BATTERY CONDITION DETERMINATION

Abstract

An energy conversion arrangement configured to convert chemical energy into electrical energy. The energy conversion arrangement comprises plural cell groups 30, 32, 34, 36, each cell group being configured to convert chemical energy into electrical energy. The energy conversion arrangement also comprises at least one measurement arrangement 38, 40, 42, 44, 46 configured to make measurements at each of the plural cell groups 30, 32, 34, 36. Each energy conversion arrangement is configured to determine a condition of at least one of: each of the plural cell groups; and the energy conversion arrangement. The condition is determined in dependence on the measurements made at each cell group and a model of each cell group.

Claims

1. An energy conversion arrangement configured to convert chemical energy into electrical energy, the energy conversion arrangement comprising: plural cell groups, each cell group being configured to convert chemical energy into electrical energy; and at least one measurement arrangement configured to make measurements at each of the plural cell groups, the energy conversion arrangement being configured to determine a condition of at least one of: each of the plural cell groups; and the energy conversion arrangement, characterised in that the condition is determined in dependence on the measurements made at each cell group and a model of each cell group.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0044] Further features and advantages of the present invention will become apparent from the following specific description, which is given by way of example only and with reference to the accompanying drawings, in which:

[0045] FIG. 1 shows an Electric Vehicle comprising an energy conversion arrangement according to the present invention during charging;

[0046] FIG. 2 is a block diagram representation of an energy conversion arrangement according to the present invention;

[0047] FIG. 3 is a block diagram representation of circuitry of the energy conversion arrangement which is located at a cell block;

[0048] FIG. 4 is a block diagram representation showing cell group models and a battery model; and

[0049] FIG. 5 illustrates a cycle of use of a battery cell.

DESCRIPTION OF EMBODIMENTS

[0050] A Battery Electric Vehicle (BEV) 10 comprising an energy conversion arrangement 12 in the form of an electric battery according to the present invention is shown in FIG. 1 while the electric battery is being charged. As can be seen from FIG. 1, the BEV 10 further comprises charging apparatus 14 which is electrically coupled to the electric battery 2 and which is operative to control the charging of the electric battery. In accordance with known practice the charging apparatus 14 is electrically connected to a vehicle charging point 16. The vehicle charging point 16 is in communication with a remote server 18. Data is communicated between the vehicle charging point 16 and the server 18 as is described further below. Plural further vehicle charging points 20 are provided at respective different locations with each further vehicle charging point 20 also being in communication with the server 18 whereby data from each of the vehicle charging points 16, 20 is conveyed to, stored in and operated upon by the server 18. FIG. 1 shows an arrangement involving charging by way of a wired coupling between the charging point 16 and the energy conversion arrangement 12. According to an alternative approach charging is wireless by way of an inductive coupling between the charging point 16 and the energy conversion arrangement 12. The design of an inductively coupled wireless charging arrangement will be within the ordinary design capabilities of the person of ordinary skill in the art. As can be seen from FIG. 1, the electric battery 12 comprises sixteen blocks 22 of lithium-ion battery cells (which each constitute a cell group) which are connected in series. Each block 22 comprises plural lithium-ion battery cells which are connected in parallel. In an alternative form each block comprises solely one battery cell it being noted that the present invention is equally applicable when solely one battery cell is used instead of plural battery cells. The electric battery 12 further comprises a battery management arrangement 24 which is described further below with reference to FIGS. 2 to 4.

[0051] A block diagram representation of four series connected blocks of lithium-ion battery cells and associated circuitry (which together constitute an energy conversion arrangement) is shown in FIG. 2. The arrangement of FIG. 2 comprises first, second, third and fourth blocks of cells 30, 32, 34, 36. The arrangement of FIG. 2 further comprises first, second, third and fourth measurement and processing circuitry 31, 33, 35, 37 which are each at a respective one of the first, second, third and fourth blocks of cells 30, 32, 34, 36. The first, second, third and fourth measurement and processing circuitry 31, 33, 35, 37 are substantially the same as one another. One of the first, second, third and fourth measurement and processing circuitry 31, 33, 35, 37 is shown in more detail in FIG. 3. The measurement and processing circuitry of FIG. 3 comprises a low value resistor 38 in series with each block which is operative to sense the current drawn from the block and which forms part of a voltage divider arrangement (not shown) which is operative to convert the drawn current to a measurable voltage. Measurement is to 12 or 14 bit accuracy with the actual accuracy to which the current is determined being less than afforded by 12 or 14 bit measurement on account of the effect of self heating. A first analogue-to-digital converter 40 is operative to convert the voltage from the voltage divider arrangement to corresponding digital data. The measurement and processing circuitry of FIG. 3 also comprises a second analogue-to-digital converter 42 which is operative to convert the voltage across the block to corresponding digital data. The voltage across the block is measured to within 1 mV. The measurement and processing circuitry of FIG. 3 also comprises a temperature sensor 44 such as a silicon bandgap temperature sensor (or Proportional To Absolute Temperature [PTAT] sensor) comprised in an integrated circuit or a discrete thermistor, which is disposed on or near the block and thus is operative to sense the temperature of the block. A third analogue-to-digital converter 46 is operative to convert the analogue signal from the temperature sensor 44 to corresponding digital data. Temperature is measured to 0.5 degrees Centigrade.

