KEY-OFF CONSUMER MONITORING OF A VEHICLE

Abstract

Current limits are established to detect anomalous current draw from a battery in an electrical system during a vehicle's key-off state. A variety of test vehicles are selected that can be operated without any anomalous current draw. A variety of time intervals are identified according to a variety of predetermined key-off load modes (KOL modes) that are sequentially activated after the key-off state is initiated. Key-off interval currents are measured in each of the test vehicles during each of the time intervals. A respective average current of the interval currents is determined for each time interval. A respective standard deviation current of the interval currents is determined for each time interval.A current limit is set for each time interval according to the respective average current plus a respective incremental value determined according to the respective standard deviation current.

Description

FIELD OF TECHNOLOGY

[0001] The present invention relates generally to battery power management in a motor vehicle, and more particularly to monitoring battery power management in and between a plurality of vehicles to identify performance anomalies and / or identify measures to protect battery reserves. BACKGROUND OF THE INVENTION

[0002] A typical automotive electrical system relies on a storage battery to supply power to electrical system loads during times when a main electrical power source is inactive / inaccessible. In a gasoline-powered vehicle, the main electrical power source typically includes an alternator / generator driven by an internal combustion engine (which is started by a starter motor, which is also powered by the storage battery). In an electric-powered vehicle, the main electrical power source typically includes a high-voltage / high-capacity battery pack (also known as a high-voltage traction battery). Typically, the high-voltage battery for vehicle propulsion provides up to several hundred volts and has an energy storage capacity of tens or hundreds of kilowatt-hours.The electric vehicle will also typically utilize a low-voltage storage battery, typically a 12-V battery (which may be identical to the storage battery used in an internal combustion engine vehicle). The storage battery has a finite charge capacity, much smaller than that of the high-voltage traction battery, and it is important to maintain sufficient energy storage to start the vehicle (e.g., to start the internal combustion engine or close the high-voltage contactors to connect a traction battery to motors for electric propulsion).

[0003] Many modern electronic vehicle systems operate continuously, even when the vehicle is parked and unattended, when the storage battery is the only available power source. Examples of electronic modules that must be powered at all times include those that perform functional operations while parked (e.g., anti-theft systems and remote entry systems), and those that require only a reduced amount of power to maintain memory contents or monitor / measure various conditions or electrical communication signals (e.g., when in a sleep mode). Other modules may continue to operate for a set time after the driver has turned off the vehicle, but after the set time, the power supply may be discontinued (e.g., interior lighting).

[0004] Because a vehicle may remain parked for long periods of time, it is important to limit battery discharge so that a sufficient battery charge is still available to restart the vehicle (e.g., start the engine or close the contactors) when the user returns. Therefore, the vehicle manufacturer specifies limits on the current drawn by various modules under each of the conditions that may occur. In particular, quiescent current limits are specified for the modules, with the limits applying during periods when the vehicle ignition switch has been turned OFF for a specified time and there has been no user activity. Some modules may be designated as having multiple stages of sleep states (e.g., from a light sleep state to a deep sleep state), gradually reducing the allowable current levels over extended periods.For example, a large display screen as part of a driver information system may take an undesirably long time to power on (e.g., boot up) if placed in a completely power-off state, forcing the driver to wait to use the device as intended when starting the vehicle. By defining progressive intermediate current levels for the display unit's KOL current, a compromise can be achieved between rapid availability of a driver information system and maintaining sufficient battery power to restart the vehicle.

