JP2016170819A - System for determining necessity of charging, on-vehicle device, and electric automobile - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system for determining necessity of charging for comprehensively determining necessity of charging so that an electric vehicle can be charged at proper timing to improve charging service.SOLUTION: A system for determining necessity of charging includes: a database 11 that stores a sample data group formed of a plurality of sample data of past cases where electric vehicles have not been charged within predetermined time, in a plurality of stratification sample data groups according to selection conditions of the past cases; a communication device 12 that collects target sample data of a target case of the electric vehicle which is a determination target for determining whether the charging is necessary and collects data on selection conditions of the target case; and a charging alarm calculation system 13 that compares a reference value of Mahalanobis distance for prescribing a range of a unit space constituted of a stratification sample data group corresponding to the selection condition of the target case among the sample data group, with a value of the Mahalanobis distance of the target sample data, to determine that the charging is required when the value of the Mahalanobis distance of the target sample data exceeds the reference value.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a charging necessity determination system, an in-vehicle device, and an electric vehicle, and more specifically, a charging necessity determination system, an in-vehicle device, and an electric device that determine whether charging is necessary so that the electric vehicle can be charged at an appropriate timing. It relates to automobiles.

An electric vehicle that travels by driving a motor with electric power supplied from a power storage device (battery) is a charge level of the power storage device (also referred to as “state of charge” or “SOC (state of charge)”).
When the battery voltage drops, it is necessary to charge in a charging facility such as a charging station. An example of a technique for improving the charging service of such an electric vehicle is described in Patent Document 1 below. According to the technology described in Patent Document 1, information on reachable charging station candidates is provided based on the current position of the host vehicle and the remaining battery level, and for the purpose of reducing waiting time at the charging station, Based on the position of other vehicles around the charging station candidate and the remaining battery level, the congestion status of the charging station candidate at the time when the host vehicle arrives at the charging station candidate is notified.

JP 2011-13893 A

Generally, whether or not a power storage device needs to be charged is determined based on the SOC of the power storage device. If the SOC decreases, not only will the number of charging stations that can be reached be reduced, but in the worst case, the SOC may bottom out before the electric vehicle reaches the charging station, making it difficult to run the electric vehicle. Can do. On the other hand, if charging is repeated frequently before the SOC decreases, the power storage device may deteriorate quickly, and the life of the power storage device may be shortened. For this reason, it is desired that the power storage device is charged at an appropriate charging timing.

  By the way, for example, even if the SOC values of the power storage devices of two electric vehicles are the same, the electric vehicle that is scheduled to travel a long distance from now on is charged more than the electric vehicle that is not scheduled to travel. The necessity is considered high. Further, for example, even when traveling to a certain destination, it is considered that charging is more necessary when a traffic jam on the way is expected than when a traffic jam is not expected. As described above, the degree of necessity of charging is influenced not only by the SOC but also by various other factors. Therefore, an appropriate charging timing should be comprehensively determined in consideration of not only the SOC but also various other factors.

  Therefore, the present invention comprehensively determines whether or not charging is necessary so that the electric vehicle can be charged at an appropriate timing, and provides a charging necessity determination system, an in-vehicle device, and an electric vehicle that can improve the charging service. It is intended to provide.

  In order to achieve the above object, the present invention provides a charging necessity determination system for determining whether or not a power storage device mounted on an electric vehicle needs to be charged, wherein the electric vehicle has not been charged within a predetermined time. Storage means for storing sample data groups consisting of sample data of cases into a plurality of stratified sample data groups according to the selection conditions of past cases, and target samples of target cases of electric vehicles that are targets for determination of whether charging is required Defines the unit space configured by the collection means that collects data and collects data on the selection conditions of the target case and the stratified sample data group corresponding to the selection conditions of the target case among the sample data group Compare the reference value of the Mahalanobis distance to the target sample data and the value of the Mahalanobis distance of the target sample data. If the value of the Mahalanobis distance of the target sample data exceeds the reference value, it is determined that charging is required. A constant means, determining means when it is determined that the main charge, to the in-vehicle device mounted in an electric vehicle of the subject case, is characterized by and a notification means for notifying the main charge.

