Failure prediction system, failure prediction method, and failure prediction program

Abstract

An acquisition unit (111) acquires running data of an electric vehicle. A prediction unit (112) predicts an aging failure of a drive circuit of a motor for driving a drive wheel of the electric vehicle, on the basis of the running data of the electric vehicle. The running data includes position data of the electric vehicle and data relating to power consumption. The prediction unit (112) predicts the aging failure of the drive circuit on the basis of an increase in the amount of power consumption of the electric vehicle over time when the electric vehicle runs on the same route.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a failure prediction system, a failure prediction method, and a failure prediction program that predict an aging failure of a switching element included in an inverter of an electric vehicle. BACKGROUND

[0002] Business vehicles such as delivery vehicles are being popularized as electric vehicles (EVs). In recent years, the travel data (battery information, movement trajectory, vehicle control information, and the like) of EVs is saved on the cloud, and an environment that can be used in many ways is being constructed.

[0003] Methods for calculating the required energy according to the travel route and charging the required amount of charge in order to make the EV travel to the destination without running out of battery are disclosed in large quantities. For example, a method is proposed that notifies the optimal amount of charge required to travel on a route used on a daily basis when a destination is set to suppress excessive charging and thus suppress battery deterioration (for example, refer to Patent Literature 1). In addition, a method is proposed that presents an optimal route with the minimum energy cost to the destination based on past travel history records and determines whether it is possible to continue traveling within the range of the amount of charge (for example, refer to Patent Literature 2). In addition, a method is proposed that continues traveling to the destination by an inverter performing control according to conditions such as the total travel distance, weight, size, resistance coefficient, speed, acceleration, history record, air temperature, terrain, and the like (for example, refer to Patent Literature 3).

[0004] In an EV, an inverter is used in order to drive a motor. The power element (for example, MOSFET (Metal-Oxide Semiconductor Field-Effect Transmitter), IGBT (Insulated Gate Bipolar Transistor)) used by the inverter deteriorates over time. The main cause of the deterioration of the power element is an increase in the contact resistance of the bonding wire. The reason for this is metal fatigue caused by thermal cycles, and the increase in the contact resistance of the bonding wire is manifested as an increase in the loss (decrease in efficiency) of the power element.

[0005] PRIOR ART DOCUMENTS

[0006] PATENT LITERATURE

[0007] Patent Literature 1: Japanese Patent Application Publication No. 2012-19627

[0008] Patent Literature 2: Japanese Patent Application Publication No. 2009-63555

[0009] Patent Literature 3: Japanese Patent Application Publication No. 2018-27012 SUMMARY

[0010] Problem to be solved by the invention

[0011] In order to predict the deterioration with age of elements other than power elements such as electrolytic capacitors, coils, and fans, dedicated sensors are respectively required. Thus, in order to predict the deterioration with age of these elements mounted on EVs, design changes of adding dedicated sensors are required.

[0012] On the other hand, as for the deterioration with age of power elements, if the progress of an increase in loss can be predicted, the prediction can be performed without adding dedicated sensors.

[0013] The present disclosure was completed in view of this situation, and aims to provide a technology of predicting the deterioration with age of a drive circuit of an electric vehicle at low cost.

[0014] Solution to problem

[0015] In order to solve the above problem, a failure prediction system of an aspect of the present disclosure includes an acquisition unit that acquires travel data of an electric vehicle, and a prediction unit that predicts an aging failure of a drive circuit of a motor that drives a drive wheel of the electric vehicle, on the basis of the travel data of the electric vehicle. Position data of the electric vehicle and data related to power consumption are included in the travel data. The prediction unit predicts the aging failure of the drive circuit on the basis of a progress of an increase in power consumption amount when the electric vehicle travels on the same route.

[0016] Furthermore, an aspect of the present disclosure obtained by transforming any combination of the above-described elements, the description of the present disclosure between an apparatus, a system, a method, a computer program, a recording medium on which a computer program is recorded, and the like is also effective.

[0017] Effects of the invention

[0018] According to the present disclosure, it is possible to predict the deterioration with age of a drive circuit of an electric vehicle at low cost. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a diagram showing an outline structure of an electric vehicle to which an embodiment is applied.

[0020] Figure 2 is a diagram showing an outline structure of a drive system of an electric vehicle.

[0021] Figure 3 is a diagram showing a structure example of a failure prediction system to which an embodiment is applied.

[0022] Figure 4(a)-(i) is a graph obtained by dividing the log data of the GPS trajectory of a certain electric vehicle over a specified period into time segments and plotting the segmented GPS trajectories as multiple curves.

[0023] Figure 5 (a)-(b) are diagrams used to illustrate the method for determining object paths.

[0024] Figure 6 (a)-(g) are diagrams illustrating specific examples of vehicle speed patterns while traveling on an object path.

[0025] Figure 7 This is a graph illustrating an example of the shift in power consumption as an object travels along a path.

