FR3154504A1 - METHOD FOR DETECTING A DEFECTIVE STORAGE ELEMENT OF A BATTERY - Google Patents

Abstract

The present invention relates to a method for detecting a defective energy storage element among a set of storage elements of a battery system comprising a device for balancing said set, the method comprising controlling a first diagnostic mode (41) consisting of detecting a defective storage element when a first parameter (P1) is greater than a first threshold (S1), determining a period of inaction of the balancing device (3, 8), and in the event of detection that the period of inaction has a duration greater than a predetermined period of inaction, controlling a second diagnostic mode (42) consisting of determining a second self-discharge parameter (P2) for each storage element of said set during the period of inaction and detecting a defective storage element when said second parameter (P2) for the period of inaction is greater than a second self-discharge threshold (S2). Figure 1.

Description

METHOD FOR DETECTING A DEFECTIVE STORAGE ELEMENT OF A BATTERY

The field of the invention relates to a method for detecting a defective energy storage element for a battery system, in particular for systems equipped with a balancing device.

Power battery systems, particularly for an electrified vehicle, typically comprise a set of storage elements. The term energy storage element refers to an electrochemical cell or an assembly of cells that can be connected in series and/or in parallel.

As a safety measure, these systems are equipped with electrical protection functions (e.g. current monitoring and electrical contactors) as well as diagnostic functions to prevent the risk of damage to the storage elements. For electromobility applications, vehicle manufacturers generally choose to store energy in lithium-ion electrochemical cells for their better energy density. It is well known that this type of cell is particularly thermally unstable, which is why protection and diagnostic functions are essential. It is therefore necessary to periodically check the health of the cells.

One criterion for a cell's health is its ability to retain its charge. Healthy cells, therefore without the risk of thermal runaway, have a low self-discharge, generally less than 2% of their total charge capacity per month, expressed in ampere-hours or charge points. Conversely, cells considered defective will generally discharge more quickly.

In this description, an electrochemical cell is said to be defective when it is in an electrochemical state affecting its self-discharge behavior. Indeed, a cell defective by degradation of the insulation between the anode and the cathode, degradation linked to an impurity, will be characterized by an additional self-discharge resistance compared to the self-discharge resistance of a healthy cell, thus generating a faster discharge. A normal cell is considered to discharge by approximately 2% per month, which can be represented by a normal self-discharge resistance of approximately 1000 ohms. A cell is considered to be defective when its resistance is less than 500 ohms, with a risk of thermal runaway when it falls below approximately 100 ohms. The phenomenon is also considered to be progressive, the resistance will gradually decrease over several weeks, as the battery is used.

Furthermore, to optimize the useful capacity of a battery system, there are balancing devices. The state of the art includes document KR101610927B1, which describes a technique for balancing the cells of a battery that consists of determining a voltage difference between a maximum voltage and an average voltage of a set of cells, and triggering either a first passive balancing mode when the difference is small, between 100 mV and 1 volt, or an active balancing mode when the difference is large, greater than 1 volt, and the vehicle is parked or stopped for a prolonged period.

Balancing techniques reduce cell voltage differences. While necessary to optimize the payload capacity of a battery system, balancing reduces the chances of detecting defective cells undergoing self-discharge. This requires fine-tuning the detection function to enable their detection while avoiding false alarms.

There is therefore a need to overcome the aforementioned problems. One objective of the invention is to improve the functions for detecting defective cells in a battery equipped with a balancing device.

More specifically, the invention relates to a method for detecting a defective energy storage element among a set of storage elements of a battery system comprising a device for balancing said set, the detection method being implemented by a control unit of the battery system and comprising the control of a first diagnostic mode consisting of detecting a defective storage element when a first parameter representative of the charge difference between the most charged storage element and the least charged storage element is greater than a first threshold. According to the invention, the method further comprises the following successive steps of determining a period of inaction of the balancing device, and in the event of detection that the period of inaction has a duration greater than a predetermined period of inaction, of controlling a second diagnostic mode consisting of determining a second self-discharge parameter for each storage element of said assembly during the period of inaction and detecting a defective storage element when said second parameter of said storage element for the period of inaction is greater than a second self-discharge threshold.

The method according to the invention may include the following additional characteristics, alone or in combination:

- The balancing device is configured to be deactivated during a shutdown of the battery system and wherein determining the period of inactivity of the balancing device comprises measuring a period of shutdown of the battery system.

- To measure the shutdown period of the battery system, a step of recording a first instant of detection of a command to stop the battery system, and a step of recording a second instant of detection of a command to start the battery system after the shutdown command.

