WO2024038672A1

WO2024038672A1 - Energy consumption estimation device, model generation device, program, and method for generating model

Info

Publication number: WO2024038672A1
Application number: PCT/JP2023/022798
Authority: WO
Inventors: 悟渡辺
Original assignee: 株式会社ブリヂストン
Priority date: 2022-08-19
Filing date: 2023-06-20
Publication date: 2024-02-22
Also published as: JP2024028051A

Abstract

Provided is an energy consumption estimation device (10) for estimating energy consumption of a vehicle (20), the energy consumption estimation device comprising a model generation unit (131) that generates a model for obtaining an objective variable from an explanatory variable. The model is a multiple regression model that includes the energy consumption of the vehicle as an objective variable, and includes vehicle travel information, vehicle load information, and vehicle tire information as explanatory variables. The model generation unit calculates a first coefficient associated with the vehicle tire information, using vehicle performance data indicating the objective variable with respect to the explanatory variables.

Description

Energy consumption estimation device, model generation device, program and model generation method

The present disclosure relates to an energy consumption estimation device, a model generation device, a program, and a model generation method. The present disclosure particularly relates to an energy consumption estimation device, a model generation device, a program, and a model generation method for estimating the energy consumption of a vehicle.

Conventionally, techniques have been proposed to estimate energy consumption such as vehicle fuel consumption. For example, in Patent Document 1, in addition to acquiring the fuel consumption amount and mileage associated with the driving history from the vehicle, the fuel consumption amount of the planned driving route is calculated based on the tendency of fuel consumption amount for each driving situation pattern of the driver. Discloses a system for estimating.

JP2009-031046A

Here, energy consumption such as fuel consumption of a vehicle changes depending on various factors such as driving conditions and equipment. Conventional technologies such as Patent Document 1 estimate fuel consumption from information on a planned travel route, but cannot estimate changes in fuel consumption due to changes in vehicle equipment, for example. Therefore, there is a need to further improve the estimation accuracy of vehicle energy consumption.

An object of the present disclosure, made in view of the above circumstances, is to provide an energy consumption estimation device, a model generation device, a program, and a model generation method that are capable of estimating the energy consumption of a vehicle with high accuracy. .

(1) An energy consumption estimation device according to an embodiment of the present disclosure is an energy consumption estimation device for a vehicle, and includes a model generation unit that generates a model for obtaining a target variable from an explanatory variable, and the model is a multiple regression model that includes the energy consumption of the vehicle as the objective variable, and includes traveling information of the vehicle, load information of the vehicle, and tire information of the vehicle as the explanatory variables; , calculating a first coefficient associated with tire information of the vehicle using performance data of the vehicle indicating the objective variable for the explanatory variable.

(2) As an embodiment of the present disclosure, in (1), the first coefficient is associated with a dynamic load radius.

(3) As an embodiment of the present disclosure, in (1) or (2), the model generation unit calculates a second coefficient associated with at least one of slope resistance and rolling resistance.

(4) As an embodiment of the present disclosure, in any one of (1) to (3), the model further includes vehicle information of the vehicle as the explanatory variable.

(5) As an embodiment of the present disclosure, in any one of (1) to (4), the vehicle includes an energy consumption estimation unit that estimates the energy consumption of the vehicle using the generated model.

(6) As an embodiment of the present disclosure, in (5), the energy consumption estimating unit is configured to reduce energy consumption of the vehicle, which is consumed by at least one of rolling resistance, air resistance, slope resistance, acceleration resistance, and kinetic energy. Estimate energy consumption.

(7) A model generation device according to an embodiment of the present disclosure is a model generation device that generates a model for estimating energy consumption of a vehicle, and is a model that generates a model for obtaining a target variable from an explanatory variable. The model includes an energy consumption amount of the vehicle as the target variable, travel information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as the explanatory variables. The model generation unit generates the model by machine learning using performance data of the vehicle indicating the objective variable for the explanatory variable.

