Operation condition calculation system, controller for internal combustion engine, and adaptation device for internal combustion engine

Abstract

To calculate a favorable operating condition in terms of a performance index.SOLUTION: An operation condition calculation system comprises: a first learning device that inputs an operation condition of an on-vehicle device installed in a vehicle and outputs a physical state that occurs when the on-vehicle device operates based on the operation condition; a second learning device that inputs the output of the first learning device, and outputs a performance index; a gradient calculation unit that calculates the gradients of the first learning device and the second learning device; and an operation change unit that changes the operation condition using the gradients.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to an operating condition calculation system, a control device for an internal combustion engine, and an adaptation device for an internal combustion engine. [Background technology]

[0002] In recent years, as the development of environmentally friendly engines progresses, it has become an issue to reduce emissions of PN (Particulate Number) and THC (Total Hydro Carbon), which are generated when the engine is started or when it is operating at low temperatures. These are particularly generated when the external atmosphere is low temperature or when the engine is started, and suppression of these is required. In addition, it is necessary to predict their generation in advance by real-time control and adjust the operating conditions. Patent Document 1 discloses a method for monitoring an engine using a cascaded neural network including multiple neural networks, the method including the steps of storing data corresponding to the cascaded neural network in a memory, inputting signals generated by multiple engine sensors to the cascaded neural network, updating a second neural network with the output of the first neural network at a first speed, the output being based on the input signal, and outputting at least one engine control signal from the second neural network at a second speed, the second speed being faster than the first speed. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-328837 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the invention described in Patent Document 1, there is room for improvement in setting the operating conditions. [Means for solving the problem]

[0005] A driving condition calculation system according to a first aspect of the present invention includes a first learning device that receives as input driving conditions of an on-board device mounted on a vehicle and outputs a physical state generated by operation of the on-board device based on the driving conditions, a second learning device that receives as input the output of the first learning device and outputs a performance index, a gradient calculation unit that calculates gradients of the first learning device and the second learning device, and a driving change unit that changes the driving conditions using the gradient. A control device for an internal combustion engine according to a second aspect of the present invention includes the operating condition calculation system described above. An adaptation device for an internal combustion engine according to a third aspect of the present invention includes the operating condition calculation system described above. Effect of the Invention

[0006] According to the present invention, it is possible to calculate preferable operating conditions from the viewpoint of a performance index. [Brief description of the drawings]

[0007] [Figure 1] Schematic diagram of an internal combustion engine [Diagram 2] FIG. 1 is a diagram showing an example of a malfunction in an internal combustion engine. [Diagram 3] Functional configuration diagram of a computing device according to a first embodiment [Figure 4] Hardware configuration diagram of the computing device [Diagram 5] 1 is a flowchart showing the processing of a computing device. [Figure 6] Functional configuration diagram of a calculation device in Modification 1 [Figure 7] Functional configuration diagram of a computing device according to a second embodiment [Figure 8] Functional configuration diagram of a computing device according to a third embodiment [Figure 9] Functional configuration diagram of a computing device according to the fourth embodiment [Figure 10] A functional configuration diagram of a computing device according to a fifth embodiment. [Figure 11] Functional configuration diagram of a computing device according to the sixth embodiment DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] --First embodiment-- Hereinafter, a first embodiment of an operating condition calculation system will be described with reference to FIGS.

[0009] 1 is a schematic diagram of an internal combustion engine 1 that is the subject of calculations by the operating condition calculation system. The internal combustion engine 1, which is mounted on a vehicle 9, includes a piston 101, a cylinder 102, and a cylinder head 104. The cylinder head 104 defines a combustion chamber 103 by a piston crown surface 101P of the piston 101 and a combustion chamber inner wall 109 of the cylinder 102. The internal combustion engine 1 also includes at least one of a direct injection injector 111 and a port injection injector 112 as a fuel injection injector.

