EP4343135A1

EP4343135A1 - Vehicle management system, vehicle, and vehicle traveling management method

Info

Publication number: EP4343135A1
Application number: EP22804366.7A
Authority: EP
Inventors: Kenji Fujita
Original assignee: Toyota Industries Corp
Current assignee: Toyota Industries Corp
Priority date: 2021-05-17
Filing date: 2022-03-22
Publication date: 2024-03-27
Also published as: WO2022244448A1; AU2022276789A1; BR112023021216A2; JP2022176528A

Abstract

Each of a vehicle (1A) and a plurality of vehicles (1B) includes sensors (4) that detect the combustion state of fuel in an engine (3). A server (23) includes a memory (232) that stores a trained model for engine control. The trained model has been trained with, on the basis of sensor data indicating detection results of the sensors (4) of each of the plurality of vehicles (1B) and gas station data for identifying a gas station where each of the plurality of vehicles (1B) has been supplied with fuel, a correction amount for correcting the combustion state of fuel in the engine (3). A processor (231) of the server (23) calculates the correction amount by applying sensor data and gas station data on the vehicle (1A) to the trained model, and transmits the calculated correction amount to the vehicle (1A). The vehicle (1A) uses the received correction amount to correct the combustion state of fuel in the engine (3) of the vehicle (1A).

Description

The present disclosure relates to a vehicle management system, a vehicle, and a vehicle traveling management method, and more specifically relates to technology for managing a vehicle with an engine mounted therein.

BACKGROUND

Japanese Patent Laying-Open No. 2005-113715 (PTL 1) discloses a control device for controlling an operation of a power device (specifically, an engine) that uses a prescribed fuel.

CITATION LIST

PATENT LITERATURE

PTL 1: Japanese Patent Laying-Open No. 2005-113715

SUMMARY

TECHNICAL PROBLEM

PTL 1 describes that in the case where fuel is supplied to a vehicle, composition information of the fuel is acquired based on position information of the vehicle. The engine control device disclosed in PTL 1 is configured to correct an operation of the engine based on composition information of the fuel.
In order for the engine control device disclosed in PTL 1 to produce an effect, namely, to appropriately correct an operation of the engine (a combustion state of fuel in the engine), it is necessary for composition information of fuel to be obtained. However, in order to obtain composition information of fuel, a complex process such as an analysis of fuel may be necessary.
The present disclosure has been made to solve the above-stated problem, and an objective of the present disclosure is to provide technology capable of appropriately correcting a combustion state of fuel in an engine, even if composition information of fuel is not obtained.

SOLUTION TO PROBLEM

(1) A vehicle management system relating to a certain aspect of the present disclosure includes a target vehicle, and a server that communicates with the target vehicle and a plurality of vehicles. Each of the target vehicle and the plurality of vehicles includes at least one sensor that detects a combustion state of fuel in an engine. The server includes a memory that stores a trained model related to engine control, and a processor that executes an arithmetic process using the trained model. In the trained model, a correction amount for correcting a combustion state of fuel in the engine is trained based on sensor data and gas station data, the sensor data showing a detection result of the at least one sensor of each of the plurality of vehicles and the gas station data being capable of specifying a gas station where each of the plurality of vehicles received a supply of fuel. The processor calculates the correction amount by providing the sensor data and the gas station data of the target vehicle to the trained model, and transmits the calculated correction amount to the target vehicle. The target vehicle corrects a combustion state of fuel in the engine of the target vehicle by using the received correction amount.
(2) The gas station data includes data related to an installation location of the gas station and data related to a fuel supply time period at the gas station.
(3) In the trained model, the correction amount is trained based on the sensor data, the gas station data, and data related to a traveling position of the plurality of vehicles. The processor calculates the correction amount of the target vehicle by providing, in addition to the sensor data and the gas station data, data related to a traveling position of the target vehicle to the trained model.
(4) The sensor data includes data related to at least one of a cylinder internal pressure of the engine, a rotational speed of the engine, a temperature of exhaust gas from the engine, a front/back pressure difference of a filter for capturing particulate matter within the exhaust gas, and a nitrogen oxide concentration within the exhaust gas.
(5) A vehicle relating to another aspect of the present disclosure is a vehicle that is capable of communicating with a server including a trained model related to engine control. In the trained model, a correction amount for correcting a combustion state of fuel in an engine is trained based on sensor data and gas station data collected by communication with a plurality of vehicles each with an engine mounted therein. The sensor data is data showing a detection result of a combustion state of fuel in the engine of each of the plurality of vehicles. The gas station data is data capable of specifying a gas station where the plurality of vehicles received a supply of fuel. The vehicle includes a communicator and a controller. The communicator transmits the sensor data and the gas station data of the vehicle to the server, and receives the correction amount calculated in the server by providing the sensor data and the gas station data from the vehicle to the trained model. The controller corrects a combustion state of fuel in the engine of the vehicle by using the received correction amount.
(6) A vehicle traveling management method relating to another additional aspect of the present disclosure manages traveling of a vehicle by using a trained model related to engine control. In the trained model, a correction amount for correcting a combustion state of fuel in an engine is trained based on sensor data and gas station data collected by communication with a plurality of vehicles each with an engine mounted therein. The sensor data is data showing a detection result of a combustion state of fuel in the engine of each of the plurality of vehicles. The gas station data is data capable of specifying a gas station where each of the plurality of vehicles received a supply of fuel. The traveling management method includes first and second steps. The first step is a step of calculating the correction amount for correcting a combustion state of fuel in the engine of a target vehicle by providing the sensor data and the gas station data from the target vehicle to the trained model. The second step is a step of correcting a combustion state of fuel in the engine of the target vehicle by using the correction amount.