(8) A program according to an embodiment of the present disclosure is a program that causes a computer to generate a model for estimating energy consumption of a vehicle, the model including the energy consumption of the vehicle as a target variable. , a machine learning model that includes running information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as explanatory variables, and the objective variable for the explanatory variable is displayed on the computer. acquiring performance data of the vehicle, and performing machine learning using the performance data.

(9) A method of generating a model according to an embodiment of the present disclosure is a method of generating a model for estimating energy consumption of a vehicle, which is executed by a computer, wherein the model uses energy of the vehicle as an objective variable. The machine learning model is a machine learning model that includes a consumption amount, and includes driving information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as explanatory variables, and the objective variable for the explanatory variable is indicated. and performing machine learning using the performance data.

According to the present disclosure, it is possible to provide an energy consumption estimation device, a model generation device, a program, and a model generation method that can estimate the energy consumption of a vehicle with high accuracy.

FIG. 1 is a diagram illustrating a configuration example of an energy consumption estimation system including an energy consumption estimation device according to an embodiment of the present disclosure. FIG. 2 is another diagram showing a configuration example of the energy consumption estimation system of FIG. 1. FIG. 3 is a flowchart illustrating an example of an energy consumption estimation method according to an embodiment of the present disclosure. FIG. 4 is a flowchart showing details of the model generation process of FIG. 3. FIG. 5 is a flowchart showing details of the energy consumption estimation process of FIG. 3.

Hereinafter, an energy consumption estimation device, a model generation device, a program, and a model generation method according to an embodiment of the present disclosure will be described with reference to the drawings. In each figure, the same or corresponding parts are given the same reference numerals. In the description of this embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate.

FIGS. 1 and 2 are diagrams showing a configuration example of an energy consumption estimation system including an energy consumption estimation device 10 according to the present embodiment. FIG. 1 is a block diagram including an example of the internal configuration of an energy consumption estimation device 10. As shown in FIG. FIG. 2 shows the overall configuration of the energy consumption estimation system.

The energy consumption estimation device 10 is a device for estimating the energy consumption of the vehicle 20. In this embodiment, the energy consumption estimation device 10 generates a model and estimates the energy consumption of the vehicle 20 using the generated model. Here, the energy consumption estimation device 10 may include only a function of generating a model, or may include only a function of estimating the energy consumption of the vehicle 20. Further, the functional part of the energy consumption estimation device 10 that generates a model may be treated as one device and referred to as a model generation device. The energy consumption amount is the amount of energy consumed by the vehicle 20 for running, and is expressed based on a certain standard so that it can be compared objectively. For example, if the vehicle 20 uses gasoline, the energy consumption may be the fuel consumption. In this embodiment, the energy consumption amount will be explained as the fuel consumption amount, but it is not limited to this. For example, if the vehicle 20 is an electric vehicle, the energy consumption may be the power consumption. Further, for example, the energy consumption may be the amount of hydrogen used if the vehicle 20 uses hydrogen.

The energy consumption estimation device 10 includes a communication section 11, a storage section 12, and a control section 13. The control unit 13 includes a model generation unit 131 and an energy consumption estimation unit 132. The energy consumption estimation device 10 may be, for example, a computer as a hardware configuration. Details of the components of the energy consumption estimation device 10 will be described later.

The energy consumption estimation device 10 may constitute an energy consumption estimation system together with the operation management device 60 connected through the network 40. The network 40 is, for example, the Internet, but may also be a LAN (Local Area Network) or the like.