- The second self-discharge parameter of each storage element is determined from a first value of the voltage of each storage element at the first instant, and from a second value of said voltage at the second instant, and from the calculation of the voltage difference between the first and second values.

- The second threshold has a value dependent on the duration of the period of inaction.

- The second threshold has a value dependent on an average value of the self-discharge parameter for all storage elements measured during the period of inaction.

- The first parameter is determined from the measurement of the maximum open-circuit voltage and the minimum open-circuit voltage among all the storage elements during a start-up of the battery system.

- The predetermined period of inaction is between 15 days and 30 days.

Further contemplated is a battery system comprising a set of energy storage elements, a device for balancing said set and a control unit, wherein the control unit is configured to implement the detection method according to any one of the preceding embodiments.

An electrified vehicle with such a battery system is also planned.

A control unit is further provided comprising means specifically configured to implement the detection method according to the invention.

Further provided is a computer program comprising instructions which, when the program is executed by a control unit of a battery system, cause the latter to implement any one of the embodiments of the detection method according to the invention.

Further provided is a computer-readable recording medium comprising instructions which, when executed by a computer, cause the computer to implement any one of the embodiments of the detection method according to the invention.

The detection method has the advantage of being a fully software-based detection solution that does not require the implementation of a specific sensor. The invention takes advantage of a prolonged period of vehicle inactivity to implement a specific diagnostic mode to avoid false detections. The invention improves the safety and protection of battery systems.

Other characteristics and advantages of the present invention will appear more clearly on reading the detailed description which follows, comprising embodiments of the invention given as non-limiting examples and illustrated by the appended drawings, in which:

FIG. 1 schematically represents a battery system intended to implement the method according to the invention.

FIG. 2 is a graph illustrating the detection strategy of the first diagnostic mode according to the invention.

FIG. 3 schematically represents by a functional algorithm sequence an embodiment of the method according to the invention implementing the second diagnostic mode.

The invention applies to electrified vehicles, i.e. vehicles comprising an electric motor machine and power electronics, 100% electric or hybrid, preferably motor vehicles, but not only such as trucks, aircraft, tractors, bicycles, ships. More generally, the invention applies to any autonomous electrical system such as drones, robots or portable devices, computers, tablets, mobile phones, consoles, cameras, scanners, which are cited as a non-exhaustive list of application examples.

The invention provides a method for detecting a defective energy storage element of a battery system. The method aims to improve detection for systems equipped with a device for balancing the energy storage elements.

In FIG. 1 , a battery system 1 configured to implement the detection method is shown schematically. The system comprises a set of energy storage elements C1, C2 to Cn electrically arranged in series so as to deliver an electrical voltage to connection terminals. Each storage element C1 to Cn may comprise at least one electrochemical energy storage cell or may consist of a group of cells connected in series and/or in parallel according to the electrical specifications of the system.

An electrochemical cell is an electrical energy accumulator with two terminals and a voltage of a few volts, most often between 2.3V and 4.2V, approximately. The cells can be of the Lithium-ion type (a lithium Nickel Manganese Cobalt oxide NMC or a lithium iron phosphate LFP can be cited as examples of positive electrode active materials), Nickel Cadmium (Ni-cd), Nickel-Metal-Hydride (Ni-MH) for example. More precisely, a Lithium-ion cell is composed mainly of a porous positive electrode, a porous negative electrode, a separator and an electrolyte (which can be liquid, polymeric or solid). The operating principle of a Lithium-ion cell is based on the reversible exchange of lithium ions between the two porous electrodes.

Furthermore, the battery system comprises a control unit 2 (designated by the acronym BMS for “Battery Management System” or TBCU for “Traction Battery Control Unit”) adapted to supervise the electrical parameters 9 specific to the battery and each energy storage element by means of current and voltage sensors, such as the state of charge SOC (“State of Charge”) which designates the level of state of charge of the battery expressed by a ratio between the quantity of energy stored at a given moment and the maximum quantity of energy storable at a given moment, the open circuit voltage OCV (“Open Circuit Voltage”) expressed in Volts, the charging current expressed in Amperes, the state of health SOH (“State of Health”), which designates the aging level parameter of the battery expressing a ratio between the maximum quantity of electricity storable at a given moment and the maximum quantity of electricity storable in the new state of the battery.