[0026] Figure 8 This is a flowchart illustrating the process of predicting aging faults of switching elements included in an inverter, performed by the fault prediction system according to the embodiment. Detailed Implementation

[0027] Figure 1 This is a diagram showing the general structure of the electric vehicle 3 according to the embodiment. In this embodiment, the electric vehicle 3 is assumed to be a pure EV without an internal combustion engine. Figure 1 The electric vehicle 3 shown is a rear-wheel drive (2WD) EV equipped with a pair of front wheels 31f, a pair of rear wheels 31r, and a motor 34 as a power source. The pair of front wheels 31f are connected via a front axle 32f, and the pair of rear wheels 31r are connected via a rear axle 32r. A transmission 33 transmits the rotation of the motor 34 to the rear axle 32r at a predetermined shift ratio. Alternatively, the electric vehicle 3 could also be a front-wheel drive (2WD) or 4WD vehicle.

[0028] The power system 40 includes a battery unit (battery section) 41 and a management unit 42. The battery unit 41 includes multiple battery cells. Lithium-ion battery cells, nickel-metal hydride battery cells, etc., can be used for the battery cells. In the following specification, we assume the use of a lithium-ion battery cell (nominal voltage: 3.6-3.7V). The management unit 42 monitors the voltage, temperature, current, SOC (State of Charge), and SOH (State of Health) of the multiple battery cells included in the battery unit 41 and transmits this information to the vehicle control unit 30 via an in-vehicle network. For example, a CAN (Controller Area Network) or LIN (Local Interconnect Network) can be used as the in-vehicle network.

[0029] The inverter 35 is a drive circuit that drives the motor 34, and converts direct-current electric power supplied from the battery unit 41 into alternating-current electric power and supplies it to the motor 34 at the time of motoring. At the time of regeneration, the inverter 35 converts alternating-current electric power supplied from the motor 34 into direct-current electric power and supplies it to the battery unit 41. At the time of motoring, the motor 34 rotates in accordance with the alternating-current electric power supplied from the inverter 35. At the time of regeneration, the motor 34 converts rotational energy generated due to deceleration into alternating-current electric power and supplies it to the inverter 35.

[0030] Figure 2 is a diagram showing an outline configuration of a drive system of the electric vehicle 3. In Figure 2 , an example is shown in which a three-phase alternating-current motor is used for driving the motor 34 of the electric vehicle 3, and the three-phase alternating-current motor 34 is driven by a three-phase inverter 35. The three-phase inverter 35 converts direct-current electric power supplied from the battery unit 41 into three-phase alternating-current electric power whose phases are offset by 120 degrees, respectively, to drive the three-phase alternating-current motor 34.

[0031] The inverter 35 includes a first arm in which a first switching element Q1 and a second switching element Q2 are connected in series, a second arm in which a third switching element Q3 and a fourth switching element Q4 are connected in series, and a third arm in which a fifth switching element Q5 and a sixth switching element Q6 are connected in series, and the first to third arms are connected in parallel to the battery unit 41.

[0032] In Figure 2 , IGBTs are used for the first to sixth switching elements Q1 to Q6. First to sixth diodes D1 to D6 are connected in anti-parallel to the first to sixth switching elements Q1 to Q6, respectively. Further, in the case where MOSFETs are used for the first to sixth switching elements Q1 to Q6, parasitic diodes formed in a direction from a source to a drain are used as the first to sixth diodes D1 to D6.

[0033] The motor controller 36 acquires an input direct-current voltage and an input direct-current current of the inverter 35 detected by an input voltage / current sensor 381, an output alternating-current voltage and an output alternating-current current of the inverter 35 detected by an output voltage / current sensor 382, and a rotational speed and a torque of the three-phase alternating-current motor 34 detected by a rotational speed / torque sensor 383. In addition, the motor controller 36 acquires an acceleration signal or a brake signal corresponding to an operation of a driver or generated by an automatic driving controller.

[0034] The motor controller 36 generates a PWM signal for driving the inverter 35 based on these input parameters, and outputs to the gate driver 37. The gate driver 37 generates a drive signal of the first to sixth switching elements Q1-Q6 based on the PWM signal input from the motor controller 36 and a prescribed carrier, and inputs to the gate terminals of the first to sixth switching elements Q1-Q6.

[0035] The motor controller 36 transmits the input direct-current voltage of the inverter 35, the input direct-current current of the inverter 35, the rotation speed of the motor 34, and the torque of the motor 34 to the vehicle control portion 30 via the in-vehicle network.

[0036] Returning to Figure 1 The vehicle control portion 30 is a vehicle ECU (Electronic Control Unit) that controls the entire electric vehicle 3, and can also be constituted by a comprehensive VCM (Vehicle Control Module), for example.

[0037] The GPS sensor 384 detects position information of the electric vehicle 3, and transmits the detected position information to the vehicle control portion 30. Specifically, the GPS sensor 384 receives radio waves including respective transmission times from a plurality of GPS satellites, and calculates the latitude / longitude of the reception site based on a plurality of transmission times included in the respective radio waves received.

[0038] The vehicle speed sensor 385 generates a pulse signal proportional to the rotation speed of the front axle 32f or the rear axle 32r, and transmits the generated pulse signal to the vehicle control portion 30. The vehicle control portion 30 detects the speed of the electric vehicle 3 based on the pulse signal received from the vehicle speed sensor 385.

