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

Abstract

An acquisition unit (111) acquires travel data of an electric vehicle. A prediction unit (112) predicts a failure due to aging in a drive circuit of a motor for driving a drive wheel of the electric vehicle on the basis of the travel data of the electric vehicle. The travel data includes position data and power-consumption-related data of the electric vehicle. The prediction unit (112) predicts a failure due to aging in the drive circuit on the basis of the increased trend of a power consumption amount obtained each time the electric vehicle travels 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 time-dependent failures of switching elements included in an inverter of an electric vehicle. [Background technology]

[0002] Electric vehicles (EVs) are becoming increasingly popular, primarily as commercial vehicles such as delivery vehicles. In recent years, EV driving data (battery information, movement trajectory, vehicle control information, etc.) is being stored on the cloud, creating an environment where this data can be utilized in a variety of ways.

[0003] Many methods have been proposed for calculating the energy required from the driving route and charging the required amount of battery power to enable an EV to travel to a destination without running out of battery power. For example, a method has been proposed in which the optimal amount of charge required to travel a commonly used route is notified when the destination is set, thereby preventing excessive charging and battery deterioration (see, for example, Patent Document 1). Another method has been proposed in which the optimal route to the destination with the lowest energy cost is presented based on past driving history, and whether the cruising distance is possible within the range of the stored battery power (see, for example, Patent Document 2). Another method has been proposed in which an inverter controls the cruising distance to the destination based on conditions such as total driving distance, weight, size, drag coefficient, speed, acceleration, history, temperature, and terrain (see, for example, Patent Document 3).

[0004] EVs use inverters to drive the motor. The power elements used in the inverter (e.g., MOSFETs (Metal-Oxide Semiconductor Field-Effect Transmitters) and IGBTs (Insulated Gate Bipolar Transistors)) deteriorate over time. A major factor in the deterioration of power elements is the increase in contact resistance of the bonding wire. This is caused by metal fatigue due to heat cycles, and the increase in contact resistance of the bonding wire manifests as increased loss (decreased efficiency) in the power elements. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-19627 [Patent Document 2] Japanese Patent Application Laid-Open No. 2009-63555 [Patent Document 3] Japanese Patent Application Publication No. 2018-27012 Summary of the Invention [Problem to be solved by the invention]

[0006] To predict the deterioration of elements other than power elements, such as electrolytic capacitors, coils, and fans, dedicated sensors are required for each element. Therefore, to predict the deterioration of these elements installed in an EV, a design change is required to add dedicated sensors.

[0007] On the other hand, the deterioration of power elements over time can be predicted without adding a dedicated sensor if the transition of loss increase can be predicted.

[0008] The present disclosure has been made in light of these circumstances, and its purpose is to provide a technology for predicting aging deterioration of a drive circuit of an electric vehicle at low cost. [Means for solving the problem]

[0009] To solve the above problem, a failure prediction system according to one aspect of the present disclosure includes an acquisition unit that acquires driving data of an electric vehicle, and a prediction unit that predicts an age-related failure of a drive circuit of a motor that drives drive wheels of the electric vehicle based on the driving data of the electric vehicle. The driving data includes position data and data related to power consumption of the electric vehicle. The prediction unit predicts an age-related failure of the drive circuit based on the increase in power consumption while the electric vehicle is traveling the same route.

[0010] Any combination of the above components, or conversion of the present disclosure into an apparatus, system, method, computer program, recording medium on which a computer program is recorded, etc., is also valid as an aspect of the present disclosure. [Effects of the Invention]

[0011] According to the present disclosure, it is possible to predict the aging deterioration of a drive circuit of an electric vehicle at low cost. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing a schematic configuration of an electric vehicle according to an embodiment; [Figure 2] FIG. 1 is a diagram showing a schematic configuration of a drive system of an electric vehicle. [Figure 3] 1 is a diagram illustrating an example of the configuration of a failure prediction system according to an embodiment; [Figure 4] 4(a)-(i) are diagrams in which log data of a GPS trajectory of a certain electric vehicle over a predetermined period is divided into time periods, and the divided GPS trajectories are plotted on multiple graphs. [Figure 5] 5(a) and 5(b) are diagrams for explaining a method for determining a target route. [Figure 6] 6(a)-(g) are diagrams showing specific examples of vehicle speed patterns when traveling along a target route. [Figure 7] FIG. 10 is a diagram showing an example of the transition of the amount of power consumption when traveling along a target route. [Figure 8] 1 is a flowchart showing a process flow for predicting age-related failure of a switching element included in an inverter, performed by a failure prediction system according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] FIG. 1 is a diagram showing a schematic configuration of an electric vehicle 3 according to an embodiment. In this embodiment, the electric vehicle 3 is assumed to be a pure EV that is not equipped with an internal combustion engine. The electric vehicle 3 shown in FIG. 1 is a rear-wheel drive (2WD) EV that has 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 by a front wheel axle 32f, and the pair of rear wheels 31r are connected by a rear wheel axle 32r. A transmission 33 transmits the rotation of the motor 34 to the rear wheel axle 32r at a predetermined conversion ratio. The electric vehicle 3 may be a front-wheel drive (2WD) or 4WD electric vehicle.