Furthermore, the battery system 1 comprises a cell balancing device comprising a control means 3 and a balancing circuit 8 for each cell or group of cells of the storage elements C1 to Cn of the assembly. Balancing consists of transferring energy in order to reduce the dispersion of the state of charge of the storage elements. The balancing device may be of the passive type and comprises circuits of switches and resistors that can be controlled so as to discharge the most charged cells to homogenize the charge of the cells. Alternatively, the balancing device may be of the active type and comprises circuits of switches and capacitors arranged so as to control energy transfers between adjacent cells, or groups of cells.

The control unit 2 is capable of detecting a period of activation/inaction of the balancing for each storage element. Preferably, the balancing device is activated only when the battery system is in the started state and when the computer of the control unit of the battery system is started. The balancing is operable in the charging, discharging, relaxation and resting phases of the cells. When the battery system is stopped, the balancing device is therefore in a period of inaction.

However, in a variant, the balancing of a storage element C1 to Cn can be activated even when the battery system 1 is stopped. It is envisaged that each storage element may comprise an individual self-powered control circuit allowing discharge of the electrochemical cell even when the computer of the battery system control unit is switched off. By default, this balancing mode when the battery system is switched off is deactivated.

Among the electrical protection and safety functions, the battery system further comprises a device 4 for detecting a defective storage element among the set C1 to Cn. The detection device 4 is capable of identifying a storage element among the set comprising a defective cell from the real-time measurement of the electrical parameters 9. More precisely, the detection device 4 is configured to implement at least one diagnostic mode. Due to the action of the balancing device, the detection device 4 implements two distinct diagnostic modes in accordance with the method according to the invention.

A first diagnostic mode 41 is executed when the battery system is started, for example when the vehicle is started. This first diagnostic mode 41 consists of measuring a first parameter P1 representative of the difference in state of charge between the most charged storage element and the least charged one of the set of storage elements C1 to Cn. If this difference is greater than a first threshold S1, then at least one storage element is reported as being defective, the least charged one. The fault is reported to the control unit of the battery system and the vehicle supervisor to signal it via a warning light and/or to prohibit starting of the vehicle.

In FIG. 2 , a graph illustrates the detection strategy of the first diagnostic mode, in which the first parameter P1 is determined from the measurement of the maximum voltage Vmax and the minimum voltage Vmin among the voltages of the storage elements C1 to C100 distributed along the x-axis. The voltages are for example the no-load voltages at the start of the battery system. The set comprises 100 storage elements in this non-limiting example. On the y-axis, the individual voltage values of each storage element are represented by points. In this example, the storage element Ck is defective because the voltage deviates significantly from the other cells. This is the sign of a self-discharge greater than the other cells. The first threshold S1, represented by a double arrow, is a value in voltage difference for example between 0.2 volts and 1 volt. The first threshold S1 is determined experimentally and is stored in the memory of the control unit 2. S1 is calibrated according to the storage element balancing strategy which is implemented by the control unit 2.

In one variant, the first parameter P1 representative of the state of charge deviation can be estimated from the difference between the average value of the voltages and the minimum voltage measured, said voltages being included among the measured voltages of all the storage elements C1 to Cn. In another variant, the first diagnostic mode can consist of comparing a maximum state of charge deviation with respect to a first threshold S1. The detection strategy is then based on state of charge estimation models based on measurements of the exchanged current. The first mode is therefore operable at any time of use of the system, while driving or charging for example.

According to the invention, a second diagnostic mode 42 is activated only when a period of inaction of the balancing device is detected and if this period of inaction is greater than a predetermined duration of inaction of between fifteen days and several weeks, for example approximately four weeks. This second diagnostic mode 42 aims to be executed punctually during the life of use of the vehicle, for example only once or twice a year, unlike the first mode which is executed at each start or periodically while driving or charging, for example. This second diagnostic mode 42 is capable of determining a second self-discharge parameter P2 for each storage element C1 to Cn for the period of inaction and of detecting a defective state of a storage element when said second parameter P2 of said storage element during the period of inaction is greater than a second self-discharge threshold S2. The second threshold S2 is expressed in volts or charge variation points. The second threshold is for example between 0.1 volts and 0.3 volts for a correspondence in state of charge deviation between 3% and 5% of SOC in self-discharge for a cell rest period of approximately four weeks. The detection strategy of the second mode 42 aims to detect a cell which has a self-discharge greater than a healthy cell. For example, S2 is delivered by a table taking as input the inaction period DR and delivering a self-discharge voltage deviation value specific to the inaction period DR. The second threshold S2 is determined experimentally and is recorded in the memory of the control unit 2. In a variant, the second mode 42 estimates an average self-discharge value of a group of storage elements or of the set C1 to Cn in its entirety and compares the second self-discharge parameter P2 of each cell with respect to this average. The second threshold S2 is then a value corresponding to the average affected by an authorized deviation.