[0010] An ignition plug 105 having an electrode 106 for igniting the air-fuel mixture is disposed directly above the combustion chamber 103. During the intake stroke, air flows into the combustion chamber 103 through an intake port 107 that communicates with the main combustion chamber and has an intake valve 113 open. Fuel is sprayed in a mist form from a direct injection injector 111 or a port injection injector 112 and sent to the combustion chamber 103, where it vaporizes and becomes an air-fuel mixture. The mixture is then compressed by the piston 101, and a main ignition signal is sent to the ignition coil 110 at the appropriate time, causing the spark plug 105 to ignite the air-fuel mixture in the combustion chamber 103 and burn the mixture in the combustion chamber 103. This causes the pressure in the combustion chamber 103 to rise, pushing down the piston 101, and the connecting rod 113A rotates the crankshaft 115 to obtain power.

[0011] FIG. 2 is a diagram showing an example of a malfunction in the internal combustion engine 1. When liquid fuel is injected from an injector, the liquid fuel may adhere to an opposing wall in a liquid state. For example, as shown in FIG. 2, when a spray 201 is injected into a combustion chamber 103 by a port injector 112, a fuel liquid film 202 is formed on a piston crown surface 101P, a combustion chamber inner wall 109, or the like. In addition, since the spray 201 is high-speed, droplets 203 may fly out into the combustion chamber 103 due to the impact when the spray 201 collides with the piston crown surface 101P, the combustion chamber inner wall 109, or the like. In addition, tip wet 204 may occur in which droplets of fuel remain and adhere around the spray port of the direct injector 111.

[0012] When the mixture in the combustion chamber 103 is ignited by the spark plug 105, if any fuel film 202 or droplets 203 remain, they will burn in a liquid state. At this time, the remaining fuel film 202, droplets 203, and tip wet 204 at the moment the flame arrives will not contribute to the engine output and will diffuse and burn. As a result, PN (Particulate Number) and THC (Total Hydro Carbon) will be generated. The same problem will occur even if the port injector 112 is used. Fuel that cannot contribute to the engine output, such as the fuel film 202, droplets 203, and tip wet 204, is called unburned fuel 205. In addition, if the generation of unburned fuel 205 reduces the fuel concentration of the mixture in the combustion chamber 103 and there is insufficient fuel around the spark plug 105, problems such as the flame not growing sufficiently when ignited by the spark plug 105 and misfire, or unstable combustion will occur.

[0013] In addition, under cold engine conditions, such as low-temperature start-up, not only is unburned fuel 205 likely to be generated, but there is also the following problem. That is, since the temperature inside the combustion chamber 103 is often low, the number of misfires increases, and a large amount of PN and THC is generated, which may lead to a deterioration in combustion stability. Furthermore, in the operation of an internal combustion engine, the internal state of the combustion chamber 103, including the amount of unburned fuel 205, is intricately related to many factors, such as external factors such as the flow inside the cylinder, the outside air temperature, the cooling water temperature, and the fuel temperature, and control variables such as the engine speed, the fuel injection timing, the fuel injection division number, and the ignition timing. In addition, since the amount of PN and THC generated is determined through complex physical phenomena such as combustion and chemical reactions, it is not easy to predict the amount of PN and THC generated from the operating conditions, etc.

[0014] In response to this, several techniques have been proposed for predicting final exhaust components from operating conditions, etc., by using machine learning, such as a neural network. However, as described above, the prediction accuracy is reduced because there are too many operating conditions on which learning is based, and conditions unrelated to the final output are also input, resulting in noise. Therefore, in this embodiment, focusing on the fact that the exhaust components from the internal combustion engine 1 and the combustibility within the internal combustion engine 1 are largely dependent on the state of the mixture ignited by the spark plug 105 in the combustion chamber 103, preferable operating conditions are calculated from the viewpoint of performance indexes, as described below.

[0015] FIG. 3 is a functional configuration diagram of the arithmetic device 800. The arithmetic device 800 is a general-purpose computer or an electronic control device. The arithmetic device 800 may be mounted on a vehicle 9 having an internal combustion engine 1, or may not be mounted on the vehicle 9. When the arithmetic device 800 is mounted on the vehicle 9, the internal combustion engine 1 can be controlled using the calculation result of the arithmetic device 800. In this case, the arithmetic device 800 can be called a "control device" of the internal combustion engine 1. When the arithmetic device 800 is not mounted on the vehicle 9, for example, in the design stage of the internal combustion engine 1, the optimal control conditions according to the design specifications of the internal combustion engine 1 are determined. In this case, the arithmetic device 800 is also called an "adaptation device" of the internal combustion engine 1. The arithmetic device 800 includes a first learning device 302, a second learning device 306, a first gradient calculation unit 307, a second gradient calculation unit 308, and an operation change unit 309.

