calculation is complete (step S20). If the arithmetic unit 44 determines that the calculation is not complete (No in step S20), it returns to step S14 and repeats the calculation process of the state of unburned gas in the exhaust. If the arithmetic unit 44 determines that the calculation is complete (Yes in step S20), it terminates this process.
[0031] The exhaust gas state calculation device 40 can calculate the state of unburned material in the exhaust gas by acquiring specific operating data obtained from the control device 30 and processing it with a learned program.
[0032] Next, we will describe the processing of the combustion control device 29, specifically the processing of the control device 30 using the calculation results of the exhaust gas state calculation device 40. Figure 3 is a flowchart showing an example of the processing of the combustion control device.
[0033] The control device 30 acquires the target operating value of the engine (step S22). The control device 30 acquires the status of unburned material in the exhaust gas from the exhaust gas state calculation device 40 (step S24). The control device 30 calculates an operating command value based on the status of unburned material in the exhaust gas, the target operating value, and other operating data (step S26). The control device 30 controls each part of the engine based on the operating command value (step S28).
[0034] The combustion control device 29 controls the engine 12 with the control device 30 based on the calculation results of the exhaust gas state calculation device 40, thereby calculating the state of unburned material in the exhaust gas without installing a physical sensor to detect the state of unburned material in the exhaust gas, and enabling operation while suppressing ammonia leakage. For example, the combustion control device 29 maintains the operating state of the engine 12 if the state of unburned material in the exhaust gas is in the first state. The control device 30 also switches the operating state of the engine 12 if the state of unburned material in the exhaust gas is in the second state. Switching the operating state of the engine 12 in this case includes at least one of the following: stopping or reducing the supply of mixed fuel to the engine 12, and switching the air surplus ratio, ignition method, and intake air temperature of the engine 12. The air surplus ratio can be adjusted by controlling the intake air amount (the controlled objects are the opening of the intake throttle valve, the opening of the turbocharger wastegate / VG nozzle, the opening of the exhaust throttle valve, and the VVT setting) and the fuel flow rate. Ignition timing is controlled by the ignition timing in the case of spark ignition, and by the fuel injection pattern (injection timing, number of injection stages, common rail pressure, injection amount) in the case of compression autoignition. Temperature control is controlled by controlling the set temperature of the discharge gas using an intercooler.
[0035] <Program creation system> Next, we will explain an example of creating a pre-trained program used to calculate the state of unburned gas in exhaust gas. Figure 4 is a block diagram showing an example of a program creation system. Figure 5 is a block diagram showing an example of a calculation program creation device.
[0036] The program creation system 100 includes an engine 12, an ammonia supply device 14, a fuel supply device 16, an air supply device 18, an exhaust pipe 20, an SCR catalyst 22, an ASC (Ammonia Slip Catalyst) 24, a combustion control device 29, a control device 30, a calculation program creation device 70, ammonia concentration sensors 102 and 104, and a temperature sensor 106. The engine 12, ammonia supply device 14, fuel supply device 16, air supply device 18, exhaust pipe 20, SCR (Selective Catalytic Reduction) catalyst 22, ASC (Ammonia Slip Catalyst) 24, combustion control device 29, and control device 30 of the program creation system 100 are the same devices as the parts of the combustion system 10.
[0037] The ammonia concentration sensor 102 measures the ammonia concentration of the exhaust gas in the exhaust pipe 20 between the engine 12 and the SCR catalyst 22. The ammonia concentration sensor 104 measures the ammonia concentration of the exhaust gas in the exhaust pipe 20 downstream of the ACS 24. The ammonia concentration sensors 102 and 104 can be replaced with a vehicle-mounted exhaust gas measurement system (PEMS) or a constant potential sensor. The temperature sensor 106 measures the temperature of the exhaust gas in the exhaust pipe 20 downstream of the ACS 24. In this embodiment, the system is equipped with ammonia concentration sensors 102 and 104 and a temperature sensor 106, but it is sufficient to measure the state of unburned material in the exhaust gas, and either the ammonia concentration sensor 102 or 104 alone may be used, or the temperature sensor 106 may not be provided.
[0038] The program creation system 100 measures the relationship between the operating conditions of the engine 12 running in the combustion system 10 and the state of unburned material in the exhaust gas, and creates a learned program by processing the measured results with the program creation device 70.