ADVANTAGEOUS EFFECTS OF INVENTION

According to the present disclosure, a combustion state of fuel in an engine can be appropriately corrected, even if composition information of fuel is not obtained.

BRIEF DESCRIPTION OF DRAWINGS

Fig. 1 is a figure schematically showing an overall configuration of a vehicle management system relating to a present embodiment.
Fig. 2 is a figure showing in more detail configurations of a vehicle and a management center.
Fig. 3 is a figure showing an example of a configuration of an engine.
Fig. 4 is a figure showing another example of a configuration of an engine.
Fig. 5 is a figure showing an example of a configuration of sensors.
Fig. 6 is a functional block diagram for showing an example of a prediction model related to a combustion state of an engine in the present embodiment.
Fig. 7 is a figure for describing an example of a combustion improvement process to correct an operation of an injection system of an engine.
Fig. 8 is a figure for describing another example of a combustion improvement process to correct an operation of an injection system of an engine.
Fig. 9 is a figure for describing an example of a combustion improvement process to correct an operation of an EGR system of an engine.
Fig. 10 is a figure for describing an example of a combustion improvement process to correct an operation of an air intake system of an engine.
Fig. 11 is a flowchart showing a procedure of processes of a combustion improvement process in the present embodiment.

DESCRIPTION OF EMBODIMENTS

Hereinafter, an embodiment of the present disclosure will be described in detail with reference to the figures. Note that, the same reference numerals are attached to same or corresponding portions in the figures, and an explanation of these portions are not repeated.

[Embodiment]

Fig. 1 is a figure schematically showing an overall configuration of a vehicle management system relating to the present embodiment. A vehicle management system 100 includes a vehicle 1A, a plurality of vehicles 1B, and a management center 2. A state in which the vehicle 1A is supplied with fuel at a gas station 9 is illustrated in Fig. 1. The vehicle 1A corresponds to a "target vehicle" relating to the present disclosure.
The vehicle 1A and the plurality of vehicles 1B have basically a common configuration. In the case where not distinguishing the vehicle 1A and the vehicles 1B from each other, they will be described as the "vehicle 1". Each of the vehicles 1 and the management center 2 are configured to perform two-way communication.
The vehicle 1 is a vehicle in which an engine 3 is mounted (refer to Fig. 3), and specifically, is a so-called conventional vehicle, hybrid electric vehicle, or plug-in hybrid electric vehicle. The type of fuel burnt by the engine 3 is not particularly limited. The fuel is, for example, gasoline fuel, diesel fuel, biofuel (ethanol or the like), or gaseous fuel (propane gas or the like). Moreover, the fuel may be hydrogen, ammonia, e-fuel (synthetic fuel of carbon dioxide and hydrogen), dimethyl ether (DME) or the like.
Fig. 2 is a figure showing in more detail configurations of the vehicle 1 and the management center 2. The vehicle 1 includes a navigation system 11, a Data Communication Module (DCM) 12, an Electronic Control Unit (ECU) 13, an in-vehicle network 14, the engine 3, and sensors 4.
The navigation system 11 includes a Global Positioning System (GPS) receiver (not illustrated), which specifies the position of the vehicle 1 based on electrical waves from an artificial satellite (not illustrated). The navigation system executes various types of navigation processes of the vehicle 1, by using position information of the vehicle 1 specified by the GPS receiver.
The DCM 12 is configured to perform wireless two-way communication with the management center 2. The management center 2 can manage a traveling history of the vehicle 1, by collecting position information of the vehicle 1. Hereinafter, the position information of the vehicle 1 will also be described as "vehicle position data".
The ECU 13 includes a processor 131 such as a Central Processing Unit (CPU), a memory 132 such as a Read Only Memory (ROM) and a Random Access Memory (RAM), and an input/output port 133 to/from which signals are input/output. The ECU 13 controls the engine 3 so that the vehicle 1 becomes a desired state, based on an input of signals from the sensors 4 and maps and programs stored in the memory 132.
The in-vehicle network 14 is, for example, a wired network such as a Controller Area network (CAN). The in-vehicle network 14 mutually connects the navigation system 11, the DCM 12, and the ECU 13.
The engine 3 outputs a driving force in order for the vehicle 1 to travel, by burning fuel in accordance with a control instruction from the ECU 13. The sensors 4 inclusively describe various types of sensors mounted on the vehicle 1. The sensors 4 in the present embodiment are mainly used for a prediction of a combustion state of the engine 3. Configurations of the engine 3 and the sensors 4 will be described in more detail in Fig. 3 and Fig. 5.
The management center 2 is configured to manage traveling of the plurality of vehicles 1. The management center 2 includes a vehicle database 21, a traveling history database 22, a server 23, a communication module 24, and an intranet 25.
The vehicle database 21 stores identification information, manufacturing information and the like related to each of the plurality of vehicles 1. The identification information includes, for example, a manufacturing number of the vehicle 1 and a communicator ID of the DCM 12. The manufacturing information includes, for example, information related to a manufacturer and a vehicle type (model, grade, model year and the like) of the vehicle 1.
The traveling history database 22 stores a traveling history of each of the plurality of vehicles 1. The traveling history includes, in addition to vehicle position data, data related to a combustion state of fuel in the engine 3 (also described as a combustion state of the engine 3). Gas station data, which will be described below, is also stored in the traveling history database 22.
The server 23 includes, similar to the ECU 13, a processor 231, a memory 232, and an input/output port 233. The server 23 may be a cloud server. One of the main processes executed by the server 23 in the present embodiment is a process for improving a combustion state of the engine 3. This process is described as a "combustion improvement process". The combustion improvement process will be described in detail below.
The communication module 24 is configured to perform wireless two-way communication with the DCM 12 mounted in the vehicle 1.
The intranet 25 mutually connects the vehicle database 21, the traveling history database 22, the server 23, and the communication module 24.