[0052] The measurement and processing circuitry of FIG. 3 yet further comprises a microcontroller 48, non-volatile RAM 50, a Real Time Clock (RTC) 52 and transceiver circuitry 54. Components of the measurement and processing circuitry of FIG. 3 other than the microcontroller 48 are constituted as a custom integrated circuit with the measurement and processing circuitry being contained within or mounted on the block. The microcontroller 48 receives digital data from the first to third analogue-to-digital converters 40, 42, 46 corresponding to current, voltage and temperature and is operative to process the received data and to determine block conditions and electrical parameters as described below. As will become apparent from the following description condition determination involves comparison with threshold values. The measurement and processing circuitry of FIG. 3 therefore comprises a stable voltage reference, such as a band-gap reference, which is calibrated against a known voltage. The determined conditions and derived electrical parameters are stored in the non-volatile RAM 50 along with a time stamp from the RTC 52 whereby data is stored for the lifetime of the block. The measurement and processing circuitry of FIG. 3 is operative to measure the voltage, current and temperature at a rate between 0.01 Hz (Le. less than once per minute) and 1 kHz depending on system activity. When there is no charging or discharging, measurement is at a very slow rate to minimise power consumption. On the other hand, measurement is at a higher rate when there is activity such as charging or discharging. The microcontroller 48 is operative to prioritise stored data and periodically deletes old data or data of less significance to thereby make efficient use of the non-volatile RAM 50. The RTC 52 is autonomous and is driven by either a quartz crystal oscillator or a timing signal from the microcontroller 48. Whatever long term drift may be present is addressed by periodically synchronising the RTC 52 to an external clock such as an Internet based clock service accessed by way of the vehicle charging point 16. The measurement and processing circuitry of FIG. 3 is operative to detect when no current is drawn by a block whereupon the measurement and processing circuitry enters a sleep state from which it awakes periodically to resume measurement.

[0053] Onward communication of data from the measurement and processing circuitry of FIG. 3 is by way of the transceiver circuitry 54 to a bus 56. The transceiver circuitry 54 is configured to provide galvanic isolation from the bus 56. Galvanic isolation is employed to address the cumulative voltage shift as the bus 56 traverses all the blocks in the battery. In a first form, the transceiver circuitry 54 is configured to communicate by way of the IEEE 802.15 Personal Area Network standard with data being conveyed by way of twisted-pair cable connected by a transformer (not shown) to each transceiver circuitry 54. In a second form, data is conveyed by way of the main electrical battery bus. The main electrical battery bus is normally noisy and therefore a robust protocol of a kind used in Power Line Communications (PLC) is employed in the second form instead of the IEEE 802.15 Personal Area Network standard.

[0054] Returning now to FIG. 2, a supervisory microcontroller 58 is connected to the bus 56. The supervisory microcontroller 58 is configured to perform supervisory operations in relation to the measurement and processing circuitry of FIG. 3 at its respective block of cells 30, 32, 34, 36. The supervisory microcontroller 58 is either one of the microcontrollers at a block of cells or a separate microcontroller. The supervisory microcontroller 58 is configured to provide for communication with an external network which is accessed when the Electric Vehicle 10 of FIG. 1 is connected to the vehicle charging point 16. Communication of data with the external network is by way of a wired channel or a wireless channel depending on whether charging is achieved by way of a wired approach or a wireless approach. As mentioned above with reference to FIG. 1 there is data communication between the vehicle charging point 16 and the server 18. The supervisory microcontroller 58 is configured to provide for communication with the external network in accordance with an Internet protocol such as TOP protocol over IPv6. Use of an Internet protocol provides for ease of communication of data between the supervisory microcontroller 58 and the server 18. The server 18 of FIG. 1 is operative to store data from a large number of BEVs over time with such stored data being of value to the like of cell and energy conversion arrangement manufacturers. The supervisory microcontroller 58 is further configured to cooperate with other on-board measurement and data recording apparatus whereby data from in-vehicle sensors and the like is received by the supervisory microcontroller 58 for onward communication to the server to provide for enhanced data recording. The supervisory microcontroller 58 is further configured to analyse recorded data and to compare an outcome of the analysis with reference data. The reference data is, for example, based on data recorded during at least one previous journey and the analysis comprises comparing efficiency of energy use between a presently completed journey and the reference data. By way of further example, the reference data may be based on data collected from several BEVs which is stored in the server 18 and subsequently communicated to the BEV which is making use of a facility to compare driving efficiency with the like of a mean standard established for a similar journey by a large number of other vehicles.

