NL2004503C2

NL2004503C2 - Method and device for charging a battery and battery charger.

Info

Publication number: NL2004503C2
Application number: NL2004503A
Authority: NL
Inventors: Crijn Bouman
Original assignee: Epyon B V
Priority date: 2010-04-02
Filing date: 2010-04-02
Publication date: 2011-10-04
Also published as: WO2011122946A3; WO2011122946A2; TW201203659A

Description

Method and device for charging a battery and battery charger

The present invention relates to a method and device for charging a battery. More in particular the method relates to a method for shortening the charging time of (li-ion) 5 batteries with a low temperature prior to charging or at the beginning of a charging process, without damaging the battery or shortening its life.

Short charging times for batteries are desired in certain applications, for example when the batteries power an electric vehicle that needs to be used. However, short charging 10 times require large currents, which have an impact on battery life. This negative effect even increases when the battery temperature is lower, resulting for example in damage to the battery caused by an effect known as lithium plating inside the battery.

It is a goal of the present invention to provide a method and device for charging a 15 battery, in particular a lithium-ion battery that decreases the charging time at a low starting temperature, and that has a minimum impact on battery life.

It is a further goal of the present invention to provide a method that is useful to improve the battery life under a certain given charging pattern or usage scenario.

20

The invention thereto proposes a method for charging a battery, comprising heating a battery if a battery temperature is below a low-threshold temperature, and charging the battery when the battery temperature is above the low-threshold temperature.

25 The low-threshold temperature depends on the battery and may be varied in dependence of the battery type used. An average value may for example be at room temperature, i.e. about 20 degrees Celcius. By heating the battery, a larger charging current may be used, and the charging process can be performed quicker, without damaging the battery. The heating may be performed by a specific heating component such as an electric spiral or 30 the like, or in hybrid cars with an combustion engine, the battery may be incorporated in the cooling system of the combustion engine, to receive heat deducted thereof.

As an alternative, heating may comprise generating dissipation energy in the battery by applying a current with at least an AC component to the battery. Dissipation then takes 2 place in the internal resistance of the battery. This has not only the advantage that the heat is generated exactly at the location where it is required, so that a high efficiency is obtained, but it also doesn’t require any additional hardware of modifications, both in the battery, and in the charger.

5

The connection between the charger and the battery is typically a conductor. This can be charging cable, for example with conductive connector or the DC bus of a vehicle (e.g. when using an on-board charger or motor controller for charging). Also an inductive coupling can be used. Such an inductive coupling already supplies AC current, that can 10 be used to heat up the battery. For that the rectifier and or a low-pass filter may be switched on or off.

The battery may be a traction battery in particular a Li-ion battery and even more specific a Lithium Iron Phosphate (LiFeP04) battery. The battery may have a BMS that 15 may be altered or temporarily switched off or changed to a different mode to accept the PWM signal. This may be done by changing the software of the BMS. The battery may be a part of a grid storage buffer used for peak-shaving, in particular associated with a vehicle charging station.

20 The heating can also occur during, at the end and/or near the end of charging to heat up the battery for use in cold conditions, such as cold weather, cold-storage warehouse, high altitude, space, in polar regions, at night or near cooling machinery such as on an artificial ice track. Also a heated battery has less losses during use.

25 The temperature of the battery may be measured or estimated by measuring cooling water or gas of the battery, in particular the difference between entering and exiting coolant. The temperature may be estimated from other temperatures such as cabin temperature (air-conditioning sensor), ambient temperature (measured by the car) or the motor drive temperature. The latter can be an indication of motor usage and therefore 30 battery usage. The temperature may alternatively be estimated from the date or time and geographic location or based on the temperature of batteries of other vehicles.

The battery temperature information can be communicated to the controller through any means. For example the information can be communicated from the vehicle or 3 battery to the charger through the charging cable. The information can be communicated through a computer network, such as a LAN or an internet connection. A vehicle can be equipped with a wireless communication device that communicates this information directly or indirectly (e.g. through an internet server) to the charger.

5

Preferably, the average current towards the battery is positive, for charging the battery during heating. This leads to an even shorter overall charging time, and to more energy effectiveness.

10 A suitable waveform for this purpose is a one-directional DC current. The actual value of the current may be changed, as will be explained in the following, but the energy flow is only toward the battery, so that no discharge (costing extra time) takes place during heating.