  According to the present invention configured as described above, sample data of past cases that have not been charged within a predetermined time, that is, past cases that have not been charged, is accumulated in the storage means. Then, depending on whether or not the target sample data of the electric vehicle that is the determination target of charging is included in the unit space configured by the sample data of the past case that was not required to be charged, the electric vehicle that is the determination target It is determined whether or not it is necessary to charge the power storage device. In the present invention, the Mahalanobis distance used for multivariate analysis in statistics is used to determine whether the target sample data is included in the unit space. The Mahalanobis distance represents a distance from the center of the distribution of the sample data group in consideration of the correlation of the parameters of the sample data. Since the Mahalanobis distance is based on the phase between the multiple variables, it is suitable for comprehensively determining whether charging is necessary in consideration of not only the SOC but also parameters of various other elements.

  By the way, the past cases where charging was not necessary include cases under various conditions. An example of the condition is the number of passengers in the electric vehicle. There may be a difference in driving tendency between the case where only one driver is on the electric vehicle and the case where there is a passenger. As a result, the range of the unit space may vary depending on the number of passengers. Therefore, rather than comparing the unit space composed of the sample data groups of all past cases with the target sample data, the unit space composed only of the sample data groups of the past cases having the same conditions as the target case Compared with the target sample data, the necessity of charging can be more accurately determined.

  Therefore, in the present invention, the sample data group including the sample data of the past cases is stratified into a plurality of stratified sample data groups according to the selection conditions of the past cases and stored in the storage unit. Here, the selection condition is, for example, the number of passengers described above. Then, whether or not charging is necessary is determined based on whether or not the target sample data is included in a unit space constituted by the stratified sample data group corresponding to the selection condition of the target case. Here, the stratified sample data group is, for example, a stratified sample data group when the number of passengers is one and a stratified sample data group when the number of passengers is two or more. As a result, the determination accuracy is improved, the necessity of charging is comprehensively determined so that the electric vehicle can be charged at an appropriate timing, and the charging service is improved.

  Preferably, in the present invention, each sample data of the past case and the target sample data of the target case include a plurality of parameters of the same item, and the parameters of the same item are stored in the electric storage device mounted on the electric vehicle. In addition to the charge level information, it includes electric vehicle travel plan information, road congestion information, and charging facility information.

  Thereby, not only the charge level of the electric vehicle but also various other factors can be taken into consideration to determine whether or not charging is necessary. As a result, the electric vehicle can be charged at an appropriate timing, and the charging service can be improved.

  In the present invention, it is preferable that the selection condition includes at least one of a driving tendency of a driver of the electric vehicle, a vehicle type of the electric vehicle, and the number of passengers of the electric vehicle.

  Not only the driving tendency of the driver but also the selection tendency such as the vehicle type and the number of passengers may cause a difference in driving tendency. For this reason, the necessity of charging is determined by comparing the target sample data with the unit space composed of the stratified sample data group stratified according to one or a combination of two or more of these selection conditions. Can be accurately determined.

  According to the charging necessity determination system, the in-vehicle device, and the electric vehicle of the present invention, it is possible to comprehensively determine the necessity of charging so that the electric vehicle can be charged at an appropriate timing, and to improve the charging service.

It is a block diagram which shows the structure of the whole system containing the charge necessity determination system by embodiment of this invention. It is a schematic diagram which shows distribution of a sample data group and object sample data. It is a schematic diagram which shows the general distance of the object sample data with respect to a sample data group. It is a graph which shows the general distance of normal data and abnormal data. It is a schematic diagram which shows the Maharabinos distance of the object sample data with respect to a sample data group. It is a graph which shows the Mahalanobis distance of normal data and abnormal data. It is a schematic diagram which shows the unit space for every stratified sample data group. It is a flowchart which shows the process of the charge necessity determination system by embodiment of this invention.