[0016] The first learning device 302 is a learning device that has been previously trained. An input of the first learning device 302 is the operating condition 301 of the internal combustion engine 1, and an output of the first learning device 302 is a physical quantity 303 indicating the internal state at the ignition time of the combustion chamber 103. The second learning device 306 is a learning device that has been previously trained. An input of the second learning device 306 is the physical quantity 303 indicating the internal state at the ignition time of the combustion chamber 103, and an output of the second learning device 306 is a performance index 305. Although not shown in FIG. 3, a part or all of the operating condition 301 may be input to the second learning device 306. The data used for the learning of the first learning device 302 and the second learning device 306 may be data obtained by numerical analysis or may be an actual measurement value. The operating condition 301 is data actually acquired from a sensor mounted on the vehicle 9, or data that can be acquired from a sensor mounted on the vehicle 9.

[0017] The operating conditions 301 include specific conditions, which are a plurality of detailed operating conditions. The specific conditions include intake air temperature, cooling water temperature, load, number of fuel injections, rotation speed, injection amount for each injection, injection start time, injection duration, fuel pressure, ignition timing, and the like. The specific conditions may further include the number of divisions at the time of split fuel injection, injection time, injection duration, and injection division ratio. The specific conditions include conditions that cannot be easily changed in the internal combustion engine 1, such as the outside air temperature (hereinafter also referred to as "external factors"), and conditions related to the control of the internal combustion engine 1 (hereinafter also referred to as "control variables"). The external factors include the outside air temperature, cooling water temperature, and load. The control variables include the number of fuel injections, injection amount for each injection, injection start time, injection duration, fuel pressure, ignition time, and the like. Various combinations of the operating conditions 301 are assumed, but for example, the rotation speed, load, intake pressure, intake air temperature, water temperature, fuel pressure, number of fuel injections, fuel injection timing, division ratio, and ignition time may be essential components.

[0018] The physical quantity 303 includes specific physical quantities that are a plurality of specific physical quantities. The specific physical quantities are, for example, the amount of the mixture, the temperature in the combustion chamber at the time of ignition, and the pressure in the combustion chamber at the time of ignition. Various combinations of the physical quantity 303 are assumed, but the temperature in the cylinder, the pressure in the cylinder, the amount of attached fuel, and the amount of rich mixture at the time of ignition may be essential components.

[0019] The performance index 305 includes one or more specific indexes. The specific indexes are, for example, the amount of PN generated, the amount of THC generated, and the combustion variation rate. The specific indexes may also include an index indicating that some malfunction has occurred in the internal combustion engine, such as the exhaust temperature and the catalyst temperature. The occurrence of PN is strongly influenced by at least the amount of unburned fuel 205, represented by fuel adhesion and floating droplets generated in the combustion chamber, the amount of a rich fuel mixture, and the temperature and pressure in the combustion chamber at the time of ignition. THC is strongly influenced by the fuel that could not be burned and the temperature of the combustion chamber 103, and the combustion variation rate is influenced by the fuel richness near the electrode 106 of the ignition plug 105, the pressure and temperature of the combustion chamber 103, and the homogeneity of the mixture in the combustion chamber 103.

[0020] The neural network constituting the first learning device 302 has nodes in the input layer as many as the specific conditions, and has nodes in the output layer as many as the specific physical quantities. The neural network constituting the second learning device 306 has nodes in the input layer as many as the specific physical quantities, and has nodes in the output layer as many as the specific indicators.

[0021] For example, it is possible to create a single learning device that inputs the operating conditions 301 and outputs the performance index 305, but by combining the first learning device 302 and the second learning device 306, the scale of each learning device can be reduced. Also, the number of calculations required in the learning device to estimate the performance index 305 can be reduced, and the estimation accuracy can be improved.