[0039] The calculation program creation device 70 includes an operation data acquisition unit 72, an unburned component state acquisition unit 74, a calculation unit 76, and a storage unit 78. The operation data acquisition unit 72 acquires operation data such as control values for the engine 12 detected and generated by the control device 30, which is acquired by the control device 30. The operation data acquired by the operation data acquisition unit 72 is the value of a parameter used to calculate the state of unburned components in the exhaust gas. The unburned component state acquisition unit 74 acquires information on a classification (the first state, second state, and third state described above) that has been determined in advance based on the operation state of the engine 12. The first state, second state, third state, etc., are classified according to the unburned ammonia concentration measured by a concentration sensor, for example. The classification result of the state of unburned components in the exhaust gas may be stored in a storage unit such as an external storage device 60, or in a storage unit not shown in the control device 30 or the calculation program creation device 70.
[0040] The arithmetic unit 76 is an arithmetic device and includes arithmetic circuits such as a CPU. The arithmetic unit 76 has a machine learning unit 80 and a data creation unit 82. The machine learning unit 80 performs deep learning using a neural network or the like. The machine learning unit 80 takes data acquired by the operation data acquisition unit 42 as input and creates a trained program that outputs data on the state of unburned material in the exhaust gas at a specific location in the exhaust pipe 20. The data creation unit 82 creates training data that associates the data acquired by the operation data acquisition unit 42 with the data on the state of unburned material in the exhaust gas at a specific location in the exhaust pipe 20.
[0041] The memory unit 78 is a memory that stores various information such as the calculation contents and programs of the arithmetic unit 76, and includes at least one of the following: RAM, main memory such as ROM, and external memory such as HDD. The memory unit 78 stores the learning program 90 and the training data 92. The learning program 90 is a machine learning program that takes specific operating data of the engine 12 as input and creates a trained program that outputs the state of unburned material in the exhaust gas.
[0042] The training data 92 is the data used for training in the learning program 90. Figure 6 is an explanatory diagram showing an example of the training data. The training data 92 is data in which operating data and the classification of the state of unburned components in the exhaust gas are stored in time series. The operating data in Figure 6 are the ammonia co-firing ratio, the cov value (coefficient of variation) of the indicated mean effective pressure (IMEP), exhaust pressure, and excess air ratio. In this embodiment, time series data is used, but any data consisting of a combination of operating data and the classification result of the state of unburned components in the exhaust gas at the corresponding point in time would suffice.
[0043] Figure 7 is a flowchart showing an example of the processing of the calculation program creation device. The calculation program creation device 70 executes each process in Figure 7 with the calculation unit 76. The calculation unit 76 acquires operating data, that is, operating data used to calculate the state of unburned components in the exhaust gas (step S40). The calculation unit 76 acquires the classification result of the state of unburned components in the exhaust gas (step S42). The calculation unit 76 creates training data (step S44). In other words, the calculation unit 76 creates data in the data creation unit 82 that associates the operating data with the classification result of the state of unburned components in the exhaust gas. The data creation unit 82 stores the created data in the storage unit 78. The calculation unit 76 performs learning with the training data and creates a trained program (step S46).
[0044] In the above embodiment, data obtained by operating the actual machine and taking measurements using ammonia concentration sensors 102 and 104 was used as training data, but training data may also be created through simulation.
[0045] In the above embodiment, a trained program was created using machine learning to calculate the state of unburned components in exhaust gas based on operating data of specific engine parameters, but the present invention is not limited thereto. The combustion control device can use a calculation rule to calculate the state of unburned components in exhaust gas based on operating data of specific engine parameters, and the calculation rule may be calculated using regression analysis.
[0046] Figure 8 is a flowchart showing another example of the processing of the calculation program creation device. The calculation program creation device 70 executes each process in Figure 8 with the calculation unit 76. The calculation unit 76 acquires operating data, that is, operating data used to calculate the state of unburned components in the exhaust gas (step S50). The calculation unit 76 labels the state of unburned components in the exhaust gas using clustering for the acquired operating data (step S52). In other words, the calculation unit 76 generates a trained program by so-called unsupervised learning. For example, the calculation unit 76 groups the acquired operating data into an arbitrary number of clusters and labels each cluster as first state, second state, third state, etc.
[0047] The technical scope of the present invention is not limited to the embodiments described above, and modifications can be made as appropriate without departing from the spirit of the invention. For example, in the embodiments described above, the state of unburned material in the exhaust gas was determined using the concentration of unburned material in the exhaust gas (a co-firing ratio that is highly correlated with the concentration), but the invention is not limited thereto. For example, the state of unburned material may be determined based on the catalyst temperature caused by the concentration of unburned material in the exhaust gas. For example, the unburned state may be determined based on the catalyst temperature detected by the catalyst temperature sensor 108 (see Figure 1).
[0048] [Effects of this embodiment] (1) An exhaust gas state calculation device for calculating the state of unburned components in exhaust gas discharged from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, comprising: a data acquisition unit that acquires one or more operational data from the operational data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; a storage unit that stores a calculation rule calculated based on the relationship between the operational data and the state of unburned components in the exhaust gas, which takes operational data as input and outputs the state of unburned components in the exhaust gas; and a calculation unit that processes the operational data acquired by the data acquisition unit with the calculation rule stored in the storage unit to calculate the state of unburned components in the exhaust gas.