[0005] Regarding the consolidated key-off load management system, which combines the current drawn from the storage battery by all loads, there may be a first (deepest) KOL mode, designated KOL-1, which allows a maximum total quiescent current of no more than 8 milliamperes (mA). KOL-1 is typically used only in a newly manufactured vehicle during vehicle transport mode when delivered to a dealer. A next deepest KOL mode, designated KOL-2, may allow a maximum quiescent current of no more than 20 mA. This can be considered a normal sleep state in which some modules are still partially powered during the key-off condition (e.g., to provide keypad recognition input capability and other normal vehicle monitoring operations).A KOL-3 mode may be provided, allowing a maximum quiescent current of up to 50 mA to enable full activation of some features. For example, luxury vehicles may operate a welcoming function, such as welcome lighting, based on detecting the user's approach to the vehicle (e.g., interrogating to detect a wireless device on the user). A KOL-4 mode may be provided, allowing a maximum quiescent current of up to 500 mA. This higher level of current draw may be used to partially power large display panels or other ECU and computer-based systems during vehicle sleep to provide rapid boot-up (e.g., a few seconds instead of 30 seconds or more for an unpowered display system).Even with improved storage battery capacities, a KOL-4 mode can typically only be maintained for about 14 to 24 hours. It should be noted that not all four of the above modes would necessarily be implemented in any particular vehicle, and a fewer number of modes may be used. For example, the KOL-3 mode and the KOL-4 mode, as described above, may be combined into a single mode (i.e., a welcome function and a quick start mode of a display system may be controlled together). Furthermore, the KOL modes may be subsets of one another, so that a higher mode may turn on certain loads at a higher current draw, but any loads other than those specifically modified for that mode will continue to operate as defined in the lower modes.

[0006] Determining maximum total quiescent current values ​​depends on balancing the electrical demands of various modules to provide the desired characteristics at the appropriate times with the need to maintain battery charge levels for the required periods. Due to the complexity of device interactions and a large number of contingencies in both the current drawn by loads and the storage capacity of individual batteries, determining quiescent current limits to detect excessive power consumption during key-offs was time- and resource-intensive. Furthermore, tracking and evaluating vehicle KOL performance was difficult and inaccurate for fleets.

[0007] Attempts by vehicle manufacturers to proactively identify potential key-off load performance issues by predicting trends in detections of events involving excessive KOL current draw have been hampered by the lack of meaningful metrics. It would be desirable to enable better detection of KOL performance issues and improved diagnosis of the root cause of such performance problems. Manufacturers want to monitor such trends because excessive current draw can cause battery wear, leading to warranty costs and dissatisfaction. SUMMARY OF THE INVENTION

[0008] In one aspect of the invention, there is a method for establishing current limits for detecting abnormal current draw from a battery in an electrical system during a key-off condition of a vehicle from among a plurality of test vehicles operable without any abnormal current draw. A plurality of time intervals are identified according to a plurality of predetermined key-off load (KOL) modes that are sequentially activated after initiation of the key-off condition. Key-off interval currents are measured during each of the time intervals in each of the test vehicles. A respective average current of the interval currents is determined at each respective time interval. A respective standard deviation current of the interval currents is determined at each respective time interval.A respective current limit is set corresponding to each time interval according to the respective average current plus a respective incremental value determined according to the respective standard deviation current.

[0009] A key-on or key-off state, as used in this document, refers to on and off vehicle states that are controlled not only using a physical key and lock, but also using a push-button switch, a keypad, voice commands, or wireless devices such as a remote key fob or a smartphone with a phone-as-a-key (PaaK) application. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a graph showing current draw limits for various KOL modes compared to the actual current draw during a key-off state. Fig. Figure 2 is a block diagram showing the interaction of a power management module and a control module. Fig. 3 is a block diagram illustrating one embodiment of an electrical architecture for distributing electrical power and multiplexing communication signals in which KOL power management is implemented. Fig. Figure 4 is a graph showing measured current consumption from a variety of test vehicles during a key-off condition. Fig. Figure 5 is a data distribution of measured currents during an exemplary time interval of the test vehicles. Fig. Figure 6 is a flowchart showing a process for determining current limits to detect excessive current draw in purchased vehicles. Fig. 7 is a block diagram illustrating one embodiment of a circuit for detecting and responding to excessive KOL current in a vehicle. Fig. Figure 8 is a flowchart showing a method for monitoring a fleet of purchased vehicles. Fig. Figure 9 is a block diagram showing a system for monitoring a fleet of vehicles. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