The operation management device 60 is a device that manages the operation of the vehicle 20. The operation management device 60 is configured, for example, by a computer different from the energy consumption estimation device 10. The vehicle 20 is not limited to a specific purpose and may be, for example, a passenger car, a taxi, a truck, etc., but will be described as a bus in this embodiment. That is, in this embodiment, the operation management device 60 is a device that manages the vehicle 20 as a bus that is a public transportation system. The operation management device 60 manages the running state of the vehicle 20, the load state, information about the tires 30 mounted on the vehicle 20, and the like. The operation management device 60 may manage the type, size, dynamic load radius, rolling resistance, etc. of the tires 30 as information about the tires 30 mounted on the vehicle 20. The dynamic load radius is the effective radius of the tire 30 in terms of the actual amount of movement of the vehicle 20. The traffic management device 60 may manage the type of the vehicle 20, the front projected area according to the type, etc. as vehicle information. Further, the operation management device 60 may manage data such as the weight of the vehicle 20, map information of the travel route (route) of the vehicle 20, and the like. The operation management device 60 may accumulate information regarding the operation of the vehicle 20 as a database and store it in an accessible storage device. Additionally, the traffic management device 60 provides information in the database to the energy consumption estimation device 10 so that the energy consumption estimation device 10 can estimate the energy consumption of the vehicle 20.

In this embodiment, the vehicle 20, which is a bus, includes a detection device 70 and a communication device 80. The detection device 70 detects vehicle operation information, which is information indicating the running state of the vehicle 20, and outputs it to the communication device 80. The detection device 70 may include, for example, an acceleration sensor, and output information on the acceleration of the vehicle 20 as vehicle operation information. The detection device 70 may include, for example, a speed sensor, and output information on the speed of the vehicle 20 as vehicle operation information. The detection device 70 includes, for example, a device having a GPS (Global Positioning System) function, and may output position information or travel distance of the vehicle 20 as vehicle operation information. In this embodiment, the detection device 70 includes an IC card reader for paying fares, and outputs the number of people getting on and off the vehicle 20 as vehicle operation information.

The communication device 80 is an in-vehicle device that outputs vehicle operation information to the operation management device 60. The communication device 80 may be configured with, for example, a wireless communication module, or may be a digital tachograph, but is not limited to a specific device. In this embodiment, the communication device 80 is configured to be able to communicate with the operation management device 60 via the network 40. The vehicle operation information output to the operation management device 60 is accumulated in the database of the operation management device 60 as performance data.

Here, the performance data (vehicle operation information output to the operation management device 60 via the communication device 80) is stored in the database in association with the actual fuel consumption of the vehicle 20 by the operation management device 60. The actual fuel consumption of the vehicle 20 may be calculated by the operation management device 60 based on the refueling information of the vehicle 20, for example. Further, the performance data may be stored in the database by the operation management device 60 in association with the type of tire 30 or the type of vehicle 20.

Hereinafter, the details of the components of the energy consumption estimation device 10 will be explained. The communication unit 11 is configured to include one or more communication modules connected to the network 40. The communication unit 11 may include a communication module compatible with mobile communication standards such as 4G (4th Generation) and 5G (5th Generation). The communication unit 11 may include, for example, a communication module compatible with a wired LAN standard (for example, 1000BASE-T). The communication unit 11 may include, for example, a communication module compatible with a wireless LAN standard (for example, IEEE802.11).

The storage unit 12 is one or more memories. The memory is, for example, a semiconductor memory, a magnetic memory, or an optical memory, but is not limited to these and may be any memory. The storage unit 12 is built in, for example, the energy consumption estimation device 10, but a configuration in which it is externally accessed by the energy consumption estimation device 10 via an arbitrary interface is also possible.

The storage unit 12 stores various data used in various calculations executed by the control unit 13. Furthermore, the storage unit 12 may store intermediate data and results of various calculations executed by the control unit 13. In this embodiment, the storage unit 12 stores the generated model.

The control unit 13 is one or more processors. The processor is, for example, a general-purpose processor or a dedicated processor specialized for specific processing, but is not limited to these and can be any processor. The control unit 13 controls the overall operation of the energy consumption estimation device 10.

Here, the energy consumption estimation device 10 may have the following software configuration. One or more programs used to control the operation of the energy consumption estimation device 10 are stored in the storage unit 12. When the program stored in the storage unit 12 is read by the processor of the control unit 13, the program causes the control unit 13 to function as the model generation unit 131 and the energy consumption estimation unit 132.