Fig. 3 is a figure showing an example of a configuration of the engine 3. The engine 3 is, for example, an inline four-cylinder spark ignition type internal combustion engine. The engine 3 includes an engine body 31. The engine body 31 includes four cylinders 311-314 lined up in one direction. Since the configuration of each of the cylinders 311-314 is the same, a configuration of the cylinder 311 will be representatively described hereinafter.
Two intake valves 321, two exhaust valves 322, an injector 323, and a spark plug 324 are provided in the cylinder 311. Moreover, an intake passage 33 and an exhaust passage 34 are connected to the cylinder 311. The intake passage 33 is opened/closed by the intake valves 321. The exhaust passage 34 is opened/closed by the exhaust valves 322. Air-fuel mixture is generated, by adding fuel to air supplied to the engine body 31 through the intake passage 33. The fuel is injected within the cylinder 311 by the injector 323, and the mixture is generated within the cylinder 311. Then, the spark plug 324 ignites the mixture within the cylinder 311. In this way, the mixture is burnt within the cylinder 311. Combustion energy produced at the time when burning the mixture within the cylinder 311 is converted into kinetic energy by a piston (not illustrated) within the cylinder 311, and is output. Note that, the fuel supply system of the engine 3 is not limited to a cylinder internal injection system, and may be a port injection system, or a cylinder internal injection system and a port injection system may be used together.
The engine 3 further includes a supercharger 35, a Waste Gate Valve (WGV) device 36, and an EGR device 37. The supercharger 35 is configured to supercharge intake air by using exhaust energy. In more detail, the supercharger 35 includes a compressor 351, a turbine 352, and a shaft 353. The compressor 351 is disposed in the intake passage 33, and the turbine 352 is disposed in the exhaust passage 34. The compressor 351 and the turbine 352 are connected to each other via the shaft 353, and are configured to integrally rotate. The turbine 352 rotates by receiving a flow of exhaust gas exhausted from the engine body 31. The rotational force of the turbine 352 is transmitted to the compressor 351 via the shaft 353, and causes the compressor 351 to rotate. By having the compressor 351 rotate, intake air toward the engine body 31 is compressed, and compressed air is supplied to the engine body 31.
An air flow meter 405 (described below) is provided at a position upstream of the compressor 351 in the intake passage 33. An intercooler 331 is provided at a position downstream of the compressor 351 in the intake passage 33. A throttle valve 332 is provided at a position downstream of the intercooler 331 in the intake passage 33. Air (intake air) flowing into the intake passage 33 is supplied to each of the cylinders 311-314 of the engine body 31 through the compressor 351, the intercooler 331, and the throttle valve 332. The intercooler 331 cools the intake air compressed by the compressor 351. The throttle valve 332 is configured to adjust a flow rate of intake air flowing inside the intake passage 33.
A catalyst device 341 and a postprocessing device 342 are provided in the exhaust passage 34. The catalyst device 341 includes, for example, a three-way catalytic converter. The catalyst device 341 oxides unburnt components (for example, hydrocarbons (HC) or carbon monoxide (CO)) included in the exhaust gas exhausted from the engine 3, and reduces oxidation components (for example, nitrogen oxide (NOx)). The postprocessing device 342 is provided at a position downstream of the catalyst device 341. The postprocessing device 342 includes a filter such as a Diesel Particulate Filter (DPF), and collects Particulate Matter (PM) exhausted from the engine 3.
The WGV device 36 is provided at a position upstream of the catalyst device 341 in the exhaust passage 34. The WGV device 36 is configured to enable exhaust gas exhausted from the engine body 31 to flow by bypassing the turbine 352, and to enable the diverted exhaust gas amount to be adjusted. The WGV device 36 includes a bypass passage 361, a WGV 362, and a WGV actuator 363.
The bypass passage 361 is connected to the exhaust passage 34, and within which exhaust gas flows by bypassing the turbine 352. Specifically, the bypass passage 361 is branched off from a position upstream of the turbine 352 (for example, between the engine body 31 and the turbine 352) in the exhaust passage 34, and joins up at a position downstream of the turbine 352 (for example, between the turbine 352 and the catalyst device 341) in the exhaust passage 34.
The WGV 362 is disposed in the bypass passage 361. The WGV 362 is a negative pressure type valve driven by the WGV actuator 363. The WGV 362 is configured to enable a flow rate of exhaust gas guided to the bypass passage 361 from the engine body 31 to be adjusted, in accordance with an opening thereof. The more the WGV 362 is closed, the more an exhaust gas flow rate guided from the engine body 31 to the bypass passage 361 decreases, and on the other hand, the more an exhaust gas flow rate flowing into the turbine 352 increases, the more a pressure of intake air (supercharging pressure) increases.
The EGR device 37 is provided in the exhaust passage 34, and causes exhaust gas to flow into the intake passage 33. The EGR device 37 includes an EGR passage 371, an EGR valve 372, and an EGR cooler 373. The EGR passage 371 takes out a part of the exhaust gas from the exhaust passage 34 as EGR gas, and guides this EGR gas to the intake passage 33, by connecting a position between the catalyst device 341 and the postprocessing device 342 in the exhaust passage 34, and a position between the compressor 351 and the air flow meter 405 in the intake passage 33. The EGR valve 372 is configured to enable a flow rate of the EGR gas flowing in the EGR passage 371 to be adjusted. The EGR cooler 373 cools the EGR gas flowing in the EGR passage 371.
Fig. 4 is a figure showing another example of a configuration of an engine. The engine 3A shown in Fig. 4 is different to the engine 3 shown in Fig. 3 for the point of including a variable nozzle device 38, instead of the WGV device 36.
The variable nozzle device 38 includes a plurality of variable nozzles (nozzle vanes) provided on an outer periphery of a turbine wheel. The variable nozzles are configured so that an opening thereof (VN opening) changes within a range from opened at a fully opened positioned to closed at a fully closed position, in accordance with a control instruction from the ECU 13. When the nozzle opening of the variable nozzles is small, the opening area of a region where exhaust gas passes through becomes small, and a flow speed of exhaust gas introduced into the turbine 352 increases. As a result, the recovered energy of the turbine wheel increases, and the supercharging pressure increases. In contrast, when the variable nozzle opening is large, the opening area of a region where exhaust gas passes through becomes large, and a flow speed of exhaust gas decreases. As a result, the recovered energy of the turbine wheel reduces, and the supercharging pressure reduces.

