SENSOR FAULT DETECTION USING SAMPLE PAIR CORRELATION

Abstract

A method for current sensor fault detection comprising: Receiving a sequence of battery voltage samples from a voltage sensor, which is configured to measure a battery voltage of a battery, and a sequence of battery current samples from a current sensor which is configured to measure a battery current of the battery; Determining a change in the battery voltage samples over a predetermined number of samples and a change in the battery current samples over the predetermined number of samples; Verify whether a ratio of the change in the battery voltage samples and the change in the battery current samples is within an expected range for (i) a resistance of the battery or (ii) a conductivity of the battery; and Detecting a fault in the current sensor based on whether the ratio is within the expected range; characterized in that determining the change in battery voltage samples and the change in battery current samples further comprises: Determining the change in the battery voltage samples as a difference between a value of the battery voltage samples at a first time step and a value of the battery voltage samples at a second time step, wherein the second time step is a first predetermined number of time steps after the first time step; and Determining the change in the battery current samples as a difference between a value of the battery current samples at a third time step and a value of the battery current samples at a fourth time step, wherein the fourth time step is the first predetermined number of time steps after the third time step, wherein the third time step is a second predetermined number of time steps before or after the first time step.

Description

REGIONThe present invention relates to a method for current sensor fault detection and to a corresponding battery system according to the preamble of independent claims 1 and 10, as known from US 2009 / 0 128 157 A1.US 2015 / 0 066 262 A1 discloses a system for current sensor fault detection which enables detection of a current sensor fault while the current sensor is operating and provides a current sensor output which is within a normal operating range. The current sensor is monitored by at least one controller that compares a change in battery state of charge error to a predetermined threshold. The change in battery state of charge error is a difference between a first change in battery state of charge calculated by integrating battery current over time and a second change in battery state of charge calculated based on the battery open-circuit voltage over time.BACKGROUNDBattery systems often include one or more sensors for monitoring parameters of the battery, such as current and voltage during operation. In many applications, reliable monitoring of battery parameters is critical to safe and efficient operation of the battery system. In certain automotive applications, on-board diagnostics (OBD) regulations require a two-sided plausibility check to detect current sensor faults. The current prior art for detecting faults in a current sensor of a battery system requires incorporating a second, redundant current sensor. The system compares values of the two different current sensors. If the values of the sensors deviate too much, an error is assumed by the system. Accordingly, it would be advantageous to provide a system and method for detecting current sensor faults in a battery system with only one current sensor.SUMMARYThe method for current sensor fault detection according to the invention is disclosed in claim 1.The battery system according to the invention is disclosed in claim 10. Preferred refinements are the subject matter of the respective dependent claims.BRIEF DESCRIPTION OF THE DRAWINGSThe foregoing aspects and other features of the system and method for detecting a sensor fault in a battery system are explained in the following description, which is to be understood in conjunction with the accompanying drawings. FIG. 1 shows a battery system according to the disclosure. FIG. 2 shows a method for detecting a current sensor fault in a battery system. FIG. 3 shows example sequences of battery voltage samples and battery current samples.DETAILED DESCRIPTIONFor the purpose of promoting an understanding of the principles of the disclosure, reference will now be made to the embodiments illustrated in the drawings and described in the following written description. It should be understood that no limitation on the scope of the disclosure is thereby intended. It is further understood that the present disclosure includes alterations and modifications to the illustrated embodiments and includes other applications of the principles as would normally occur to one skilled in the art to which this disclosure pertains.FIG. 1 shows a battery system 100 according to the disclosure. In at least one embodiment, battery system 100 is integrated into a vehicle, such as an electric vehicle or a full / mild / micro hybrid electric vehicle. The battery system 100 includes a battery 102 operatively connected and configured to provide power to loads 104. In at least one embodiment, battery 102 includes a plurality of individual battery cells 106 connected together in series and / or in parallel to provide an output voltage of battery 102 (e.g., 12 volts, 48 volts, 200+ volts) between a positive battery terminal 108 and a negative battery terminal 110. The battery cells 106 may include any of various types of battery cells, such