[0014] The power supply system 40 includes a battery unit (battery section) 41 and a management section 42. The battery unit 41 includes a plurality of cells. The cells may be lithium-ion battery cells, nickel-metal hydride battery cells, or the like. In the following description, an example is assumed in which lithium-ion battery cells (nominal voltage: 3.6-3.7V) are used. The management section 42 monitors the voltage, temperature, current, SOC (State of Charge), and SOH (State of Health) of the plurality of cells included in the battery unit 41, and transmits these to the vehicle control section 30 via an in-vehicle network. For example, a CAN (Controller Area Network) or a LIN (Local Interconnect Network) may be used as the in-vehicle network.

[0015] The inverter 35 is a drive circuit that drives the motor 34, and during power running, converts DC power supplied from the battery unit 41 into AC power and supplies it to the motor 34. During regeneration, the inverter 35 converts AC power supplied from the motor 34 into DC power and supplies it to the battery unit 41. During power running, the motor 34 rotates in accordance with the AC power supplied from the inverter 35. During regeneration, the motor 34 converts rotational energy generated by deceleration into AC power and supplies it to the inverter 35.

[0016] Fig. 2 is a diagram showing a schematic configuration of a drive system for electric vehicle 3. Fig. 2 shows an example in which a three-phase AC motor is used as the motor 34 that drives electric vehicle 3, and the three-phase AC motor 34 is driven by a three-phase inverter 35. The three-phase inverter 35 converts DC power supplied from battery unit 41 into three-phase AC power, the phases of which are shifted by 120 degrees, and drives the three-phase AC motor 34.

[0017] 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.

[0018] 2, IGBTs are used for the first switching element Q1 to the sixth switching element Q6. The first diode D1 to the sixth diode D6 are connected in anti-parallel to the first switching element Q1 to the sixth switching element Q6, respectively. If MOSFETs are used for the first switching element Q1 to the sixth switching element Q6, parasitic diodes formed in the source-to-drain direction are used as the first diode D1 to the sixth diode D6.

[0019] The motor controller 36 acquires the input DC voltage and input DC current of the inverter 35 detected by an input voltage / current sensor 381, the output AC voltage and output AC current of the inverter 35 detected by an output voltage / current sensor 382, ​​and the rotation speed and torque of the three-phase AC motor 34 detected by a rotation speed / torque sensor 383. The motor controller 36 also acquires an accelerator signal or a brake signal in response to the driver's operation or generated by the automatic driving controller.

[0020] Based on these input parameters, the motor controller 36 generates a PWM signal for driving the inverter 35 and outputs it to the gate driver 37. The gate driver 37 generates drive signals for the first switching element Q1 to the sixth switching element Q6 based on the PWM signal input from the motor controller 36 and a predetermined carrier wave, and inputs them to the gate terminals of the first switching element Q1 to the sixth switching element Q6.

[0021] The motor controller 36 transmits the input DC voltage of the inverter 35, the input DC current of the inverter 35, the rotation speed of the motor 34, and the torque of the motor 34 to the vehicle control unit 30 via the in-vehicle network.

[0022] Returning to Fig. 1, the vehicle control unit 30 is a vehicle ECU (Electronic Control Unit) that controls the entire electric vehicle 3, and may be configured as, for example, an integrated VCM (Vehicle Control Module).

[0023] The GPS sensor 384 detects the position information of the electric vehicle 3 and transmits the detected position information to the vehicle control unit 30. Specifically, the GPS sensor 384 receives radio waves, including the respective transmission times, from a plurality of GPS satellites, and calculates the latitude and longitude of the reception point based on the plurality of transmission times included in the plurality of received radio waves.

[0024] The vehicle speed sensor 385 generates a pulse signal proportional to the rotation speed of the front wheel shaft 32f or the rear wheel shaft 32r, and transmits the generated pulse signal to the vehicle control unit 30. The vehicle control unit 30 detects the speed of the electric vehicle 3 based on the pulse signal received from the vehicle speed sensor 385.

