Torque estimation device for internal combustion engine

Abstract

Provided is a torque estimation device for internal combustion engines with which it is possible to accurately learn variation in output torque of internal combustion engines that is caused by individual differences in internal combustion engines and by changes over time. A torque estimation device (100) for internal combustion engines calculates a first estimated torque (τest1) using torque characteristic data pertaining to the first estimated torque, calculates a second estimated torque (τest2) on the basis of an angular acceleration rate (αd), learns a deviation (Δτest2) between the second estimated torque (τest2) and a preset second-estimated-torque reference value (τest2ref) as an error learning value (Δτlrn), and calculates a value obtained by correcting the first estimated torque (τest1) using the error learning value (Δτlrn) as a third estimated torque (τest3).

Description

[Technical Field]

[0001] The present application relates to a torque estimation device for an internal combustion engine. [Background technology]

[0002] The technology of Patent Document 1 estimates the output torque of an internal combustion engine using the torque characteristics of the internal combustion engine based on the operating conditions that affect the output torque, such as the throttle opening, intake air amount, and fuel injection amount. Also, the technology of Patent Document 1 estimates the output torque of the internal combustion engine based on the characteristics of the torque converter, and learns the torque estimation error. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-291174 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology in Patent Document 1 learns the output torque variations caused by individual differences and aging of internal combustion engines by utilizing the characteristics of a torque converter, which is an external device to the internal combustion engine. However, the characteristics of torque converters also vary, and the operating range in which they can be learned is limited.

[0005] Therefore, an object of the present application is to provide a torque estimation device for an internal combustion engine that can accurately learn variations in output torque of an internal combustion engine that occur due to individual differences in the internal combustion engine and changes over time. [Means for solving the problem]

[0006] The torque estimation device for an internal combustion engine according to the present application comprises: a rotation information detection unit that detects the rotation speed and angular acceleration of a crankshaft of the internal combustion engine; a first estimated torque calculation unit that uses torque characteristic data for a first estimated torque, in which a relationship between an operating state for torque characteristic data of at least one of a throttle opening, an in-cylinder intake air amount, a fuel injection amount, and the rotational speed and a first estimated torque as an output torque of the internal combustion engine is preset, and calculates the first estimated torque corresponding to a current operating state for the torque characteristic data; a second estimated torque calculation unit that calculates a second estimated torque, which is an estimated value of the torque of the crankshaft, based on the angular acceleration; and a third estimated torque calculation unit that learns the deviation between the second estimated torque and a predetermined reference value of the second estimated torque as an error learning value, and calculates the value obtained by correcting the first estimated torque using the error learning value as a third estimated torque. [Effects of the Invention]

[0007] According to the torque estimation device for an internal combustion engine according to the present application, by comparing the second estimated torque calculated based on the actually detected angular acceleration with a reference value of the second estimated torque, it is possible to learn torque variations caused by individual variations of the internal combustion engine, aging, etc. as an error learned value. Then, by correcting the first estimated torque calculated based on torque characteristic data in a preset reference state using the error learned value to calculate the third estimated torque, it is possible to calculate an accurate torque estimated value that reflects torque variations caused by individual variations of the internal combustion engine, aging, etc. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic configuration diagram of an internal combustion engine and a torque estimation device (control device) according to a first embodiment. [Figure 2] 1 is a block diagram of a torque estimation device (control device) according to a first embodiment. [Figure 3] 1 is a hardware configuration diagram of a torque estimation device (control device) according to a first embodiment. [Figure 4] 4 is a time chart illustrating the processing of the rotation information detection unit according to the first embodiment. [Figure 5] FIG. 3 is a diagram illustrating the relationship between each cylinder and each stroke according to the first embodiment. [Figure 6] FIG. 4 is a diagram illustrating torque characteristic data for a first estimated torque according to the first embodiment. [Figure 7] FIG. 4 is a diagram illustrating calculation of the maximum value of angular acceleration of each cylinder according to the first embodiment. [Figure 8] FIG. 4 is a diagram illustrating torque characteristic data for reference values ​​according to the first embodiment. [Figure 9] FIG. 10 is a block diagram of a torque estimation device (control device) according to a second embodiment. [Figure 10] 10 is a time chart for explaining error learning according to the second embodiment. [Figure 11] 10 is a time chart for explaining error learning according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] 1. First Embodiment A torque estimation device 100 for an internal combustion engine according to a first embodiment (hereinafter simply referred to as the torque estimation device 100) will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an internal combustion engine 1 and the torque estimation device 100 according to this embodiment, and FIG. 2 is a block diagram of the torque estimation device 100 according to this embodiment. In this embodiment, the torque estimation device 100 is incorporated into a control device 50 for the internal combustion engine. Note that the torque estimation device 100 may be separate from the control device 50 for the internal combustion engine.

