Deposit amount estimation device

Abstract

To improve the estimation accuracy when estimating the deposit amount. [Solution] A deposit amount estimation device used in an engine to estimate the amount of deposits accumulated in the engine's intake system, which uses the engine's position information, a fuel injection parameter that reflects an increased injection amount relative to the stoichiometric air-fuel ratio with respect to the intake air amount, the overlap amount between the intake valve and the exhaust valve, and a temperature parameter that reflects the temperature of the engine's combustion chamber as input variables, and a pre-trained model obtained by machine learning with the deposit amount as the output variable, and estimates the deposit amount based on the position information, the fuel injection parameter, the overlap amount, and the temperature parameter.

Description

Technical Field

[0001] The present disclosure relates to a deposit amount estimation device.

Background Art

[0002] Conventionally, as this type of deposit amount estimation device, there has been proposed one that is used in an engine and estimates the deposit amount in the intake valve and its vicinity (see, for example, Patent Document 1). In this device, the deposit amount is estimated using the overlap amount between the intake valve and the exhaust valve, the intake pressure, and the intake air amount.

Prior Art Documents

Patent Documents

[0003] [[ID=XX]] [[ID=XX]]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the above deposit amount estimation device, further improvement in the estimation accuracy when estimating the deposit amount is recognized as an important issue.

[0005] The deposit amount estimation device of the present disclosure has as its main object to improve the estimation accuracy when estimating the deposit amount.

Means for Solving the Problems

[0006] The deposit amount estimation device of this disclosure employs the following means to achieve the above-mentioned main objective. The deposit amount estimation device of this disclosure is used in an engine and estimates the deposit amount as the amount of deposit accumulated in the intake system of the engine, and the gist of the device is a trained model obtained by machine learning, which takes the engine's position information, a fuel injection amount that reflects an increased injection amount relative to the stoichiometric air-fuel ratio with respect to the intake air amount, the overlap amount between the intake valve and the exhaust valve, and a temperature parameter that reflects the temperature of the engine's combustion chamber as input variables and the deposit amount as an output variable, and estimates the deposit amount based on the position information, the fuel injection parameter, the overlap amount, and the temperature parameter. By having such a configuration, the deposit amount estimation device of this disclosure can improve the estimation accuracy when estimating the deposit amount. [Brief explanation of the drawing]

[0007] [Figure 1] A schematic diagram of the automobile 20 according to the embodiment of this disclosure. [Figure 2] An explanatory diagram illustrating how to create a pre-trained model. [Figure 3] A flowchart showing an example of a decision routine executed by ECU70. [Modes for carrying out the invention]

[0008] Embodiments of this disclosure will be described with reference to the drawings. Figure 1 is a schematic diagram showing the configuration of an automobile 20 equipped with a deposit amount estimation device according to an embodiment of this disclosure. As shown in the figure, the automobile 20 of the embodiment includes an engine 22, an EGR device 150, a navigation system 94, a transmission 60 that changes the power of the engine 22 and transmits it to the drive wheels 64a and 64b via a differential gear 62, and an electronic control unit (hereinafter referred to as "ECU") 70. In this embodiment, the ECU 70 corresponds to the "deposit amount estimation device".

[0009] Engine 22 is configured as an internal combustion engine that uses gasoline as fuel to produce power. This engine 22 draws air from the intake port 52 into the air cleaner 23, and the air cleaned by the air cleaner 23 is circulated through the intake manifold 24, while gasoline is injected from the fuel injector 26. The mixture of air and fuel is then drawn into the combustion chamber via the intake valve 28, and is combusted by an electric spark from the spark plug 30, converting the reciprocating motion of the piston 32, which is pushed down by the energy, into the rotational motion of the crankshaft 33. The exhaust from the combustion chamber is discharged into the outside air via a purification device 34 that has a purification catalyst (three-way catalyst) that purifies harmful components such as carbon monoxide (CO), hydrocarbons (HC), and nitrogen oxides (NOx). In addition, exhaust and unburned gases generated in the combustion chamber are not only discharged into the outside air, but also leak out into the space inside the crankcase 31 through the gap between the piston 32 and the cylinder 37 (hereinafter, the exhaust and unburned gases that leak out in this way are called "blow-by gas"). The engine 22 is provided with a blow-by gas pipe 90 that connects the intake port 52 side of the throttle valve 25 in the intake manifold 24 to the crankcase 31. A blow-by valve 92 is attached to the blow-by gas pipe 90, which directs the blow-by gas in one direction (from the crankcase 31 towards the intake manifold 24), and the blow-by gas is returned to the intake manifold 24 after passing through the blow-by gas pipe 90. The engine 22 is operated and controlled by the ECU 70.