Hereinafter, an embodiment of a charging necessity determination system according to the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram showing a configuration of the entire system including a charging necessity determination system according to an embodiment of the present invention. FIG. 1 shows a host controller 10 as a charging necessity determination system, an electric vehicle (EV) 20 as a charging necessity determination target, and a charging station 30 as an overall system.

  The host controller 10 is a charging necessity determination system that determines whether or not a power storage device mounted on the electric vehicle 20 is charged, and includes a database (DB) 11 as a storage unit, and a communication device that also serves as a collection unit and a notification unit 12 and a charging alarm calculation system 13 as determination means.

  First, the database 11 will be described. The database 11 stores a sample data group from sample data of a plurality of past cases in which the electric vehicle has not been charged within a predetermined time. The past case in which charging was not performed within a predetermined time is a case in which charging was not required when determining whether charging was necessary. Any suitable value can be set for the predetermined time. When the predetermined time is set to a time within the range of 48 hours, for example, only the past cases that did not need to be charged within 48 hours from the determination of the necessity of charging are included in the sample data group. On the other hand, for example, a case that must be charged the next day is a case that should be determined to be charged at the time of determination, and is not included in the sample data group.

  The sample data includes parameters of electric vehicle travel plan information, road congestion information, and charging facility information, in addition to the charge level information of the power storage device mounted on the electric vehicle. Each parameter is appropriately quantified in order to calculate the correlation between parameters.

The travel plan information may include a departure point, a destination point, and a departure date and time as the contents of the travel schedule of past cases. The destination point is input to the in-vehicle device 21 including the car navigation system of the electric vehicle 20. Further, the position of the electric vehicle 20 at the time of inputting the destination may be set as the departure point. The position of the electric vehicle 20 is determined by GPS (global positioning syste
m: Global positioning system). Furthermore, the date and time when the destination is input to the car navigation system may be set as the departure date and time. The travel schedule may greatly affect the necessity of charging at the time when the destination is input, for example.

  The road congestion information may include calendar information and weather information that can affect the road congestion in addition to traffic information. The traffic information may include prediction information on future road congestion from that time in addition to information on road congestion at that time.

  The calendar information includes information on seasons and days of the week that affect road congestion. For example, during a tourist season or a holiday, there is a general tendency for the degree of traffic congestion on the road to increase.

  In the weather information, in the case of bad weather, generally, the degree of traffic congestion on the road tends to increase. Further, the weather information may include not only the current weather information but also a weather forecast for the next day, for example, about two weeks ahead. For example, when bad weather is expected next weekend, the degree of congestion on the road in this weekend's resort area tends to increase. Thus, the weather information may include not only the current day but also forecasts for the following day.

  The charging facility information may include information indicating the charging level of the charging facility and the degree of deterioration of the battery of the charging facility in addition to the position of the charging facility. For example, when there is no charging facility near the destination, the necessity of charging at the time of departure tends to be higher than when there is a charging facility near the destination. In addition, even if there is a charging facility near the destination, even if the in-vehicle battery is deteriorating and it is difficult to achieve a sufficient amount of charge, the need for charging at the time of departure tends to increase. is there. Data on the degree of deterioration of the battery is transmitted from the in-vehicle device 21 to the host controller 10.

  Thus, the sample data of each past case includes the parameters of the items of the storage device charging level information, the electric vehicle travel plan information, the road congestion information, and the charging facility information. These parameters are correlated with each other.

  Further, in the database 11, the sample data group is stored in a plurality of stratified sample data groups according to the selection conditions of the past cases. Here, the selection condition may be one of the driving tendency of the driver of the electric vehicle, the vehicle type of the electric vehicle, and the number of passengers of the electric vehicle, or a combination of two or more. Depending on such selection conditions, the driving tendency may differ.

  The selection condition for the driving tendency of the driver is classified as, for example, whether there is a tendency to go directly to the destination set in the car navigation system or a tendency to take a detour without going directly to the set destination. Good. The selection conditions for the vehicle type may be classified into, for example, sports cars and other vehicle types. The tendency of speed varies depending on the vehicle type. In addition, the selection conditions for the number of passengers may be classified into, for example, a case of one person and a case of two or more persons. Note that the selection conditions may be classified into three or more per item.