[0010] 1-1. Configuration of internal combustion engine 1 First, the configuration of the internal combustion engine 1 will be described. As shown in Fig. 1, the internal combustion engine 1 has a combustion chamber 7 (hereinafter also referred to as a cylinder 7) that burns a mixture of air and fuel. The internal combustion engine 1 has an intake pipe 23 that supplies air to the combustion chamber 7, and an exhaust pipe 17 that discharges exhaust gas burned in the combustion chamber 7. The internal combustion engine 1 is a gasoline engine. The internal combustion engine 1 has a throttle valve 4 that opens and closes the intake pipe 23. The throttle valve 4 is an electronically controlled throttle valve that is driven to open and close by an electric motor controlled by a control device 50. The throttle valve 4 is provided with a throttle opening sensor 19 that outputs an electric signal according to the opening of the throttle valve 4.

[0011] An air flow sensor 3 is provided in the intake pipe 23 upstream of the throttle valve 4, and outputs an electric signal corresponding to the amount of intake air taken into the intake pipe 23. The internal combustion engine 1 is equipped with an exhaust gas recirculation device 20. The exhaust gas recirculation device 20 has an EGR flow path 21 that recirculates exhaust gas from the exhaust pipe 17 to the intake manifold 12, and an EGR valve 22 that opens and closes the EGR flow path 21. The intake manifold 12 is the portion of the intake pipe 23 downstream of the throttle valve 4. The EGR valve 22 is an electronically controlled EGR valve that is driven to open and close by an electric motor controlled by the control device 50. The exhaust pipe 17 is equipped with an air-fuel ratio sensor 18 that outputs an electric signal corresponding to the air-fuel ratio of the exhaust gas in the exhaust pipe 17.

[0012] The intake manifold 12 is provided with a gas pressure sensor 8 that outputs an electric signal corresponding to the pressure inside the intake manifold 12. An injector 13 that injects fuel is provided downstream of the intake manifold 12. The injector 13 may be provided so as to inject fuel directly into the cylinders 7. The internal combustion engine 1 is provided with an atmospheric pressure sensor 33 that outputs an electric signal corresponding to the atmospheric pressure. The internal combustion engine 1 is provided with a water temperature sensor 34 that detects the temperature of cooling water.

[0013] A spark plug that ignites the air-fuel mixture and an ignition coil 16 that supplies ignition energy to the spark plug are provided at the top of the combustion chamber 7. Also provided at the top of the combustion chamber 7 are an intake valve 14 that adjusts the amount of intake air drawn into the combustion chamber 7 from an intake pipe 23, and an exhaust valve 15 that adjusts the amount of exhaust gas discharged from the combustion chamber 7 to an exhaust pipe 17. The intake valve 14 is provided with an intake variable valve timing mechanism that varies the valve opening and closing timing. The exhaust valve 15 is provided with an exhaust variable valve timing mechanism that varies the valve opening and closing timing. The variable valve timing mechanisms 14 and 15 have electric actuators.

[0014] The internal combustion engine 1 has a plurality of combustion chambers 7 (four in this example). A piston 5 is provided inside each combustion chamber 7. The piston 5 of each combustion chamber 7 is connected to the crankshaft 2 via a connecting rod 9 and a crank 32. The crankshaft 2 is rotated by the reciprocating motion of the piston 5. The combustion gas pressure generated in each combustion chamber 7 presses against the top surface of the piston 5, rotating the crankshaft 2 via the connecting rod 9 and the crank 32. The crankshaft 2 is connected to a power transmission mechanism that transmits driving force to wheels, a generator, etc. The power transmission mechanism is made up of a transmission, etc.

[0015] The internal combustion engine 1 includes a rotor 31 that rotates integrally with the crankshaft 2. The rotor 31 has a plurality of teeth at a plurality of predetermined crank angles. In this embodiment, the rotor 31 has teeth arranged at intervals of 20 degrees. The teeth of the rotor 31 have missing portions where some of the teeth are missing. The internal combustion engine 1 includes a crank angle sensor 11 that is fixed to the engine block 24 and detects the teeth of the rotor 31.