[0022] The first gradient calculation unit 307 calculates the relationship between the input and output in the first learning device 302 as a first gradient 321. Specifically, the first gradient calculation unit 307 outputs a control variable having a maximum gradient for each specific physical quantity that is the output of the first learning device 302, and the value of the gradient. The second gradient calculation unit 308 calculates the relationship between the input and output in the second learning device 306 as a second gradient 322. Specifically, the second gradient calculation unit 308 outputs a specific physical quantity having a maximum gradient for each performance index that is the output of the second learning device 306, and the value of the gradient. Therefore, the operation change unit 309 can acquire the influence of the physical quantity 303 on each performance index 305 and the influence of the operating condition 301 on each physical quantity 303 as gradient values.

[0023] The operation change unit 309 receives the operation conditions 301, the first gradient 321, and the second gradient 322, and outputs new operation conditions 310. However, the new operation conditions 310 only have different names and symbols to distinguish them from the operation conditions 301 that are initially read, and the first learning device 302 reads the new operation conditions 310 in the same way as the operation conditions 301. In calculating the new operation conditions 310, the operation change unit 309 utilizes, for example, the Newton method using the gradient vector obtained from the first gradient calculation unit 307.

[0024] In the case where the performance index 305 includes a plurality of indexes, the index to be improved may be specified in advance, or a priority order may be determined, in the operation change unit 309. The operation change unit 309 calculates new operation conditions 310 in which the control variables are changed so as to improve the performance index 305. For example, the new operation conditions 310 increase the injection time from the current "100 ms" by "10 ms" to "110 ms".

[0025] FIG. 4 is a hardware configuration diagram of the arithmetic device 800. The arithmetic device 800 includes a CPU 81 which is a central arithmetic unit, a ROM 82 which is a read-only storage device, a RAM 83 which is a readable and writable storage device, and an input / output device 84 which is a user interface. The CPU 81 loads a program stored in the ROM 82 into the RAM 83 and executes it to perform various calculations such as the first learning device 302 and the operation change unit 309. The arithmetic device 800 may be realized by a field programmable gate array (FPGA) which is a rewritable logic circuit or an application specific integrated circuit (ASIC) which is an integrated circuit for a specific application, instead of the combination of the CPU 81, the ROM 82, and the RAM 83. The arithmetic device 800 may also be realized by a combination of different configurations, for example, a combination of the CPU 81, the ROM 82, the RAM 83, and an FPGA, instead of the combination of the CPU 81, the ROM 82, and the RAM 83.

[0026] 5 is a flowchart showing the processing of the arithmetic device 800. In step S301, the first learning device 302 reads the initial value of the operating condition 301. In the following step S302, the first learning device 302 and the second learning device 306 calculate the performance index 305 based on the operating condition 301. Specifically, the first learning device 302 calculates the physical quantity 303 based on the operating condition 301, and the second learning device 306 calculates the performance index 305 based on the physical quantity 303.

[0027] In the next step S303, the operation change unit 309 judges whether or not the performance index 305 calculated in step S302 is a desirable value. For example, the operation change unit 309 makes the judgment in this step by comparing a predetermined threshold value with the performance index 305. If the operation change unit 309 judges that the performance index 305 is a desirable value, it ends the process shown in Fig. 5, and if it judges that the performance index 305 is not a desirable value, it proceeds to step S304.

[0028] In step S304, the second gradient calculation unit 308 identifies an index influencing physical quantity, which is a physical quantity that has a strong influence on the performance index 305. In step S305, the first gradient calculation unit 307 identifies an index influencing operating condition, which is an operating condition that has a strong influence on the index influencing physical quantity. In the following step S306, the operation change unit 309 updates the operating condition 301 and returns to step S302. Note that when returning from step S306 to step S302, the operating condition 301 updated in step S306 is used in step S302.