Fig. 5 is a figure showing an example of a configuration of the sensors 4. The sensors 4 include a vehicle speed sensor 401, an accelerator opening sensor 402, an engine rotational speed sensor 403, a knock sensor 404, an air flow meter 405, a throttle opening sensor 406, a cylinder internal pressure sensor 407, a coolant temperature sensor 408, a supercharging pressure sensor 409, a supercharging temperature sensor 410, a turbine rotational speed sensor 411, an exhaust temperature sensor 412, an air/fuel ratio sensor 413, an NOx sensor 414, and a differential pressure sensor 415.
The vehicle speed sensor 401 detects a speed of the vehicle 1. The accelerator opening sensor 402 detects a stepped amount (accelerator opening) of an accelerator pedal (not illustrated) by a driver. The engine rotational speed sensor 403 detects a rotational angle (crank angle) of a crank shaft (not illustrated) and a rotational speed of the engine 3. The knock sensor 404 detects an occurrence of knocking (vibration of the engine body 31) in the engine 3. The air flow meter 405 detects an intake air amount to the intake passage 33.
The throttle opening sensor 406 detects an opening (throttle opening) of the throttle valve 332. The cylinder internal pressure sensor 407 detects a pressure within a combustion chamber (not illustrated). The coolant temperature sensor 408 detects a temperature (coolant temperature) of coolant circulating in a water jacket (not illustrated) provided in the engine body 31. The supercharging pressure sensor 409 detects a supercharging pressure (pressure downstream of the compressor 351) by the supercharger 35. The supercharging temperature sensor 410 detects a temperature (supercharging temperature) of intake air cooled by the intercooler 331.
The turbine rotational speed sensor 411 detects a rotational speed of the turbine 352 of the supercharger 35. The exhaust temperature sensor 412 detects a temperature of the exhaust gas. The air/fuel ratio sensor 413 detects an oxygen concentration within the exhaust gas. The NOx sensor 414 detects a nitrogen oxide (NOx) concentration within the exhaust gas. The differential pressure sensor 415 detects a front/back pressure difference of a filter (not illustrated) included in the postprocessing device 342. Each of the above sensors outputs a signal showing a detection result thereof to the ECU 13.