[0016] The internal combustion engine 1 is equipped with a camshaft connected to the crankshaft by a chain. The camshaft drives the opening and closing of intake valves 14 and exhaust valves 15. The camshaft rotates once for every two rotations of the crankshaft 2. The internal combustion engine 1 is equipped with a cam rotor that rotates integrally with the camshaft. The cam rotor has a plurality of teeth at a plurality of predetermined camshaft angles. The internal combustion engine 1 is equipped with a cam angle sensor 30 (see FIG. 3) that is fixed to the engine block 24 and detects the teeth of the cam rotor.

[0017] Based on two types of output signals from the crank angle sensor 11 and the cam angle sensor 30, the control device 50 detects the crank angle relative to the top dead center (TDC) of each piston 5 and determines the stroke of each combustion chamber 7. The internal combustion engine 1 is a four-stroke engine having an intake stroke, a compression stroke, a combustion stroke, and an exhaust stroke.

[0018] The crank angle sensor 11 and the cam angle sensor 30 output electrical signals in response to changes in the distance between each sensor and the teeth due to the rotation of the crankshaft 2. The output signal of each angle sensor 11, 30 is a square wave that turns on and off depending on whether the sensor is close to the teeth or far from them. Each angle sensor 11, 30 may be, for example, an electromagnetic pickup type sensor.

[0019] The configuration of the internal combustion engine is not limited to the configuration described with reference to FIG. 1, and internal combustion engines of various configurations may be used.

[0020] 1-2. Configuration of the torque estimation device 100 (control device 50) Next, the torque estimation device 100 will be described. The torque estimation device 100 is incorporated into a control device 50 of an internal combustion engine. As shown in FIG. 2, the control device 50 includes processing units such as a rotation information detection unit 51, a first estimated torque calculation unit 52, a second estimated torque calculation unit 53, and a third estimated torque calculation unit 54. The processing units 51 to 54 of the control device 50 are realized by processing circuits included in the control device 50. Specifically, as shown in FIG. 3, the control device 50 includes, as processing circuits, an arithmetic processing device 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 connected to the arithmetic processing device 90 via a signal line such as a bus, an input circuit 92 that inputs external signals to the arithmetic processing device 90, and an output circuit 93 that outputs signals from the arithmetic processing device 90 to the outside.

[0021] The arithmetic processing device 90 may be an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), various logic circuits, various signal processing circuits, etc. Furthermore, a plurality of the same or different types of arithmetic processing device 90 may be provided, and each process may be shared and executed.

[0022] The storage device 91 includes volatile and non-volatile storage devices such as RAM (Random Access Memory), ROM (Read Only Memory), EEPROM (Electrically Erasable Programmable ROM), etc. The input circuit 92 is connected to various sensors and switches and includes an A / D converter and the like that inputs output signals from these sensors and switches to the arithmetic processing device 90. The output circuit 93 is connected to electrical loads and includes a drive circuit and the like that outputs control signals from the arithmetic processing device 90 to these electrical loads.

[0023] The functions of the processing units 51 to 54 of the control device 50 are realized by an arithmetic processing device 90 executing software (programs) stored in a storage device 91 such as a ROM or EEPROM, in cooperation with other hardware of the control device 50, such as the storage device 91, an input circuit 92, and an output circuit 93. Setting data used by the processing units 51 to 54, such as torque characteristic data for the first estimated torque, the moment of inertia Icrk, the load torque τload, and torque characteristic data for reference values, is stored in the storage device 91, such as a ROM or EEPROM. Data such as calculated values ​​and detected values ​​calculated by the processing units 51 to 54, such as the rotational speed Ne, angular velocity ωd, angular acceleration αd, first estimated torque τest1, maximum angular acceleration αmax, second estimated torque τest2, and error learning value Δτlrn, is stored in a rewritable storage device 91, such as a RAM.

[0024] In this embodiment, the input circuit 92 is connected to the crank angle sensor 11, cam angle sensor 30, water temperature sensor 34, air flow sensor 3, throttle opening sensor 19, gas pressure sensor 8, atmospheric pressure sensor 33, air-fuel ratio sensor 18, accelerator position sensor 26, etc. The output circuit 93 is connected to the throttle valve 4 (electric motor), EGR valve 22 (electric motor), injector 13, ignition coil 16, intake variable valve timing mechanism 14, exhaust variable valve timing mechanism 15, etc. Note that various sensors, switches, actuators, etc. (not shown) are connected to the torque estimation device 100. The control device 50 detects the operating conditions of the internal combustion engine 1, such as the throttle opening θth, the amount of intake air in the cylinder, the pressure in the intake manifold, the atmospheric pressure, and the air-fuel ratio, based on the output signals of the various sensors.