[0025] The wireless communication unit 39 performs signal processing for wirelessly connecting to a network via an antenna 39a. Examples of wireless communication networks that can be used by the electric vehicle 3 include a mobile phone network (cellular network), a wireless LAN, V2I (Vehicle-to-Infrastructure), V2V (Vehicle-to-Vehicle), an ETC system (Electronic Toll Collection System), and DSRC (Dedicated Short Range Communications).

[0026] While the electric vehicle 3 is traveling, the vehicle control unit 30 can transmit traveling data in real time to a cloud server for data accumulation or the company's own server using the wireless communication unit 39. The traveling data includes position data (latitude and longitude) of the electric vehicle 3, the vehicle speed of the electric vehicle 3, the voltage, current, temperature, SOC, SOH of the multiple cells included in the battery unit 41, the input DC voltage and input DC current of the inverter 35, and the rotation speed and torque of the motor 34. The vehicle control unit 30 samples this data periodically (for example, every 10 seconds) and transmits it to the cloud server or the company's own server each time.

[0027] The vehicle control unit 30 may store driving data of the electric vehicles 3 in an internal memory and transmit the driving data stored in the memory in a batch at a predetermined timing. For example, the vehicle control unit 30 may transmit the driving data stored in the memory in a batch to a terminal device at a sales office after the business hours of the day have ended. The terminal device at the sales office transmits the driving data of multiple electric vehicles 3 to a cloud server or the company's own server at a predetermined timing.

[0028] Furthermore, when charging from a charger equipped with a network communication function, the vehicle control unit 30 may transmit all of the driving data stored in the memory to the charger via the charging cable. The charger then transmits the received driving data to a cloud server or its own server. This example is effective for an electric vehicle 3 that is not equipped with a wireless communication function.

[0029] FIG. 3 is a diagram showing an example of the configuration of a failure prediction system 10 according to an embodiment. The failure prediction system 10 is constructed with one or more servers. For example, the failure prediction system 10 may be constructed with a single in-house server installed in a data center or in-house facility. The failure prediction system 10 may also be constructed with a cloud server used based on a cloud service. The failure prediction system 10 may also be constructed with multiple in-house servers distributed and installed at multiple bases (data centers, in-house facilities). The failure prediction system 10 may also be constructed with a combination of a cloud server used based on a cloud service and an in-house server. The failure prediction system 10 may also be constructed with multiple cloud servers based on contracts with multiple cloud service providers.

[0030] The failure prediction system 10 includes a processing unit 11 and a recording unit 12. The processing unit 11 includes a driving data acquisition unit 111, a target route determination unit 112, and a failure prediction unit 113. The functions of the processing unit 11 can be realized by a combination of hardware resources and software resources, or by hardware resources alone. As hardware resources, a CPU, ROM, RAM, GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array), and other LSIs can be used. As software resources, programs such as an operating system and applications can be used.

[0031] The recording unit 12 includes a running data holding unit 121. The recording unit 12 includes a non-volatile recording medium such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and records various data.

[0032] The travel data acquisition unit 111 acquires travel data of the electric vehicle 3 via a network and stores the acquired travel data in the travel data storage unit 121. The target route determination unit 112 reads out the travel data of the electric vehicle 3 from the travel data storage unit 121 and extracts a movement trajectory of the electric vehicle 3 from changes in the position data of the electric vehicle 3. Based on the extracted movement trajectory of the electric vehicle 3, the target route determination unit 112 determines a frequently used route that is always used (hereinafter referred to as a target route).

[0033] 4(a)-(i) are diagrams in which log data of the GPS trajectory of a certain electric vehicle 3 over a predetermined period is time-divided and the divided GPS trajectories are plotted on multiple graphs. In the example shown in Fig. 4(a)-(i), the GPS trajectories are simply plotted on a graph with the horizontal axis representing longitude and the vertical axis representing latitude, but the GPS trajectories may also be plotted superimposed on an actual map.

[0034] 5(a)-(b) are diagrams for explaining a method for determining a target route. In the example shown in FIG. 5(a), the target route determination unit 112 generates a two-dimensional kernel density distribution (two-dimensional frequency distribution) based on the GPS trajectory of the electric vehicle 3 over a predetermined period. When generating the two-dimensional kernel density distribution, it is desirable to plot only position data when the vehicle speed is equal to or greater than a set value (e.g., 10 km / h). If position data when traveling at low speeds or when stopped is also plotted, the number of plots of position data when traveling at low speeds or when stopped increases, and the generated density distribution may deviate from the density distribution that reflects reality.