  The stratified sample data groups stratified according to the selection conditions may be stored as separate files in separate storage areas, or all common sample data groups are stored with a flag for each selection condition. May be.

  When a flag is given, a separate flag may be given for each classification of each item of the selection condition. For example, in the item of the driving tendency of the driver, a flag of “0” is given when the driver does not take a detour, and a flag of “1” is given when the driver tends to take a detour, and “0” when the vehicle type is a sports car. ", It may be given a flag of" 1 "in cases other than sports cars. In addition, a flag of “0” may be given when the number of passengers is one, and “1” may be given when there are two or more passengers.

  The selection condition is, for example, that the driving tendency of the driver is “take a detour” “1”, the vehicle type is “sport car” “0”, and the number of passengers is “two or more” “1”. In this case, from the sample data group, the sample data in which the flag “1” is assigned to the driving tendency, the flag “0” is assigned to the vehicle type, and the flag “1” is assigned to the number of passengers. It is good to select and stratify and to configure the stratified sample data group. When the driving condition of the driver is “take a detour” or “1”, all the sample data with the driving tendency flag “1” is selected.

Next, the communication device 12 of the host controller 10 will be described.
First, the communication device 12 collects target sample data of a target case of the electric vehicle 20 that is a determination target of whether or not charging is necessary as a collection unit. The target sample data also includes the above-described parameters of the same items as the sample data of each past case.

  The communication device 12 also collects data on selection conditions for the target case. Here, the selection condition may be one of the driving tendency of the electric vehicle driver, the vehicle type of the electric vehicle, and the number of passengers of the electric vehicle, as in the selection condition in the database described above, or two or more. A combination may be used.

First, a case where one item of the selection condition is the driving tendency of the driver will be described. The driving tendency of the driver of the electric vehicle 20 to be determined whether charging is necessary may be set for each electric vehicle, or may be set for each driver who actually drives the electric vehicle.
When the driving tendency is set for each electric vehicle, the driving tendency data is registered in the memory of the in-vehicle device 21 of the electric vehicle 20, and the driving tendency data is stored in the controller 10 together with the identification signal of the electric vehicle. To send to. The driving tendency data may correspond to a flag in the database, for example, by sending a signal of “0” in the case of a driving tendency that does not take a detour, and “1” in the case of a driving tendency that takes a detour.

  In addition, when the driving tendency is set for each driver who actually drives the electric vehicle, the driver's identification data is registered in the driver's ignition key (or card) so that the driver can be identified. The driver may be identified, or the registered driver may be identified by image recognition of the driver's head. Furthermore, the driving tendency may be set for each driver in the memory of the in-vehicle device 21 of the electric vehicle 20. The electric vehicle 20 may transmit the driving tendency data corresponding to the identified driver to the controller 10 together with the identification signal of the electric vehicle.

  Next, a case where one item of the selection condition is a vehicle type will be described. The electric vehicle 20 may register the vehicle type data of the own vehicle in the memory of the in-vehicle device 21 and transmit the vehicle type data to the controller 10 together with the identification signal of the electric vehicle. The vehicle type data may correspond to a flag in the database, for example, by sending a signal of “0” when the vehicle type is a sports car and “1” when the vehicle type is other than a sports car.

  Next, a case where one item of selection conditions is the number of passengers will be described. The number of passengers of the electric vehicle 20 to be determined whether charging is necessary may be determined by providing a seating sensor on each seat in the passenger compartment and detecting the presence or absence of seating for each seat, or imaging the passenger compartment with a camera. The determination may be made by recognizing the captured image. Then, the electric vehicle 20 may transmit data of the determined number of passengers to the controller 10 together with the identification signal of the electric vehicle. As for the data on the number of passengers, for example, a signal of “0” may be transmitted when the number of passengers is one, and “1” when the number of passengers is two or more.