[0025] As a basic control, the control device 50 calculates the fuel injection amount, ignition timing, etc. based on input output signals from various sensors, and controls the injector 13, the ignition coil 16, etc. The control device 50 calculates the output torque of the internal combustion engine 1 required by the driver based on the output signal from the accelerator position sensor 26, etc., and controls the throttle valve 4, etc., so as to obtain an intake air amount that realizes the required output torque. Specifically, the control device 50 calculates a target throttle opening and controls the operation of the electric motor of the throttle valve 4 so that the throttle opening θth detected based on the output signal from the throttle opening sensor 19 approaches the target throttle opening. The control device 50 also calculates a target opening of the EGR valve 22 based on input output signals from various sensors, and controls the operation of the electric motor of the EGR valve 22. The control device 50 calculates the target opening / closing timing of the intake valve and the target opening / closing timing of the exhaust valve based on the output signals of various sensors input thereto, and controls the operation of the intake and exhaust variable valve timing mechanisms 14, 15 based on the target opening / closing timing.

[0026] 1-2-1. Rotation information detection unit 51 The rotation information detection unit 51 detects the rotation speed Ne and angular acceleration αd of the crankshaft of the internal combustion engine. In this embodiment, the rotation information detection unit 51 detects the crank angle θd based on the output signal of the crank angle sensor 11, and calculates an angular velocity ωd, which is the time rate of change of the detected crank angle θd, and an angular acceleration αd, which is the time rate of change of the angular velocity ωd. The rotation information detection unit 51 detects the rotation speed Ne of the crankshaft based on the output signal of the crank angle sensor 11. Note that the rotation speed Ne corresponds to the angular velocity ωd, but in this embodiment, it is an average speed per stroke period.

[0027] 4, in this embodiment, the rotation information detection unit 51 detects the crank angle θd and also detects the detection time Td at which the crank angle θd is detected based on the output signal of the crank angle sensor 11. Then, the rotation information detection unit 51 calculates the angle interval Δθd and the time interval ΔTd corresponding to the detected angle θd, based on the detected crank angle θd and the detection time Td.

[0028] For example, the rotation information detection unit 51 determines the crank angle θd when it detects the falling edge (or rising edge) of the output signal (rectangular wave) of the crank angle sensor 11. Using a known method, the rotation information detection unit 51 detects the crank angle θd relative to the top dead center (TDC) of the piston of the first cylinder based on two types of output signals from the crank angle sensor 11 and the cam angle sensor 30, and determines the stroke of each cylinder 7. Figure 5 shows the strokes of the first to fourth cylinders.

[0029] <Calculation of angular velocity ωd and angular acceleration αd> The rotation information detection unit 51 calculates the angular velocity ωd based on each crank angle θd and the detection time Td at which each crank angle θd was detected. For example, as shown in the following equation, the rotation information detection unit 51 calculates the angular velocity ωd(n) of the currently detected angle based on the angle interval Δθd(n) between the currently detected crank angle θd(n) and the previously detected crank angle θd(n-1) and the time interval ΔTd(n) between the currently detected time Td(n) and the previously detected time Td(n-1). Note that various other well-known methods may be used to calculate the angular velocity ωd.

number

[0030] The rotation information detection unit 51 calculates the angular acceleration αd based on the angular velocity ωd. For example, as shown in the following equation, the rotation information detection unit 51 calculates the angular acceleration αd(n) of the currently detected angle based on the angular velocity ωd(n) calculated for the currently detected angle, the angular velocity ωd(n-1) calculated for the previously detected angle, and the time interval ΔTd(n) of the currently detected angle. Note that various other well-known methods may also be used to calculate the angular acceleration αd.

number

[0031] 1-2-2. First estimated torque calculation unit 52 The first estimated torque calculation unit 52 calculates a first estimated torque τest1 corresponding to the current operating state for the torque characteristic data, using torque characteristic data for the first estimated torque, in which the relationship between the operating state for at least one of the torque characteristic data, i.e., the throttle opening θth, the in-cylinder intake air amount, the fuel injection amount, and the rotation speed Ne, and the first estimated torque τest1 as the output torque of the internal combustion engine, is preset. The first estimated torque τest1 is the torque output from the crankshaft to the outside of the internal combustion engine, and is the average output torque during the stroke period.