The first threshold S1 and the second self-discharge threshold S2 have differentiated values. For example, S1 can be a fixed value or a value calibrated according to the activation rate of the balancing device over an observed period and S2 is a value which is a function of the duration of the inaction period DR.

This second self-discharge threshold S2, differentiated from the first threshold S1, makes it possible to avoid false detections because the detection strategy is carried out individually for each storage element and is not affected by the measurement scatter that would result from the dispersion of measurements between the storage elements of the battery. In addition, this strategy makes it possible to calibrate the detection more sensitively because it only occurs after a period of inaction of the balancing device and individually for each cell.

The second parameter P2 representing self-discharge can be an estimate of the variation in state of charge, expressed in volts, ampere-hours or in total capacity points. The state of charge or voltage can be determined from the measurement of the cell voltage, for example using a sensor specific to the cell or a group of cells.

Furthermore, it is provided that the battery system 1 is intended to communicate with a communication bus 7, for example of the CAN (“Controller Area Network”) type in order to exchange data with external systems, for example a supervisor 5. In one embodiment, the period of inactivity can be determined from time parameters delivered by a clock 6 of the supervisor 5. Furthermore, the battery system 1 is capable of signaling the fault state of a storage element, or an alert and diagnostic message to the supervisor 5, or even ordering the temporary shutdown of a system powered by the battery system 1.

Some or all of the functions of the battery system 1 are controlled by the control unit 2. The control unit 2 may include at least one processor that executes instructions stored in a computer-readable medium such as non-volatile memory. The control unit 2 may also be multiple computing devices that control individual components or subsystems of the battery system 1 in a distributed manner. The processor may be any conventional processor, such as a commercially available central processing unit. Alternatively, the processor may be a dedicated device such as an application-specific integrated circuit (ASIC) or other hardware processor.

In FIG. 3 , an embodiment of the detection method is schematically represented by a sequence of a functional algorithm. In a first step E1, the battery system is in operation. The control unit of the battery system operates the first diagnostic mode aimed at detecting a defective cell. In this first diagnostic mode, the control unit measures, at the start of the battery system, the charge (or voltage) difference between the most charged storage element and the least charged one of all the storage elements of the battery system, and compares the difference with the first threshold S1. More precisely, this first mode consists of measuring the maximum voltage Vmax and the minimum voltage Vmin among the measured voltages of all the storage elements. The measurement of the voltage values is easily obtained by the battery system at its start. The first threshold S1 is for example between 0.2 volts and 1 volt. When the difference is greater than S1, an alert is signaled. Alternatively, the charge gap can be estimated from battery system models and estimators.

In parallel, during step E1, the control unit periodically performs a balancing of the electrical storage elements based on checks carried out periodically during its operation, for example at each start-up of the battery system.

In a second step E2, the control unit detects a request to shut down the battery system. The battery system is configured so that balancing is not active during a battery shutdown.

In a third step E3, the control unit implements steps for detecting a period of inactivity DR of the balancing system during the shutdown of the battery system, and the duration of this inactivity. For this purpose, the control unit executes the recording of a first instant t1 of the command to shut down the battery system, for example the date or any time information provided by a clock or a navigation system of the vehicle. In addition, the control unit records a first value of an electrical parameter representative of the state of charge for each storage element at this first instant t1. The electrical parameter is for example the value of the voltage of the cells at the time the vehicle is stopped, or an estimate of the state of charge.

The battery system is commanded to be off. The balancing device is configured to be inactive during the off period.

At a fourth step E4, following the system shutdown period, a start command is detected.

Then, in a fifth step E5, the control unit executes the recording of a second instant t2 of the start command of the battery system after the stop command. It also measures a second value representative of the state of charge of each storage element at the second start instant. The electrical parameter is for example the value of the voltage of the cells at the time of starting the vehicle or an estimate of the state of charge.

In a sixth step E6, the control unit checks whether the system shutdown period DR, and therefore the inaction period of the balancing device, is greater than a predetermined inaction duration D1. The duration D1 is between fifteen days and several weeks, for example thirty days. The duration is also sufficient to discriminate between a cell having a self-discharge anomaly.