[0039] The wireless communication portion 39 performs signal processing for wirelessly connecting with a network via the antenna 39a. As a wireless communication network with which the electric vehicle 3 can wirelessly connect, a mobile phone network (cellular network), wireless LAN, V2I (Vehicle-to-Infrastructure), V2V (Vehicle-to-Vehicle), ETC system (Electronic Toll Collection System), and DSRC (Dedicated Short Range Communications), for example, can be used.

[0040] During running of the electric vehicle 3, the vehicle control portion 30 can transmit running data to a data storage cloud server or a company server in real time using the wireless communication portion 39. The running data includes position data (latitude / longitude) of the electric vehicle 3, vehicle speed of the electric vehicle 3, voltage, current, temperature, SOC, SOH of the plurality of battery cells included in the battery unit 41, input direct current voltage of the inverter 35, input direct current of the inverter 35, rotation speed of the motor 34, torque of the motor 34. The vehicle control portion 30 samples these data periodically (for example, at intervals of 10 seconds), and transmits them to the cloud server or the company server each time.

[0041] Further, the vehicle control portion 30 can also store the running data of the electric vehicle 3 in an internal memory, and transmit the running data stored in the memory all at once at a prescribed timing. For example, the vehicle control portion 30 can also transmit the running data stored in the memory all at once to a terminal device of a business office after the business of the day is over. The terminal device of the business office transmits the running data of a plurality of electric vehicles 3 to the cloud server or the company server at a prescribed timing.

[0042] In addition, it can also be that, when charging is performed from a charger having a network communication function, the vehicle control portion 30 transmits the running data stored in the memory to the charger via the charging cable. The charger transmits the received running data to the cloud server or the company server. This example is effective for an electric vehicle 3 that does not have a wireless communication function mounted.

[0043] Figure 3 FIG. 1 is a diagram showing an example of the structure of a failure prediction system 10 according to the embodiment. The failure prediction system 10 is constructed by one or more servers. For example, the failure prediction system 10 can be constructed by one company server provided in a data center or a company facility. In addition, the failure prediction system 10 can be constructed by a cloud server used based on a cloud service. In addition, the failure prediction system 10 can be constructed by a plurality of company servers provided in a plurality of sites (data centers, company facilities) in a decentralized manner. In addition, the failure prediction system 10 can be constructed by a combination of a cloud server used based on a cloud service and a company server. In addition, the failure prediction system 10 can be constructed by a plurality of cloud servers based on contracts with a plurality of cloud service businesses.

[0044] The failure prediction system 10 has a processing section 11 and a recording section 12. The processing section 11 includes a travel data acquisition section 111, an object path decision section 112, and a failure prediction section 113. The functions of the processing section 11 can be realized by cooperation of hardware resources and software resources, or by hardware resources alone. As the hardware resources, a CPU, a ROM, a RAM, a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), other LSIs, or the like can be used. As the software resources, an operating system, programs such as applications, or the like can be used.

[0045] The recording section 12 includes a travel data holding section 121. The recording section 12 includes a nonvolatile recording medium such as an HDD (Hard Disk Drive), an SSD (Solid State Drive), or the like, and is used to record various data.

[0046] The travel data acquisition section 111 acquires travel data of the electric vehicle 3 via a network, and saves the acquired travel data in the travel data holding section 121. The object path decision section 112 reads out the travel data of the electric vehicle 3 from the travel data holding section 121, and extracts a movement trajectory of the electric vehicle 3 from the progress of the position data of the electric vehicle 3. The object path decision section 112 decides a frequently used path (hereinafter referred to as an object path) having a high frequency of use, on the basis of the extracted movement trajectory of the electric vehicle 3.

[0047] Figure 4 (a)-(i) of FIG. 10 are graphs in which log data of GPS trajectories for a prescribed period of a certain electric vehicle 3 is time-divided, and the divided plurality of GPS trajectories are plotted as a plurality of graphs. In (a)-(i) of FIG. 10, the GPS trajectories are simply marked in graphs in which the horizontal axis is set to the longitude and the vertical axis is set to the latitude, but the GPS trajectories can be plotted on an actual map. Figure 4

[0048] Figure 5 (a)-(b) of FIG. 11 are graphs for explaining a method of deciding an object path. In (a)-(b) of FIG. 11, the GPS trajectories of the electric vehicle 3 are plotted on an actual map. Figure 5 ​In the example of (a), the object path determining section 112 generates a two-dimensional kernel density distribution (two-dimensional frequency distribution) based on the GPS trajectory of the electric vehicle 3 for a prescribed period. When generating the two-dimensional kernel density distribution, it is desirable to plot only the position data when the vehicle speed is equal to or higher than a set value (for example, 10 km / h). When the position data when traveling at low speed or when parked is also plotted, the plotted amount of the position data when traveling at low speed or when parked increases, and sometimes a density distribution that deviates from the density distribution according to the actual situation is generated.