[0035] The target route determination unit 112 uses a predetermined kernel function (e.g., a Gaussian function) and selects a predetermined bandwidth to generate a kernel density curve of plotted latitudes and a kernel density curve of plotted longitudes. The target route determination unit 112 identifies the highest peak position and the second highest peak position from the plotted kernel density curve of latitudes. Similarly, the target route determination unit 112 identifies the highest peak position and the second highest peak position from the plotted kernel density curve of latitudes. The target route determination unit 112 determines the most frequently used route as the target route from among routes connecting the intersection of the route with 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. Figure 5(b) shows the target route determined by the target route determination unit 112 from the GPS trajectory for the predetermined period shown in Figure 5(a).

[0036] The method for determining the target route is not limited to the method based on a two-dimensional kernel density distribution. For example, the target route determination unit 112 may determine, as the target route, a route designated by an administrator who is familiar with the road conditions in the area in which the electric vehicle 3 travels. It is desirable to select a target route such that the environmental conditions when the electric vehicle 3 travels are as constant as possible. For example, it is desirable to select a route that is flat, has few curves, has few traffic lights, has little congestion, and is frequently used.

[0037] The failure prediction unit 113 reads out multiple pieces of travel data when the electric vehicle 3 traveled along a target route from the travel data storage unit 121, and generates an increase trend in the amount of power consumption when traveling along the target route. The failure prediction unit 113 predicts an aging failure of the switching elements Q1-Q6 included in the inverter 35 based on the increase trend in the amount of power consumption.

[0038] The deterioration of the switching elements Q1-Q6 over time can be estimated from the increase in loss (decrease in efficiency) of the switching elements Q1-Q6. In order to accurately predict the failure of the inverter 35 over time from the change in power consumption while traveling on the target route, it is desirable to eliminate the influence of factors other than the factors that cause the increase in loss of the inverter 35 as much as possible. First, it is desirable to use only data that have similar vehicle speed patterns while traveling on the target route as data to be analyzed.

[0039] 6(a)-(g) are diagrams showing specific examples of vehicle speed patterns when traveling along a target route. FIGS. 6(a)-(g) show seven vehicle speed patterns when traveling along the target route. The failure prediction unit 113 extracts similar vehicle speed patterns from the vehicle speed patterns when traveling along the target route extracted from the travel data. Various methods can be used to extract similar vehicle speed patterns. For example, extracting a vehicle speed pattern in which the cumulative high-speed traveling time is equal to or greater than a predetermined time (e.g., extracting a vehicle speed pattern in which the period when traveling at 50 km / h or greater is equal to or greater than half of the traveling time), extracting a vehicle speed pattern in which the number of stops is equal to or less than a predetermined number, or extracting a vehicle speed pattern in which the timing of stopping or acceleration / deceleration is close to each other can be considered. Furthermore, methods such as pattern matching and correlation coefficients may also be used. In the example shown in FIGS. 6(a)-(g), the failure prediction unit 113 extracts the vehicle speed patterns of FIGS. 6(a), (d), and (e), in which the cumulative high-speed traveling time is equal to or greater than a predetermined time.

[0040] The failure prediction unit 113 extracts the SOC of the battery unit 41 at the start point of the target route and the SOC of the battery unit 41 at the finish point from the driving data when driving the target route, and based on the difference between these, can calculate the amount of power consumed when driving the target route.

[0041] Furthermore, the failure prediction unit 113 can also extract the logs of the input DC voltage V and input DC current I of the inverter 35 from the driving data when the target route is driven, from the start point to the finish point of the target route, and calculate the amount of power consumed when the target route is driven by integrating the input power of the inverter 35 over the driving time of the target route, as shown in the following (Equation 1). Power consumption=∫(V·I)dt / 1000[kWh] (Formula 1)

[0042] The electricity consumption can be calculated by dividing the distance of the target route by the calculated amount of power consumption. The amount of power consumption calculated as above takes into account not only the losses of the inverter 35 (mainly losses due to the contact resistance of the bonding wires of the switching elements Q1-Q6) but also the effects of mechanical drive losses that occur in the process of transmitting the rotational force of the motor 34. Drive losses include losses due to driving friction of the drive shaft, sliding friction of the differential, deformation of the rubber of the tires, friction between the tires and the road surface, etc.

[0043] The failure prediction unit 113 extracts from the driving data the log of the rotation speed rpm of the motor 34 and the rotational torque Nm of the motor 34 from the start point to the finish point of the target route, and can calculate the cumulative shaft output when driving the target route by integrating the shaft output of the motor 34 over the driving time of the target route as shown in the following (Equation 2). Accumulated shaft output = ∫(rpm·Nm·2π / 60)dt / 1000[kWh] (Equation 2)

[0044] The failure prediction unit 113 can calculate the amount of power consumption from which the influence of mechanical driving loss has been removed by subtracting the cumulative shaft output calculated using the above formula (2) from the amount of power consumption calculated using the above formula (1).