15 To obtain a controllable one directional current, Pulse Width Modulation (PWM) may be applied. This waveform can usually be supplied by a standard battery charger, and it causes relatively high dissipation, thus contributing to a quick temperature rise, and as a result to a short charging time. The duty cycle of the PWM and thus the average current may then be controlled as a function of the battery temperature.

20

It is advantageous when the edges of the PWM pulses are rounded to avoid or at least reduce electromagnetic interference due to harmonics. Rounding the pulses may be done by making use of (parasitic or internal) capacitances or inductances, but more sophisticated (digital) charging equipment may be programmable to generate any 25 desired waveform.

Besides pulse PWM, also Pulse Density Modulation (PDM) can be applied, and it is even thinkable that sinusoidal pulses are used, in which again PWM (which then for example causes a phase cut signal) or PDM can be applied. The choice for a waveform 30 may be dependent on the charger used, and the sensitivity to specific harmonics of the environment of the battery to be charged.

The average current may be dependent on the temperature of the battery, which may be measured, or estimated based on a starting environment temperature and an estimated 4 cumulative dissipation. Both the desired (average) current, and a desired (maximum) temperature as an estimated cumulative dissipation may further be obtained from an external source, such as a vehicle or battery management system, or an external database with battery characteristics.

5

The dependence of the average current on the temperature may be adapted for each specific battery, but preferably at least in a temperature interval, the average current is doubled for every 5-15, and in particular for about every 10 degrees of temperature increase.

10

When the battery is above the threshold temperature, constant current charging or declining (continuous / non switched) current charging may be applied, the declining current charging for example obtained by constant voltage charging.

15 A constant-current charging at a level that is only limited by the battery of charger or grid specifications leads to the quickest energy transfer possible. Constant current charging may take place at the peak value of the PWM charging, but it is also possible that the peaks of the PWM charging are chosen above the constant current level. This is possible, since with a PWM signal, that comprises current-free intervals, components, in 20 particular those of the charger, may be operated above their maximum values, which are usually determined for constant currents, voltages or power.

According to the invention, current reduction may be applied when the battery temperature exceeds a high-threshold temperature or a predetermined voltage or a 25 predetermined state of charge, or a predetermined charging time. Current reduction usually leads to decreasing dissipation within the battery. Herewith, overtemperature is avoided during the charging process. Such a current reduction can be achieved by applying constant voltage (CV) charging or constant temperature charging.

30 The invention further proposes a battery charging apparatus, configured to perform a method as described above. The charging apparatus may be a programmable or controllable charging apparatus.

5

When applying PWM, the apparatus may be operated above its nominal level, since PWM comprises low-current timeslots, in which the components can cool down, enabling them to withstand higher peaks in the timeslots wherein current is conducted.

5 It is also possible to configure the battery charging apparatus to charge a plurality of batteries at a time, wherein a PWM signal for a second vehicle is set high when the PWM signal for a first vehicle is low. The output of the charger is then switched between the batteries to be charged. The actual current level that can be reached in this situation depends on the total duty cycle of the charger, when charging the plurality of 10 batteries. It is also possible to switch between a charging current of a battery, and a heating element for said battery. Herewith, an optimum charging time can be reached.

The invention will now be elucidated into more detail with reference to the following non limitative figures, wherein: 15 - Figures la and lb show charging currents for a battery according to the present invention;

Figures 2a and 2b show schematic overviews of a system for applying the present invention;

Figures 3a-3f show embodiments of a PWM converter according to the present 20 invention; - Figures 4a-4f show process flows of a system making use of the present invention.

Figure la shows a schematic charging curve of a current during battery charge 25 according to the present invention. The charging curve comprises the three following modes. A low temperature PWM mode A, wherein the battery is heated. Figure la schematically shows that during mode A, the current waveform does not necessarily need to be a square wave. A normal charging mode B wherein a steady charge current is used not to generate more losses than necessary. A high temperature mode C wherein 30 the battery current is reduced to prevent the battery from overheating. A steady (declining) charge current is used not to generate more losses than necessary.