In order to implement a fuel efficiency improvement and a drivability improvement of the vehicle 1A, it is desirable to improve a combustion state of the engine 3. It can be considered that the vehicle 1A acquires composition information of fuel, and improves a combustion state of the engine 3 based on this composition information (for example, refer to PTL 1). However, a complex process such as an analysis of fuel may be necessary in order to obtain composition information of fuel. Accordingly, in the present embodiment, the configuration described below is adopted, instead of a configuration that uses composition information of fuel.
The vehicle 1 (ECU 13) transmits data acquired by the signals from each of the sensors included in the sensors 4 to the management center 2 (server 23). In this way, the management center 2 can acquire or estimate a combustion state of fuel in the engine 3 (cylinders 311-314). Specifically, the management center 2 can acquire a combustion state of fuel in the cylinders 311-314, based on a signal (waveform signal of a cylinder internal pressure) by the cylinder internal pressure sensor 407. Moreover, the management center 2 can estimate a combustion state of the engine 3, based on a signal (a signal representing noise or vibrations occurring in the cylinders 311-314) detected by the knock sensor 404. In addition, it is also possible for the management center 2 to estimate a combustion state of the engine 3, based on signals detected by the exhaust temperature sensor 412, the NOx sensor 414, and/or the differential pressure sensor 415. Hereinafter, the data transmitted from the vehicle 1 to the management center 2 and collected by the management center 2 will also be inclusively called "sensor data".
The sensors included in the sensors 4 shown in Fig. 5 are described in a confirmational manner as being merely exemplifications. Namely, the sensor data is not limited to data acquired from these sensors. The sensor data may be data acquired from a part of these sensors, or may include data acquired from other sensors not shown in Fig. 5.
The vehicle 1 (vehicle 1A and vehicles 1B) transmits, in addition to the sensor data, data for specifying a gas station where the vehicle 1 is supplied with fuel to the server 23. Hereinafter, data that enables a gas station to be specified will also be called "gas station data". Specifically, the gas station data includes vehicle position data at the time of supplying fuel to the vehicle 1. Since an installation location of a gas station is already known by the management center 2 from map information, a gas station can be specified from vehicle position data at the time of being supplied with fuel.
Moreover, even at the same gas station, the properties of supplied fuel can change in accordance with a fuel supply time period (for example, the season). Therefore, it is preferable for the gas station data to include data related to a fuel supply time period of the vehicle 1, in addition to vehicle position data at the time of fuel supply of the vehicle 1. The properties of supplied fuel can also be associated with time, by having the gas station data include data related to a fuel supply time period (for example, data related to a date and a time). However, gas station data may not include data related to a fuel supply time period.
A prediction model constructed by machine training and/or data mining related to a combustion state of various fuels is stored in the memory 232 of the server 23 installed in the management center 2. A prediction model is constructed for each vehicle (model, grade, model year and the like). It is desirable for the management center 2 to update each prediction model so as to correspond to the latest fuel properties, by continuing training of each prediction model by using sensor data and gas station data collected from the vehicles 1B. When receiving sensor data and gas station data from the vehicle 1A, the management center 2 inputs these pieces of data to a prediction model corresponding to the vehicle type of the vehicle 1A and obtains an output from the prediction model. In this way, it becomes possible to improve a combustion state of the engine 3 of the vehicle 1A, even if not using composition information of a specific fuel.