[0010] Fig. Figure 1 shows a variation in the current drawn by the combined electrical loads from a storage battery over a period of time when main electrical power is unavailable (i.e., the vehicle is in a key-off condition). A curve 10 represents the total current draw from the storage battery following a gradual decrease beginning with the turning off of an ignition key (IGN Off), which causes curve 10 to drop at 11 to a reduced level 12 during an initial key-off mode. The current draw 12 occurs during a period (from IGN Off to a first delay time Delay 1) known as the accessory load delay. During the accessory load delay, several systems and modules continue to operate to temporarily assist the occupants (e.g., interior lighting, active power outlets, and power window operation).During the accessory load delay, active systems can continue to draw power from the battery to operate various devices according to time periods and capabilities determined within each system. For the purpose of detecting undesirably high current draw, a dashed line 13 represents a maximum current draw specified during the accessory load delay. As used in this document, a "no power state" includes both 1) a sleep state, in which a module suspends most activity but still uses a small amount of power to monitor for a wake-up signal, and 2) a completely off condition.

[0011] At a subsequent time, shown as Delay 1, a power management controller determines that a global accessory delay period has expired. The accessory delay period may be a predetermined amount of time after IGN off or may also depend on a moment when it is detected that the occupants are exiting the vehicle. At Delay 1, many functions are disabled and the expected current draw falls below a level indicated at 15 because some systems / modules are without power. After this time, a new KOL mode begins, during which the aggregate battery discharge is expected to be below a current limit 14. At subsequent times Delay 3 and Delay 4, other KOL modes begin during which additional systems / modules are shut down and stricter current limits are imposed.

[0012] Fig. 2 illustrates portions of a vehicle electrical system involved in key-off load management. A body control module (BCM) 16 is configured (e.g., via software and / or firmware programming) to implement a key-off load management strategy in response to a vehicle ignition status determined in response to data messages from a multiplexed bus. The BCM 16 is coupled to a battery monitoring system (BMS) 17 connected to a storage battery (not shown). An electronic control module 18 represents any electronic module (e.g., a lighting module, an HVAC module, an audio entertainment module, or a wireless communication module) that receives power from the storage battery during a key-off state and may have several different power modes monitored by the BCM 16.The BCM 16 receives a measured current i from the BMS 17. The BCM 16 exchanges various control and data signals with the representative control module 18, such as commands to establish power, enter a sleep mode, and initiate a power-off state. The module 18 can transmit various response signals to the BCM 16, such as a status / acknowledgement signal and a data message identifying a period of time for which the module 18 has remained continuously awake.

[0013] Fig. 3 shows a more complete electrical architecture with a storage battery 20 connected to a battery monitoring system (BMS) 21. The BMS 21 may be a conventional component that, among other things, measures battery current flow from the battery 20 to the electrical loads. When part of an electric vehicle, the low-voltage storage battery 20 may be recharged from a high-voltage battery pack 22 via a DC / DC converter 23. When included in a gasoline-powered or hybrid vehicle, the storage battery 20 may be recharged by an alternator 24 driven by an internal combustion engine 25.

[0014] A body control module (BCM) 26 and a BMS 21 are both included in and communicate via a bus network (not shown), such as a CAN bus, that connects the electronic modules within a grouping 27. The bus network for grouping 27 is connected to a gateway 28, which is further connected to additional bus network groups 29 and 30, which may operate using different protocols. Gateway 28 reformats and forwards messages between networks so that modules in different bus networks can exchange communication signals, as known in the art. A power bus 31 distributes an output of battery 20 to various modules, including the BCM 26 and many other modules, including a number of modules 32-34 interconnected in bus group 27.

[0015] A module 35 and a powertrain control module (PCM) 36 are also connected within bus group 27. Power to modules 35 and 36 is controlled by the BCM 26 as a subnetwork 37 (e.g., their power is managed by the BCM 26 as a single unit). Modules 35 and 36 receive a supply voltage directly via power bus 31, but each receive shared control signals from the BCM 26 to turn power on or off to modules 35 and 36 simultaneously. Control messages from the BCM 26 may further include commands that control power delivery to lower-level components of a control module, such as sensors 38 and actuators 41, that receive power through the PCM 36.