[0049] The object path determining section 112 selects a prescribed bandwidth using a prescribed kernel function (for example, a Gaussian function), and generates a kernel density curve of the plotted latitude and a kernel density curve of the plotted longitude. The object path determining section 112 determines the highest peak position and the second highest peak position from the kernel density curve of the plotted latitude. Similarly, the object path determining section 112 determines the highest peak position and the second highest peak position from the kernel density curve of the plotted longitude. The object path determining section 112 determines the path having the highest frequency of use among the paths connecting the intersection of the longitude of the highest peak position and the latitude of the highest peak position and the intersection of the longitude of the second highest peak position and the latitude of the second highest peak position as the object path. Figure 5 (b) shows the object path determined by the object path determining section 112 from the kernel density curves of the plotted latitude and the plotted longitude. Figure 5 The object path determined by the object path determining section 112 from the GPS trajectory for the prescribed period shown in (a).

[0050] Further, the method of determining the object path is not limited to the method of determining based on the two-dimensional kernel density distribution. For example, the object path determining section 112 can determine the path specified by the manager who is familiar with the road conditions of the travel region of the electric vehicle 3 as the object path. With respect to the object path, it is desirable to select a path in which the environmental conditions when the electric vehicle 3 travels are as fixed as possible. For example, it is desirable to select a path that is flat, has few curves, has a small number of traffic lights, has little congestion, and has a high frequency of use.

[0051] The failure prediction section 113 reads out a plurality of travel data of the electric vehicle 3 when traveling on the object path from the travel data holding section 121, and generates the progress of the increase in the power consumption amount when traveling on the object path. The failure prediction section 113 predicts the aging failure of the switching elements Q1-Q6 included in the inverter 35 based on the progress of the increase in the power consumption amount.

[0052] The deterioration with age of the switching elements Q1-Q6 can be estimated from the increase in the loss (decrease in efficiency) of the switching elements Q1-Q6. In order to predict the aging failure of the inverter 35 with high accuracy from the progress of the power consumption amount when traveling on the object path, it is desirable to exclude the influence of factors other than the increase in the loss of the inverter 35 as much as possible. First, it is desirable to use only data similar to the vehicle speed pattern when traveling on the object path as analysis target data.

[0053] Figure 6 Fig. 7 is a diagram showing specific examples of the vehicle speed pattern when traveling on the object path. Figure 6 Figs. 7(a) to 7(g) show seven vehicle speed patterns when traveling on the object path. The failure prediction portion 113 extracts a similar vehicle speed pattern from the vehicle speed pattern when traveling on the object path extracted from the travel data. As a method of extracting a similar vehicle speed pattern, various methods can be used. For example, a vehicle speed pattern in which the cumulative time of high-speed travel is extracted to be equal to or more than a prescribed time (for example, a vehicle speed pattern in which the period of traveling at 50 km / h or more is extracted to be equal to or more than half of the travel time), a vehicle speed pattern in which the number of stops is extracted to be equal to or less than a prescribed number, a vehicle speed pattern in which the timing of stopping or accelerating / decelerating is close, and the like are considered. In addition, a method such as pattern matching, a correlation coefficient, or the like can be used. In the examples shown in Figs. 7(a) to 7(g), the failure prediction portion 113 extracts a vehicle speed pattern in which the cumulative time of high-speed travel is equal to or more than a prescribed time, and a vehicle speed pattern in which the number of stops is equal to or less than a prescribed number. Figure 6 Figure 6 The vehicle speed patterns of Figs. 7(a), 7(d), and 7(e).

[0054] The failure prediction portion 113 can extract the SOC of the battery unit 41 at the starting point of the object path and the SOC of the battery unit 41 at the target point from the travel data when traveling on the object path, and calculate the power consumption amount when traveling on the object path based on the difference therebetween.

[0055] In addition, the failure prediction portion 113 can also extract the log of the input DC voltage V and the input DC current I of the inverter 35 from the starting point of the object path to the target point from the travel data when traveling on the object path, and calculate the power consumption amount when traveling on the object path by integrating the input power of the inverter 35 with the travel time of the object path as shown in the following (Formula 1).

[0056] Power consumption amount = ∫(V · I)dt / 1000 [kWh] (Formula 1)

[0057] If the distance of the object path is divided by the calculated power consumption amount, the fuel consumption can be calculated. With respect to the power consumption amount calculated as described above, in addition to the loss of the inverter 35 (mainly the loss generated by the contact resistance of the joint line of the switching elements Q1 to Q6), the influence of the mechanical drive loss generated in the transmission process of the rotational force of the motor 34 is also considered. In the drive loss, the loss generated by the drive friction of the drive shaft, the slip friction of the differential, the rubber deformation of the tire, the friction between the tire and the road surface, and the like are included.

[0058] ​The failure prediction unit 113 can extract logs of the rotation speed rpm of the motor 34 and the rotational torque Nm of the motor 34 from the start point to the target point of the object path from the travel data, and calculate the cumulative shaft output at the time of travel on the object path by integrating the shaft output of the motor 34 with respect to the travel time of the object path as in the following (Formula 2).

[0059] Cumulative shaft output = ∫(rpm · Nm · 2π / 60)dt / 1000 [kWh]... (Formula 2)

[0060] The failure prediction unit 113 subtracts the cumulative shaft output calculated by the above (Formula 2) from the power consumption calculated by the above (Formula 1), and thereby can calculate the power consumption from which the influence of the mechanical drive loss is removed.