[0049] The exhaust gas state calculation device can accurately calculate the state of unburned components in the exhaust gas based on operating data. Furthermore, by estimating the state of unburned components in the exhaust gas rather than estimating the concentration, the processing load can be reduced, and the calculation speed can be improved. In addition, a smaller number of datasets are required when learning the calculation rules, and hyperparameter tuning becomes easier.
[0050] (2) The calculation rule is a pre-trained program calculated using machine learning, as described in (1) for the exhaust gas state calculation device. With this configuration, it is possible to determine the unburned state without installing concentration sensors or the like.
[0051] (3) The data acquisition unit is an exhaust gas state calculation device as described in (1) or (2) that acquires one or more operating data from among multiple operating data that have low correlation with each other. With this configuration, the state of unburned material in the exhaust gas can be estimated with high accuracy by acquiring operating data that have low correlation with each other from among multiple operating data.
[0052] (4) The data acquisition unit is an exhaust gas state calculation device according to any one of (1) to (3) that acquires at least one of the following as operating data: the coefficient of variation of the indicated mean effective pressure of a dual-fuel engine, the excess air ratio, the exhaust pressure, and the temperature difference before and after the catalyst. With this configuration, the state of unburned material in the exhaust gas can be accurately estimated by acquiring at least one of the following as operating data: the coefficient of variation of the indicated mean effective pressure of a dual-fuel engine, the excess air ratio, the exhaust pressure, and the temperature difference before and after the catalyst.
[0053] (5) An exhaust gas state calculation device according to any one of (1) to (4), wherein the state of unburned components in the exhaust gas includes a first state that maintains the operating state of the dual-fuel engine, a second state that switches the operating state of the dual-fuel engine, and a third state that transitions between the first and second states. With this configuration, by defining the state of unburned components in the exhaust gas as a first state, a second state, and a third state, the processing load can be reduced compared to estimating the concentration, and the calculation speed can be improved.
[0054] (6) The exhaust gas state calculation device according to (5), wherein the switching of the operating state of the dual-fuel engine in the second state includes at least one of stopping or reducing the supply of mixed fuel, and switching the excess air ratio, ignition method, and intake air temperature. With this configuration, when the state of unburned material in the exhaust gas becomes the second state, the supply of mixed fuel can be stopped or reduced, and the excess air ratio, ignition method, and intake air temperature can be switched, thereby enabling an appropriate operating state.
[0055] (7) An exhaust gas state calculation device according to any one of (1) to (6), wherein the first fuel is ammonia and the second fuel is diesel fuel. With this configuration, when ammonia is used as the first fuel and diesel fuel is used as the second fuel, the state of unburned material in the exhaust gas can be appropriately calculated based on operating data, and the processing burden can be reduced.
[0056] (8) A combustion control device comprising an exhaust gas state calculation device as described in any of (1) to (7), and a control device for controlling engine combustion, wherein the control device controls the operation of the engine based on the state of unburned material in the exhaust gas calculated by the exhaust gas state calculation device. With this configuration, the dual-fuel engine can be brought into an appropriate operating state based on the state of unburned material in the exhaust gas.
[0057] (9) The combustion control device described in (8) which controls the operation of the dual-fuel engine to maintain the operating state of the dual-fuel engine or to switch the operating state of the dual-fuel engine based on the state of unburned material in the exhaust gas. With this configuration, the dual-fuel engine can be brought into an appropriate operating state by switching the operating state of the dual-fuel engine based on the state of unburned material in the exhaust gas.
[0058] (10) An exhaust gas state calculation method for calculating the state of unburned components in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, comprising the steps of: acquiring one or more operational data from the operational data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; acquiring a calculation rule that takes operational data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operational data and the state of unburned components in the exhaust gas; and processing the acquired operational data with the acquired calculation rule to calculate the state of unburned components in the exhaust gas. With this configuration, the state of unburned components in the exhaust gas can be appropriately calculated based on operational data. Furthermore, by estimating the state of unburned components in the exhaust gas rather than estimating the concentration, the processing load can be reduced and the calculation speed can be improved. In addition, the number of datasets required when learning the calculation rule can be reduced, and the adjustment of hyperparameters can be simplified.