as nickel metal hydride or lithium ions.The loads 104 are connected to the battery 102 and configured to receive power from the battery 102 during operation. In some embodiments, the loads 104 are connected to the positive battery terminal 108 via a high-side switch 112 and to the negative battery terminal 110 via a low-side switch 114. The switches 112 and 114 may include electromechanical switches such as relays or contactors, or electronic switches such as power metal oxide semiconductor field effect transistors (power MOSFETs) or insulated gate bipolar transistors (IGBTs). In some embodiments, the loads 104 may include a vehicle traction motor or a vehicle electrical system. In some embodiments, the battery system 100 may include various power electronics elements (not shown) connected between the battery 102 and make or all of the loads 104, such as a DC-DC converter configured to down-convert the battery voltage to that of the vehicle electrical system.The battery system 100 further includes a current sensor 116, a voltage sensor 118, and a temperature sensor 120. The current sensor 116 is configured to measure a battery current I flowing through the battery 102. In one embodiment, current sensor 116 includes a shunt resistor that is disposed in series with battery 102 and that provides a voltage that is proportional to battery current I. In one embodiment, current sensor 116 includes a Hall effect sensor disposed in series with battery 102 and configured to measure battery current I. The voltage sensor 118 is connected in parallel with the battery 102 and is configured to measure a battery voltage U across the positive and negative battery terminals 108 and 110 of the battery 102. In some embodiments, voltage sensor 118 is further configured to measure voltages of individual cells 106 of battery 102. The temperature sensor 120 is configured to measure a battery temperature T of the battery 102. In one embodiment, the temperature sensor 120 includes several individual temperature sensors and the measured battery temperature T may be a minimum, a maximum, or an average of some measurements from the different sensors.The battery system 100 further includes a controller 122 operatively connected to the current sensor 116, the voltage sensor 118, and the temperature sensor 120. The controller 122 is configured to receive measurements of the battery voltage U and the battery current I from the current sensor 116 and the voltage sensor 118. The controller 122 is configured to detect at least sensor faults of the current sensor 116. The controller 122 generally includes at least one processor and at least one associated memory having stored therein program instructions executed by the at least one processor to achieve the described functionalities. It will be understood by one of ordinary skill in the art that a "controller" or "processor" includes any hardware system, mechanism, or component that processes data, signals, or other information. The controller 122 may include a system having a central processing unit, multiple processing units, or dedicated circuitry to achieve particular functionality. In some embodiments, the controller 122 is a battery management system or component thereof configured to provide other functions in addition to sensor fault detection. In some embodiments, the controller 122 is operatively connected to the switches 112 and 114 and configured to command the switches 112 and 114 to open and close. In some embodiments, the controller 122 is operatively connected to an output device 124 and configured to operate the output device 124 to generate an audible or visual alarm. The output device 124 may include a speaker, a light, a display screen, or the like.Various methods for detecting a sensor fault in the battery system are described below. In describing the methods, statements that the method performs any task or function refer to a controller or general purpose processor executing program instructions stored in non-transitory computer readable storage media operatively connected to the controller or processor to manipulate data or operate one or more components in the system 100 for performing the task or function. In particular, the above controller 122 may be such a controller or such a processor, and the executed program instructions may be stored in a memory. In addition, the steps of the method may be performed in any feasible chronological order, regardless of the order shown in the figures or the order in which the steps are described.FIG. 2 shows a method 200 for detecting a sensor fault in a battery system, such as the battery system 100. The method 200, referred to herein as a "sample pair correlation" method, judges a degree of correlation between two sensor signals whose relationship within small regions is approximately linear by comparing their relative behavior (i.e., a relative change of the signals with time, as opposed to a relative magnitude). In this way, the method of detecting a gain error in the one of the sensors is most helpful. In addition, the method 200 is described herein as being directed to a battery system having a single current sensor and a voltage sensor and for the purpose of detecting a boost fault in a current sensor. However, the method can equally be applied to other applications with sensors measuring correlated signals.The method 200 begins with a step of receiving a sequence of battery voltage samples and a sequence of battery current samples (block 210). In