  As described above, the communication device 12 collects the target sample data of the target case of the electric vehicle, which is a target for determining whether charging is necessary, and the data of the selection conditions of the target case as the collecting means.

  The communication device 12 also transmits a charging alarm signal as a notification means in order to notify the charging of the target case to the electric vehicle of the target case when it is determined that charging is required as will be described later. .

Next, the charging alarm calculation system 13 of the host controller 10 will be described.
The charging alarm calculation system 13 includes a Mahalanobis distance reference value that defines a unit space range constituted by a stratified sample data group corresponding to a selection condition of a target case, and a Mahalanobis distance of the target sample data. Compare the value of. If each sample data of the entire sample data group has a flag indicating the selection condition, the sample data to which the selection condition flag of the target case is given is selected from the entire sample data group and Another sample data group is formed. Then, the filling alarm calculation system 13 determines that charging is required when the value of the Mahalanobis distance of the target sample data of the target case exceeds the reference value.

Here, the Mahalanobis distance will be described with reference to FIGS.
First, FIG. 2 schematically shows the distribution of the sample data group and the target sample data. FIG. 2 shows, for convenience, two parameters among a large number of parameters of the sample data, that is, the correlation between the weather condition and the traffic condition condition. The weather parameters are digitized based on an appropriate index, with “sunny” being good and “rain” being bad. In addition, the traffic condition parameters are also digitized based on an appropriate index, with “congested” being bad and “no traffic” being good.

In FIG. 2, sample data of past cases is plotted with black circles. The sample data of the past cases are cases in which charging is not performed within a predetermined time, and is referred to as normal data for convenience. On the other hand, in FIG. 2, sample data of two cases that are charged within a predetermined time are plotted as abnormal data A and B by white triangle marks, respectively.
The sample data group indicated by black circles plotted in FIG. 2 has a distribution extending from the lower left to the upper right, indicating that there is a positive correlation between the weather and traffic conditions. In addition, the actual distribution of the sample data group depends on the correlation among parameters of a larger number of items.

  Here, it is considered to distinguish between normal data indicated by black circles in FIG. 2 and abnormal data indicated by white triangles. First, FIG. 3 shows a general distance L1 from the center O of the distribution of the sample data group of normal data plotted with black circles. Normal data D1 and abnormal data A and B are located on the circumference of the radial distance L1. That is, as shown in FIG. 4, the general distance from the center O of the normal data D1 and the abnormal data A and B is all L1. For this reason, it is difficult to distinguish normal data D1 and abnormal data A and B by a general distance.

  Next, FIG. 5 shows the Mahalanobis distance from the center O of the distribution of the sample data group of normal data plotted with black circles. The Mahalanobis distance is a distance from the center of the distribution of the sample data group in consideration of the correlation with the parameter. FIG. 5 shows ellipses of Mahalanobis distances M1, M2 and M3 from the center O of the distribution of the sample data group of black circles, taking into account the correlation between the parameters of the weather conditions and the parameters of the traffic conditions. . As shown in FIG. 5, the Mahalanobis distance of the normal data D1 is M2, which is located inside the elliptical unit space of the Mahalanobis distance M3. On the other hand, the abnormal data A and B are located outside the elliptical unit space having the Mahalanobis distance M3. That is, as shown in FIG. 6, the Mahalanobis distance of normal data D1 is smaller than M3, while the Mahalanobis distance of abnormal data A and B is larger than M3. Therefore, normal data and abnormal data can be distinguished using the Mahalanobis distance M3 as a threshold value.

Next, the unit space of the stratified sample data group will be described.
FIG. 7 schematically shows the distribution of the stratified sample data group. Also in FIG. 7, as in FIGS. 2, 3, and 5, among the many parameters of the sample data, two parameters are shown for convenience, that is, the correlation between the weather condition and the traffic condition condition. In FIG. 7, the first stratified sample data group constituting the first unit space S1 is plotted with black circles, and the second stratified sample data group constituting the second unit space S2 is plotted with white circles. Has been. The first stratified sample data group is stratified as a selection condition that the driving tendency of the driver is a driving tendency to “take a detour”, while the second stratified sample data group is The driving tendency of the driver is stratified as a selection condition that the driving tendency is “do not take a detour”. The reference value of the Mahalanobis distance that defines the outer edge of the first unit space S1 may be equal to or different from the reference value of the Mahalanobis distance that defines the outer edge of the second unit space S2.