The low temperature PWM mode A is typically at the beginning or at early stage of the charging process. It is intended to heat up the battery. The battery is also charged during 6 this phase. Arrhenius’s law states that chemical processes in batteries are strongly dependent on temperature. A battery that can be charged in 30 minutes at room temperature cannot be charged that fast at lower temperatures. Attempting to do so can result in damage to the battery. Heating up the battery at the beginning of charge 5 therefore enables faster charging. In particular an average charging current that is divided by roughly two for every 10°C drop in temperature gives good results. For example: at 20°C the battery is charged at 16A. At 10°C it is charged at 8A, at 0°C at 4A, at -10°C at 2A and at -20°C at 1A and so forth. The function may be approximated by using other functions, such as steps or polynomial, linear or goniometrie functions.

10

The benefit of using low temperature PWM with compared to using an AC current to heat up the battery is that this mode can be performed using components that are normally used in a charger. Therefore it requires minimal change to the system.

15 The benefit of using low temperature PWM with compared to using a large charging current to heat up the battery it that the damage to the battery will be lower when using a PWM charging.

When using a large current, the internal battery voltages will rise, damaging the battery. 20 Using the PWM charging will result in lower voltages because the average current is lower. This gives less strain on the battery when heating the battery at the same moment.

In an embodiment the pulses are rounded to reduce electro-magnetic interference. The 25 pulses may be sinusoidal. The roundness of the pulses may be generated by controlling the charge current rather than relying on a low pass filter.

In another embodiment the height of the charging pulses can be made higher than the specifications of the components (e.g. power converter, cables, charging connector), 30 because the components have a resting period to cool down due to the PWM. That way a 125A charger can be used at a 50% duty cycle at 175A, without wearing down the components more than necessary.

7

Because it can be expected that this situation will arise at low ambient temperatures, the charger can also be used at an (even higher) rating, because ambient air temperature is lower.

5 This charging mode can also be used to heat up the components of the charger or other components of the vehicle, such as the charging cable making it more flexible and easier to handle in cold weather.

It is further possible to use a charger device to switch charging current between the 10 batteries. If the batteries each only have half the capacity of one big battery, the charging current is also half. Therefore the charging connector, cables and charger can be smaller.

The frequency of the duty cycle is usually in the range of 0.01Hz - 100kHz. For 15 Lithium-ion batteries a low frequency is beneficial to prevent (paracitic) capacitance and use real Ohmic resistance as much as possible. A too low frequency will result in a too high voltage rise, damaging the battery.

The frequency of the duty cycle (the time of ton and t0ff) can also be chosen to match the 20 response time of the charger. For instance of the ramp-up time of the charger is one second a frequency of 0.1 Hz can be used.

The low temperature PWM mode is usually followed by a mode B wherein the charge current is not limited by temperature. In a typical case a steady charge current is used 25 not to generate more losses than necessary. In this mode the current may still rise with temperature, but the duty cycle is absent. This is particularly useful when the current is large enough to heat up the battery and a PWM in his case would result in too high temperature or a too high temperature for a later stage of the charging process.

30 In a third mode C the current is gradually reduced, to stop generating heat in the battery. With respect to using a PWM to reduce (average) charging current is that PWM introduces unnecessary losses in the battery, while the goal of reducing the current is to reduce losses. When temperature is the major limiting factor for the charging system, mode A can be followed by mode C directly.

8 A (part of a) charging output filter of a charger may be reconfigured when using mode A charging to enable better PWM performance. The reconfiguration can be done by bypassing parts of the filter. This can be done by using switches (e.g. relays, transistors) 5 that shunt or disconnect inductors and/or capacitors.

Another embodiment has an isolation detection circuit. This circuit detects if galvanic isolation is maintained. The isolation circuit itself is not that important, because these are readily available components. What is important: The isolation detection in the 10 vehicle or the charger is (temporarily) disabled during charge. This is because the isolation monitor may be confused by the pulsing currents.

Another embodiment uses the battery voltage, impedance or frequency response as a way to regulate the PWM. The impedance can be determined by applying an AC current 15 or AC current component to the battery. In particular the PWM itself can be used as an impedance measurement.

For example at the PWM on-phase the voltage rise is monitored. When the voltage crosses a certain threshold the current is reduced or set to zero. The voltage will then 20 fall. After a period of time or when the voltage drops below a threshold the current is switched back on again.

Figure lb shows a situation, wherein the temperature drops during the charging (for example due to a low environmental temperature), and the charging switches back to 25 PWM, as is indicated with D in curve 1’.