The model generation unit 131 generates a model used to estimate the energy consumption (fuel consumption in this embodiment) of the vehicle 20. The model is not limited to a specific model as long as it is configured to obtain the objective variable from the explanatory variables. In this embodiment, the model is a multiple regression model (multiple regression equation) that has coefficients for multiple explanatory variables. The target variable is the fuel consumption amount of the vehicle 20. Further, the explanatory variables include travel information of the vehicle 20, load information of the vehicle 20, and tire information of the vehicle 20. As described above, the performance data accumulated in the database of the operation management device 60 is associated with the actual fuel consumption of the vehicle 20. In other words, the performance data indicates fuel consumption with respect to travel information, load information, and tire information of the vehicle 20. The model generation unit 131 generates a model by determining coefficients using performance data of the vehicle 20. Here, the explanatory variables may further include vehicle information of the vehicle 20.

Here, the travel information of the vehicle 20 may include at least one of the acceleration and speed of the vehicle 20 in the performance data. Further, the traveling information of the vehicle 20 may include position information of the vehicle 20. Based on the position information, the distance traveled by the vehicle 20, etc. can be calculated. Further, based on the position information and map information managed by the traffic management device 60, the slope or altitude of the location where the vehicle 20 is traveling can be calculated.

Furthermore, the load information of the vehicle 20 may include information on the weight of the vehicle 20 managed by the operation management device 60. In this embodiment, the load information of the vehicle 20 is determined in consideration of the number of people getting on and off the vehicle 20. The weight of the vehicle 20 is not fixedly determined only by the weight of the vehicle 20 itself, but the weight of the vehicle 20 is accurately calculated by taking into account variations based on the number of passengers getting on and off the vehicle.

Additionally, the tire information of the vehicle 20 includes at least the type of tire 30 and the dynamic load radius. The tire information of the vehicle 20 may further include rolling resistance, size, and the like. The relationship between the dynamic load radius and the kinetic energy of the vehicle 20 will be described later.

Further, the vehicle information of the vehicle 20 includes at least the type of the vehicle 20 and the front projected area. The front projected area is a cross-sectional area when the vehicle 20 is viewed from the front, and an air resistance value can be obtained by multiplying this by an air resistance coefficient. The front projected area differs depending on the type of vehicle 20 (model, class, etc.). Here, when the model includes vehicle information as an explanatory variable, it becomes possible to take into account differences in air resistance depending on the vehicle 20, and it becomes possible to estimate the energy consumption of the vehicle 20 with even higher accuracy. .

Here, the model generation unit 131 may classify the performance data used for model generation, for example, by the type of vehicle 20. The model generation unit 131 may generate a plurality of models using each of the classified performance data. Further, when the vehicle 20 is a bus as in this embodiment, the model generation unit 131 may classify the performance data used for model generation by service system. For example, when performance data is classified by system X, system Y, or system Z, which are operating systems, there is a model corresponding to the vehicle 20 traveling on system X, a model corresponding to the vehicle 20 traveling on system Y, and a model corresponding to the system A model corresponding to the vehicle 20 traveling in Z may be generated. The model generation unit 131 may cause the storage unit 12 to store the generated model.

The energy consumption estimation unit 132 estimates the energy consumption (fuel consumption in this embodiment) of the vehicle 20 using the model generated by the model generation unit 131. The energy consumption estimation unit 132 executes the estimation of the fuel consumption when receiving an estimation execution instruction and estimation conditions (input data) from outside the energy consumption estimation device 10 via the communication unit 11. good. When a plurality of models are generated by the model generation unit 131, the energy consumption estimation unit 132 may select a model according to input data. For example, when the input data specifies the vehicle 20 of a specific model, the energy consumption estimation unit 132 selects a model corresponding to the specific model and estimates (calculates) the energy consumption. It's fine.