particular, with respect to the embodiments described herein, the current sensor 116 is configured to generate a sequence of measurement samples with respect to a battery current I flowing through the battery 102. Similarly, voltage sensor 118 is configured to generate a sequence of measurement samples with respect to a battery voltage U across battery terminals 108 and 110 of battery 102. The controller 122 is configured to receive the sequence of battery current samples from the current sensor 116 and the sequence of battery voltage samples from the voltage sensor 118. In some embodiments, the current sensor 116 and voltage sensor 118 may instead be configured to generate analog measurement signals that are sampled by the controller 122 to generate the sequence of battery current samples and the sequence of battery voltage samples.FIG. 3 shows example sequences of battery voltage and current samples. In particular, the graph 302 shows a sequence of battery voltage samples 304 generated by the voltage sensor 118 and received by the controller 122. Similarly, graph 306 shows a sequence of battery current samples 308 generated by current sensor 116 and received by controller 122. As shown, the battery voltage samples 304 and the battery current samples 308 span multiple time steps from t=0 to t=18. in at least one embodiment, the specific duration of time between time steps is a function of the sampling rate of the current sensor 116 and the voltage sensor 118. In some embodiments, current sensor 116 and voltage sensor 118 are configured with a common sampling rate and the time steps coincide with the common sampling period of current sensor 116 and voltage sensor 118. In other embodiments, one of the sensors 116 and 118 has a sampling rate that is a multiple of the other of the sensors 116 and 118, and the time step may match the longer of the sampling periods of the sensors 116 and 118 or otherwise match a common multiple of the sampling periods, such as the smallest common multiple.Returning to FIG. 2, the method 200 continues with a step of determining a change in battery voltage samples and a change in battery current samples (block 220). In particular, with respect to the embodiments described in detail herein, the controller 122 is configured to determine a change in the battery voltage ΔU over a predetermined number of samples and a change in the battery current ΔI over the predetermined number of samples. More specifically, in at least one embodiment, the controller 122 is configured to calculate the change in battery voltage ΔU based on the expression ΔU=U s- U s-w where s is the index for current time step and w is the predetermined width of the sampling window that is a positive integer of one or greater. Similarly, in at least one embodiment, controller 122 is configured to calculate the change in battery current ΔI based on the expression ΔI=I s-d- I s-w-d where d is a delay of current relative to voltage to compensate for any phase difference between battery current I and battery voltage U, which is a positive or negative integer or zero. For example, d=1 will correlate voltage sample U s with previous current sample I s-1 and d=-2 will correlate voltage sample U s with a subsequent current sample I s+2. The phase difference between the battery current I and the battery voltage U may be a result of small differences in latency between the current sensor 116 and the voltage sensor 118, as well as any reactances in the battery circuit. FIG. 3 illustrates an example battery voltage window 310 and battery current window 312, where current time step s=t=10 and window width w=2. Also, as shown, d = -3 to compensate for a delay in battery current I by three time steps relative to battery voltage U.The method 200 continues with a step of checking whether a ratio of the change in battery voltage samples and the change in battery current samples is within an expected range for (i) a resistance of the battery or (ii) a conductivity of the battery (block 230). In particular, with respect to the embodiments described in detail herein, the controller 122 is configured to calculate a ratio of the change in the battery voltage ΔU and the change in the battery current ΔI. In at least one embodiment, controller 122 is configured to calculate the ratio according to expression, but may also calculate the inverse ratio according to expression.The controller 122 is configured to compare the ratio to an estimated resistance R of the battery 102 if the ratio is formed or an estimated conductivity of the battery 102 if the ratio is formed. In some embodiments, the controller 122 is configured to receive a temperature measurement T from the temperature sensor 120 and determine an estimated resistance R of the battery 102 based on the measured temperature T using a mathematical model of the battery 102 or a resistance-temperature lookup table stored in an associated memory.If the ratio is not sufficiently similar to the estimated resistance or conductivity of the battery 102, then a fault in either of the sensors 116 and 118 may be suspected. In particular, the relative behavior of the battery current and the battery voltage is mainly determined by Ohm's law. Accordingly, when the sensors 116 and 118 are functioning properly, the ratio is expected to correspond to the internal resistance or conductivity of the battery 102. In some embodiments, the controller 122 is configured to determine whether the ratio is within a predetermined range of the