[0061] For the power consumption calculated by the above (Formula 1), the influence of various fluctuation factors is considered in addition to the influence of the loss of the inverter 35 and the mechanical drive loss. For example, the influence of the weather, the air pressure of the tires, the load, the internal resistance of the battery, the vehicle speed (initial speed) at the start point of the object path, the elevation difference of the object path, the number of stops at the time of travel on the object path, and the like is considered. In a case where the elevation difference of the object path is significantly different between the outbound and the return, the power consumptions of the outbound and the return become different values. The number of stops at the time of travel on the object path affects the regeneration amount of the battery cell 41.

[0062] Figure 7 is a graph showing an example of the progress of the power consumption at the time of travel on the object path. The power consumption can be roughly explained by the sum of the power consumptions caused by the loss of the inverter 35 and the mechanical drive loss. In addition, the power consumption is also affected by other fluctuation factors. The failure prediction unit 113 plots a plurality of power consumptions at the time of travel on the object path. The failure prediction unit 113 calculates a first regression straight line L1 based on the plurality of power consumptions plotted. Similarly, the failure prediction unit 113 calculates a second regression straight line L2 based on a plurality of cumulative shaft outputs at the time of travel on the object path.

[0063] In a case where the other fluctuation factors are assumed to be fixed, the failure prediction unit 113 can estimate the difference between the first regression straight line L1 and the second regression straight line L2 as the loss of the inverter 35. The first regression straight line L1 and the second regression straight line L2 also extend in the future direction, and thus the failure prediction unit 113 can predict the future loss of the inverter 35.

[0064] The failure prediction unit 113 can predict the period when the statistically shortest life of the inverter 35 comes based on the predicted future loss of the inverter 35. As the preventive maintenance, the manager of the electric vehicle 3 can replace the inverter 35 before the statistically shortest life of the inverter 35 comes.

[0065] In addition, the failure prediction section 113 can also predict the timing of the statistical average life of the inverter 35 based on the predicted future deterioration of the inverter 35. As predictive maintenance, the manager of the electric vehicle 3 can replace the inverter 35 before the statistical average life of the inverter 35 arrives. In this case, it is possible to minimize the downtime and effectively utilize the inverter 35 in use.

[0066] The failure prediction section 113 can apply various corrections to the power consumption amount so as to make other fluctuating factors close to fixed. The processing section 11 of the failure prediction system 10 can also be provided with a weather information acquisition section (not shown). The weather information acquisition section acquires weather data for the date and time of travel on the subject route from a weather information database server on the network.

[0067] The failure prediction section 113 can also estimate the influence of wind on the power consumption amount of the electric vehicle 3 based on the wind direction, wind volume, and travel direction of the electric vehicle 3, and correct the power consumption amount in a manner that normalizes this influence. In addition, the failure prediction section 113 can also estimate the friction coefficient of the road surface based on the amount of rain, estimate the influence of the friction coefficient of the road surface on the power consumption amount of the electric vehicle 3, and correct the power consumption amount in a manner that normalizes this influence.

[0068] In addition, the failure prediction section 113 can also estimate the power consumption amount due to the use of the air conditioner based on the temperature, estimate the influence of the power consumption amount of the air conditioner on the power consumption amount of the electric vehicle 3, and correct the power consumption amount in a manner that normalizes this influence. Furthermore, in the case where the power consumption amount is calculated based on the input voltage / input current of the inverter 35 without regard to the SOC of the battery cell 41, the power consumption amount of the air conditioner does not need to be taken into account.

[0069] In the case where the log of the tire air pressure sensor is included in the travel data, the failure prediction section 113 can also estimate the influence of the tire air pressure on the power consumption amount of the electric vehicle 3, and correct the power consumption amount in a manner that normalizes this influence. In the case where the log of the load sensor is included in the travel data, the failure prediction section 113 can also estimate the influence of the load on the power consumption amount of the electric vehicle 3, and correct the power consumption amount in a manner that normalizes this influence.

[0070] Further, in a case where the log of the load sensor is not included, the failure prediction section 113 can also estimate the load according to the type of the delivery vehicle in which the electric vehicle 3 is used. For example, in a case where the electric vehicle 3 is a pickup-type delivery vehicle, the failure prediction section 113 uses a load model in which the load gradually becomes heavy from a pickup start time in the morning to a pickup end time in the evening. In a case where the electric vehicle 3 is a delivery-type delivery vehicle, the failure prediction section 113 uses a load model in which the load gradually becomes light from a delivery start time in the morning to a delivery end time in the evening. In a case where the electric vehicle 3 is a between-branch delivery-type delivery vehicle, the failure prediction section 113 estimates that the load variation is fixed. The failure prediction section 113 can estimate the type of the delivery vehicle according to the power consumption at different time periods.

[0071] The failure prediction section 113 can also estimate the influence of the initial speed of the object path on the amount of power consumption of the electric vehicle 3 based on the vehicle speed at the departure location of the object path, and correct the amount of power consumption in a manner that normalizes the influence.