[0029] According to the above-described first embodiment, the following advantageous effects can be obtained. (1) A calculation device 800, which can also be called an operating condition calculation system, includes a first learning device 302 that receives operating conditions 301 of an internal combustion engine 1, which is an on-vehicle device, as input, and outputs a physical quantity 303 indicating a physical state generated by the operation of the internal combustion engine 1 based on the operating conditions 301, a second learning device 306 that receives the output of the first learning device 302 as input, and outputs a performance index 305, a first gradient calculation unit 307 and a second gradient calculation unit 308 that are gradient calculation units that calculate the gradients of the first learning device 302 and the second learning device 306, and an operation change unit 309 that changes the operating conditions 301 using the first gradient 321 and the second gradient 322. Therefore, it is possible to calculate an operating condition 301 that is preferable from the viewpoint of the performance index 305.

[0030] (2) The performance index 305 is at least one of the exhaust components and the combustion variation rate from the internal combustion engine 1. The operation change unit 309 changes at least one of the injection time, the injection split ratio, the fuel pressure, and the ignition timing during split injection of fuel.

[0031] (3) The first learning unit 302 receives at least the rotation speed, load, intake pressure, intake air temperature, water temperature, fuel pressure, number of fuel injections, fuel injection timing, split ratio, and ignition time, and outputs at least the in-cylinder temperature, in-cylinder pressure, amount of fuel attached, and amount of rich mixture at the ignition time.

[0032] (4) At least the in-cylinder temperature immediately before ignition, the in-cylinder pressure, the amount of fuel attached, the amount of rich mixture, and some or all of the inputs to the first learning unit are input to the second learning unit 306. The second learning unit 306 outputs at least the exhaust components generated after combustion and the combustion variation rate.

[0033] (Variation 1) FIG. 6 is a functional configuration diagram of a calculation device 800 in the first modified example. This diagram is different from FIG. 3 in the first embodiment in that a fuel variable 301C is added to the operating condition 301. The fuel variable 301C is, for example, the properties of the fuel supplied to the internal combustion engine 1, such as viscosity and mass, information on the type of fuel, the time when the fuel was put into the fuel tank, the time when the fuel injection device was replaced, and the like. In the combustion of the internal combustion engine 1, the combustion form changes depending on the fuel used, and the amount of PN, THC, and the like emitted also changes. Furthermore, since the fuel properties change depending on the deterioration state of the fuel, by including the fuel state in the operating condition 301, it is possible to respond to the case where the fuel is refueled or the type of fuel changes.

[0034] According to the first modification, the following advantageous effects can be obtained. (5) The first learning unit 302 receives the properties of the fuel fed to the internal combustion engine 1. Therefore, the operating conditions 301 according to the properties of the fuel can be calculated.

[0035] (6) The type of fuel fed to the internal combustion engine 1 and the time when the fuel was fed to the fuel tank are input to the first learning unit 302. Therefore, it is possible to calculate the operating conditions 301 according to more detailed fuel properties, such as the type of fuel and the time when the fuel was fed to the fuel tank.

[0036] (Variation 2) The calculation by the calculation device 800 can be applied not only to the internal combustion engine 1 but also to various parts mounted on many vehicles 9. For example, it can also be applied to an electric motor as described below. For example, in an electric motor, the current value applied to the motor corresponds to the operating condition 301, the generated magnetic flux and the interlinked magnetic flux to the core correspond to the physical quantity 303, and the torque, speed, and efficiency correspond to the performance index 305.

[0037] (Variation 3) In the first embodiment described above, the learning of the neural networks included in the first learner 302 and the second learner 306 is completed in advance. However, the first learner 302 and the second learner 306 may perform additional learning using measurable values ​​in parallel with inference.

[0038] --Second embodiment-- A second embodiment of the operating condition calculation system will be described with reference to FIG. 7. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that the first learning device 302 is composed of multiple learning devices.

[0039] FIG. 7 is a functional configuration diagram of a calculation device 800A in the second embodiment. FIG. 7 differs from FIG. 3 in the first embodiment in that a first learning device 302A is provided instead of the first learning device 302. The first learning device 302A includes one or more learning devices that estimate only one specific physical quantity. FIG. 7 shows a case where the first learning device 302A estimates three specific physical quantities, and the first learning device 302A includes a 1-1 learning device 302-1, a 1-2 learning device 302-2, and a 1-3 learning device 302-3. The neural networks that constitute each of the 1-1 learning device 302-1, the 1-2 learning device 302-2, and the 1-3 learning device 302-3 have only a node that indicates any one of the specific physical quantities in the output layer.