[0016] In another example of power management, the BCM 26 is connected to a relay 39 that receives power from the power bus 31 and selectively transfers power to a module 40. The relay 39 may, for example, consist of an ignition relay. The module 40 is further connected to the bus grouping 27. The relay 39 may be controlled by a direct signal connection to the BCM 26 or, alternatively, via a multiplexed message. However, instead of having a sleep mode, the module 40 is either fully powered or not powered according to the ON / OFF state of the relay 39. On the other hand, the modules 35 and 36 are powered by the power bus 31 at all times, but each invoke a respective reduced-power mode, such as a sleep state.

[0017] Due to the Fig. Through the connections shown in Figure 3, the BCM 26 may respond to vehicle conditions to operate various loads (e.g., modules) at selected power levels (KOL modes) according to a key-off schedule and then obtain a measured and aggregated key-off current flowing from the storage battery during each of the key-off time intervals. If a measured key-off current is greater than a prescribed limit, the BCM 26 may record the event for diagnostic purposes and take corrective action to maintain the storage battery's state of charge.

[0018] The present invention is directed to selecting appropriate current limits and monitoring compliance with the current limits by a fleet of vehicles. A key-off load metric is derived that can better identify when and why problems with key-off loads occur, enabling the prediction and prevention of excessive draw for key-off loads. The metric follows a temporal profile of acceptable current draw from the battery (e.g., according to time intervals corresponding to the elapsed time since the ignition key was turned off). An exemplary profile has seven time intervals.The metric may provide an overall score that can be used to evaluate the performance of (A) a single key-off cycle of a vehicle, (B) all key-off cycles for a single vehicle, (C) all key-off cycles in a given week for a given vehicle program or model, or (D) any other combination of key-off cycles of interest.

[0019] When the vehicle is turned off, different processes occur (e.g., via messaging on the CAN bus) while the vehicle is shut down. Some modules turn off within a few minutes, while others are allowed to draw current for a longer period. Based on the shutdown pattern, a profile can define an acceptable current draw for each respective time interval. To develop the profile, a representative sample of vehicles with fault-free performance (i.e., nominal test vehicles) is identified. For these well-functioning vehicles, which have no known problems from key-off consumers, a history of the current drawn during a variety of key-off cycles is acquired in a process to determine target currents. An average current is measured within each time interval of a key-off cycle.The average current can consist of a mean (µ) of periodic current measurements during the interval. Average currents are obtained during each of the seven time intervals. In one example, the time intervals were 0-6 minutes, 6-12 minutes, 12-30 minutes, 30-60 minutes, 60-90 minutes, 90 minutes to 3 hours, and more than 3 hours. A standard deviation (σ) of the measured current during each time interval is also determined (using known statistical methods).

[0020] According to some embodiments, if a measured (average) interval current (i) for a single time interval is less than one standard deviation above the mean (µ + σ), the vehicle is in a green zone for that time interval. If a measured interval current is between one and two standard deviations above the mean [i.e., (µ + σ) < i < (µ + 2σ)], it is in a yellow zone. If the measured interval current is more than two standard deviations above the mean [i.e., i > (µ + 2σ)], it is in a red zone. To evaluate an entire ignition-off cycle, a root-sum-of-squares evaluation method compares the actual vehicle currents during each respective time interval to a normalized acceptable rating that uses a standard error to account for sample size.Thus, the more key-off cycles used for evaluation, the closer the averaged currents from the operating vehicles are to the previously derived target currents (developed from the ideal / well-performing vehicles).

[0021] If a group of key-off cycles has a final rating of less than or equal to 1 using this rating system, it is operating as expected (i.e., it would pass). If a group of key-off cycles scores more than 1, it is drawing more current than allowed, and those vehicles should receive more attention.