[0010] The EGR device 150 comprises an EGR pipe 152, an EGR valve 154, an EGR cooler 156, and a stepping motor (not shown). The EGR pipe 152 connects the downstream side of the exhaust pipe 35 beyond the purification device 34 to the intake pipe 24. The EGR valve 154 is installed in the EGR pipe 152 and is driven by a stepping motor (not shown) controlled by the ECU 70. The EGR cooler 156 is installed in the EGR pipe 152 and cools the exhaust gas passing through the EGR pipe 152. The EGR device 150 adjusts the return flow rate of exhaust gas from the exhaust pipe 35 to the intake pipe 24 by adjusting the opening degree of the EGR valve 154 using the stepping motor, thereby returning the exhaust gas to the intake pipe 24.

[0011] The ECU70 is configured as a microprocessor centered around the CPU72, and in addition to the CPU72, it includes a ROM74 for storing processing programs, a RAM76 for temporarily storing data, a flash memory78 for storing data, and input / output ports (not shown). Signals from various sensors are input to the ECU70 via the input ports. Examples of signals input to the ECU70 include the crank angle θcr from the crank position sensor40 which detects the rotational position of the crankshaft 33, the cam position Pc from the cam position sensor44 which detects the rotational position of the camshaft that opens and closes the intake valve 28 and exhaust valve 29 that perform intake and exhaust to the combustion chamber, and the throttle opening TH from the throttle valve position sensor46 which detects the position of the throttle valve 25. Other signals include the air-fuel ratio AF from the air-fuel ratio sensor 35a, the oxygen signal from the oxygen sensor 35b, the coolant temperature Tw from the coolant temperature sensor 42 that detects the temperature of the coolant in the engine 22, the intake air volume Qa from the airflow meter 48 attached to the intake manifold 24, the intake pressure Pa from the pressure sensor 49 attached to the intake manifold 24 that detects the pressure inside the intake manifold 24, the blow-by gas pressure Ppg from the pressure sensor 91 that detects the pressure inside the blow-by gas pipe 90, the road surface information Infl at the vehicle's current location from the navigation system 94, the fuel pressure Pf from the fuel pressure sensor 26a that detects the fuel pressure of the fuel supplied to the fuel injector 26, and the humidity H from the humidity sensor 96 that detects the humidity outside the vehicle. Furthermore, the ignition signal from the ignition switch 80, the shift position SP from the shift position sensor 82 which detects the operating position of the shift lever 81, the accelerator opening Acc from the accelerator pedal position sensor 84 which detects the amount the accelerator pedal 83 is pressed, the brake pedal position BP from the brake pedal position sensor 86 which detects the amount the brake pedal 85 is pressed, and the vehicle speed V from the vehicle speed sensor 88 can also be mentioned. The navigation system 94 is equipped with a storage device, a display device, and a processing device. The storage device stores various programs, map information, road surface information, etc. The display device stores location information etc.The system displays various information. The processing unit performs various processes. For example, when a destination is set by the occupant, the processing unit sets a route to the destination based on the destination, the vehicle's current location, and map information, and displays the set route on the display unit to provide route guidance. Location information includes the latitude and longitude of the current location and road surface information Infl. Road surface information Infl includes pavement information (whether it is a paved or unpaved road) as information indicating the condition of the road surface.

[0012] The ECU 70 outputs various control signals for operating the engine 22 via its output ports. Examples of signals output from the ECU 70 include a drive signal to the fuel injector 26, a drive signal to the throttle motor 36 that adjusts the position of the throttle valve 25, and a control signal to the ignition coil 38 which is integrated with the igniter. The ECU 70 also outputs control signals to the stepping motor of the EGR device 150 and to the transmission 60 via its output ports.

[0013] The ECU 70 calculates the rotational speed of the crankshaft 33, i.e., the rotational speed of the engine 22 (engine speed) Ne, at predetermined intervals (for example, every few milliseconds) based on the crank angle θcr from the crank position sensor 40. The ECU 70 also calculates the load ratio (the ratio of the volume of air actually inhaled in one cycle to the stroke volume per cycle of the engine 22, the engine load ratio) KL based on the calculated rotational speed of the engine 22 Ne and the intake air volume Qa from the airflow meter 48. Furthermore, the ECU 70 calculates the blow-by gas volume Vp, which is the volume of blow-by gas introduced into the intake manifold 24, from the intake pressure Pa from the pressure sensor 49 and the blow-by gas pressure Ppg from the pressure sensor 91.