  Both the first and second stratified sample data groups are made up of sample data of past cases that were not charged within a predetermined time. However, as shown in FIG. 7, the range of the first unit space S1 and the range of the second unit space S2 are greatly different. For this reason, the target sample data C indicated by a white triangle is located outside the first unit space S1 and at the same time inside the second unit space S2.

  Whether the target sample data C is included in the first unit space S <b> 1 when the selection condition of the target case of the target sample data C is, for example, a driving tendency of the driver “take a detour”. Whether or not charging is necessary is determined. In other words, whether or not charging is necessary is determined based on whether or not the Mahalanobis distance of the target sample data C exceeds the reference value of the Mahalanobis distance to the outer edge of the first unit space S1. On the other hand, when the selection condition of the target case of the target sample data C is, for example, the driving tendency of the driver “do not take a detour”, the target sample data C is in the second unit space S2. Whether or not charging is necessary is determined depending on whether or not it is included. In other words, whether or not charging is necessary is determined based on whether or not the Mahalanobis distance of the target sample data C exceeds the reference value of the Mahalanobis distance to the outer edge of the second unit space S2.

  Therefore, when the driving tendency of the driver in the selection condition of the target case is a driving tendency to “take a detour”, it is determined that charging is necessary because the target sample data C is located outside the first unit space S1. Is done. On the other hand, when the driving tendency of the driver in the selection condition of the target case is a driving tendency that “does not take a detour”, the target sample data C is located inside the second unit space S2, so that charging is performed. Judged as unnecessary.

  As described above, if the stratified sample data group to be compared with the target sample data C is different, the determination result of the necessity of charging may be different. Therefore, it is possible to more accurately determine the necessity of charging by comparing the target sample data with a unit space constituted only by sample data groups of past cases having the same selection conditions as the target case.

Next, with reference to the flowchart of FIG. 8, the process of the charge necessity determination system by embodiment of this invention is demonstrated.
In the present embodiment, the processing is started when the destination is input to the car navigation system of the electric vehicle 20 in which the necessity of charging is determined.

  When the destination is input, the selection conditions for the target case of the electric vehicle 20 to be determined are collected by the communication device 12 of the host controller 10 (S81). Here, the selection condition is a combination of the three items of the driving tendency of the driver of the electric vehicle, the vehicle type of the electric vehicle, and the number of passengers of the electric vehicle.

  Subsequently, the charging alarm calculation system 13 of the host controller 10 selects and reads a stratified sample data group corresponding to the selection condition of the target case from the sample data group (S82). If each sample data of the entire sample data group has a flag indicating the selection condition, the sample data to which the selection condition flag of the target case is given is selected from the entire sample data group and Another sample data group is formed.

  When the selection condition is, for example, that the driver's driving tendency is “take a detour”, the vehicle type is “sports car”, and the number of passengers is “two or more”, these selections are made from the database 11. Sample data corresponding to the combination of conditions is selected, and a stratified sample data group is configured. In addition, in the database 11, when the selection condition flag is assigned to each sample data, the sample data to which the combination flag corresponding to the combination of the selection conditions of the target case is assigned is selected.

  Subsequently, the communication device 12 of the host controller 10 collects target sample data of a target case of the electric vehicle 20 that is a determination target of whether or not charging is required (S83). The target sample data includes parameters having the same items as the past case sample data stored in the database 11. Here, in addition to the charge level information of the power storage device mounted on the electric vehicle 20, the travel plan information, road congestion information, and charging facility information of the electric vehicle 20 are included as parameters.