Fig. 6 is a functional block diagram showing an example of a prediction model (combustion prediction model) related to a combustion state of the engine 3 in the present embodiment. When receiving sensor data and gas station data, the combustion prediction model sets an index indicated by the sensor data and gas station data as a characteristic amount, and outputs a correction amount in accordance with a calculation rule obtained by machine training.
The combustion prediction model is a trained model, for example, a neural network model constructed by unsupervised training or reinforced training. The neural network model includes an input layer, a hidden layer, and an output layer. An input variable to the input layer is described as y_i (i is a natural number). An output variable from the hidden layer to the output layer is described as θ_j (j is a natural number). An output variable taken out from the output layer to the outside is described as Z_p. A weight (weighting) from an i^th node of the input layer to a j^th node of the hidden layer is described as W_ji. A weight from a j^th node of the hidden layer to a node of the output layer is described as W_j'.
The input variable y_i is sensor data and gas station data collected from the vehicle 1. In the example shown in Fig. 6, position information of the vehicles 1B (vehicle position data) acquired by the navigation system 11 is also included in the input variable y_i. The output variable Z_p is a correction amount of a control instruction to the engine 3. Machine training is performed by using this correction amount as teaching data.
The correction amount can be classified into, for example, a correction amount of a control instruction to an injection system, a correction amount of a control instruction to an air intake system, and a correction amount of a control instruction to an EGR system. More specifically, the correction amount of a control instruction to an injection system is an amount of change from a standard amount of an ignition amount of fuel from the injector 323, an amount of change from a standard timing of an injection timing of fuel from the injector 323, an amount of change from a standard timing of an ignition timing of the spark plug 324 or the like. The correction of a control instruction to an air intake system is an amount of change from a standard amount of a throttle opening, an amount of change from a standard amount of opening of the WGV 262 (or the variable nozzle device 38) or the like. The correction amount of a control instruction to an EGR system is an amount of change from a standard amount of an opening of the EGR valve 372 or the like. These correction amounts are transmitted to the vehicle 1A by the communication module 24 provided in the management center 2.
It is desirable for vehicle position data to be used, in addition to sensor data and gas station data, for the training of a prediction model. Various information (information related to weather, temperature, precipitation, gradient, speed limit, or a presence or absence of traffic congestion) related to a traveling point of the vehicles 1B is reflected in the vehicle position data. Therefore, it becomes possible to construct a prediction model that is trained with more accuracy, by using the vehicle position data.
Note that, the neural network model is merely an example of a technique that can be adopted for machine training of a combustion state of the engine 3, and the technique of machine training is not limited to this. The server 23 may use other machine training models such as a deep learning model, a logistic regression model, or a support vector machine.
Moreover, the server 23 may analyze sensor data and gas station data collected by the management center 2 by a data mining technique for so-called big data. A publicly known technique such as follows can be adopted as a data mining technique. The server 23, for example, uses an injection timing of fuel from the injector 323 as a target variable and uses other parameters as explanatory variables, and selects a plurality of explanatory variables to be used for a correction of the injection timing of fuel, for example, by discovering a relationship between the explanatory variables by clustering. Also, in order to predict how the target variable (in this example, an injection timing of fuel) changes in the case where a selected explanatory variable changes, the server 23 extracts a regularity established between the explanatory variable and the target variable. The extracted regularity is represented as a prediction model or a relation expression of the target variable (statistical modeling). As an example, in the case where performing regression analysis (for example, multiple regression analysis) using a plurality of explanatory variables, the server 23 can calculate parameters of a prediction model used for regression analysis by parameter adaptation (fitting).

Fig. 7 is a figure for describing an example of a combustion improvement process to correct an operation of an injection system of the engine 3. In Fig. 7 and Fig. 8 to Fig. 10, which will be described below, the horizontal axis represents a crank angle of the engine 3. The vertical axis represents a cylinder internal pressure of the cylinders 311-314 of the engine 3.
There is the possibility that the cylinder internal pressure deviates from a target cylinder internal pressure (an aimed-for cylinder internal pressure) by the properties of fuel (refer to the cylinder internal pressure before correction within the figure). In such a case, the cylinder internal pressure can be brought close to the target cylinder internal pressure, for example, by correcting an operation of an injection system of the engine 3 by using a correction amount (refer to the cylinder internal pressure after correction). An example of retarding an injection timing of fuel is shown in Fig. 7. An air intake amount, exhaust amount (difference between a target value and an actual value of an exhaust amount), unburnt components within the exhaust gas (hydrocarbons or the like) and/or a concentration of nitrogen oxide, an ignition timing of mixture, a combustion noise or the like can be used for the construction of a combustion prediction model (training of a model) in this example.
Fig. 8 is a figure for describing another example of a combustion improvement process to correct an operation of an injection system of the engine 3. In the case where the ignitability of fuel supplied to the engine 3 is good or in the case where bad, compared to a typical ignitability of fuel, there is the possibility that the cylinder internal pressure deviates from a target cylinder internal pressure. An ignition amount of fuel from the injector 323 being adjusted by changing a pulse size (pulse height and/or pulse width) output from the ECU 13 to the injector 323. By correcting the pulse size by using a correction amount, the cylinder internal pressure can be brought close to the target cylinder internal pressure. It is desirable to correct an injection amount of fuel for the present combustion in multi-injection.
Fig. 9 is a figure for describing an example of a combustion improvement process to correct an operation of an air intake system of the engine 3. Another cause of the cylinder internal pressure deviating from a target cylinder internal pressure is an air intake amount to the engine 3. In such a case, the cylinder internal pressure can be brought close to the target cylinder internal pressure, by correcting an air intake amount by an opening adjustment of the throttle valve 332 and/or the WGV 362 (variable nozzle device 38) using a correction amount.
Fig. 10 is a figure for describing an example of a combustion improvement process to correct an operation of an EGR system of the engine 3. The quantity of EGR gas can also become a cause of the cylinder internal pressure deviating from a target cylinder internal pressure. In such a case, the cylinder internal pressure can be brought close to the target cylinder internal pressure, by correcting an EGR gas amount by an opening adjustment of the EGR valve 372 using a correction amount.
Note that, while examples have been described in Fig. 7 to Fig. 10 in which the cylinder internal pressure is brought close to a target cylinder internal pressure, a combustion state of the engine 3 may be improved by using a correction amount, so that another parameter (for example, a heat generation rate of the engine 3) is brought close to a target value.