If the shutdown duration DR is greater than D1, then the control unit operates, in a seventh step E7, the second diagnostic mode which consists in determining the second parameter P2 representative of the self-discharge for each storage element of said entire battery system for the inaction period DR and in detecting a defective state of a storage element when said second parameter P2 is greater than a second threshold S2. The second parameter P2 representative of the self-discharge for each cell is the voltage difference between the measured voltage V1 at the shutdown time t1 and the measured voltage V2 at the startup time t2. The second self-discharge threshold S2 is a voltage difference equivalent to twice the voltage difference resulting from the self-discharge of a healthy cell. For example, S2 is delivered by a table taking as input the inaction duration DR and delivering a self-discharge voltage difference value specific to the inaction duration DR. If a faulty cell is detected, the battery control unit issues an alert signal, for example by controlling a warning light and/or shutting down the system.

The second diagnostic mode takes advantage of the detection of an inactivity situation of the balancing device for a sufficiently long period to observe the self-discharge of each storage element individually and to compare the observed self-discharge with the second threshold. It also makes it possible to detect several defective cells because the diagnosis is carried out individually for each cell. It also has the advantage of using a more precise detection calibration than detection based on the difference between the maximum voltage and the minimum voltage of all the cells. Preferably, the second threshold S2 has a value lower than the first threshold S1. The second mode is controlled only if a sufficiently long inactivity period is detected, for example approximately four weeks. The second diagnostic mode is operated once for each storage element at the start-up time, then the first diagnostic mode is activated again in nominal operating mode.

Furthermore, if the inaction period DR is less than the predetermined inaction time D1, then the first diagnostic mode is executed at startup.

The detection method applies to the battery system comprising a device for balancing the state of charge of the storage elements, and is preferably implemented for an electrified vehicle battery system, in particular a motor vehicle. The battery system is the traction battery system powering at least one electric motor machine.

The invention is described in the above by way of example. It is understood that the person skilled in the art is able to produce different variant embodiments of the invention by associating, for example, the different characteristics above taken alone or in combination, without departing from the scope of the invention.

Claims (10)

Method for detecting a defective energy storage element among a set of storage elements (C1-Cn) of a battery system (1) comprising a balancing device (3, 8) for said set, the detection method being implemented by a control unit (2) of the battery system (1) and comprising the control (E1) of a first diagnostic mode (41) consisting of detecting a defective storage element when a first parameter (P1) representative of the charge difference between the most charged storage element and the least charged storage element is greater than a first threshold (S1), the method being characterized in that it further comprises the following successive steps:
- the determination (E6) of a period of inaction (DR) of the balancing device (3, 8),
- in the event of detection that the period of inaction (DR) has a duration greater than a predetermined period of inaction (D1), the control (E7) of a second diagnostic mode (42) consisting of determining a second self-discharge parameter (P2) for each storage element of said set during the period of inaction (DR) and detecting a defective storage element when said second parameter (P2) of said storage element for the period of inaction (DR) is greater than a second self-discharge threshold (S2). Detection method according to claim 1, wherein the balancing device (3, 8) is configured to be deactivated during a shutdown of the battery system (1) and wherein the determination (E6) of the inaction period (DR) of the balancing device (3, 8) consists of measuring a shutdown period of the battery system (1). Detection method according to claim 2, comprising, for measuring the shutdown period of the battery system (1), a step (E3) of recording a first instant (t1) of detection (E2) of a command to stop the battery system (1), and a step (E5) of recording a second instant (t2) of detection (E4) of a command to start the battery system (1) after the shutdown command. Detection method according to claim 3, in which the second self-discharge parameter (P2) of each storage element (C1-Cn) is determined from a first value (V1) of the voltage of each storage element at the first instant (t1), and from a second value (V2) of said voltage at the second instant (t2), and from the calculation of the voltage difference between the first (V1) and the second value (V2). Detection method according to any one of claims 1 to 4, in which the second threshold (S2) has a value dependent on the duration of the period of inaction (DR). Detection method according to any one of claims 1 to 4, in which the second threshold (S2) has a value dependent on an average value of the self-discharge parameter for all of the storage elements measured during the inaction period (DR). Detection method according to any one of claims 1 to 6, in which the first parameter (P1) is determined from the measurement of the maximum open-circuit voltage (Vmax) and the minimum open-circuit voltage (Vmin) among all the storage elements during a start-up of the battery system (1). Detection method according to any one of claims 1 to 7, in which the predetermined duration of inaction (D1) is between 15 days and 30 days. Battery system (1) comprising a set of energy storage elements (C1, Cn), a balancing device (3, 8) for the storage elements (C1-Cn) of said set and a control unit (2), characterized in that the control unit (2) is configured to implement the detection method according to any one of claims 1 to 8. An electrified vehicle comprising a battery system according to claim 9.