Figure 2a shows a first schematic overview 2 of a system according to the present invention. The simplest embodiment of this system comprises a charger 4 which is connected to a vehicle 3 and can communicate with said vehicle and more specifically 30 the BMS of the battery by means of some channel such as a wireless communications system, through the charge cable 5 (either separate contacts or communication over the same pins as the charge current) or any other system for said communication. It can then obtain the battery temperatures from a BMS, and based on these temperatures modify the charge current. In this embodiment the charger has all details and algorithms 9 concerning this low-temperature stored locally, and can autonomously decide to apply these. For other tasks such as selecting the standard charge currents and prioritizing of vehicles it can still connect to a remote server.

5 It is also possible that the charger 4 has reduced local intelligence and data storage. The decision to use the low-temperature charging algorithm is made by a remote decision and storage facility 6 (shown in dashed lines), and also all parameters for this algorithm are calculated and decided by this facility. This has as advantages that the charger can be made much cheaper, as less logics and data storage is required. Besides, performance 10 of many chargers in the field can be upgraded in one go, as the code and parameters for the algorithm only have to be replaced in one place, and not all chargers have to be upgraded separately.

Furthermore, the charger may be coupled to a temperature sensing device 7, which may 15 be integrated in the battery, or form part of a battery management system. In another embodiment, the optional temperature sensing device lacks, and the charger is unable to obtain the actual temperature from the BMS of the vehicle. In this embodiment (which might work independently from the presence of a remote server) the temperature of the battery will have to be estimated or measured based on other methods. These methods 20 may be one or a combination of the following methods: • Use the ambient temperature measured by the charger. Preferably here the estimation is backed up by statistical data which correlates environmental temperature with the batteries temperature for increased reliability. A precise model which models the battery temperature change caused by charging current is required to keep the 25 estimate accurate during the charge process; • Use some infra-red sensing device to measure the heat radiated by the battery section of a car; • If no local temperature measurement is available for whatever reason, weather measurements or forecast which could be available from a remote server can be used for 30 a temperature estimation. Also for this solution it would be preferable to have some sort of statistical relation between ambient temperature and battery temperature; • A relation can be made between internal resistance of a battery and its temperature. As soon as the type of battery is known, its temperature can be estimated preferably when its age and SOH (State of Health) are known by applying a certain 10 predetermined current and measuring the instant voltage rise (or drop) to serve as a basis for temperature estimation; • When the BMS of the car returns multiple temperature values, an average, a maximum or a minimum of these values may be used as input for the charging 5 algorithm. When the difference between these multiple temperature values is known, the algorithm may be devised in such a way that it will not overheat the warmest parts of the battery, even when the low temperature of other parts is still limiting the average charge current; • When more is known about the construction of the battery pack and the relative 10 position of the different temperatures the average mentioned above can be a weighted average, where the weighting can be based on knowledge of the sensor positions, optimality knowledge of the charging algorithm or other prior knowledge. This battery data can also be combined with ambient temperature values, for instance for averaging one temperature value from the centre of the pack with the ambient temperature to 15 obtain the average temperature in the pack.

The method to effectuate the low frequency PWM signal on the battery may be implemented in many ways, while leaving the essence of the invention intact. The following list contains possible embodiments of the PWM system: 20 • The output of a power converter with a DC output can be connected and disconnected from the battery by means of a relay.

• To save on wear said relay can be replaced by a solid-state relay, MOSFET, IGBT or other semiconductor technology.

• As said power converter may be susceptible to damage caused by 25 abovementioned switching devices the PWM signal can be generated by applying a low-power PWM signal to the control input(s) of the power converter.

• When the used power converter is embodied by an on-board charger, the idea of the invention can still be applied by switching on and off the AC current fed to the onboard charger. Care has to be taken that the on-board charger does not apply some 30 temperature correction itself as this might unnecessarily increase the charge time.

If none of the above methods for obtaining a temperature measurement are present, it is possible to derive the temperature indirectly, since the I-Y characteristic of a battery changes with its temperature. This embodiment skips the measurement of the 11 temperature and just uses a high charge current. Due to higher internal resistance, a certain voltage threshold will be reached earlier. When the power converter is switched off at this threshold, and a certain time delay is implemented before it is switched on again, the idea of temperature dependent PWM pulse charging is still intact, but 5 temperature sensing is eliminated. For determination of the current, determination of the time delay and actuation of the current ideas from aforementioned embodiments can be used without fundamentally altering the idea of the invention.