[0014] The ECU 70 controls the operation of the engine 22 based on the accelerator opening Acc and vehicle speed V, including intake air volume control, fuel injection control, and ignition control. For intake air volume control, the required air volume Qa* is set based on the accelerator opening Acc and vehicle speed V, and the throttle valve 25 is controlled so that the difference between the intake air volume Qa and the required air volume Qa* cancels out. For fuel injection control, a basic injection volume Qfbs is set as the fuel injection volume that achieves the stoichiometric air-fuel ratio for the intake air volume Qa, an increment coefficient kf is set to increase the injection volume from the basic injection volume Qfbs, the required injection volume Qf* is set as the product of the basic injection volume Qfbs and the increment coefficient kf, and the fuel injector 26 is controlled so that the required injection volume Qf* of fuel is injected. For ignition control, a target ignition timing Ti* is set based on the rotational speed Ne and the required air volume Qa*, and the spark plug 30 is controlled to perform ignition based on the target ignition timing Ti*. The ECU 70 controls the intake valve 28 and exhaust valve 29 to open them simultaneously and overlap their opening periods when transitioning from the exhaust stroke to the intake stroke. This utilizes the differential pressure generated between the intake and exhaust sides of the engine 22 to scavenge residual gases remaining in the combustion chamber into the exhaust pipe 35, improving the efficiency of intake air charging. The ECU 70 calculates the EGR rate Re based on the intake air volume Qa from the airflow meter 48, the engine speed Ne of the engine 22, and the opening degree EV of the EGR valve. The EGR rate Re is the ratio of the EGR amount Qe to the sum of the intake air volume Qa from the airflow meter 48 and the EGR amount Qe, which is the amount of exhaust gas recirculated into the intake pipe 24.

[0015] The ECU70 stores a trained model obtained through machine learning in the ROM74, with road surface information Infl, fuel enrichment coefficient kf, overlap amount θo, coolant temperature (temperature parameter) Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine speed Ne of the engine 22, load ratio KL, EGR rate Re, and blow-by gas amount Vp as input variables (inputs to the input layer), and deposit amount Vdepo, which represents the amount of deposits accumulated in the intake system of the engine 22 (air cleaner 23, intake pipe 24, throttle valve 25, and intake valve 28), as an output variable (output to the output layer). In this machine learning approach, the tester drives the car 20 multiple times on roads, test courses, and chassis dynamometers to obtain multiple datasets that include road surface information Infl, an inflation coefficient kf, overlap amount θo, coolant temperature Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine speed Ne of the engine 22, load ratio KL, EGR rate Re, blow-by gas amount Vp, and deposit amount Vdepo. Then, when the ECU70 acquires the necessary number of datasets for constructing a highly accurate neural network, it creates a trained model using a neural network that takes road surface information Infl, increment coefficient kf, overlap amount θo, coolant temperature Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine speed Ne, load ratio KL, EGR rate Re, and blow-by gas amount Vp as inputs to the input layer, and outputs deposit amount Vdepo from the output layer, as shown in Figure 2, and stores it in the ROM74. The reason for using road surface information Infl as an input to the input layer is that the amount of dust in the outside air differs when driving on unpaved roads and when driving on paved roads, and the amount of dust taken in from the outside air differs, so the road surface information Infl affects the deposit amount Vdepo. The reason for using the fuel enrichment coefficient kf as input to the input layer is that the intake air composition of the engine 22 differs depending on the degree of fuel enrichment relative to the stoichiometric air-fuel ratio, and therefore the fuel enrichment coefficient kf affects the deposit amount Vdepo.The overlap amount θo is used as input to the input layer because, during the overlap period, gas is blown back from the cylinder into the intake system, and this gas contains deposits, thus the overlap amount θo affects the deposit amount Vdepo. The coolant temperature Tw is used as input to the input layer because the coolant temperature Tw reflects the temperature inside the combustion chamber, and the combustion state in the combustion chamber differs depending on the temperature inside the combustion chamber, and the temperature inside the combustion chamber, i.e., the coolant temperature Tw, affects the deposit amount Vdepo. The intake air amount Qa is used as input because, depending on the intake air amount Qa, the amount of dust taken in from the outside air differs, thus the intake air amount Qa affects the deposit amount Vdepo. The fuel pressure Pf is used as input to the input layer because, depending on the fuel pressure Pf, the amount of deposit detachment in the intake system (especially around the fuel injector 26 and the fuel injector 26 in the intake manifold 24) differs, thus the fuel pressure Pf affects the deposit amount Vdepo. The reason for using humidity H as input to the input layer is that the amount of moisture in the intake air differs depending on the amount of moisture in the outside air, and therefore humidity H affects the deposit amount Vdepo. The reason for using the rotational speed Ne of the engine 22 as input to the input layer is that the degree to which deposits are formed differs depending on the rotational speed Ne, and therefore rotational speed Ne affects the deposit amount Vdepo. The reason for using the load factor KL as input to the input layer is that the degree to which deposits are formed differs depending on the load factor KL, and therefore the load factor KL affects the deposit amount Vdepo. The reason for using the EGR rate Re as input to the input layer is that the components of the deposits in the intake air differ depending on the amount of exhaust gas returned from the exhaust pipe 35 to the intake pipe 24, and therefore the EGR rate Re affects the deposit amount Vdepo. The reason for using the blow-by gas amount Vp as input to the input layer is that the components of the deposits in the intake air differ depending on the blow-by gas amount Vp, and therefore the blow-by gas amount Vp affects the deposit amount Vdepo. Therefore, the trained model created in this way is a highly accurate model.