  The charge level information and travel plan information of the power storage device of the electric vehicle 20 are transmitted from the in-vehicle device 21 of the electric vehicle 20 to the communication device 12 of the host controller 10. Further, as road congestion information, traffic conditions, weather conditions, and calendar information are transmitted from the ground station (not shown) to the communication means 12 of the host controller 10, for example. The charging facility information is transmitted from the communication device 32 of the charging station 30 to the communication device 12 of the host controller 10.

Subsequently, the charging alarm calculation system 13 of the host controller 10 calculates the Mahalanobis distance of the target sample data of the target case with respect to the selected stratified sample data group based on the correlation of each parameter (S84). For the calculation of the Mahalanobis distance based on the correlation of parameters, a calculation method usually used in multivariate analysis in statistics can be used.
In calculating the Mahalanobis distance, it is desirable to remove specific sample data such as accidents.

  Subsequently, the charging alarm calculation system 13 compares the reference value of the Mahalanobis distance that defines the outer edge of the unit space of the selected stratified sample data group with the Mahalanobis distance of the target sample data of the target case (S85). That is, it is determined whether the target sample data is included in the unit space of the stratified sample data group.

  For example, when the unit space of the selected stratified sample data group is the first unit space S1 shown in FIG. 7 and the target sample data is the target sample data C shown in FIG. Is located outside the unit space S1, and therefore, it is determined that charging is necessary. On the other hand, when the unit space of the selected stratified sample data group is the second unit space S2 shown in FIG. 7, since the target sample data C is located inside the unit space S2, charging is performed. Judged as unnecessary.

  When the Mahalanobis distance of the target sample data exceeds the reference value, that is, when the target sample data is not included in the unit space of the stratified sample data group (“yes” in S86), the charging alarm The calculation system 13 further determines the necessity of charging based on the Mahalanobis distance of the target sample data (S87). Specifically, a second reference value is provided, and when the Mahalanobis distance of the target sample data exceeds the second reference value, the necessity of charging is determined as “high” and does not exceed It may be determined that the necessity of charging is “low”. Further, more reference values may be set to determine the necessity of charging stepwise, or the necessity may be increased continuously according to the Mahalanobis distance of the target sample data.

  Subsequently, a charging alarm signal corresponding to the necessity of charging is transmitted from the communication device 12 of the controller 10 toward the target electric vehicle 20 (S88). In the electric vehicle 20 that has received the charging alarm signal, the in-vehicle device 21 displays a charging alarm on, for example, an instrument panel to notify the driver that charging is necessary. It is desirable that the charging alarm display reflects the degree of necessity for charging.

  If the Mahalanobis distance of the sample data is equal to or less than the reference value, the charging alarm calculation system 13 determines that charging is not necessary (“no” in S86). In this case, the target sample data of the case that is the current determination target is added to the database 11 as the sample data of the past case (S89).

  In this way, whether or not charging is necessary depends on whether or not the target sample data of the determination target case is included in the unit space configured by the stratified sample data group corresponding to the selection condition of the target case. Is determined. As a result, the determination accuracy is improved, the necessity of charging is comprehensively determined so that the electric vehicle can be charged at an appropriate timing, and the charging service is improved.

  In the above-mentioned embodiment, although the example which comprised this invention on the specific conditions was demonstrated, this invention can perform a various change and combination, and is not limited to this. For example, in the above-described embodiment, an example in which a plurality of selection conditions are combined to select a stratified sample data group has been described. However, in the present invention, one selection condition may be used.