[0022] Fig. Figure 4 shows a current consumption of key-off loads for a plurality of test vehicles. A key-off state of a vehicle is divided into seven time intervals, labeled T1 to T7. At the beginning of each time interval, a specific set of electrical loads (e.g., modules) is disabled, so that the aggregate total current falls accordingly. The solid line, the dashed line, and the dotted line each represent a current consumption for a key-off interval for a respective one of the test vehicles. Since the current consumption in a particular vehicle may not be constant, a measured current for an interval may, for example, consist of an average current during the interval. Multiple key-off cycles can be characterized for each of a plurality of test vehicles, and the resulting data can be processed, as in Fig. 5, to determine an average current for each respective interval as well as a standard deviation of the current values ​​in the sample. In Fig. In Figure 5, each vertical bar shows the number of occurrences of each current value in the sample. An arithmetic mean is shown along with one standard deviation (σ) and two standard deviations (2σ). Based on the data from the well-performing vehicles, a current limit can be set at an appropriate level above the mean. For example, a current limit can be set to the mean plus twice the standard deviation.

[0023] A preferred method for deriving current limits is described in Fig. 6. In step 42, a plurality of nominal test vehicles are selected. Test vehicles known to operate without abnormal current draw during the key-off state are selected. In particular, the test vehicles may include prototype vehicles and / or early production vehicles. Based on the key-off load (KOL) modes utilized in the test vehicles, a plurality of time intervals for monitoring KOL current draw are defined in step 43. The time intervals may each have a start time and an end time measured in respective elapsed times since entering the key-off state.

[0024] In step 44, KOL performance data is collected for multiple key-off cycles of each of the test vehicles. In particular, a key-off current is measured at each time interval in each test vehicle. Preferably, a single current value is derived for each time interval, where the single value may consist of (1) an average of the current draw over the time interval, (2) a maximum current draw during the time interval, or (3) any other representative current value. Using the measured currents during the key-off interval, an average value (e.g., mean) and a standard deviation of all measured values ​​for each respective time interval are determined in step 45.

[0025] A current limit to be used for each respective time interval is derived based on the mean (µ) and the standard deviation (σ). When monitoring vehicles based on current limits, measured current draws below the mean plus the standard deviation (µ + σ) would typically be considered acceptable. For current draws just above the mean plus the standard deviation, there may be a small probability of excessive draw and an "amber" condition may be detected. Measured current draws above an even higher threshold may indicate a "red" condition, which may result in a higher probability of depleted charge in the storage battery. A preferred value for a higher threshold may be, for example, the mean plus twice the standard deviation (µ + 2σ).The calculated values ​​for µ, σ, and any threshold values ​​are loaded into the vehicles to be monitored in step 46. The vehicles to be monitored can be in-service vehicles manufactured for sale or other transfer to customers. As used in this document, "purchased vehicles" refers to any vehicles transferred to a customer, whether through sale, lease, or otherwise.

[0026] The current limits derived in this document can be used by the purchased vehicle for in-vehicle monitoring so that excessive KOL current draw can be detected and corrected. Fig. Figure 7 shows a monitoring circuit 50, wherein a measured current i xfor a time interval x is coupled to an inverting input of a comparator 51 and to a non-inverting input of a comparator 52. A first threshold value (e.g., µ + σ) is coupled to the non-inverting input of the comparator 51. A second threshold value (e.g., µ + 2σ) is coupled to the inverting input of the comparator 52. The outputs of the comparators 51 and 52 are coupled to a statistical analysis unit 54. If the value of the current i x is less than the first threshold, the output of comparator 51 goes high (a logic 1) to indicate a green condition (i.e., no excessive draw). At the same time, the output of comparator 52 is low (logic 0) because the current i x is also less than the second threshold. The statistical analysis unit 54 is configured to take no action in response to the green condition.