[0072] The failure prediction section 113 can also estimate the influence of the elevation difference of the object path on the amount of power consumption of the outbound trip and the amount of power consumption of the inbound trip, and correct the amounts of power consumption in a manner that normalizes the influence. Further, the failure prediction section 113 can also extract only the amount of power consumption of the outbound trip or the inbound trip among the plurality of amounts of power consumption at the time of traveling the object path, to estimate the progress of the increase in the amount of power consumption.

[0073] The failure prediction section 113 can also deduct, from the amount of power consumption, the amount of regenerative electric power estimated based on the number of stops at the time of traveling the object path, by the calculation method of the amount of power consumption. Further, in a case where the amount of power consumption is calculated according to the input voltage / input current of the inverter 35 without according to the SOC of the battery cell 41, the amount of regenerative electric power does not need to be considered.

[0074] The failure prediction section 113 can also estimate the internal resistance of the battery cell 41 based on the SOC, SOH, temperature of the battery cell 41, estimate the influence of the internal resistance of the battery cell 41 on the amount of power consumption of the electric vehicle 3, and correct the amount of power consumption in a manner that normalizes the influence. Further, in a case where the amount of power consumption is calculated according to the input voltage / input current of the inverter 35 without according to the SOC of the battery cell 41, the internal resistance of the battery cell 41 does not need to be considered.

[0075] In a case where the log of the driver is included in the travel data, the failure prediction section 113 can also extract only the amount of power consumption based on the driving of the same driver among the plurality of amounts of power consumption at the time of traveling the object path, to estimate the progress of the increase in the amount of power consumption.

[0076] Figure 8is a flowchart showing a flow of processing of predicting an aging failure of the switching elements Q1-Q6 included in the inverter 35 by the failure prediction system 10 according to the embodiment. The subject path deciding section 112 reads out the travel data of the electric vehicle 3 from the travel data holding section 121, and extracts the moving track of the electric vehicle 3 from the progress of the position data of the electric vehicle 3 (S10). The subject path deciding section 112 generates a two-dimensional kernel density distribution based on the moving track of the electric vehicle 3 (S11). The subject path deciding section 112 decides the subject path based on the generated two-dimensional kernel density distribution (S12). Further, in a case where the subject path is decided in a manner not using the two-dimensional kernel density distribution, the subject path deciding section 112 time-divides the moving track of the electric vehicle 3 by a prescribed period as necessary.

[0077] The failure prediction section 113 extracts similar speed patterns from among the plurality of speed patterns when the electric vehicle 3 travels on the subject path (S13). The failure prediction section 113 calculates each power consumption amount when the electric vehicle 3 travels on the subject path from the travel data of the extracted speed patterns (S14). The failure prediction section 113 predicts the period when the switching elements Q1-Q6 included in the inverter 35 generate an aging failure based on the change in the power consumption amount with time (S15).

[0078] As explained above, according to the present embodiment, it is possible to predict the deterioration with years of the switching elements Q1-Q6 included in the inverter 35 of the electric vehicle 3 at low cost. If the travel data of the electric vehicle 3 is acquired and held, it is not necessary to add new components (for example, a failure detection sensor of the switching elements Q1-Q6) to the electric vehicle 3. By merely analyzing the log data, it is possible to predict the failure of the switching elements Q1-Q6 with high accuracy and at low cost.

[0079] By predicting the failure of the switching elements Q1-Q6 from the prediction of the annual increase in the loss of the inverter 35, it is possible to notify and urge the user to perform the replacement repair of the inverter 35 in advance. Thereby, it is possible to avoid the inconvenience of being unable to travel due to a sudden failure of the inverter 35.

[0080] By estimating the annual change in the power consumption amount when traveling under a load condition substantially identical, it is possible to predict the failure of the switching elements Q1-Q6 in advance. In order to collect the power consumption amount under a load condition substantially identical, the subject path frequently used with a high frequency of use is decided based on the two-dimensional kernel density distribution. In addition, by subtracting the value [kWh] obtained by multiplying the time integral of the shaft output (torque x rotation speed) of the motor 34 by a prescribed coefficient from the input power amount [kWh] of the inverter 35, it is possible to exclude the drive loss from the output shaft of the motor 34 to the drive wheels (rear wheels 31r), and it is possible to grasp only the increase in the loss of the inverter 35.

[0081] In addition, the influence of various fluctuation factors can be eliminated by various other corrections. For example, by estimating the load fluctuation from the chronological use history of the electric vehicle 3, the influence of the fluctuation factor caused by the load can be eliminated. By these processes, the failure period can be predicted with higher accuracy compared to the case where the failure period of the switching elements Q1-Q6 is predicted based on the measured raw power consumption amount.

[0082] The above describes the present disclosure based on the embodiments. As will be readily appreciated by those skilled in the art, the embodiments are illustrative, and various modifications can be made to the respective constituent elements and combinations of the respective processes, and such modifications are also within the scope of the present disclosure.