Here, the model generation unit 131 generates a multiple regression model having coefficients for a plurality of explanatory variables, as described above. In the present embodiment, the model generation unit 131 calculates a first coefficient associated with tire information of the vehicle 20 using performance data of the vehicle 20, and calculates a first coefficient associated with at least one of slope resistance and rolling resistance. 2. Calculate the coefficient. That is, the model generation unit 131 generates a model used for estimating the energy consumption amount of the vehicle 20 by determining the first coefficient and the second coefficient of the multiple regression model. The concept of fuel consumption estimation in this embodiment will be explained below in relation to the first coefficient and the second coefficient.

In the vehicle 20, energy generated by fuel is converted into rotational motion of the tires 30. The energy generated by the fuel is determined by dividing the kinetic energy of the vehicle 20 by the energy conversion efficiency. Therefore, the energy consumption (fuel consumption in this embodiment) of the vehicle 20 can be estimated with high accuracy by accurately determining the kinetic energy of the vehicle 20. The kinetic energy E[J] of the vehicle 20 is calculated using the following formula.

Here, M is the weight [kg] of the vehicle 20. a is the acceleration [m/s ² ] of the vehicle 20. L is the moving distance [m] of the vehicle 20. R is the running resistance [N] of the vehicle 20. Running resistance can generally be broken down into Rr, Ra, Rg, and Rac. Rr is rolling resistance [N]. Ra is air resistance [N]. Rg is gradient resistance [N]. Rac is acceleration resistance [N]. Rr, Ra, Rg and Rac are calculated using the following formulas.

Here, g is gravitational acceleration [m/s ² ]. RRC is rolling resistance coefficient [-]. ρ is the air density [N·s ² /m ⁴ ]. Cd is the air resistance coefficient [-]. A is the front projected area [m ² ] of the vehicle 20. V is the speed [m/s] of the vehicle 20. θ is the gradient angle [°] of the road on which the vehicle 20 travels. Mr is the weight [kg] equivalent to the rotating part of the vehicle 20.

Here, in calculating the kinetic energy E, the weight (M) of the vehicle 20 is obtained from the load information of the vehicle 20 stored in the database. The acceleration (a), travel distance (L), and speed (V) of the vehicle 20 are obtained from travel information of the vehicle 20 stored in the database. The slope angle (θ) can be obtained based on travel information of the vehicle 20 stored in the database. The rolling resistance coefficient (RRC) can be obtained based on the tire information of the vehicle 20 stored in the database. For the front projected area (A) of the vehicle 20, a general size of a bus (7 [m ² ] as an example) may be used, or it may be calculated from the vehicle information when the vehicle information of the vehicle 20 is stored in the database. Good to get it. Constants may be used for the gravitational acceleration (g), air density (ρ), and air resistance coefficient (Cd). However, the weight equivalent to the rotating part (Mr) of the vehicle 20 cannot be obtained as a measured value. Furthermore, when calculating the weight equivalent to the rotating parts of the vehicle 20, values related to the internal mechanical structure of the vehicle 20, such as the gear ratio of the transmission, are required. Values regarding the internal mechanical structure of the vehicle 20 are not disclosed to the public. Therefore, it is difficult to accurately determine the kinetic energy E according to the calculation formula for the kinetic energy E. However, it is known that the weight equivalent to the rotating part of the vehicle 20 changes according to (1/r ² ), where r is the dynamic load radius [m] of the tire 30.

In this embodiment, for the portion of the kinetic energy E related to the weight equivalent to the rotating part, a multiple regression model is adopted in which the explanatory variable is the dynamic load radius (first term). Further, for other parts of the kinetic energy E, a term (second term) is used in which the explanatory variable is at least one of gradient resistance and rolling resistance. The coefficient of the first term is the first coefficient, and the coefficient of the second term is the second coefficient. Here, for the portion corresponding to the second term, measured values can be obtained as described above, but generally the measured values may include errors. For example, in the case of a gradient resistance, the gradient angle may include an error. Additionally, rolling resistance is sometimes given as a theoretical value, which may differ from the actual value. Therefore, parts other than the weight equivalent to the rotating part are also analyzed as the second term of the multiple regression model, and correction is performed using the multiple regression coefficient. The multiple regression equation used in this embodiment is shown as follows, for example. y is the energy consumption amount of the vehicle 20. a1 and a2 are the first coefficient and the second coefficient, respectively.