[0059] (11) An exhaust gas state calculation program for calculating the state of unburned components in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel and a second fuel different from the first fuel, wherein the program causes a computer to perform the following steps: acquire one or more operational data from the operational data of the dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel; acquire a calculation rule that takes the operational data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operational data and the state of unburned components in the exhaust gas; and process the acquired operational data using the acquired calculation rule to calculate the state of unburned components in the exhaust gas. With this configuration, the state of unburned components in the exhaust gas can be appropriately calculated based on the operational data. Furthermore, by estimating the state of unburned components in the exhaust gas rather than estimating the concentration, the processing load can be reduced and the calculation speed can be improved. In addition, the number of datasets required when learning the calculation rule can be reduced, and hyperparameter tuning can be simplified. [Explanation of Symbols]
[0060] 10 Combustion System 12 Engines 14. Ammonia supply system 16 Fuel supply system 18. Air supply device 20 Exhaust piping 22 SCR catalyst 24 ASC 29, 29a, 29b, 29c Combustion control device 30 Control device 40 Exhaust gas condition calculation device 42 Operation data acquisition unit 44 Storage section 46 Memory section 48 Trained Programs 60 External storage device 70 Calculation program creation device 72 Operation Data Acquisition Unit 74 Unburned component status acquisition unit 76 Arithmetic section 78 Memory section 80 Machine Learning Department 82 Data Creation Department 90 Learning Programs 92 Training Data 100 Program Creation Systems
Claims
1. An exhaust gas state calculation device for calculating the state of unburned components in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel with a second fuel different from the first fuel, A data acquisition unit that acquires one or more operational data from the operational data of a dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel. A storage unit that stores a calculation rule calculated based on the relationship between the aforementioned operating data and the state of unburned material in the exhaust gas, which takes the aforementioned operating data as input and outputs the state of unburned material in the exhaust gas, A calculation unit that processes the operating data acquired by the data acquisition unit using the calculation rule stored in the storage unit to calculate the state of unburned material in the exhaust gas. An exhaust gas condition calculation device equipped with the following features.
2. The aforementioned calculation rule is a trained program calculated using machine learning. The exhaust gas state calculation device according to claim 1.
3. The data acquisition unit acquires one or more of the operation data from among the multiple operation data that have low correlation with each other. The exhaust gas state calculation device according to claim 1.
4. The data acquisition unit acquires at least one of the following as operating data: the coefficient of variation of the indicated mean effective pressure of the dual-fuel engine, the excess air ratio, the exhaust pressure, and the temperature difference before and after the catalyst. The exhaust gas state calculation device according to claim 1.
5. The state of unburned material in the exhaust gas includes a first state in which the operating state of the dual-fuel engine can be maintained, a second state in which the operating state of the dual-fuel engine is switched, and a third state in which the state transitions between the first and second states. The exhaust gas state calculation device according to claim 1.
6. Switching the operating state of the dual-fuel engine in the second state includes at least one of stopping or reducing the supply of the mixed fuel, and switching the excess air ratio, ignition method, and intake air temperature. The exhaust gas state calculation device according to claim 5.
7. The first fuel is ammonia, The second fuel is diesel fuel. The exhaust gas state calculation device according to claim 1.
8. An exhaust gas state calculation device according to any one of claims 1 to 7, The dual-fuel engine includes a control device for controlling combustion, The control device controls the operation of the dual-fuel engine based on the state of unburned material in the exhaust gas calculated by the exhaust gas state calculation device. Combustion control device.
9. The control device performs control to maintain the operating state of the dual-fuel engine or to switch the operating state of the dual-fuel engine based on the state of unburned material in the exhaust gas. The combustion control device according to claim 8.
10. A method for calculating the state of unburned material in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel with a second fuel different from the first fuel, A step of acquiring one or more operational data from the operational data of a dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel, A step of obtaining a calculation rule that takes the operating data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operating data and the state of unburned components in the exhaust gas, The steps include: processing the acquired operating data according to the acquired calculation rule to calculate the state of unburned material in the exhaust gas; A method for calculating exhaust gas conditions, including the exhaust gas state.
11. An exhaust gas state calculation program for calculating the state of unburned components in exhaust gas emitted from a dual-fuel engine that burns a mixed fuel obtained by mixing a first fuel with a second fuel different from the first fuel, A process for acquiring one or more operational data from the operational data of a dual-fuel engine, including at least the mixing ratio of the first fuel and the second fuel, or the flow rates of the first fuel and the second fuel. A process to obtain a calculation rule that takes the operating data as input and outputs the state of unburned components in the exhaust gas, calculated based on the relationship between the operating data and the state of unburned components in the exhaust gas, The process involves processing the acquired operating data according to the acquired calculation rule to calculate the state of unburned material in the exhaust gas. A program that uses a computer to calculate exhaust gas conditions.
Citation Information
Patent Citations
METHOD FOR CONTROLLING FUEL SUPPLY TO HIGH PRESSURE GAS INJECTION ENGINE AND ITS ENGINE
JP3432098B2