[0083] To make the prediction of the increase in the loss of the inverter 35 more accurate, travel data for an interval of a prescribed period or more apart from the sampling interval of the GPS data can also be excluded from the analysis target data. For example, in a travel interval where there are many tunnels, the absence of GPS data increases. In addition, travel data for days with poor weather conditions (e.g., snowy days) can also be excluded from the analysis target data.

[0084] The failure prediction system 10 according to the embodiments can also be used to predict the failure period of the switching elements Q1-Q6 included in the inverter 35 mounted on a hybrid vehicle (HV) or a plug-in hybrid vehicle (PHV). The prediction can be made based on the travel data for the motor travel period among the motor travel period and the engine travel period.

[0085] In addition, in the above-described embodiments, a four-wheel electric vehicle using the inverter 35 is assumed as the electric vehicle 3. In this regard, it can also be an electric motorcycle (electric scooter), an electric bicycle. In addition, the electric vehicle includes not only a full-spec electric vehicle but also a low-speed electric vehicle such as a golf cart, a land car (in Japanese) used in shopping centers, amusement facilities, and the like.

[0086] Further, the embodiments can be determined by the following items.

[0087] [Item 1]

[0088] A failure prediction system (10) characterized by comprising:

[0089] an acquisition unit (111) that acquires travel data of an electric vehicle (3); and

[0090] a prediction unit (113) that predicts an aging failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3) based on the travel data of the electric vehicle (3),

[0091] wherein the travel data includes position data of the electric vehicle (3) and data related to power consumption,

[0092] The prediction unit (113) predicts the aging failure of the drive circuit (35) based on a progression of an increase in the amount of power consumption of the electric vehicle (3) when traveling on the same route.

[0093] Thus, the annual degradation of the drive circuit (35) can be predicted at low cost.

[0094] [Item 2]

[0095] The failure prediction system (10) according to Item 1, characterized in that

[0096] The travel data includes vehicle speed,

[0097] The prediction unit (113) generates the progression of the increase in the amount of power consumption based on travel data similar to a pattern of vehicle speed when the electric vehicle (3) travels on the route.

[0098] Thus, the prediction accuracy of the annual degradation of the drive circuit (35) can be improved.

[0099] [Item 3]

[0100] The failure prediction system (10) according to Item 1 or 2, characterized in that

[0101] The travel data includes input voltage of the drive circuit (35), input current of the drive circuit (35), rotation speed of a motor (34) driven by the drive circuit (35), and rotational torque of the motor (34),

[0102] The prediction unit (113) estimates a loss of a switching element (Q1-Q6) included in the drive circuit (35) from an amount of power consumption obtained by integrating input power of the drive circuit (35) based on input voltage and input current of the drive circuit (35) with respect to travel time on the route, minus cumulative shaft output of the motor (34) obtained by integrating shaft output of the motor (34) based on rotation speed and rotational torque of the motor (34) with respect to travel time on the route.

[0103] Thus, the influence of mechanical drive loss can be excluded from the prediction of the annual degradation of the switching element (Q1-Q6) included in the drive circuit (35).

[0104] [Item 4]

[0105] The failure prediction system (10) according to any one of Items 1 to 3, characterized in that

[0106] The drive circuit (35) is an inverter (35),

[0107] The prediction unit (113) predicts an aging failure of a switching element (Q1-Q6) included in the inverter (35).

[0108] Thus, it is possible to predict the deterioration with age of the switching element (Q1-Q6) included in the inverter (35) at low cost.

[0109] [Item 5]

[0110] The failure prediction system (10) according to any one of items 1 to 4, characterized by

[0111] Further provided is a route decision unit (112) that extracts a movement trajectory of the electric vehicle (3) from a change in position data of the electric vehicle (3) and generates a two-dimensional frequency distribution to decide the route.

[0112] Thus, it is possible to decide the target route, which is a basis for sampling the power consumption amount, with high accuracy.

[0113] [Item 6]

[0114] The failure prediction system (10) according to item 5, characterized by

[0115] The two-dimensional frequency distribution generated by the route decision unit (112) is a two-dimensional kernel density distribution.

[0116] Thus, it is possible to decide the target route, which is a basis for sampling the power consumption amount, with high accuracy based on a density function.

[0117] [Item 7]

[0118] The failure prediction system (10) according to item 5, characterized by

[0119] The route decision unit (112) determines a target route with a high frequency of use based on the extracted movement trajectory of the electric vehicle (3),

[0120] The prediction unit (113) predicts an aging failure of the drive circuit (35) based on an increase in the power consumption amount during travel when traveling on the target route.

[0121] Thus, it is possible to predict the deterioration with age of the drive circuit (35) at low cost based on travel data of the target route with high accuracy.

[0122] [Item 8]

[0123] The failure prediction system (10) according to item 5 or 6, characterized by

[0124] the travel data includes vehicle speed,

[0125] The route determining section (112) excludes position data at the time of travel when the vehicle speed is less than a set value, to generate the two-dimensional frequency distribution.

[0126] Thus, the two-dimensional frequency distribution can be generated with high accuracy.