[0027] The outputs of the comparators 51 and 52 are each coupled to a respective inverting input of an AND gate 53, the output of which is coupled to the statistical analysis unit 54. If the value of the current i x becomes greater than the first threshold and less than the second threshold, the low outputs from comparators 51 and 52 are inverted at the input of AND gate 53, so that the output of AND gate 53 goes high to indicate a yellow condition. When the value of current i xis greater than the second threshold, the output of comparator 52 goes high to indicate a red condition (i.e., excessive draw). Statistical analysis unit 54 is configured to monitor the yellow and red conditions over time to determine appropriate actions to mitigate excessive KOL current draw (e.g., driver notification and / or changing the permitted loads during respective time intervals).

[0028] The current limits, as derived in this document, can be used by the purchased vehicle to detect KOL conditions during which data on excessive KOL current draw should be reported outside the vehicle to facilitate monitoring of a fleet of vehicles to which the purchased vehicle belongs (e.g., all vehicles of a specific car or truck model). Once a purchased, in-service vehicle is operated based on the current limits as described above, a procedure as described in Fig. 8, can be used to monitor such a fleet of vehicles. In step 60, each vehicle determines a score for each key-off cycle. The score may be based on examining an aggregation of all discrepancies between the respective power limits and the actual power consumption during the respective time intervals. For example, a root sum of squares normalized according to a standard error may be used to calculate a score value for a key-off cycle in each individual vehicle. In step 61, the scores obtained from multiple cycles are combined (e.g., collected in a data packet and / or further processed to characterize the performance of the individual vehicle). For example, a score for a time interval may consist of a number reflecting how many standard deviations the actual power is above the power limit.A score of 1 or less would indicate no problems. A score above 1 may be sufficient to initiate a report or other action. In step 62, each individual vehicle reports the scores by transmitting the data packet to a central server. A manufacturer may maintain a central server on a network accessible to the fleet of vehicles through wireless communication, such as a cellular data connection.

[0029] The received ratings are collected and analyzed by the manufacturer using the central server. In step 63, the manufacturer may initiate an investigation if a certain number of ratings or a certain combination of ratings exceeds a threshold (i.e., an anomaly is detected). If ratings exist that exceed the threshold, an identification of the specific population of vehicles (e.g., vehicle models and / or trim levels of a specific model) is performed in step 64. In step 65, the ratings and any supporting data are analyzed to determine which time intervals are affected by excessive power draw.Based on the modules known to be active during various time intervals, the identity or identities of the modules that may be responsible for the excessive current draw can be traced in step 66. Based on the source of the excessive current draw, a solution or other responsive action can be identified and taken by the manufacturer.

[0030] The reported scores can also be used to track the progress of a vehicle fleet over time to see if it improves as a result of the actions taken. If certain vehicles have a score greater than 1 at any given time, the data can be reviewed in more detail to identify which vehicles are experiencing an issue and when it occurs in the KOL shutdown process. Once the manufacturer knows when excessive withdrawal occurs in the shutdown process, they can attempt to replicate the issue in engineered vehicles and find a solution.

[0031] Fig.Figure 9 shows a schematic diagram of an overall implementation of the invention in which test vehicles 70 are operated without abnormal current draw and from which actual current draw data is collected by recording the data during key-off trials. An engineering function 71 has performed the data collection and analyzes the data to generate respective current limits for use in vehicle production, and the current limits are used in a manufacturing process 72 that produces in-service vehicles 73. During normal operation, the vehicles 73 monitor the evaluations for key-off loads, and if necessary, the evaluations are combined and transmitted via data links 74 to a central server 75. Data from the central server 75 is used by the engineering function 71 as described above.