FIG. 3 is a flowchart illustrating an example of an energy consumption estimation method executed by the energy consumption estimation device 10 according to the present embodiment. The model generation unit 131 executes a model generation step (step S1) to generate a model for estimating the fuel consumption of the vehicle 20. Furthermore, the energy consumption estimation unit 132 executes an energy consumption estimation step (step S2) of estimating the fuel consumption of the vehicle 20 using the generated model. The energy consumption estimation step may be performed continuously with the model generation step, or may be performed after a certain period of time has elapsed after the model generation step.

FIG. 4 is a flowchart showing details of the model generation process in FIG. 3. The model generation unit 131 acquires performance data accumulated in the database of the operation management device 60 (step S11). The model generation unit 131 classifies the acquired performance data (step S12). In the present embodiment, the model generation unit 131 classifies performance data according to at least the type of vehicle 20. For example, if the vehicle models are the same (if the same vehicle model buses are used), step S12 may be omitted. Then, the model generation unit 131 generates a model (step S13). In this embodiment, the model generation unit 131 generates a model by determining coefficients of a multiple regression model. As mentioned above, in this embodiment, a first factor is associated with the dynamic load radius. As a result, it is possible to accurately estimate the acceleration resistance related to the weight equivalent to the rotating part, which is difficult to obtain from a calculation formula, based on the dynamic load radius, and to estimate the energy consumption of the vehicle 20 with high precision. becomes possible. Also, a second coefficient is associated with at least one of grade resistance and rolling resistance. This makes it possible to correct unrelated resistance and errors in measurement values, making it possible to accurately estimate kinetic energy and more accurately estimating the energy consumption of the vehicle 20. Become. Furthermore, based on the determined coefficients, the contribution rate of each explanatory variable to the amount of fuel consumption can be determined. For example, the second term above includes various types of resistance, but in the process of determining the second coefficient, the contribution rate to the kinetic energy of each of the rolling resistance, air resistance, gradient resistance, etc. associated with the second coefficient is calculated. Calculated. Therefore, it is possible to understand the relationship between a specific resistance and energy consumption based on each contribution rate.

FIG. 5 is a flowchart showing details of the energy consumption estimation process in FIG. 3. Upon receiving an instruction to perform estimation from outside the energy consumption estimation device 10, the energy consumption estimation unit 132 obtains input data (step S21) and obtains a model from the storage unit 12 (step S22). The input data is data indicating the estimation conditions as described above. The energy consumption estimation unit 132 may select a model to be used for estimation based on input data when a plurality of models are generated. The energy consumption estimating unit 132 then estimates the fuel consumption using the model and outputs the estimation result (step S23). The energy consumption estimating unit 132 can estimate the energy consumption of the vehicle 20 with high accuracy using a model that takes into account the kinetic energy related to the weight equivalent to the rotating part as described above and also corrects errors in measurement values.

Here, the energy consumption estimation unit 132 can estimate the energy consumption of the vehicle 20, which is consumed by at least one of rolling resistance, air resistance, slope resistance, acceleration resistance, and kinetic energy. As described above, since the relationship between a specific resistance and energy consumption is grasped in the process of determining the first coefficient and the second coefficient, the energy consumption estimation unit 132 calculates the 20 energy consumption can be estimated. In other words, energy consumption due to some factors can be estimated with high accuracy.