Figure 2b shows a second schematic view 8 of a system for performing the method 10 according to the present invention, wherein a DC power supply (charger) 9 is coupled to a grid or DC-Bus 10. A BMS 12 of the battery 11 senses the temperature of the battery and passes this information on to the charger 9. This charger 9 is augmented with a temperature charging controller 13. The TCC 13 identifies the BMS 12 and sets the DC power supply current and voltage. The TCC 13 decides, based on the temperature and 15 the battery type what the PWM duty cycle should be, and controls the DC power supply 9 to switch off and on accordingly. A PWM duty cycle is determined, based on an exponential decay of charge current with relation to battery temperature.

In an embodiment where multiple batteries are charged, two or more vehicles can be 20 charged with a single power source if the power for one vehicle is “on” while for the other vehicles it is “off’.

Figure 3a shows yet another schematic view 14 of a system according to the present invention. A controller 15 can be a microcontroller a (web)server, a part of the power 25 converter 16 (Such as a semi-conductor driver or electronics used to control the semiconductor drivers). The controller 15 can be an analog or digital circuit, such as a multivibrator. The multivibrator can be controlled by changing the time contant and the amplitude by for instance changing the resistor(s) or capacitor(s) value(s) depending on (battery) temperature. In particular one or more of the resistors of the multivibrator can 30 be an NTC, PTC or transistor (e.g. field effect transistor) or one or more of the capacitors can be a variable capacitor (varicap).

The PTC and NTC can be used to change the behaviour of the multivibrator dependent on (battery) temperature (for instance because the PTC or NTC is placed near or inside 12 the battery. The transistor and varicap are voltage dependent and can be used to change the behaviour of the multivibrator dependent on battery voltage, which can be an indication of the battery temperature. For that, the battery voltage may pass through an AM detection circuit. These components can be near or inside the battery 17, while the 5 rest of the circuit is near the charger (that may be outside the vehicle). The battery 17 has a temperature 18 which is determined by a temperature sensor 19 that sends the measured value to the controller 15.

Figure 3b shows how the controller 15 can be a fuzzy logic circuit or a logic circuit with 10 fuzzy logic programming. Depending on memberships of the parameters (e.g.

Temperature, voltage, Set current) a different or combined control algorithm can be implemented, for example comprising the modes A, B and C from figures la and lb.

The controller 15 may be trained using training data or real life situations. The 15 controller 15 could therefore best comprise an adaptive system, such as a neural network. The controller can be a part of a cloud computing network. It can make its own decisions and/or leave solutions to partial decisions to other computers. The controller 15 can comprise or use a database that holds information about vehicles, batteries and environmental parameters.

20

Figure 3c shows an embodiment 20 with a hardware or software structure for a system that can have blocks for the different charging modes/phases. An event (e.g. change of temperature) can make the system switch to operate in one or more of the other modes.

25 Figure 3d shows another embodiment 21 with a current control that gets input from (optional) hardware or software that changes the current based on various factors. For instance a low-temperature module can change the duty cycle of the current. Under normal temperature the duty cycle is 100% (full current) but at lower temperatures the duty cycle becomes smaller. The high temperature control may lower the current value 30 (amplitude) in case of high temperature. An high voltage control module may reduce the charging current in case of high voltage (constant voltage charging). A time target module may reduce te current if the vehicle has no urgency to leave soon. The modules for low and high temperature control may be a single module for temperature control.

13

The controller can be connected to the power converter using a network connection (e.g TCP/IP). The connection between the controller and the power converter can be analog, such as a direct PWM control signal or parametric (0-1OV for amplitude and 0-10V for duty cycle and/or frequency). The connection between the controller and the power 5 converter can be digital control such as RS232, RS485,12C, CAN or SPI. The connection between the controller and the power converter can be a PWM control that gives parametric signals (0-100% for amplitude and 0-100% for duty cycle). The controller can directly control a switch (e.g. relay or semiconductor) control signal that switches the input or output current of the power converter. This may be implemented 10 in the vehicle while the rest of the power converter is inside or outside the vehicle.

The connection between the controller and the power converter can be optical (e.g. using LDR or optocoupler). This is in particular interesting to maintain galvanic isolation. The controller may control a switch that switches the power converter between two vehicles. This way a single power converter can be used to charge multiple 15 vehicles.

The power converter can be inside a vehicle or in a charging station or another type of charger. The power converter may be a regular power supply, such as a switch-mode power supply. The power supply may be a phase control dimmer.