DESCRIPTION OF SYMBOLS 10 Host controller 11 Database 12 Communication apparatus 20 Electric vehicle 21 In-vehicle apparatus 30 Charging station 31 Battery management apparatus 32 Communication apparatus

Claims (14)

A charging necessity determination system for determining whether or not a power storage device mounted on an electric vehicle is necessary,
Storage means for storing a sample data group composed of sample data of a plurality of past cases that the electric vehicle did not charge within a predetermined time, and stratified into a plurality of stratified sample data groups according to the selection conditions of the past cases;
Collecting means for collecting target sample data of a target case of an electric vehicle that is a determination target of charging and collecting data of selection conditions of the target case;
Among the sample data groups, a Mahalanobis distance reference value defining a unit space range constituted by a stratified sample data group corresponding to the selection condition of the target case, and a Mahalanobis distance value of the target sample data In comparison, when the value of the Mahalanobis distance of the target sample data exceeds the reference value, a determination unit that determines that charging is required,
When the determination means determines that charging is required, a notification means for notifying the charging required for the in-vehicle device mounted on the electric vehicle of the target case;
A charging necessity determination system comprising: Each sample data of the past case and target sample data of the target case include a plurality of parameters of the same item,
2. The charging according to claim 1, wherein the parameters of the same item include travel plan information, road congestion information, and charging facility information of an electric vehicle in addition to charging level information of a power storage device mounted on the electric vehicle. Necessity judgment system. 3. The charging necessity determination system according to claim 1, wherein the selection condition includes at least one of a driving tendency of a driver of the electric vehicle, a vehicle type of the electric vehicle, and a number of passengers of the electric vehicle. An in-vehicle device mounted on an electric vehicle,
Register the charge level information of the electric vehicle and the travel plan information,
An in-vehicle device, which is transmitted to the charging necessity determination system according to claim 2. An in-vehicle device mounted on an electric vehicle,
Register the driving tendency data of the driver of the electric vehicle, the data of the vehicle type, and the data of the number of passengers,
An in-vehicle device, which is transmitted to the charging necessity determination system according to claim 3. An in-vehicle device mounted on an electric vehicle,
An in-vehicle device characterized by displaying a charging alarm on an instrument panel of an electric vehicle and notifying a driver of the required charging when a charging is notified from the charging necessity determination system according to any one of claims 1 to 3. .   The electric vehicle carrying the vehicle-mounted apparatus of Claims 4-6. A charging necessity determination system for determining whether or not a power storage device mounted on an electric vehicle is necessary,
Storage means for storing a sample data group composed of sample data of a plurality of past cases that the electric vehicle did not charge within a predetermined time, and stratified into a plurality of stratified sample data groups according to the selection conditions of the past cases;
Collecting means for collecting target sample data of a target case of an electric vehicle that is a determination target of charging and collecting data of selection conditions of the target case;
A reference value of a distance defining a unit space range constituted by a stratified sample data group corresponding to the selection condition of the target case among the sample data group is compared with a distance value of the target sample data. Determining means for determining that charging is required when the distance value of the target sample data exceeds the reference value;
When the determination means determines that charging is required, a notification means for notifying the charging required for the in-vehicle device mounted on the electric vehicle of the target case;
A charging necessity determination system comprising: Each sample data of the past case and target sample data of the target case include a plurality of parameters of the same item,
9. The charging according to claim 8, wherein the parameters of the same item include travel plan information, road congestion information, and charging facility information of an electric vehicle in addition to charging level information of a power storage device mounted on the electric vehicle. Necessity judgment system. 10. The charging necessity determination system according to claim 8, wherein the selection condition includes at least one of a driving tendency of a driver of the electric vehicle, a vehicle type of the electric vehicle, and a number of passengers of the electric vehicle. An in-vehicle device mounted on an electric vehicle,
Register the charge level information of the electric vehicle and the travel plan information,
It transmits to the charge necessity determination system of Claim 9. The vehicle-mounted apparatus characterized by the above-mentioned. An in-vehicle device mounted on an electric vehicle,
Register the driving tendency data of the driver of the electric vehicle, the data of the vehicle type, and the data of the number of passengers,
It transmits to the charge necessity determination system of Claim 10. The vehicle-mounted apparatus characterized by the above-mentioned. An in-vehicle device mounted on an electric vehicle,
An in-vehicle device characterized by displaying a charging alarm on an instrument panel of an electric vehicle and notifying a driver of the required charging when a charging is notified from the charging necessity determination system according to claims 8 to 10. .   An electric vehicle equipped with the in-vehicle device according to claim 11.

Images