For example, when the input data requests estimation of fuel consumption when the type A tires 30 installed on the vehicle 20 are changed to type B tires 30, the energy consumption estimation unit 132 , the estimation can be made as follows. After estimating the energy consumption of the vehicle 20 equipped with the type A tire 30, the energy consumption estimating unit 132 calculates the difference (eg, ratio) in rolling resistance between the type A and type B tires 30. The energy consumption estimation unit 132 specifies the contribution rate of rolling resistance to the total kinetic energy based on, for example, the second coefficient. The energy consumption estimating unit 132 then adjusts the estimated energy consumption based on the difference in rolling resistance and the contribution rate. The estimation results may be transmitted over the network 40 and displayed on a display that can be viewed by the person requesting the fuel consumption estimation (eg, a bus operator). For example, the results of a plurality of fuel consumption calculations for tires 30 of type A and tires 30 of type B may be displayed for comparison. Here, the conditions specified by the input data are not limited to the type of tire 30, and may be, for example, the type of vehicle 20. For example, when the types of vehicles 20 are different, the energy consumption estimation unit 132 may adjust the energy consumption by calculating the difference in air resistance and acceleration resistance. Further, when estimation of a change in fuel consumption on a slope is requested, the energy consumption estimation unit 132 may calculate the difference in slope resistance and adjust the energy consumption.

As described above, the energy consumption estimation device 10, model generation device, program, and model generation method according to the present embodiment can estimate the energy consumption of the vehicle 20 with high accuracy through the above configuration and steps. enable.

Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art can easily make various changes or modifications based on the present disclosure. It should therefore be noted that these variations or modifications are included within the scope of this disclosure. For example, the functions included in each component or each step can be rearranged to avoid logical contradictions, and multiple components or steps can be combined or divided into one. It is. Embodiments according to the present disclosure can also be realized as a storage medium recording a program executed by a processor included in the device. It is to be understood that these are also encompassed within the scope of the present disclosure.

For example, the configurations of the energy consumption estimation device 10 and the energy consumption estimation system shown in FIGS. 1 and 2 are merely examples, and are not limited to the configurations shown in FIGS. 1 and 2. For example, the energy consumption estimation system may have a configuration in which the energy consumption estimation device 10 and the operation management device 60 are integrated. In this case, the energy consumption estimation device 10 may also perform the function of the operation management device 60, and the energy consumption estimation device 10 may function alone as an energy consumption estimation system. Further, for example, when the energy consumption estimation system includes the traffic management device 60, the traffic management device 60 may be composed of not one computer but a plurality of computers. For example, the traffic management device 60 is configured of a plurality of computers that can communicate with each other, including at least one computer that manages location information of the vehicle 20 and another computer that manages information on the number of people getting on and off the vehicle 20. It's fine.

The model generation unit 131 and the energy consumption estimation unit 132 may be included in different computers. For example, the model generation unit 131 may be included in another computer that can communicate with the energy consumption estimation device 10 and also have access to the storage unit 12. In this case, the energy consumption estimation device 10 may estimate the energy consumption of the vehicle 20 using a model generated by another computer and stored in the storage unit 12. The model may be generated in the same manner as above. That is, another computer may acquire performance data of the vehicle 20 and generate a model using the performance data.

Also, in the above embodiments, a multiple regression model was used, but a machine learning model may be used. That is, the model may be a machine learning model that includes the energy consumption amount of the vehicle 20 as an objective variable, and includes travel information of the vehicle 20, load information of the vehicle 20, and tire information of the vehicle 20 as explanatory variables. The tire information of the vehicle 20 has at least a dynamic load radius. The machine learning method may be, but is not limited to, a neural network, for example. In this case, the model generation unit 131 acquires the track record data of the vehicle 20, performs machine learning using the track record data, and generates a model. For example, a model capable of estimating the energy consumption of the vehicle 20 with high accuracy is generated by performing machine learning using appropriate performance data, such as through classification processing by the model generation unit 131. In addition, highly accurate estimation becomes possible using the model. Here, even when a machine learning model is used, the model generation unit 131 and the energy consumption estimation unit 132 may be included in different computers. That is, another computer may generate a machine learning model, and the energy consumption estimation device 10 may estimate the energy consumption of the vehicle 20 using the model generated by the other computer and stored in the storage unit 12. .