20

Figure 3e shows how using such a dimmer provides a basic way to change the frequency of the pulsed current (by skipping some of the phase pulses) and the amplitude.

25 An instable power supply such as aforementioned dimmer, a welding transformer or a low-cost power supply can be used to generate the PWM signal. Optionally a stabilisation stage that can be switched on or off can be added to stabilise the power when no PWM is required. The system can include a stable power supply for stable current and a signal generator of any sort to create the PWM signal. The PWM can be 30 generated by the motor controller, that may function as or as a part of the charging circuit.

14

Figure 3f shows an embodiment 22 wherein the PWM can be coupled in from the net through a capacitor or an alternative 23 with a transformer that can be switched on or off.

5 Figures 4a-4f show process flows of a system making use of the present invention.

Figure 4a shows an embodiment wherein the system can optimize the charge speed through application of a data processing methods which make trade offs between battery life, charge speed and battery temperature. A data processing device in this 10 specific embodiment decides that a battery should be heated first and then fast charged because this will deliver a faster charge than without preheating

In figure 4a the following text belongs to the corresponding boxes: 1. A vehicle connects to the energy exchange station.

15 2. The energy exchange station communicates with the vehicle and receives battery data such as state-of-charge and temperature of the battery which is very low.

3. The user of the vehicle indicates via a data entry interface that he wants to receive the fastest possible charge.

4. A battery knowledge base contains relationships between the temperature of the 20 battery, the fastest possible charge rate and the battery life and other battery related parameters.

The battery knowledge base can be a large internet based storage system, but it can also be a small software program or register in the BMS or computer of the vehicle, or a small software program somewhere in the charger.

25 5. The data processing device calculates the best charging method for this situation: it requests the relationships between battery life, charge speed and temperature from a battery knowledge base. Based on this information it decides that the battery should be heated first before commencing fast charging. It tells the energy exchange station that it should preheat the battery by applying a pulsed current for a defined period of time and 30 afterwards applying a fast charge at a defined level for several minutes.

6. The energy exchange station applies a pulsed current for several minutes followed by a fast charge at a defined level for several minutes 15

Figure 4b shows an embodiment wherein the system can optimize the battery life through application of a data processing methods which make trade offs between battery life, charge speed and battery temperature. A data processing device in this specific embodiment decides that a battery should be preheated first and then fast 5 charged because this will enable a longer battery life than without preheating

In figure 4b the following belongs to the corresponding boxes: 1. A vehicle connects to the energy exchange station.

10 2. The energy exchange station communicates with vehicle 1 and receives battery data such as state-of-charge and temperature of the battery which is very low.

3. Via a software application or database or other input method the system receives information which indicates that the battery should be treated such that it will reach a battery life of at least 3000 charge cycles.

15 4. A battery knowledge base contains relationships between the temperature of the battery, the fastest possible charge rate and the battery life and other battery related parameters.

5. The data processing device calculates the best charging method for this situation: it requests the relationships between battery life, charge speed and temperature from a 20 battery knowledge base. Based on this information it decides that the battery should be heated first before commencing fast charging because it will enable a longer battery life. It tells the energy exchange station that it should preheat the battery by applying a pulsed current for a defined period of time and afterwards applying a fast charge at a defined level for several minutes.

25 6. The energy exchange station applies a pulsed current for several minutes followed by a fast charge at a defined level for several minutes.

Figure 4c shows an embodiment wherein the system can optimize the charge speed through application of a data processing methods which make trade offs between 30 battery life, charge speed and battery temperature. A data processing device in this specific embodiment initially decides that a slow charge is enough but when input from a service provider is received it decides to speed up charging by initially applying a short battery pre-heat and then an accelerated fast charge 16

In figure 4c the following belongs to the corresponding boxes: 1. A vehicle connects to the energy exchange station.

2. The energy exchange station communicates with the vehicle and receives battery data 5 such as state-of-charge and temperature of the battery which is low.

3. Via a customer input the system receives information that the user does not need a really fast charge: one hour charge time is requested.

4. A battery knowledge base contains relationships between the temperature of the battery, the fastest possible charge rate and the battery life and other battery related 10 parameters.

5. The data processing device calculates the best charging method for this situation: it requests the relationships between battery life, charge speed and temperature from a battery knowledge base. Based on this information it decides that the battery does not need pre-heating: the temperature of the battery is low but because 1 hour charging is 15 requested it can charge at a slow speed.