Fig. 11 is a flow chart showing a procedure of processes of a combustion improvement process in the present embodiment. The processes shown in this flow chart are executed at the time of establishing a predetermined condition (for example, for each prescribed control cycle within an operation of the engine 3 during traveling of the vehicle 1A or the like). The processes executed in the vehicle 1A are shown on the left side of the figure, and the processes executed in the management center 2 are shown on the right side. While each step is implemented by software processes by the vehicle 1A (ECU 13) or the management center 2 (server 23), they may be implemented by hardware (electronic circuits) disposed within the ECU 13 or the server 23.
In this example, the ECU 13 uses a flag to manage a transmission control to the management center 2 related to sensor data, vehicle position data, and gas station data. This flag is described as a "data transmission flag".
In S11, the ECU 13 determines whether or not fuel of the vehicle 1A has been supplied. The ECU 13 can determine that fuel has been supplied in the case where an increase of fuel within a fuel tank (not illustrated) has been detected by a fuel meter (not illustrated). In the case where fuel has been supplied (YES in S11), the ECU 23 turns ON the data transmission flag (S12). Then, the ECU 23 transmits gas station data to the management center 2 (S13). The gas station data is stored in the traveling history database 22. Note that, in the case where fuel of the vehicle 1A has not been supplied (NO in S11), the processes of S12 and S13 are skipped, and the data transmission flag is maintained as OFF.
In S14, the ECU 13 determines whether or not the data transmission flag is turned ON. In the case where the data transmission flag is ON (YES in S14), the ECU 13 transmits sensor data and vehicle position data to the management center 2 (S15). Since the series of processes on the left side of the figure are executed each time a prescribed condition is established (for example, for each control cycle), the process of S15 is continuously executed. Namely, sensor data and vehicle position data are continuously transmitted, until a collection of data of a necessary amount is completed (described below). The sensor data and vehicle position data are stored in the traveling history database 22 in association with the gas station data 13 collected in S13. In the case where the data transmission flag is OFF (NO in S14), transmission of the sensor data and vehicle position data from the vehicle 1A to the management center 2 is not performed.
In S21, the server 23 determines whether or not data of a necessary amount for calculating a correction amount by using a combustion prediction model has been collected. In the case where a collection of necessary data has been completed (YES in S21), the server 23 calculates a correction amount by providing the collected data (sensor data, vehicle position data, and gas station data) to a combustion prediction model (S22). Then, the server 23 transmits the calculated correction amount to the vehicle 1A (S23). Note that, in the case where the correction amount is outside a normally assumed range, there is the possibility that an unsuitable fuel (for example, a non-standard fuel) is supplied. In this case, the server 23 may transmit a warning to the vehicle 1A, instead of or in addition to a correction amount.
In S16, the ECU 13 determines whether or not a correction amount has been received from the management center 2. When a correction amount is received (YES in S16), The ECU 13 switches the data transmission flag from ON to OFF. In addition, the ECU 13 corrects an operation of an injection system, an EGR system, and/or an air intake system of the engine 3, by using a correction amount, as described in Fig. 7 to Fig. 10 (S18). In this way, a fuel state of the engine 3 is improved.
As described above, in the present embodiment, a correction amount for correcting a combustion state of the engine 3 of the vehicle 1A is calculated, from sensor data, vehicle position data, and gas station data of the vehicle 1A, by using a combustion prediction model that is a prediction model related to a combustion state of fuel. The gas station data is data for specifying a gas station, and is not data related to an analysis result of a supplied fuel composition. According to the present embodiment, a combustion state of the engine 3 can be calculated, even if composition information of the fuel is not obtained. In addition, the occurrence of combustion defects (for example, misfiring or abnormal combustion noise) in the engine 3 can be suppressed.
The properties of fuel within the fuel tank can be changed with each supply of fuel. However, the combustion prediction model is constructed by using big data that represents a detection result of a combustion state of various fuels, collected from the plurality of vehicles 1B. Therefore, according to the present embodiment, an appropriate correction of a combustion state of the engine 3 is possible for a large number of various fuels, and not just for a specified fuel.
In addition, generally, the arithmetic processing capability of a vehicle ECU is limited. By using the server 23 installed in the management center 2, an advanced arithmetic processing capability can be easily secured, and an arithmetic load of the ECU 13 can be reduced. Moreover, by having the server 23 train a correction amount, it is possible for an arithmetic load of the server 23 to also be reduced.
The embodiment currently disclosed is an exemplification for all points, and should be considered to be not limited. The range of the present disclosure is shown not by the above-stated description of the embodiment, but by the scope of the claims, and is intended to include all modifications within the meaning and scope equivalent to the description of the claims.