estimated resistance R, or alternatively, whether the ratio is within a predetermined range of the estimated conductivity. More specifically, in one embodiment, the controller 122 is configured to determine whether the ratio is within an estimated range for the resistance R according to the expression where δ n is the error of the estimated resistance R in the negative direction (e.g., -50%), and where δ p is the error of the estimated resistance R in the positive direction (e.g., 59%). Alternatively, if the ratio is formed, the controller 122 is configured to determine whether the ratio is within an estimated range for the conductivity according to the expression. In one embodiment, the controller 122 is configured to periodically adjust the error δ n and / or the error δ p over time as a function of the measured temperature T, a performance of the temperature sensor 120, cell aging of the battery 102, a pulse profile, the state of charge of the battery 102, a polarity of the battery current I, and / or an uncompensated phase difference between the battery voltage U and the battery current I. Additionally, in some embodiments, the controller 122 is configured to adjust the error δ n and / or the error δ p as a function of the manufacturing variability of the battery 102 as a constant or one-time adjustment.In some embodiments, the controller 122 is configured to detect a fault in one of the current sensor 116 or the voltage sensor 118 in response to the ratio being outside the estimated range for the resistance R or the conductivity. In at least one embodiment, faults of voltage sensor 118 are detected using other detection processes and controller 122 is configured to detect a fault of current sensor 116 in response to ratio being outside of estimated range for resistance R or conductivity. In such embodiments, the controller 122 may be further configured to perform any type of ameliorating operation in response to detecting the sensor fault. In one embodiment, the controller 122 is configured to actuate the switches 112 and 114 to open in response to detecting the fault, thereby disconnecting the terminals 108 and 110 of the battery 102 from the loads 104. In one embodiment, the controller 122 is configured to operate the output device 124 to generate an audible or visual alarm in response to detecting the fault, thereby alerting a user of the detected sensor fault. In one embodiment, the controller 122 is configured to transmit a signal indicative of the detected sensor fault to a higher level controller for further processing, such as on-board diagnostics (OBD) evaluations using a monitoring performance ratio during use (IUMPR) or other real-time monitoring techniques. In one embodiment, the controller 122 is configured to perform the on-board diagnostic (OBD) evaluations themselves based on the detection of the sensor fault.However, in some embodiments, the method 200 continues with a step of storing whether the ratio is outside the expected range of the one of (i) the resistance of the battery and (ii) the conductivity of the battery (block 240). In particular, with respect to the embodiments described in detail herein, the controller 122 is configured to store whether the ratio is outside the estimated range for the resistance R or the conductivity. In one embodiment, the controller 122 is configured to increment a counter (or skip a step of incrementing a counter) in response to the ratio being outside the estimated range for the resistance R or the conductivity. In this way, the controller 122 is configured to count a number of times that the check fails and / or succeeds. In one embodiment, the controller 122 is configured to store the result of the comparison (e.g., pass, fail, inside or outside) in association with the respective time step.The method 200 proceeds by repeating the steps of determining the change in battery voltage samples and the change in battery current samples (block 220), checking whether the ratio of the change in battery voltage samples and the change in battery current samples is within the expected range for the one of (i) the resistance of the battery and (ii) the conductivity of the battery (block 230), and storing whether the ratio is outside the expected range of the one of (i) the resistance of the battery and (ii) the conductivity of the battery (block 240) for successive time steps of the battery voltage and battery current samples. In particular, the controller 122 is configured for a plurality of consecutive repetitions to increase the time step s by at least one time step and compare a ratio of the change in the battery voltage and the change in the battery current with a known resistance or a known conductivity. In particular, in at least one embodiment, controller 122 is configured for multiple consecutive iterations to increase time step s by one and re-assess expression or its inverse, as discussed above. After each repetition, the controller 122 is configured to increment a counter based on the result or otherwise store the result in association with the respective time step.After a predetermined number of repetitions, the method 200 continues with a step of determining a number of times that the ratio was outside the expected range during the predetermined number of repetitions (block 250). Specifically, after a predetermined number of consecutive repetitions, the controller 122 is configured to determine a number of times that the ratio is outside the estimated range for the resistance