[0032] For example, a case will be described in which the throttle opening θth and the rotation speed Ne are used as the operating conditions for the torque characteristic data. As shown in Fig. 6, the torque characteristic data for the first estimated torque is map data in which the relationships between the throttle opening θth, the rotation speed Ne, and the first estimated torque τest1 are preset. The first estimated torque calculation unit 52 refers to the torque characteristic data for the first estimated torque and calculates the first estimated torque τest1 corresponding to the current throttle opening θth and the rotation speed Ne. Here, the amount of intake air in the cylinder or the amount of fuel injection may be used instead of the throttle opening θth. In other words, parameters correlated with the rotation speed Ne and the amount of fuel injection may be used as the operating conditions for the torque characteristic data.

[0033] The operating conditions for the torque characteristic data may include various operating conditions of the internal combustion engine that correlate with the output torque, such as the EGR rate, ignition timing, intake valve opening / closing timing, and exhaust valve opening / closing timing.The torque characteristic data for the first estimated torque may be a combination of multiple map data, or may be derived from a neural network or the like.

[0034] 1-2-3. Second estimated torque calculation unit 53 The second estimated torque calculation unit 53 calculates a second estimated torque τest2, which is an estimated value of the torque of the crankshaft, based on the angular acceleration αd.

[0035] 7, the second estimated torque calculation unit 53 calculates the maximum value αmax of the angular acceleration αd during the combustion stroke, and calculates the second estimated torque τest2 based on the maximum value αmax of the angular acceleration. According to this configuration, by calculating the maximum value αmax of the angular acceleration during the combustion stroke, the second estimated torque τest2 can be calculated using the maximum value αmax of the angular acceleration increased by combustion, and the maximum value of the instantaneous torque increased mainly by combustion can be calculated as the second estimated torque τest2.

[0036] In this embodiment, the internal combustion engine 1 has four combustion chambers 7 (cylinders 7). The second estimated torque calculation unit 53 calculates maximum values ​​αmax1, αmax2, αmax3, and αmax4 of angular acceleration during the combustion stroke for each of the four combustion chambers 7, and calculates the second estimated torque τest2 based on the average value αmaxave of the maximum values ​​of angular acceleration for the four combustion chambers, as shown in the following equation. The average value αmaxave is calculated every two rotations of the crankshaft.

number

[0037] According to this configuration, even if there is variation in combustion torque between cylinders, it is possible to calculate the maximum value of the average instantaneous combustion torque between cylinders.

[0038] The second estimated torque calculation unit 53 calculates the second estimated torque τest2 by multiplying the angular acceleration αd (in this example, the inter-cylinder average value αmaxave of the maximum angular acceleration) by the moment of inertia Icrk. The moment of inertia Icrk is the total value of the moments of inertia of the rotating members that rotate integrally with the crankshaft, and is set in advance.

number

[0039] In this embodiment, the second estimated torque calculation unit 53 calculates the second estimated torque τest2 by subtracting the load torque τload from the product of the angular acceleration αd and the moment of inertia Icrk. The load torque τload is a load torque applied to the crankshaft from outside the internal combustion engine and is usually a negative value. For example, a preset value is used for the load torque τload. Alternatively, because the load torque τload usually does not fluctuate significantly over the stroke period, the second estimated torque calculation unit 53 may use, as the load torque τload, the torque obtained by multiplying the angular acceleration αd of the angle near the top dead center (TDC) of the piston by the moment of inertia Icrk. The angle near the top dead center (TDC) may be set within a range from 10 degrees before the top dead center to 10 degrees after the top dead center, for example, at TDC. Near the top dead center, the connecting rod and the crank are aligned, and the force of the internal cylinder pressure pushing the piston does not generate any axial torque, so the load torque τload can be calculated. According to this configuration, the load torque τload applied to the crankshaft from outside the internal combustion engine can be excluded, the increase in instantaneous torque due to combustion can be calculated, and the calculation accuracy of the second estimated torque τest2 can be improved.

[0040] The average angular acceleration during the stroke period may be used to calculate the second estimated torque τest2.

[0041] 1-2-4. Third estimated torque calculation unit 54 The third estimated torque calculation unit 54 learns the deviation Δτest2 between the second estimated torque τest2 and a predetermined reference value τest2ref of the second estimated torque as an error learning value Δτlrn, and calculates the value obtained by correcting the first estimated torque τest1 with the error learning value Δτlrn as the third estimated torque τest3.