REFERENCE SIGNS LIST

100 vehicle management system; 1 (1A, 1B) vehicle; 11 navigation system; 111 GPS receiver; 13 ECU; 131 processor; 132 memory; 133 input/output port; 14 in-vehicle network; 2 management center; 21 vehicle database; 22 traveling history database; 23 server; 231 processor; 232 memory; 233 input/output port; 24 communication module; 25 intranet; 3, 3A engine; 31 engine body; 311-314 cylinder; 321 intake valve; 322 exhaust valve; 323 injector; 324 spark plug; 33 intake passage; 331 intercooler; 332 throttle valve; 34 exhaust passage; 341 catalyst device; 342 postprocessing device; 35 supercharger; 351 compressor; 352 turbine; 353 shaft; 36 WGV device; 361 bypass passage; 362 WGV; 363 WGV actuator; 37 EGR device; 371 EGR passage; 372 EGR valve; 373 EGR cooler; 38 variable noise device; 4 sensors; 401 vehicle speed sensor; 402 accelerator opening sensor; 403 engine rotational speed sensor; 404 knock sensor; 405 air flow meter; 406 throttle opening sensor; 407 cylinder internal pressure sensor; 408 coolant temperature sensor; 409 supercharging pressure sensor; 410 supercharging temperature sensor; 411 turbine rotational speed sensor; 412 exhaust temperature sensor; 413 air/fuel ratio sensor; 414 sensor; 415 differential pressure sensor.

Claims

A vehicle management system, comprising:
a target vehicle; and

a server that communicates with the target vehicle and a plurality of vehicles, wherein

each of the target vehicle and the plurality of vehicles includes at least one sensor that detects a combustion state of fuel in an engine,

the server includes

a memory that stores a trained model related to engine control, and

a processor that executes an arithmetic process using the trained model,

in the trained model, a correction amount for correcting a combustion state of fuel in the engine is trained based on sensor data and gas station data, the sensor data showing a detection result of the at least one sensor of each of the plurality of vehicles, the gas station data being capable of specifying a gas station where each of the plurality of vehicles received a supply of fuel,

the processor calculates the correction amount by providing the sensor data and the gas station data of the target vehicle to the trained model, and transmits the calculated correction amount to the target vehicle, and

the target vehicle corrects a combustion state of fuel in the engine of the target vehicle by using the received correction amount.
The vehicle management system according to claim 1, wherein the gas station data includes data related to an installation location of the gas station and data related to a fuel supply time period at the gas station.
The vehicle management system according to claim 1 or 2, wherein in the trained model, the correction amount is trained based on the sensor data, the gas station data, and data related to a traveling position of the plurality of vehicles, and
the processor calculates the correction amount of the target vehicle by providing, in addition to the sensor data and the gas station data, data related to the traveling position of the target vehicle to the trained model.
The vehicle management system according to any one of claims 1 to 3, wherein the sensor data includes data related to at least one of a cylinder internal pressure of the engine, a rotational speed of the engine, a temperature of exhaust gas from the engine, a front/back pressure difference of a filter for capturing particulate matter within the exhaust gas, and a nitrogen oxide concentration within the exhaust gas.
A vehicle that is capable of communicating with a server including a trained model related to engine control, wherein
in the trained model, a correction amount for correcting a combustion state of fuel in an engine is trained based on sensor data and gas station data collected by communication with a plurality of vehicles each with the engine mounted therein,

the sensor data is data showing a detection result of a combustion state of fuel in the engine of each of the plurality of vehicles,

the gas station data is data capable of specifying a gas station where the plurality of vehicles received a supply of fuel, and

the vehicle comprises:
a communicator that transmits the sensor data and the gas station data of the vehicle to the server, and receives the correction amount calculated in the server by providing the sensor data and the gas station data from the vehicle to the trained model; and

a controller that corrects a combustion state of fuel in the engine of the vehicle by using the received correction amount.
A vehicle traveling management method for managing traveling of a vehicle by using a trained model related to engine control, wherein
in the trained model, a correction amount for correcting a combustion state of fuel in an engine is trained based on sensor data and gas station data collected by communication with a plurality of vehicles each with the engine mounted therein,

the sensor data is data showing a detection result of a combustion state of fuel in the engine of each of the plurality of vehicles,

the gas station data is data capable of specifying a gas station where the plurality of vehicles received a supply of fuel, and

the traveling management method comprises:
calculating the correction amount for correcting a combustion state of fuel in the engine of a target vehicle by providing the sensor data and the gas station data from the target vehicle to the trained model; and

correcting a combustion state of fuel in the engine of the target vehicle by using the correction amount.