Claims

[1] A method for establishing current limits for detecting abnormal current draw from a battery in an electrical system during a key-off condition of a vehicle, the method comprising the following steps: Selecting a variety of test vehicles that can be operated without any abnormal power draw; Identifying a plurality of time intervals according to a plurality of predetermined key-off load (KOL) modes that are sequentially activated after initiation of the key-off condition; Measuring key-off interval currents during each of the time intervals in each of the test vehicles; Determining a respective average current of the interval currents in each respective time interval; Determining a respective standard deviation current of the interval currents in each respective time interval; and Setting a respective current limit corresponding to each time interval according to the respective average current plus a respective incremental value determined according to the respective standard deviation current. [2] The method of claim 1, wherein each respective average current consists of an average of the respective interval currents of each respective time interval. [3] The method of claim 1, wherein each respective incremental value consists of twice the respective standard deviation current. [4] The method of claim 1, further comprising the following steps: Measuring key-off interval currents during each of the time intervals in a purchased vehicle not included in the plurality of test vehicles; Comparing the measured key-off interval currents with respective current limits; and Detect excessive current draw when any of the measured key-off interval currents is greater than the respective current limit. [5] A method for monitoring a fleet of vehicles for excessive key-off load (KOL) current draw while in a key-off state, the method comprising the following steps: Selecting a variety of test vehicles that can be operated without any abnormal power draw; identifying a plurality of time intervals according to a plurality of predetermined KOL modes that are sequentially activated after initiation of the key-off state; Measuring key-off interval currents during each of the time intervals in each of the test vehicles; Determining a respective average current of the interval currents in each respective time interval; Determining a respective standard deviation current of the interval currents in each respective time interval; Setting a respective current limit corresponding to each time interval according to the respective average current plus a respective incremental value determined according to the respective standard deviation current; Transferring the respective power limits to each of the fleet of vehicles; Measuring key-off interval currents during each of the time intervals in each of the vehicles in the fleet of vehicles; Comparing the measured key-off interval currents with the respective current limits to determine respective key-off consumer ratings for each vehicle in the fleet of vehicles; Combining key consumer ratings from the fleet of vehicles; and Detecting a vehicle anomaly in the fleet of vehicles when the combined key-off consumer ratings exceed a predetermined threshold. [6] A method according to claim 5, wherein each respective average current consists of an average of the respective interval currents of each respective time interval. [7] The method of claim 5, wherein each respective incremental value consists of twice the respective standard deviation current. [8] A method according to claim 5, wherein each respective key-off load rating consists of a root sum of squares formed from differences between respective measured key-off interval currents and respective current limits. [9] The method of claim 5, wherein each time interval is associated with a respective first current limit and a respective second current limit, and wherein each respective key-off load rating is determined according to a collection of respective time intervals in which the respective measured key-off interval current is greater than the respective first current limit and in which the respective measured key-off interval current is greater than the respective second current limit. [10] The method of claim 9, wherein each first current limit consists of the respective average current plus the respective standard deviation current, and wherein each second current limit consists of the respective average current plus twice the respective standard deviation current. [11] The method of claim 5, further comprising the following step: When the anomaly is detected, comparing the time intervals and the measured key-off interval currents with electronic modules in the fleet of vehicles capable of malfunctioning to generate the measured key-off interval currents. [12] Vehicle comprising: a battery configured to supply electrical power during a key-off state of the vehicle; a plurality of electrical loads, each configurable to receive the electrical current flowing from the battery during the key-off state in dependence on predetermined key-off load (KOL) modes active according to respective time intervals; a battery monitor that measures key-off interval currents during each of the time intervals; a controller configured to (1) compare the measured key-off interval currents to respective current limits and (2) indicate excessive current draw in response to one of the measured key-off interval currents being greater than the respective current limit; where the respective current limits are obtained from: Selecting a variety of test vehicles that can be operated without any abnormal power draw; Measuring key-off interval currents during each of the time intervals in each of the test vehicles; Determining a respective average current of the interval currents in each respective time interval; Determining a respective standard deviation current of the interval currents in each respective time interval; and Setting a respective current limit corresponding to each time interval according to the respective average current plus a respective incremental value determined according to the respective standard deviation current. [13] A vehicle according to claim 12, wherein each respective average current consists of an average of the respective key-off interval currents of each respective time interval. [14] A vehicle according to claim 12, wherein each respective incremental value consists of twice the respective standard deviation current.

Images