[0042] According to this configuration, by comparing the second estimated torque τest2 calculated based on the actually detected angular acceleration αd with the reference value τest2ref of the second estimated torque, it is possible to learn torque variations caused by individual variations, aging, etc. of the internal combustion engine 1 as the error learned value Δτlrn. Then, by correcting the first estimated torque τest1 calculated based on torque characteristic data in a preset reference state using the error learned value Δτlrn to calculate the third estimated torque τest3, it is possible to calculate an accurate torque estimated value that reflects torque variations caused by individual variations, aging, etc. of the internal combustion engine 1.

[0043] The control device 50 performs torque control to control the output torque of the internal combustion engine based on the third estimated torque τest3. For example, the control device 50 calculates a target throttle opening so that the third estimated torque τest3 approaches the required output torque. The control device 50 also transmits the third estimated torque τest3 to an external control device such as a vehicle control device or a motor control device, and reflects it in the torque control of the external control device.

[0044] As shown in FIG. 2, the third estimated torque calculation unit 54 includes a reference value calculation unit 541, a learning value calculation unit 542, and an estimated torque correction unit 543.

[0045] <Reference Value Calculation Unit 541> The reference value calculation unit 541 uses torque characteristic data for a reference value in which the relationship between the operating state for at least one reference value of the throttle opening θth, the in-cylinder intake air amount, the fuel injection amount, and the rotational speed Ne and the reference value τest2ref of the second estimated torque is preset, and calculates the reference value τest2ref of the second estimated torque corresponding to the current operating state for the reference value.

[0046] In this embodiment, the reference value τest2ref of the second estimated torque is the maximum value or increase in instantaneous torque due to combustion, and therefore is a value different from the first estimated torque τest1, which is the average output torque during the stroke period. Therefore, the torque characteristic data for the reference value is set to a value different from the torque characteristic data for the first estimated torque.

[0047] For example, a case will be described in which a throttle opening θth and an engine speed Ne are used as the operating conditions for the reference value. As shown in FIG. 8, the torque characteristic data for the reference value is map data in which the relationships among the throttle opening θth, the engine speed Ne, and the reference value τest2ref of the second estimated torque are preset. The reference value calculation unit 541 refers to the torque characteristic data for the reference value and calculates the reference value τest2ref of the second estimated torque corresponding to the current throttle opening θth and engine speed Ne. Here, the amount of intake air in the cylinder or the amount of fuel injection may be used instead of the throttle opening θth. In other words, parameters correlated with the engine speed Ne and the amount of fuel injection may be used as the operating conditions for the reference value.

[0048] The operating conditions for the reference value may be various operating conditions of the internal combustion engine that correlate with the output torque, such as the EGR rate, ignition timing, intake valve opening / closing timing, and exhaust valve opening / closing timing. The torque characteristic data for the reference value may be a combination of multiple map data, or may be derived from a neural network or the like.

[0049] <Learning Value Calculation Unit 542> The learning value calculation unit 542 updates the error learning value Δτlrn based on the deviation Δτest2 between the second estimated torque τest2 and the reference value τest2ref of the second estimated torque.

[0050] The learning value calculation unit 542 performs statistical processing on the deviation Δτest2 to update the error learning value Δτlrn. For example, low-pass filter processing is performed as the statistical processing. The learning value calculation unit 542 updates the error learning value Δτlrn using the following equation: where (j) indicates the calculated value for the current update period, (j-1) indicates the calculated value for the previous update period, and Kflt is a filter gain. In this embodiment, the update period is set to a period in which the crankshaft rotates twice.

number

[0051] The error learning value Δτlrn may be set and updated for each range of at least one of the operating conditions of the throttle opening θth, the cylinder intake air amount, the fuel injection amount, and the rotation speed Ne.

[0052] <Estimated Torque Correction Unit 543> The estimated torque correcting unit 543 calculates the third estimated torque τest3 by correcting the first estimated torque τest1 using the error learning value Δτlrn. For example, the estimated torque correcting unit 543 calculates the third estimated torque τest3 using the following equation.

number

[0053] Alternatively, the estimated torque correction unit 543 may calculate the correction coefficient Kτ based on operating conditions for at least one of the correction coefficients, namely, the throttle opening θth, the in-cylinder intake air amount, the fuel injection amount, and the rotation speed Ne, and may calculate the third estimated torque τest3 by adding the product of the error learning value Δτlrn and the correction coefficient Kτ to the first estimated torque τest1, as shown in the following equation.

number

[0054] Since the degree of influence of the error learning value Δτlrn on the estimated torque varies depending on the operating state, the correction accuracy can be improved by multiplying the error learning value Δτlrn by the correction coefficient Kτ. Also, since the proportional coefficient of the average output torque during the stroke period to the maximum value or increase in instantaneous torque increased by combustion varies depending on the operating state, the correction coefficient Kτ is calculated based on the operating state for the correction coefficient, and the correction accuracy can be improved by multiplying the error learning value Δτlrn by the correction coefficient Kτ.