R or the conductivity during the predetermined number of consecutive repetitions. In embodiments where a counter was incremented in response to each time the ratio was outside the estimated range for resistance R or conductivity, controller 122 is configured to read a value from the counter to determine the number of times. In other embodiments, the controller 122 is configured to read the results associated with the previous time steps of the predetermined number of consecutive repetitions from the memory and count a number of times that the ratio was outside the estimated range for the resistance R or the conductivityIf the determined number of times exceeds a predetermined threshold, the method 200 continues with a step of detecting a fault in the current sensor in response thereto (block 260). In particular, the controller 122 is configured to detect a fault in one of the current sensor 116 or the voltage sensor 118 in response to the determined number of times the ratio was outside the estimated range for the resistance R or the conductivity exceeding a predetermined threshold (e.g., 20%). In at least one embodiment, faults of voltage sensor 118 are detected using other detection processes, and controller 122 is configured to detect a fault of current sensor 116 in response to the determined number of times exceeding predetermined threshold (e.g., 20%). In one embodiment, the controller 122 is configured to periodically adjust the predetermined threshold over time as a function of the measured temperature T, a performance of the temperature sensor 120, cell aging of the battery 102, a pulse profile, the state of charge of the battery 102, a polarity of the battery current I, and / or an uncompensated phase difference between the battery voltage U and the battery current I. Additionally, in some embodiments, the controller 122 is configured to adjust the predetermined threshold as a function of manufacturing variability of the battery 102, as a constant or one-time adjustment.In some embodiments, in response to detecting the sensor fault, the controller 122 may be further configured to perform any type of improving operation. In one embodiment, the controller 122 is configured to actuate the switches 112 and 114 to open in response to detecting the fault, thereby disconnecting the terminals 108 and 110 of the battery 102 from the loads 104. In one embodiment, the controller 122 is configured to operate the output device 124 to generate an audible or visual alarm in response to detecting the fault, thereby alerting a user of the detected sensor fault. In one embodiment, the controller 122 is configured to transmit a signal indicative of the detected sensor fault to a higher level controller for further processing, such as on-board diagnostics (OBD) evaluations using a monitoring performance ratio during use (IUMPR) or other real-time monitoring techniques. In one embodiment, the controller 122 is configured to perform the on-board diagnostic (OBD) evaluations themselves based on the detection of the sensor fault.After detecting or not detecting a fault in the current sensor, the method 200 returns to the process of repeating the steps of determining the change in battery voltage samples and the change in battery current samples (block 220), checking whether the ratio of the change in battery voltage samples and the change in battery current samples is within the expected range for the one (i) of the resistance of the battery and (ii) the conductivity of the battery (block 230), and storing whether the ratio is outside the expected range of the one (i) of the resistance of the battery and (ii) the conductivity of the battery (block 240) for successive time steps of the battery voltage and battery current samples until a predetermined number of repetitions have been made.In some embodiments, the method 200 further includes a step of determining whether the following boundary conditions are met: (1) the battery temperature exceeds a minimum temperature threshold, (2) the change in battery voltage samples exceeds a minimum voltage change threshold, and / or (3) the current value of the battery current samples exceeds a minimum current threshold (block 270). In particular, before checking whether the ratio is outside the estimated range for the resistance R or the conductivity, the controller 122 is configured to determine whether certain boundary conditions are met. If the boundary conditions are not satisfied, the controller 122 is configured to skip the step of checking whether the ratio is out of the estimated range for the resistance R or the conductivity, and simply proceed to the next time step of the process. In one embodiment, a boundary condition is that the current battery temperature T is greater than a minimum temperature threshold T min or in other words, the expression T>T min needs to be satisfied to avoid failure due to larger changes in internal cell resistance at lower temperatures. In one embodiment, a boundary condition is that the change in battery voltage ΔU is greater than a minimum change in battery voltage threshold ΔU min or, in other words, the expression ΔU>ΔU min needs to be satisfied in order to avoid checks with minimum changes in battery voltage. In one embodiment, a boundary condition is that the current battery current I s is greater than a minimum battery current threshold I min or in other words, the expression I s > I min needs to be satisfied to avoid low current checks if necessary. In one