[0040] The neural networks constituting the 1-1st learning device 302-1, the 1-2nd learning device 302-2, and the 1-3rd learning device 302-3 may have different numbers of nodes in the input layer. That is, each neural network may input only specific conditions that are closely related to the specific physical quantity to be estimated, or may input conditions excluding specific conditions that are weakly related to the specific physical quantity to be estimated from all specific conditions.

[0041] According to the above-described second embodiment, the following advantageous effects can be obtained. (7) The first learning device 302A has one machine learning device for each specific physical quantity, which is a specific physical quantity. All or part of the operating conditions are input to each of the 1-1 learning device 302-1, the 1-2 learning device 302-2, etc. Therefore, the size of each neural network constituting the first learning device 302A can be reduced, and the accuracy of the output result can be expected to be improved.

[0042] --Third embodiment-- The third embodiment of the operating condition calculation system will be described with reference to FIG. 8. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. The present embodiment differs from the first embodiment mainly in that the second learning device 306 is composed of multiple learning devices.

[0043] Fig. 8 is a functional configuration diagram of a calculation device 800B in the third embodiment. The difference from Fig. 3 in the first embodiment is that a second learning device 306A is provided instead of the second learning device 306. The second learning device 306A includes one or more learning devices that estimate only one specific indicator. Fig. 8 shows a case where the second learning device 306A estimates three specific indicators, and the second learning device 306A includes a 2-1 learning device 306-1, a 2-2 learning device 306-2, and a 2-3 learning device 306-3.

[0044] The neural networks constituting the 2-1st learning device 306-1, the 2-2nd learning device 306-2, and the 2-3rd learning device 306-3 each have only a node indicating one of the specific indices in the output layer. Specifically, the 2-1st learning device 306-1, the 2-2nd learning device 306-2, and the 2-3rd learning device 306-3 each output a first index 305-1, a second index 305-2, and a third index 305-3.

[0045] The neural networks constituting the 2-1st learning device 306-1, the 2-2nd learning device 306-2, and the 2-3rd learning device 306-3 may have different numbers of nodes in the input layer. That is, each neural network may receive only a specific physical quantity that is closely related to the specific indicator to be estimated as an input, or may receive conditions excluding specific physical quantities that are weakly related to the specific indicator to be estimated from all specific conditions as an input.

[0046] According to the above-described third embodiment, the following advantageous effects can be obtained. (8) The second learning device 306A has one machine learning device for each performance index. All or part of the output of the first learning device 302 is input to each of the 2-1 learning device 306-1, the 2-2 learning device 306-2, etc. This makes it possible to reduce the size of each neural network constituting the second learning device 306A, and is expected to improve the accuracy of the output results.

[0047] --Fourth embodiment-- A fourth embodiment of the operating condition calculation system will be described with reference to FIG. 9. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that the first learning device 302 and the second learning device 306 are each composed of multiple learning devices.

[0048] Fig. 9 is a functional configuration diagram of a calculation device 800C in the fourth embodiment. The difference from Fig. 3 in the first embodiment is that a first learning device 302A is provided instead of the first learning device 302, and a second learning device 306A is provided instead of the second learning device 306. The first learning device 302A is as described in the second embodiment, and the second learning device 306A is as described in the third embodiment.

[0049] The first learning device 302A includes one or more learning devices that estimate only one specific physical quantity. Fig. 7 shows a case where the first learning device 302A estimates three specific physical quantities, and the first learning device 302A includes a 1-1 learning device 302-1, a 1-2 learning device 302-2, and a 1-3 learning device 302-3. The neural networks that constitute the 1-1 learning device 302-1, the 1-2 learning device 302-2, and the 1-3 learning device 302-3 each have only a node that indicates one of the specific physical quantities in the output layer.

[0050] --Fifth embodiment-- A fifth embodiment of the operating condition calculation system will be described with reference to FIG. 10. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that the operating state is detected.