[0045] The power consumption calculated using the above formula (1) takes into account the effects of various variables in addition to the effects of inverter 35 loss and mechanical drive loss. For example, the effects of weather, tire pressure, load, battery internal resistance, vehicle speed (initial speed) at the start of the target route, elevation change on the target route, and the number of stops while traveling on the target route are taken into account. If the changes in elevation change on the target route differ significantly between the outbound and return journeys, the power consumption on the outbound and return journeys will be different values. The number of stops while traveling on the target route affects the amount of regeneration to the battery unit 41.

[0046] FIG. 7 is a diagram showing an example of the transition of power consumption when traveling along a target route. The power consumption can be roughly explained by the sum of power consumption due to inverter 35 loss and mechanical driving loss. Note that the power consumption is also affected by other variable factors. The failure prediction unit 113 plots multiple power consumption values ​​when traveling along the target route. Based on the plotted multiple power consumption values, the failure prediction unit 113 calculates a first regression line L1 using the least squares method or the like. Similarly, the failure prediction unit 113 calculates a second regression line L2 based on multiple integrated axle outputs when traveling along the target route.

[0047] Assuming that other variable factors are constant, the failure prediction unit 113 can estimate the difference between the first regression line L1 and the second regression line L2 as the loss of the inverter 35. Because the first regression line L1 and the second regression line L2 also extend into the future, the failure prediction unit 113 can predict future losses of the inverter 35.

[0048] The failure prediction unit 113 can predict when the statistically shortest lifespan of the inverter 35 will arrive, based on the predicted future loss of the inverter 35. The manager of the electric vehicle 3 can replace the inverter 35 as preventive maintenance before the statistically shortest lifespan of the inverter 35 arrives.

[0049] Furthermore, the failure prediction unit 113 can also predict when the statistical average lifespan of the inverter 35 will end, based on the predicted future losses of the inverter 35. The manager of the electric vehicle 3 can replace the inverter 35 as a predictive maintenance measure before the statistical average lifespan of the inverter 35 arrives. In this case, the inverter 35 currently in use can be effectively utilized while minimizing downtime.

[0050] The failure prediction unit 113 can make various corrections to the amount of power consumption in order to make other variable factors closer to a constant. The processing unit 11 of the failure prediction system 10 may also include a weather information acquisition unit (not shown). The weather information acquisition unit acquires weather data for the date and time when the target route was traveled from a weather information database server on the network.

[0051] The failure prediction unit 113 may estimate the effect of wind on the amount of power consumed by the electric vehicle 3 based on the wind direction, wind volume, and traveling direction of the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized. Furthermore, the failure prediction unit 113 may estimate the friction coefficient of the road surface based on the amount of rainfall, estimate the effect that the friction coefficient of the road surface has on the amount of power consumed by the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized.

[0052] Furthermore, the failure prediction unit 113 may estimate the amount of power consumed by the air conditioner based on the temperature, estimate the effect of the amount of power consumed by the air conditioner on the amount of power consumed by the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized. Note that if the amount of power consumed is calculated from the input voltage and input current of the inverter 35 instead of the SOC of the battery unit 41, it is not necessary to take the amount of power consumed by the air conditioner into consideration.

[0053] If the driving data includes a log from a tire pressure sensor, the failure prediction unit 113 may estimate the effect of the tire pressure on the amount of power consumed by the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized. If the driving data includes a log from a load sensor, the failure prediction unit 113 may estimate the effect of the load on the amount of power consumed by the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized.

[0054] If the load sensor log is not included, the failure prediction unit 113 may estimate the load based on the type of delivery vehicle using the electric vehicle 3. For example, if the electric vehicle 3 is a collection-type delivery vehicle, the failure prediction unit 113 uses a load model in which the load becomes heavier from the collection start time in the morning to the collection end time in the evening. If the electric vehicle 3 is a delivery-type delivery vehicle, the failure prediction unit 113 uses a load model in which the load becomes lighter from the delivery start time in the morning to the delivery end time in the evening. If the electric vehicle 3 is an inter-base delivery vehicle, the failure prediction unit 113 estimates that the load fluctuation is constant. The failure prediction unit 113 can estimate the type of delivery vehicle from the electricity consumption by time period.

[0055] The failure prediction unit 113 may estimate the effect that the initial speed of the target route has on the amount of power consumption of the electric vehicle 3 based on the vehicle speed at the start point of the target route, and correct the amount of power consumption so that the effect is standardized.