embodiment, the controller 122 is configured to determine whether each of the boundary conditions T>T min, ΔU>ΔU min and I s>I min are met, and check whether the ratio is outside the estimated range for the resistance R or the conductivity only in response to all of the boundary conditions being met. In one embodiment, the controller 122 is configured to periodically adjust the minimum temperature threshold T min, the minimum change in battery voltage threshold ΔU min and / or the minimum battery current threshold I min over time as a function of the measured temperature T, a performance of the temperature sensor 120, cell aging of the battery 102, a pulse profile, the state of charge of the battery 102, a polarity of the battery current I, and / or an uncompensated phase difference between the battery voltage U and the battery current I. Additionally, in some embodiments, the controller 122 is configured to adjust the minimum temperature threshold T min, the minimum change in the battery voltage threshold ΔU min and / or the minimum battery current threshold I min as a function of a manufacturing variability of the battery 102 as a constant or one-time adjustment. The sample pair correlation method described herein improves the operation of the battery system 100 by enabling the controller 122 to detect faults of the current sensor 116 without the need for a secondary current sensor. Furthermore, in battery systems that include a secondary current sensor, the sample pair correlation method allows the controller 122 to provide further redundancy by detecting errors, particularly gain errors, of the current sensor 116. In contrast to making other methods, the sample pair correlation method disclosed herein requires only a small amount of data to begin producing results (e.g., two samples), which is advantageous for real-time applications such as fault detection in a battery system. Furthermore, the low computational cost of each correlation check allows the method to run continuously on low performance, low cost hardware. The sample pair correlation method is easily calibrated due to the intuitive nature of calibration variables (e.g., the predetermined pass / fail threshold or values for parameters δ n, δ p, T min, ΔU min and / or I min, discussed above) and the low interdependency between calibration variables. The sample pair correlation method advantageously comprises logic that is more easily implemented because each type of calculation is performed either for each new sample (for which the boundary conditions are fulfilled) or each time the evaluation is performed at a higher level. This is in contrast to algorithms in which some calculations are performed or not performed depending on the result of other calculations.

Claims

A method for current sensor fault detection, comprising: receiving a sequence of battery voltage samples from a voltage sensor configured to measure a battery voltage of a battery and a sequence of battery current samples from a current sensor configured to measure a battery current of the battery; determining a change in the battery voltage samples over a predetermined number of samples and a change in the battery current samples over the predetermined number of samples; checking whether a ratio of the change in the battery voltage samples and the change in the battery current samples is within an expected range for (i) a resistance of the battery or (ii) a conductivity of the battery; and detecting a fault in the current sensor based on whether the ratio is within the expected range; The method of determining the change in the battery voltage samples and the change in the battery current samples further comprises: determining the change in the battery voltage samples as a difference between a value of the battery voltage samples at a first time step and a value of the battery voltage samples at a second time step, wherein the second time step is a first predetermined number of time steps after the first time step; and determining the change in the battery current samples as a difference between a value of the battery current samples at a third time step and a value of the battery current samples at a fourth time step, wherein the fourth time step is the first predetermined number of time steps after the third time step, wherein the third time step is a second predetermined number of time steps before or after the first time step.The method of claim 1, further comprising: receiving a battery temperature measurement from a temperature sensor configured to measure a temperature of the battery; and determining the expected range based on the battery temperature measurement.The method of claim 1, wherein the detecting further comprises: detecting the error based on the ratio being outside the expected range.The method of claim 1, further comprising: repeating the steps of determining and checking for successive time steps of the battery voltage samples and battery current samples; storing for each repetition of the check whether the ratio is outside the expected range; determining after a predetermined number of repetitions of the check a number of times the predetermined number of repetitions that the ratio was outside the expected range; and detecting the fault in response to the number of times exceeding a predetermined threshold.The method of claim 4, further comprising: adjusting the expected range and / or the predetermined threshold based on (i) a cell aging of the battery, (ii) a state of charge of the