[0051] Fig. 10 is a functional configuration diagram of a calculation device 800D in the fifth embodiment. The difference from Fig. 3 in the first embodiment is that the calculation device 800D includes a driving state detection unit 331 and a sensor 332. The sensor 332 measures a value related to a performance index 305 of a vehicle equipped with the calculation device 800D. The sensor 332 is, for example, a PM sensor or a THC measurement device. The driving state detection unit 331 receives a measured value from the sensor 332 and receives the performance index 305 from the second learning device 306.

[0052] The performance index 305 is estimated based on the output of the sensor 332. The driving state detection unit 331 compares the estimated performance index 305 with the performance index 305 output by the second learning unit 306, and outputs an abnormality detection signal 333 when there is a deviation between the two by a predetermined threshold or more. The abnormality detection signal 333 may be notified to an occupant of the vehicle 9, or may be read by a device (not shown) mounted on the vehicle 9. For example, when there is a deviation between the THC calculated by the driving state detection unit 331 and the THC calculated by the second learning unit 306, there may be a malfunction such as an increase in the amount of fuel attached due to a problem in the fuel injection device or poor ignition due to wear of the ignition plug 105. In other words, the present invention can be treated as an index for checking whether the internal combustion engine 1 is operating normally.

[0053] (9) An operating state detection unit 331 is provided that can detect the operating state of the internal combustion engine 1 from the output of the sensor 332 mounted on the vehicle 9 and the performance index calculated by the second learning device 306. Therefore, it is possible to easily determine whether the internal combustion engine 1 is operating normally.

[0054] -Sixth embodiment- A sixth embodiment of the driving condition calculation system will be described with reference to FIG. 11. In the following description, the same components as those in the first embodiment are given the same reference numerals, and differences will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that it includes a display unit that displays the calculation results. Also, in this embodiment, the calculation device is not mounted on the vehicle.

[0055] Fig. 11 is a functional configuration diagram of a calculation device 800E in the sixth embodiment. Fig. 11 differs from Fig. 3 in the first embodiment in that a user interface 350 is provided. The user interface 350 functions as an input unit and an output unit. The user interface 350 outputs the calculation results of the first learning device 302, the second learning device 306, and the operation change unit 309. The user interface 350 accepts input of the operation conditions 301 by the user.

[0056] The user interface 350 has an initial operating condition input field 351, an internal combustion engine state display field 352, a performance index display field 353, and a next operating condition display field 354. The user inputs the operating conditions 301 into the initial operating condition input field 351 using a mouse or a keyboard included in the input / output device 84. The value input into the initial operating condition input field 351 is input to the first learning device 302. The first learning device 302 in this embodiment also outputs the calculated physical quantity 303 to the user interface 350, and the user interface 350 displays the physical quantity 303 in the internal combustion engine state display field 352.

[0057] The second learning device 306 in this embodiment also outputs the calculated performance index 305 to the user interface 350, and the user interface 350 displays the performance index 305 in a performance index display field 353. The operation change unit 309 in this embodiment also outputs the calculated new operating conditions 310 to the user interface 350, and the user interface 350 displays the new operating conditions 310 in a next operating condition display field 354.

[0058] When the user inputs the operating conditions 301, the estimation results by the first learning device 302 and the second learning device 306 are displayed on the user interface 350. By checking these, the user can determine whether or not the input operating conditions can be adopted. In addition, if the user wants to further improve the performance index 305, the user can refer to the next operating conditions display field 354.

[0059] According to the above-described sixth embodiment, the following advantageous effects can be obtained. (10) The system includes a user interface 350 that functions as an input unit for inputting the operating conditions 301, and a user interface 350 that also functions as a display unit for displaying the output of the first learning device 302, the output of the second learning device 306, and the output of the operation change unit 309. Therefore, appropriate operating conditions 301 can be easily calculated.

[0060] In each of the above-mentioned embodiments and modifications, the configuration of the functional blocks is merely an example. Some functional configurations shown as separate functional blocks may be integrated, or a configuration shown in one functional block diagram may be divided into two or more functions. In addition, some of the functions of each functional block may be provided by other functional blocks.