6. The energy exchange system applies a charge at relatively low rate.

7. New information enters the system through a service parameter input such as a fleet management system connection. This input tells the system that the battery must be charged as quickly as possible because another vehicle will arrive within a few minutes.

20 8. Based on the new information the data processing device decides that charging has to be changed: first the battery has to be heated by a pulsed current or by using an onboard heating device and then the highest possible charge rate has to be applied.

9. The energy exchange system changes the charging method according to 8, heats the battery and applies the highest possible charge rate.

25

Figure 4d shows an embodiment wherein the system can optimize the charge speed through application of a data processing methods which make trade offs between battery life, charge speed and battery temperature. A data processing device in this specific embodiment initially decides that a slow charge is enough but when input from 30 a utility grid parameter is received it decides to speed up charging by initially applying a short battery pre-heat and then an accelerated fast charge

In figure 4d the following belongs to the corresponding boxes: 17 1. A vehicle connects to the energy exchange station.

2. The energy exchange station communicates with the vehicle and receives battery data such as state-of-charge and temperature of the battery which is low.

3. Via a customer input the system receives information that the user does not need a 5 really fast charge: one hour charge time is requested. Also the customer input tells the system that the system should optimize according to the lowest price for electricity.

4. A battery knowledge base contains relationships between the temperature of the battery, the fastest possible charge rate and the battery life and other battery related parameters.

10 5. The data processing device calculates the best charging method for this situation: it requests the relationships between battery life, charge speed and temperature from a battery knowledge base. Based on this information it decides that the battery does not need pre-heating: the temperature of the battery is low but because 1 hour charging is requested it can charge at a slow speed.

15 6. The energy exchange system applies a charge at relatively low rate.

7. New information enters the system through a grid interface parameter. The utility grid tells the system that the price of electricity has changed to the lowest possible rate and therefore it is beneficial to use as much as possible of this low cost electricity.

8. Based on the new information the data processing device decides that charging has to 20 be changed to the fastest possible rate: first the battery has to be heated by a pulsed current or by using an onboard heating device and then the highest possible charge rate has to be applied.

25

Figure 4e shows an embodiment wherein the system can optimize the charge speed through application of a data processing methods which make trade offs between battery life, charge speed and battery temperature. A data processing device in this specific embodiment initially decides that a slow charge is enough but when input from 30 a utility grid parameter is received it decides to speed up charging by initially applying a short battery pre-heat and then an accelerated fast charge

In figure 4e the following belongs to the corresponding boxes: 18 1. A vehicle connects to the energy exchange station.

3. Via a grid input it is indicated to the system that the grid can only deliver a certain 5 low amount of power at this point in time.

5. The data processing device calculates the best charging method for this situation: it 10 decides that the grid input overrules everything: the charging is done at a low speed.

6. The energy exchange system applies a charge at low rate.

7. New information enters the system through a grid interface parameter. The utility grid tells the system that the grid now can deliver a large amount of power.

8. Based on the new information the data processing device now calculates the best 15 charging method for this situation: it requests the relationships between battery life, charge speed and temperature from a battery knowledge base. Based on this information it decides that the battery has to be preheated in order to be able to absorb the maximum charge power.

9. The energy exchange system changes the charging method according to 8, heats the 20 battery and applies the highest possible charge rate.

Figure 4f shows an embodiment wherein the energy exchange station has more than one output and can optimize the charging on both outputs based on application of a data processing methods which make trade offs between battery life, charge speed and 25 battery temperature.

In figure 4f the following belongs to the corresponding boxes: 1. A vehicle 1 (containing battery 1) connects to the energy exchange station.

30 2. The energy exchange system communicates with vehicle 1 and receives battery data such as state-of-charge and temperature of the battery which is in this example 15 degrees C.

3. The user of the vehicle 1 indicates via a data entry interface that he wants to receive a fast charge in 30 minutes.

19 4. A battery knowledge base contains relationships between the temperature of the battery, the fastest possible charge rate and the battery life and other battery related parameters.

5. The data processing device calculates the best charging method for this situation: it 5 requests the relationships between battery life, charge speed and temperature from a battery knowledge base. Based on this information it decides that the battery should be heated first before commencing fast charging because it will enable a 30 minute charge.

6. The energy exchange station applies a short preheating and then a fast charge.

7. A vehicle 2 (containing battery 2) connects to the energy exchange station.

10 8. The energy exchange system also receives the information from vehicle 2.