[0056] The failure prediction unit 113 may estimate the effect of elevation differences on the target route on the amount of power consumed on the way thereto and the amount of power consumed on the way back, and correct the amount of power consumed so that this effect is standardized. Note that the failure prediction unit 113 may extract only the amount of power consumed on the way thereto or the amount of power consumed on the way back from among the multiple amounts of power consumed when traveling the target route, and estimate the increase trend in the amount of power consumed.

[0057] Depending on the method for calculating the power consumption, the failure prediction unit 113 may subtract the amount of regenerative power estimated from the number of stops made while traveling on the target route from the amount of power consumption. Note that if the power consumption is calculated from the input voltage and input current of the inverter 35 instead of the SOC of the battery unit 41, it is not necessary to take the amount of regenerative power into consideration.

[0058] The failure prediction unit 113 may estimate the internal resistance of the battery unit 41 based on the SOC, SOH, and temperature of the battery unit 41, estimate the effect that the internal resistance of the battery unit 41 has on the amount of power consumed by the electric vehicle 3, and correct the amount of power consumed so that this effect is standardized. Note that, if the amount of power consumed is calculated from the input voltage and input current of the inverter 35 instead of the SOC of the battery unit 41, it is not necessary to take the internal resistance of the battery unit 41 into consideration.

[0059] If the driving data includes a driver's log, the failure prediction unit 113 may extract only the amount of power consumed by the same driver from among multiple amounts of power consumed when driving the target route, and estimate the increase in the amount of power consumed.

[0060] 8 is a flowchart showing the flow of processing performed by the failure prediction system 10 according to the embodiment to predict aging failures of the switching elements Q1-Q6 included in the inverter 35. The target route determination unit 112 reads out travel data of the electric vehicle 3 from the travel data storage unit 121, and extracts a movement trajectory of the electric vehicle 3 from changes in the position data of the electric vehicle 3 (S10). The target route determination unit 112 generates a two-dimensional kernel density distribution based on the movement trajectory of the electric vehicle 3 (S11). The target route determination unit 112 determines a target route based on the generated two-dimensional kernel density distribution (S12). When determining a target route without using the two-dimensional kernel density distribution, the target route determination unit 112 time-divides the movement trajectory of the electric vehicle 3 into predetermined periods as necessary.

[0061] The failure prediction unit 113 extracts similar vehicle speed patterns from a plurality of vehicle speed patterns when the electric vehicle 3 travels along a target route (S13). The failure prediction unit 113 calculates the amount of power consumption when traveling along the target route from the travel data of the extracted vehicle speed patterns (S14). The failure prediction unit 113 predicts when an age-related failure will occur in the switching elements Q1-Q6 included in the inverter 35 based on the change over time in the amount of power consumption (S15).

[0062] As described above, according to this embodiment, it is possible to predict at low cost the aging deterioration of the switching elements Q1-Q6 included in the inverter 35 of the electric vehicle 3. If the traveling data of the electric vehicle 3 is acquired and stored, there is no need to add new components (for example, sensors for detecting failures in the switching elements Q1-Q6) to the electric vehicle 3. By simply analyzing the log data, it is possible to predict failures in the switching elements Q1-Q6 with high accuracy and at low cost.

[0063] By predicting failure of the switching elements Q1-Q6 from a prediction of the aging increase in loss of the inverter 35, it is possible to notify the user in advance and urge them to replace or repair the inverter 35. This makes it possible to avoid the inconvenience of being unable to drive due to a sudden failure of the inverter 35.

[0064] By estimating the secular change in power consumption during driving under roughly consistent load conditions, it is possible to predict failures of switching elements Q1-Q6 in advance. To collect power consumption data under roughly consistent load conditions, a frequently used target route is determined based on a two-dimensional kernel density distribution. Furthermore, by subtracting the value [kWh] obtained by multiplying the time integral of the shaft output (torque x rotation speed) of motor 34 by a predetermined coefficient from the input power [kWh] of inverter 35, it is possible to eliminate the drive loss from the output shaft of motor 34 to the drive wheels (rear wheels 31r), and only the increase in loss in inverter 35 can be identified.

[0065] Furthermore, the influence of various fluctuation factors can be eliminated by various other corrections. For example, the influence of fluctuation factors due to the load can be eliminated by estimating the load fluctuation from the temporal usage history of the electric vehicle 3. These processes enable the prediction of the time of failure of the switching elements Q1-Q6 with higher accuracy than when the time of failure is predicted based on the original measured power consumption.

[0066] The present disclosure has been described above based on examples. The examples are merely illustrative, and it will be readily understood by those skilled in the art that various modifications are possible in the combination of the respective components and treatment processes, and that such modifications are also within the scope of the present disclosure.