[0061] In the above-described embodiments and modifications, the program is stored in the ROM 82, but the program may be stored in a non-volatile storage device (not shown). The arithmetic device 800 may have an input / output interface (not shown), and the program may be read from another device via the input / output interface and a medium available to the arithmetic device 800 when necessary. The medium here refers to, for example, a storage medium that is detachable from the input / output interface, or a communication medium, i.e., a network such as a wired, wireless, or optical network, or a carrier wave or digital signal that propagates through the network. Some or all of the functions realized by the program may be realized by a hardware circuit or an FPGA.

[0062] The above-mentioned embodiments and modifications may be combined with each other. Although various embodiments and modifications have been described above, the present invention is not limited to these. Other aspects that are conceivable within the scope of the technical idea of ​​the present invention are also included in the scope of the present invention. [Explanation of symbols]

[0063] 1: Internal combustion engine 9: Vehicle 84: Input / Output device 301: Operating conditions 302: First learning unit 303:Physical quantity 305:Performance index 306: Second learning unit 307: First gradient calculation unit 308: Second gradient calculation unit 309: Operation Change Division 310: New operating conditions 321: First gradient 322: Second gradient 331: Driving state detection unit 332: Sensor 350: User Interface 800, 800A to 800E: Calculation unit

Claims

1. a first learning machine that receives an operating condition of an in-vehicle device mounted on a vehicle and outputs a physical state that occurs when the in-vehicle device operates based on the operating condition; A second learning device that receives an output of the first learning device as an input and outputs a performance index; a gradient calculation unit that calculates gradients of the first learning device and the second learning device; and an operation change unit that changes the operating conditions by using the gradient.

2. The operating condition calculation system according to claim 1, The on-board device is an internal combustion engine.

3. The operating condition calculation system according to claim 2, the performance index is at least one of an exhaust component from the internal combustion engine and a combustion variation rate, The operation change unit changes at least one of an injection time, an injection split ratio, a fuel pressure, and an ignition timing during split injection of fuel.

4. The operating condition calculation system according to claim 2, An operating condition calculation system in which at least the engine speed, load, intake pressure, intake air temperature, water temperature, fuel pressure, number of fuel injections, fuel injection timing, split ratio, and ignition time are input to the first learning unit, and the in-cylinder temperature, in-cylinder pressure, amount of fuel adhesion, and amount of rich mixture at least at the ignition time are output.

5. The operating condition calculation system according to claim 2, An operating condition calculation system in which at least the in-cylinder temperature, in-cylinder pressure, amount of fuel adhered, amount of rich mixture, and some or all of the inputs to the first learning unit just before ignition are input to the second learning unit, and the second learning unit outputs at least exhaust components generated after combustion and combustion fluctuation rate.

6. The operating condition calculation system according to claim 2, the first learning device includes one machine learning device for each physical quantity, An operating condition calculation system, wherein all or a part of the operating conditions are input to each of the machine learning machines.

7. The operating condition calculation system according to claim 2, The second learning device includes one machine learning device for each performance index, An operating condition calculation system, wherein all or a part of the output of the first learning device is input to each of the machine learning devices.

8. The operating condition calculation system according to claim 4, An operating condition calculation system, wherein the first learning unit receives an input of properties of fuel to be input to the internal combustion engine.

9. The operating condition calculation system according to claim 4, An operating condition calculation system, wherein the first learning unit receives input of the type of fuel fed to the internal combustion engine and the time when the fuel was fed to a fuel tank.

10. The operating condition calculation system according to claim 2, A driving condition calculation system comprising a driving state detection unit that calculates the performance index using output from a sensor mounted on the vehicle and detects an abnormality in the vehicle by comparing the performance index calculated by the second learning device.

11. The operating condition calculation system according to claim 2, An input unit for inputting the operating conditions; a display unit that displays an output of the first learning unit, an output of the second learning unit, and an output of the operation change unit.

12. The operating condition calculation system according to claim 2, The performance index is an index indicating the condition of components and devices of the vehicle, A driving condition calculation system, wherein the driving conditions are sensor information acquired from a sensor mounted on the vehicle.

13. A control device for an internal combustion engine comprising the operating condition calculation system according to claim 2.

14. An adaptation device for an internal combustion engine comprising the operating condition calculation system according to claim 2.

Images