[0055] The estimated torque correction unit 543 calculates the correction coefficient Kτ corresponding to the current operating state for the correction coefficient using correction coefficient setting data for the reference value, which is a preset relationship between the operating state for the correction coefficient and the correction coefficient Kτ. The correction coefficient setting data is configured in the same manner as the torque characteristic data for the reference value.

[0056] 2. Second Embodiment A torque estimation device 100 according to a second embodiment will be described with reference to the drawings. Description of components similar to those of the first embodiment will be omitted. The basic configuration of the torque estimation device 100 according to this embodiment is similar to that of the first embodiment. In this embodiment, part of the processing of the third estimated torque calculation unit 54 differs from that of the first embodiment.

[0057] In this embodiment, as shown in FIG. 9, the third estimated torque calculation unit 54 includes a reference value calculation unit 541, a learning value calculation unit 542, and an estimated torque correction unit 543, as well as a learning condition determination unit 544.

[0058] A learning condition determination unit 544 determines whether a learning condition is met based on at least one of the operating conditions for learning determination, namely, the throttle opening θth, the in-cylinder intake air amount, the fuel injection amount, and the rotation speed Ne. When the learning condition is met, a learning value calculation unit 542 updates the error learning value Δτlrn based on the deviation Δτest2.

[0059] According to this configuration, when a preset learning condition is met, learning is performed, thereby improving the learning accuracy.

[0060] In this embodiment, as shown in Figures 10 and 11, the learning condition determination unit 544 determines that the learning condition is met when the rotation speed Ne passes through a predetermined determination rotation speed Thne while the rotation speed Ne is increasing.

[0061] When the rotation speed Ne is accelerating and decelerating, the torque of the internal combustion engine is increasing and decreasing, which is a transient state, and learning during this transient state is likely to result in a decrease in learning accuracy. When the rotation speed Ne is continuously increasing, the torque increases and decreases little and is stable, which improves learning accuracy. In particular, in the case of a two-wheeled vehicle, the output of the internal combustion engine is smaller and the inertia is lower than in a four-wheeled vehicle, so learning accuracy is likely to decrease due to fluctuation factors. However, the above configuration can suppress the decrease in learning accuracy.

[0062] For example, the learning condition determination unit 544 determines that the learning condition is met when the rotation speed Ne increases continuously during the determination period Tjd and the rotation speed Ne passes the determination rotation speed Thne.

[0063] By setting the judgment period Tjd, it is possible to judge with higher accuracy a stable state in which torque fluctuations are small, and to improve learning accuracy.

[0064] Furthermore, the learning condition determination unit 544 may determine that the learning condition is met when the rotation speed Ne increases continuously within a predetermined increase rate range during the determination period Tjd and the rotation speed Ne passes the determination rotation speed Thne.

[0065] By setting a range for the rate of increase of the rotation speed Ne, it is possible to more accurately determine a stable state where torque fluctuations are small, thereby improving learning accuracy.

[0066] The behavior during learning will be described using Figures 10 and 11. First, Figure 10 will be described. In Figure 10, there is no individual variation or aging in the internal combustion engine, the second estimated torque τest2 and the reference value τest2ref of the second estimated torque match, and the deviation Δτest2 is close to zero. The throttle opening θth increases, the torque increases, and the rotation speed Ne increases continuously. While the rotation speed Ne is increasing, at time t01, the rotation speed Ne passes through the judgment rotation speed Thne, and the learning condition is met. The deviation Δτest2 when the learning condition is met is learned as the error learning value Δτlrn.

[0067] Next, Fig. 11 will be described. In Fig. 11, the internal combustion engine is in a state where there are individual variations and aging changes, the second estimated torque τest2 exceeds the reference value τest2ref of the second estimated torque, and the deviation Δτest2 is smaller than zero. The throttle opening θth increases, the torque increases, and the rotation speed Ne increases continuously. While the rotation speed Ne is increasing, at time t01, the rotation speed Ne passes through the judgment rotation speed Thne, and the learning condition is met. The deviation Δτest2 when the learning condition is met is learned as the error learning value Δτlrn.