[0016] Next, the operation of the automobile 20 equipped with the deposit amount estimation device of the embodiment configured in this way, in particular, the operation when estimating the deposit amount Vdepo, will be described. Figure 3 is a flowchart of an example of a determination routine executed by the ECU 70. This routine is executed at predetermined intervals while the automobile 20 is running.

[0017] When this routine is executed, the CPU 72 of the ECU 70 receives input for road surface information Infl, enrichment coefficient kf, overlap amount θo, coolant temperature Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine speed Ne, load ratio KL, EGR rate Re, and blow-by gas amount Vp (S100).

[0018] Then, using the pre-trained model stored in ROM74, the road surface information Infl input in S100, the inflation coefficient kf, the overlap amount θo, the coolant temperature Tw, the intake air amount Qa, the fuel pressure Pf, the humidity H, the engine speed Ne of engine 22, the load ratio KL, the EGR rate Re, and the blow-by gas amount Vp, the deposit amount Vdepo is estimated (S110), and this routine ends. By estimating the deposit amount Vdepo using a pre-trained model created by machine learning in this way, the estimation accuracy when estimating the deposit amount Vdepo can be improved.

[0019] According to the vehicle 20 equipped with the deposit amount estimation device of the present embodiment described above, using a learned model created using a neural network with road surface information Infl, increase coefficient kf, overlap amount θo, coolant water temperature Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine 22 rotation speed Ne, load factor KL, EGR rate Re, and blow-by gas amount Vp as inputs to the input layer and deposit amount Vdepo as the output from the output layer, and estimating the deposit amount Vdepo using the road surface information Infl, increase coefficient kf, overlap amount θo, coolant water temperature Tw, intake air amount Qa, fuel pressure Pf, humidity H, engine 22 rotation speed Ne, load factor KL, EGR rate Re, and blow-by gas amount Vp, it is possible to improve the estimation accuracy when estimating the deposit amount Vdepo.

[0020] In the above-described embodiment, the road surface information Infl may be, for example, the current position of the vehicle 20 from the navigation system 94. Also, the coolant water temperature Tw may be, for example, the temperature in the combustion chamber or the temperature of the cylinder 37.

[0021] In the above-described embodiment, at least the road surface information Infl, increase coefficient kf, overlap amount θo, and coolant water temperature Tw may be inputs to the input layer. For the intake air amount Qa, fuel pressure Pf, humidity H, engine 22 rotation speed Ne, load factor KL, EGR rate Re, and blow-by gas amount Vp, at least one of these may be an input to the input layer, or not all of these need to be inputs to the input layer.

[0022] As described above, the embodiments for implementing the present disclosure have been described using the embodiments. However, the present disclosure is not limited to such embodiments, and it is of course possible to implement it in various forms without departing from the gist of the present disclosure.

Industrial Applicability

[0023] The present disclosure can be used in the manufacturing industry of deposit amount estimation devices and the like.

Explanation of Reference Numerals

[0024] 20 Automotive, 70 Electronic Control Units (ECUs).

Claims

[Claim 1] A deposit amount estimation device used in an engine, which estimates the amount of deposits accumulated in the intake system of the engine, A trained model obtained by machine learning takes the engine's position information, a fuel injection parameter that reflects an increased injection amount relative to the stoichiometric air-fuel ratio for the intake air amount, the overlap amount between the intake valve and the exhaust valve, and a temperature parameter that reflects the temperature of the engine's combustion chamber as input variables, and the deposit amount as an output variable, and estimates the deposit amount based on the position information, the fuel injection parameter, the overlap amount, and the temperature parameter. Deposit amount estimation device.

Images