[0067] To further improve the accuracy of prediction of loss increases in the inverter 35, travel data for sections where the sampling interval of GPS data is greater than a predetermined period may be excluded from the data to be analyzed. For example, travel sections with many tunnels will have many missing GPS data. In addition, travel data for days with poor weather conditions (for example, snowy days) may also be excluded from the data to be analyzed.

[0068] The failure prediction system 10 according to the embodiment can also be used to predict when a failure will occur in the switching elements Q1-Q6 included in the inverter 35 mounted on a hybrid vehicle (HV) or a plug-in hybrid vehicle (PHV). Of the motor running period and the engine running period, the prediction can be made based on the running data during the motor running period.

[0069] In the above-described embodiment, a four-wheeled electric vehicle using inverter 35 is assumed as power source vehicle 3. In this regard, an electric motorcycle (electric scooter) or an electric bicycle may also be used. Furthermore, electric vehicles include not only full-scale electric vehicles but also low-speed electric vehicles such as golf carts and land cars used in shopping malls, entertainment facilities, etc.

[0070] The embodiment may be specified by the following items.

[0071] [Item 1] an acquisition unit (111) that acquires traveling data of an electric vehicle (3); a prediction unit (113) that predicts an age-related failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3) based on running data of the electric vehicle (3); Equipped with The traveling data includes position data and power consumption data of the electric vehicle (3), The prediction unit (113) predicts an age-related failure of the drive circuit (35) based on an increase in power consumption while the electric vehicle (3) is traveling along the same route. A failure prediction system (10). This makes it possible to predict the deterioration of the drive circuit (35) over time at low cost. [Item 2] The driving data includes a vehicle speed, The prediction unit (113) generates the increase trend of the amount of power consumption based on travel data in which the vehicle speed pattern when the electric vehicle (3) travels along the route is similar. Item 1. A failure prediction system (10) according to item 1. This improves the accuracy of predicting deterioration over time of the drive circuit (35). [Item 3] The travel data includes an input voltage of the drive circuit (35), an input current of the drive circuit (35), a rotation speed of the motor (34) driven by the drive circuit (35), and a rotation torque of the motor (34); The prediction unit (113) estimates losses in switching elements (Q1-Q6) included in the drive circuit (35) by subtracting an integrated shaft output of the motor (34), which is obtained by integrating a shaft output of the motor (34) based on a rotation speed and a rotation torque of the motor (34) over a running time of the path, from an amount of power consumption, which is obtained by integrating an input power of the drive circuit (35) based on an input voltage and an input current of the drive circuit (35) over a running time of the path. 3. A failure prediction system (10) according to item 1 or 2. This makes it possible to eliminate the influence of mechanical drive loss from the prediction of deterioration over time of the switching elements (Q1-Q6) included in the drive circuit (35). [Item 4] The drive circuit (35) is an inverter (35), The prediction unit (113) predicts aging failure of switching elements (Q1-Q6) included in the inverter (35). 4. A failure prediction system (10) according to any one of items 1 to 3. This makes it possible to predict the deterioration over time of the switching elements (Q1-Q6) included in the inverter (35) at low cost. [Item 5] a route determination unit (112) that extracts a movement trajectory of the electric vehicle (3) from the transition of the position data of the electric vehicle (3), generates a two-dimensional frequency distribution, and determines the route; The failure prediction system (10) according to any one of items 1 to 4, further comprising: This makes it possible to determine with high accuracy the target route that is the basis for sampling the amount of power consumption. [Item 6] The two-dimensional frequency distribution generated by the path determination unit (112) is a two-dimensional kernel density distribution. Item 5. A failure prediction system (10) according to item 5. This allows the target route, which is the basis for sampling the amount of power consumption, to be determined with high precision based on the density function. [Item 7] The route determination unit (112) identifies a target route that is frequently used based on the extracted movement trajectory of the electric vehicle (3), the prediction unit (113) predicts an age-related failure of the drive circuit (35) based on an increase in power consumption during travel along the target route. Item 5. A failure prediction system (10) according to item 5. This makes it possible to predict the deterioration of the drive circuit (35) over time at low cost based on highly accurate travel data of the target route. [Item 8] The driving data includes a vehicle speed, the route determination unit (112) generates the two-dimensional frequency distribution by excluding position data when the vehicle is traveling at a speed less than a set value. 7. A failure prediction system (10) according to item 5 or 6. This makes it possible to generate a two-dimensional frequency distribution with high accuracy. [Item 9] A step of acquiring driving data of an electric vehicle (3); and predicting, based on running data of the electric vehicle (3), an age-related failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3), The traveling data includes position data and power consumption data of the electric vehicle (3), The predicting step predicts an age-related failure of the drive circuit (35) based on an increase in power consumption while the electric vehicle (3) is traveling along the same route. A failure prediction method comprising: This makes it possible to predict the deterioration of the drive circuit (35) over time at low cost. [Item 10] A process of acquiring driving data of an electric vehicle (3); and a process of predicting an age-related failure of a drive circuit (35) of a motor (34) that drives a drive wheel (31R) of the electric vehicle (3) based on running data of the electric vehicle (3); The traveling data includes position data and power consumption data of the electric vehicle (3), The prediction process predicts an age-related failure of the drive circuit (35) based on an increase in power consumption while the electric vehicle (3) is traveling along the same route. A failure prediction program characterized by: This makes it possible to predict the deterioration of the drive circuit (35) over time at low cost. [Explanation of symbols]