[0068] Note that the average value of the deviation Δτest2 obtained when the learning condition is met multiple times may be learned as the error learning value Δτlrn.

[0069] In each of the above embodiments, an example has been described in which a four-cylinder internal combustion engine is used. However, an internal combustion engine with any number of cylinders (for example, one, two, three, or six) may be used.

[0070] Although various exemplary embodiments and examples are described in this application, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are contemplated within the scope of the technology disclosed in this specification. For example, this includes cases where at least one component is modified, added, or omitted, or where at least one component is extracted and combined with components of another embodiment. [Explanation of symbols]

[0071] 1 internal combustion engine, 2 crankshaft, 7 combustion chamber (cylinder), 51 rotation information detection unit, 52 first estimated torque calculation unit, 53 second estimated torque calculation unit, 54 third estimated torque calculation unit, 100 torque estimation device, Kτ correction coefficient, Ne rotation speed, Thne judgment rotation speed, Tjd judgment period, τest1 first estimated torque, τest2 second estimated torque, τest2ref reference value of second estimated torque, τest3 third estimated torque, Δτest2 deviation, Δτlrn error learning value, αd angular acceleration, αmax maximum angular acceleration, θth throttle opening

Claims

1. a rotation information detection unit that detects the rotation speed and angular acceleration of a crankshaft of an internal combustion engine; a first estimated torque calculation unit that uses torque characteristic data for a first estimated torque, in which a relationship between an operating state for at least one of torque characteristic data, which is a throttle opening, an in-cylinder intake air amount, a fuel injection amount, and the rotation speed, and a first estimated torque as an output torque of the internal combustion engine, is preset, and calculates the first estimated torque corresponding to a current operating state for the torque characteristic data; a second estimated torque calculation unit that calculates a second estimated torque, which is an estimated value of the torque of the crankshaft, based on the angular acceleration; a third estimated torque calculation unit that learns a deviation between the second estimated torque and a preset reference value of the second estimated torque as an error learned value, and calculates a value obtained by correcting the first estimated torque with the error learned value as a third estimated torque.

2. 2. The torque estimation device for an internal combustion engine according to claim 1, wherein the third estimated torque calculation unit determines whether or not a learning condition is satisfied based on at least one of the operating states for learning judgment, which are the throttle opening, the in-cylinder intake air amount, the fuel injection amount, and the rotational speed, and updates the error learning value based on the deviation when it is determined that the learning condition is satisfied.

3. 3. The torque estimation device for an internal combustion engine according to claim 1, wherein the third estimated torque calculation unit calculates a correction coefficient based on operating conditions for a correction coefficient of at least one of the throttle opening, the in-cylinder intake air amount, the fuel injection amount, and the rotational speed, and calculates, as the third estimated torque, a value obtained by correcting the first estimated torque by a multiplication value of the error learning value and the correction coefficient.

4. 3. The torque estimation device for an internal combustion engine according to claim 1, wherein the second estimated torque calculation unit calculates a maximum value of the angular acceleration during a combustion stroke, and calculates the second estimated torque based on the maximum value.

5. The internal combustion engine includes a plurality of combustion chambers, 3. The torque estimation device for an internal combustion engine according to claim 1, wherein the second estimated torque calculation unit calculates a maximum value of the angular acceleration during a combustion stroke for each of the plurality of combustion chambers, and calculates the second estimated torque based on an average value of the maximum values ​​for the plurality of combustion chambers.

6. 3. The torque estimation device for an internal combustion engine according to claim 1 or 2, wherein the third estimated torque calculation unit uses torque characteristic data for a reference value in which a relationship between an operating state for a reference value of at least one of the throttle opening, the in-cylinder intake air amount, the fuel injection amount, and the rotational speed and a reference value of the second estimated torque is preset, and calculates a reference value of the second estimated torque corresponding to a current operating state for the reference value.

7. 3. The torque estimation device for an internal combustion engine according to claim 1, wherein the third estimated torque calculation unit determines that a learning condition is satisfied when the rotation speed passes a predetermined judgment rotation speed while the rotation speed is increasing, and updates the error learned value based on the deviation.

8. 8. The torque estimation device for an internal combustion engine according to claim 7, wherein the third estimated torque calculation unit determines that the learning condition is satisfied when the rotation speed increases continuously during a determination period and the rotation speed passes the determination rotation speed.

Images