9. The user of the vehicle 2 indicates via a data entry interface that he wants to receive a fast charge in 15 minutes.

10. Based on the new information the system decides that battery 2 also needs quick preheating. Therefore it stops charging of vehicle 1 to make all power available to 15 create a powerful pulsed charge pattern for battery 2. After a short period of pulsation it restarts charging of battery 1 and also starts fast charging battery 2.

11. The energy exchange system performs all commands under 10.

All embodiments are to be interpreted as examples only. The scope of the present 20 invention is defined in the following claims.

Claims

A method for charging a battery, comprising: - heating a battery when the battery temperature is below a lower temperature limit value; - charging the battery when the battery temperature is above the lower temperature limit value.

2. Method as claimed in claim 1, wherein the heating comprises generating dissipation energy in the battery by applying a current with at least one alternating current component to the battery.

3. Method as claimed in claim 2, wherein during the heating the average current towards the battery is positive, for charging the battery during heating.

The method of claim 1, wherein during heating, the current is a one-way direct current.

The method of any one of claims 2-4, wherein the current is a current with pulse width modulation.

6. Method according to claim 5, wherein the relative operating time of the pulse width modulation and thus of the average current is controlled as a function of the battery temperature.

A method according to any of claims 5 or 6, wherein the pulse flanks of the pulse width modulation are rounded off to prevent electromagnetic interference due to harmonics. 30

The method of claim 7, wherein the pulses are sinusoidal.

The method of any one of claims 5-8, wherein in at least one temperature interval, the average current is doubled for every 5-15 degrees, and in particular for every 10 degrees temperature increase.

A method according to any one of the preceding claims, wherein the charging of the battery, when the temperature of the battery is above the temperature limit value, comprises at least: - charging with constant charging current, or - charging with decreasing charging current, for example obtained by charging at 10 constant voltage.

The method of claim 8, wherein charging at constant charging current takes place at the peak value of charging with pulse width modulation.

A method according to claim 10 or 11, wherein charging with decreasing charging current, for example obtained by charging with constant charging voltage, is applied when the battery temperature exceeds a given upper temperature limit value or a given voltage or a given charging state or a given charging time. 20

A method according to any one of the preceding claims, wherein the current value is increased during charging.

14. A method according to any one of the preceding claims, wherein the battery temperature is measured, or estimated, based on an ambient or vehicle temperature.

A method according to any one of the preceding claims, wherein the battery temperature is measured, or estimated, based on monitoring a response of the battery to a charging or discharging current, a power interruption, or to a sequence of 30 different currents.

A method according to any one of the preceding claims, wherein the battery temperature is obtained via a data communication channel or a computer network.

17. Method as claimed in any of the foregoing claims, wherein a decision to heat the battery and the heating duration of the battery are furthermore based on a maintenance-related parameter, such as the desired maximum time for obtaining a certain predefined charge state of the battery .

18. Method as claimed in any of the foregoing claims, wherein a decision to heat the battery and the duration and other parameters for heating the battery are furthermore based on a parameter related to the mains, such as the power available at a certain point in time or a parameter related to the electricity price.

A method according to any one of the preceding claims, wherein a decision to heat the battery and the heating time of the battery are furthermore based on a parameter related to the battery, such as a desired battery life (or aspect of the life of the battery). battery, such as a desired number of charging cycles).

A method according to any one of the preceding claims, wherein a decision to heat the battery and the heating time of the battery are furthermore based on a power conversion related parameter, such as the maximum power or the maximum pulse power of a power converter.

A battery charger, adapted for carrying out a method according to any one of the preceding claims.

A battery charger according to any one of claims 5-20 and 21, wherein the pulses have a higher peak value than specified for an applied bathing device. 30

A battery charger according to claim 21 or 22, arranged for simultaneously charging a plurality of batteries, wherein a signal for pulse width modulation for a second vehicle becomes high when the signal for pulse width modulation for a first vehicle is low.

24. Battery charger according to claim 23, adapted for supplying a heating element or a battery when the pulse width modulation signal for the battery is low.

25. Battery charger as claimed in any of the claims 21-24, adapted for reading a charging profile for a battery from an external device, such as a device for vehicle or battery management, or a database with charging profiles.

A vehicle or device for energy transfer, comprising a charging device according to any one of claims 21-25.