[0127] [Item 9]

[0128] A failure prediction method characterized by comprising the steps of:

[0129] acquiring travel data of an electric vehicle (3); and

[0130] predicting, based on the travel data of the electric vehicle (3), an aging failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3),

[0131] the travel data including position data of the electric vehicle (3) and data related to power consumption,

[0132] wherein, in the step of making the prediction, the aging failure of the drive circuit (35) is predicted based on a progression of an increase in the amount of power consumption of the electric vehicle (3) when traveling on the same route.

[0133] Thus, the annual deterioration of the drive circuit (35) can be predicted at low cost.

[0134] [Item 10]

[0135] A failure prediction program characterized by causing a computer to execute the following processing:

[0136] acquiring travel data of an electric vehicle (3); and

[0137] predicting, based on the travel data of the electric vehicle (3), an aging failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3),

[0138] the travel data including position data of the electric vehicle (3) and data related to power consumption,

[0139] wherein, in the processing of making the prediction, the aging failure of the drive circuit (35) is predicted based on a progression of an increase in the amount of power consumption of the electric vehicle (3) when traveling on the same route.

[0140] Thus, the annual deterioration of the drive circuit (35) can be predicted at low cost.

[0141] Reference Signs List

[0142] 3: electric vehicle; 10: failure prediction system; 11: processing portion; 111: travel data acquisition portion; 112: object path decision portion; 113: failure prediction portion; 12: recording portion; 121: travel data holding portion; 30: vehicle control portion; 31f: front wheel; 31r: rear wheel; 32f: front wheel axle; 32r: rear wheel axle; 33: transmission; 34: motor; 35: inverter; 36: motor controller; 37: gate driver; 381: input voltage / current sensor; 382: output voltage / current sensor; 383: rotation speed / torque sensor; 384: GPS sensor; 385: vehicle speed sensor; 39: wireless communication portion; 39a: antenna; 40: power supply system; 41: battery cell; 42: management portion; Q1-Q6: switching element; D1-D6: diode.

Claims

1. A fault prediction system, characterized in that, have: The acquisition unit acquires driving data from electric vehicles; and The prediction unit predicts aging failures in the drive circuit of the motor that drives the drive wheels of the electric vehicle based on the vehicle's driving data. The driving data includes the electric vehicle's location data and power consumption-related data. The prediction unit predicts aging failures of the drive circuit based on the increasing power consumption of the electric vehicle while traveling on the same path.

2. The fault prediction system according to claim 1, characterized in that, The driving data includes vehicle speed. The prediction unit generates the shift in the increase in power consumption based on driving data similar to the speed pattern of the electric vehicle traveling on the path.

3. The fault prediction system according to claim 1 or 2, characterized in that, The driving data includes the input voltage of the drive circuit, the input current of the drive circuit, the speed of the motor driven by the drive circuit, and the rotational torque of the motor. The prediction unit estimates the losses of the switching elements included in the drive circuit by subtracting the cumulative shaft output of the motor obtained by integrating the input power of the drive circuit based on the input voltage and input current of the drive circuit with the travel time of the path, which is obtained by integrating the input power of the drive circuit based on the input voltage and input current of the drive circuit with the travel time of the path, from the cumulative shaft output of the motor obtained by integrating the shaft output of the motor based on the motor speed and rotational torque with the travel time of the path.

4. The fault prediction system according to claim 1 or 2, characterized in that, The drive circuit is an inverter. The prediction unit predicts aging failures of the switching elements included in the inverter.

5. The fault prediction system according to claim 1 or 2, characterized in that, It also includes a path determination unit, which extracts the movement trajectory of the electric vehicle from the shift of the electric vehicle's position data and generates a two-dimensional frequency distribution to determine the path.

6. The fault prediction system according to claim 5, characterized in that, The two-dimensional frequency distribution generated by the path-determining unit is a two-dimensional kernel density distribution.

7. The fault prediction system according to claim 5, characterized in that, The path determination unit determines the paths of frequently used objects based on the extracted movement trajectories of the electric vehicles. The prediction unit predicts aging failures of the drive circuit based on the increase in power consumption during travel on the object path.

8. The fault prediction system according to claim 5, characterized in that, The driving data includes vehicle speed. The path determination unit excludes the position data during driving when the vehicle speed is less than a set value in order to generate the two-dimensional frequency distribution.

9. A fault prediction method, characterized in that, Includes the following steps: Acquire driving data of electric vehicles; as well as Based on the driving data of the electric vehicle, the aging failure of the drive circuit of the motor that drives the drive wheels of the electric vehicle is predicted. The driving data includes the electric vehicle's location data and power consumption-related data. In the prediction step, the aging failure of the drive circuit is predicted based on the increase in power consumption of the electric vehicle while traveling on the same path.

10. A computer-readable storage medium storing a fault prediction program that causes a computer to perform the following processes: Acquire driving data of electric vehicles; and Based on the driving data of the electric vehicle, the aging failure of the drive circuit of the motor that drives the drive wheels of the electric vehicle is predicted. in, The driving data includes the electric vehicle's location data and data related to power consumption. In the process of making the prediction, the aging failure of the drive circuit is predicted based on the increase in power consumption of the electric vehicle when traveling on the same path.

11. A computer program product comprising a fault prediction program that causes a computer to perform the fault prediction method according to claim 9.

Images