battery, (iii) a polarity of the battery current, and / or (iv) a phase difference between the battery current and the battery voltage.The method of claim 1, wherein the checking further comprises: performing the checking only if (i) a battery temperature of the battery exceeds a minimum temperature threshold, (ii) the change in the battery voltage samples exceeds a minimum voltage change threshold, and / or (iii) a current value of the battery current samples exceeds a minimum current threshold.The method of claim 6, further comprising: adjusting the minimum temperature threshold, the minimum voltage change threshold, and / or the minimum current threshold based on (i) cell aging of the battery, (ii) a state of charge of the battery, (iii) polarity of the battery current, and / or (iv) phase difference between the battery current and a battery voltage.The method of claim 1, further comprising: operating an output device to generate (i) an audible alarm or (ii) a visual alarm in response to detecting the fault.The method of claim 1, further comprising: actuating at least one switch in response to detecting the fault to disconnect the battery from at least one load.A battery system comprising: a battery operatively connected to provide power to at least one load; a voltage sensor configured to measure a battery voltage of the battery; a current sensor configured to measure a battery current of the battery; and a controller operatively connected to the voltage sensor and the current sensor, the controller configured to: receive a sequence of battery voltage samples from the voltage sensor and a sequence of battery current samples from the current sensor; determine a change in the battery voltage samples over a predetermined number of samples and a change in the battery current samples over the predetermined number of samples; checking whether a ratio of the change in the battery voltage samples and the change in the battery current samples is within an expected range for (i) a resistance of the battery or (ii) a conductivity of the battery; and detecting the fault in the current sensor based on whether the ratio is within the expected range; characterized in that the controller is further configured to: determine the change in the battery voltage samples as a difference between a value of the battery voltage samples at a first time step and a value of the battery voltage samples at a second time step, wherein the second time step is a first predetermined number of time steps after the first time step; determining the change in the battery current samples as a difference between a value of the battery current samples at a third time step and a value of the battery current samples at a fourth time step, wherein the fourth time step is the first predetermined number of time steps after the third time step, wherein the third time step is a second predetermined number of time steps after the first time step.The battery system of claim 10, further comprising: a temperature sensor configured to measure a temperature of the battery, wherein the controller is further configured to: receive a battery temperature measurement from the temperature sensor; and determine the expected range based on the battery temperature measurement.The battery system of claim 10, wherein the controller is further configured to: detect the fault in response to the ratio being outside the expected range.The battery system of claim 10, wherein the controller is further configured to: repeat the determination and the checking for successive time steps of the battery voltage samples and battery current samples; store, for each repetition of the checking, whether the ratio is outside the expected range; determine, after a predetermined number of repetitions of the checking, a number of times of the predetermined number of repetitions that the ratio was outside the expected range; and detect the fault in response to the number of times exceeding a predetermined threshold.The battery system of claim 10, wherein the controller is further configured to: adjust the expected range and / or the predetermined threshold based on (i) a cell aging of the battery, (ii) a state of charge of the battery, (iii) a polarity of the battery current, and / or (iv) a phase difference between the battery current and a battery voltage.The battery system of claim 10, wherein the controller is further configured to: perform the check only if (i) a battery temperature of the battery exceeds a minimum temperature threshold, (ii) the change in the battery voltage samples exceeds a minimum voltage change threshold, and / or (iii) a current value of the battery current samples exceeds a minimum current threshold.The battery system of claim 15, wherein the controller is further configured to: adjust the minimum temperature threshold, the minimum voltage change threshold, and / or the minimum current threshold based on (i) cell aging of the battery, (ii) a state of charge of the battery, (iii) a polarity of the battery current, and / or (iv) a phase difference between the battery current and a battery voltage.The battery system of claim 10, wherein the controller is further configured to: operate an output device to generate (i) an audible alarm or (ii) a visual alarm in response to detecting the fault.The battery system of claim 10, wherein the controller is further configured to: actuate at least one switch in response to detecting the fault to disconnect the battery from the at least one load.

Images