[0072] 3 Electric vehicle, 10 Failure prediction system, 11 Processing unit, 111 Driving data acquisition unit, 112 Target route determination unit, 113 Failure prediction unit, 12 Recording unit, 121 Driving data storage unit, 30 Vehicle control unit, 31f Front wheels, 31r Rear wheels, 32f Front wheel axle, 32r Rear wheel axle, 33 Transmission, 34 Motor, 35 Inverter, 36 Motor controller, 37 Gate driver, 381 Input voltage and current sensor, 382 Output voltage and current sensor, 383 Rotation speed and torque sensor, 384 GPS sensor, 385 Vehicle speed sensor, 39 Wireless communication unit, 39a Antenna, 40 Power supply system, 41 Battery unit, 42 Management unit, Q1-Q6 Switching elements, D1-D6 Diodes.

Claims

1. an acquisition unit that acquires driving data of an electric vehicle; a prediction unit that predicts an age-related failure of a drive circuit of a motor that drives drive wheels of the electric vehicle based on running data of the electric vehicle, The traveling data includes position data of the electric vehicle and data related to power consumption, the prediction unit predicts an age-related failure of the drive circuit based on an increase in power consumption while the electric vehicle is traveling along the same route. A failure prediction system comprising:

2. The driving data includes a vehicle speed, the prediction unit generates the increase trend of the amount of power consumption based on travel data having a similar vehicle speed pattern when the electric vehicle travels along the route.

2. The failure prediction system according to claim 1.

3. the driving data includes an input voltage of the drive circuit, an input current of the drive circuit, a rotation speed of the motor driven by the drive circuit, and a rotation torque of the motor; the prediction unit estimates the loss of a switching element included in the drive circuit by subtracting an integrated shaft output of the motor, which is obtained by integrating a shaft output of the motor based on the rotation speed and rotation torque of the motor, over the running time of the path, from an amount of power consumption, which is obtained by integrating an input power of the drive circuit based on an input voltage and an input current of the drive circuit, over the running time of the path; 3. The failure prediction system according to claim 1 or 2.

4. the drive circuit is an inverter, the prediction unit predicts an aging failure of a switching element included in the inverter.

4. The failure prediction system according to claim 1, wherein:

5. a route determination unit that extracts a movement trajectory of the electric vehicle from a transition of position data of the electric vehicle, generates a two-dimensional frequency distribution, and determines the route; The failure prediction system according to any one of claims 1 to 4, further comprising:

6. the two-dimensional frequency distribution generated by the path determination unit is a two-dimensional kernel density distribution; 6. The failure prediction system according to claim 5.

7. the route determination unit identifies a target route that is frequently used based on the extracted movement trajectory of the electric vehicle; the prediction unit predicts an age-related failure of the drive circuit based on an increase in power consumption during travel along the target route.

6. The failure prediction system according to claim 5.

8. The driving data includes a vehicle speed, the route determination unit generates the two-dimensional frequency distribution by excluding position data when the vehicle is traveling at a speed less than a set value.

7. The failure prediction system according to claim 5 or 6.

9. A failure prediction method in which a failure prediction system predicts aging failures of an electric vehicle, comprising: The failure prediction system comprises: acquiring driving data of the electric vehicle; and predicting, based on running data of the electric vehicle, an age-related failure of a drive circuit of a motor that drives drive wheels of the electric vehicle, The traveling data includes position data of the electric vehicle and data related to power consumption, the predicting step predicts an age-related failure of the drive circuit based on an increase in power consumption when the electric vehicle travels along the same route. A failure prediction method comprising:

10. A process of acquiring driving data of an electric vehicle; and a process of predicting an age-related failure of a drive circuit of a motor that drives a drive wheel of the electric vehicle based on the running data of the electric vehicle; The traveling data includes position data of the electric vehicle and data related to power consumption, the prediction process predicts an age-related failure of the drive circuit based on an increase in power consumption when the electric vehicle travels along the same route. A failure prediction program characterized by:

Images