Furthermore, although the detection device 70 and the communication device 80 have been described as being devices mounted on the vehicle 20, they are not limited to such a configuration. For example, the detection device 70 and the communication device 80 may be realized by an application on a mobile terminal (for example, a smartphone) owned by a passenger of the vehicle 20. An application on the mobile terminal may control various sensors and communication modules included in the mobile terminal to output vehicle operation information so as to be stored in a database as performance data. Further, part of the vehicle operation information may be detected and output by the mobile terminal, and the remaining vehicle operation information may be detected and output by the detection device 70 and communication device 80 that are different from the mobile terminal. In this case, the application of the mobile terminal may detect and output the acceleration, speed, position information of the vehicle 20, the gradient of the place where the vehicle is traveling, the distance traveled, the number of people getting on and off the vehicle, and the like. Here, even when the detection device 70 and the communication device 80 are realized by an application of a mobile terminal owned by a passenger of the vehicle 20, the energy consumption estimation device 10 and the operation management device 60 are configured to be integrated. It's fine. That is, the energy consumption estimation device 10 may be configured to receive vehicle operation information from a mobile terminal and manage it in a database.

10 Energy consumption estimation device 11 Communication unit 12 Storage unit 13 Control unit 20 Vehicle 30 Tire 40 Network 60 Operation management device 70 Detection device 80 Communication device 131 Model generation unit 132 Energy consumption estimation unit

Claims

A vehicle energy consumption estimation device,
Equipped with a model generation unit that generates a model for obtaining the objective variable from the explanatory variables,
The model is a multiple regression model that includes the energy consumption amount of the vehicle as the objective variable, and includes travel information of the vehicle, load information of the vehicle, and tire information of the vehicle as the explanatory variables,
The model generation unit is an energy consumption estimation device that calculates a first coefficient associated with tire information of the vehicle using performance data of the vehicle indicating the target variable for the explanatory variable.
The energy consumption estimation device according to claim 1, wherein the first coefficient is associated with a dynamic load radius.
The energy consumption estimation device according to claim 1 or 2, wherein the model generation unit calculates a second coefficient associated with at least one of slope resistance and rolling resistance.
The energy consumption estimation device according to any one of claims 1 to 3, wherein the model further includes vehicle information of the vehicle as the explanatory variable.
The energy consumption estimation device according to any one of claims 1 to 4, further comprising an energy consumption estimation unit that estimates the energy consumption of the vehicle using the generated model.
The energy consumption amount according to claim 5, wherein the energy consumption estimator estimates the energy consumption amount of the vehicle consumed by at least one of rolling resistance, air resistance, slope resistance, acceleration resistance, and kinetic energy. Estimation device.
A model generation device that generates a model for estimating energy consumption of a vehicle, comprising:
Equipped with a model generation unit that generates a model for obtaining the objective variable from the explanatory variables,
The model includes machine learning, including energy consumption of the vehicle as the objective variable, travel information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as the explanatory variables. is a model,
The model generation unit is a model generation device that generates the model by machine learning using performance data of the vehicle indicating the objective variable for the explanatory variable.
A program that causes a computer to generate a model for estimating energy consumption of a vehicle, the program comprising:
The model is a machine learning model that includes energy consumption of the vehicle as an objective variable, and includes traveling information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as explanatory variables. can be,
to the computer;
obtaining performance data of the vehicle indicating the objective variable for the explanatory variable;
A program that executes machine learning using the performance data.
A method of generating a model for estimating energy consumption of a vehicle, the method being executed by a computer, comprising:
The model is a machine learning model that includes energy consumption of the vehicle as an objective variable, and includes traveling information of the vehicle, load information of the vehicle, and tire information of the vehicle having at least a dynamic load radius as explanatory variables. can be,
obtaining performance data of the vehicle indicating the objective variable for the explanatory variable;
A method for generating a model, the method comprising: performing machine learning using the performance data.