[0055] A unique address is stored in memory local to each microcontroller whereby each block can be uniquely identified. The unique address has a format in accordance with an Internet protocol such as TOP protocol over IPv6 to thereby provide for ease of communication of data by way of the bus 56 between each block and an Internet connected server 18.

[0056] The processing of measurements within the first to fourth measurement and processing circuitry 31, 33, 35, 37 will now be considered in more detail with reference to FIG. 4. FIG. 4 is a block diagram representation of cell group models and a battery model. FIG. 4 shows the first and second measurement and processing circuitry 31, 33 for their respective blocks. Components in common with FIGS. 2 and 3 are indicated by like reference numerals. Also shown in FIG. 4 is a first cell group model 60, a second cell group model 62, a first battery model 64 and a second battery model 66. The first cell group model 60 is stored in memory local to a first microcontroller of the first measurement and processing circuitry 31 for one block and the second cell group model 62 is stored in memory local to a second microcontroller of the second measurement and processing circuitry 33 for the other block. The first battery model 64 is stored in memory local to the first microcontroller and the second battery model 66 is stored in memory local to the second microcontroller. There is therefore a cell group model and a battery model at each block of battery cells. The same battery model is present at each block of battery cells such that in the present example the second battery model 66 is a copy of the first battery model 64. As will become apparent from the following description, each battery model 64, 66 makes use of data produced by both of the cell group models 60, 62. Data from each cell group model 60, 62 is therefore conveyed by way of the bus 56 to both battery models 64, 66.

[0057] Operation of a cell group model will now be described. As described above voltage, current and temperature data is received by the microcontroller. The microcontroller is then operative to determine block conditions and derived electrical parameters in dependence on the measured voltage, current and temperature data with reference to a cell group model. More specifically block conditions and derived electrical parameters include the like of State of Charge (SOC), State of Health (SOH), Depth of Discharge (DOD), capacity, internal resistance and internal impedance. Block conditions further include events such as SOA and NOA infringements and charge/discharge cycles. An example of a model for a cell group or an individual battery cell which is employed in condition determination is provided below. Outputs from the cell group model are then conveyed to the battery model which is operative to determine the like of overall SOC, overall SOH, overall DOD, overall capacity, overall internal resistance and overall temperature for the battery in dependence on the outputs received from cell group models 60, 62. A microcontroller at a block is then further operative to act in dependence on the battery level determinations. As a consequence of the presence of a battery model in each of the first to fourth measurement and processing circuitry 31, 33, 35, 37, each block is capable of independent operation in respect of the battery as a whole, For example each microcontroller is operative to determine the requirement for and then to initiate and control a charge balancing operation.

[0058] Charge balancing is often an important function for an electric battery, As described above each cell group determines its own SOC and receives, amongst other things, SOC data for other cell groups comprised in the electric battery whereby an SOC for the battery as a whole is determined. The battery is configured such that each cell group is operative to determine whether or not the cell group should perform passive charge balancing in dependence on the cell group's own SOC and the SOC for the battery. Thus each cell group in the battery makes a determination in respect of charge balancing either independently of the other cell groups or in cooperation with the other cell groups. The battery further comprises charge balancing apparatus which is operative in dependence on a determination being made in respect of charge balancing. The form and function of appropriate charge balancing apparatus for use herein is described in Battery Management Systems far Large Lithium-ion Battery Packs, Davide Andrea, 2010, published by Artech House, Norwood Mass. 02062, USA.

[0059] A cell group or an individual battery cell is modelled by way of the following algorithm. The inputs to the algorithm are: [0060] 1. The nominal full cell capacity, CapCell. [0061] 2. The present cell Depth of Discharge (DOD) in Ah. If the cell is full then DOD 32 0. If the cell is empty then DOD=CapCell. [0062] 3. The nominal cell resistance, Rcell_nom, [0063] 4. The cell open circuit voltage (OCV) at four appropriate points in the State of Charge (SOC) versus OCV curve for the cell. For example, the four appropriate points are: the voltage at SOCempty 0% =Vempty; the voltage at SOCbottom 15% =Vbottom; the voltage at SOCtop 95% =Vtop; and the voltage at SOCfull 100% =Vfull. The OCV is the cell terminal voltage when no current is drawn and when the cell has had time to relax.

[0064] The algorithm has two independent loops, Loop 1 and Loop 2.

[0065] Loop 1: [0066] When the drawn current changes, calculate the cell resistance, Rcell, on the basis of Rcell=(V1−V2)/(I2 -I1), where V1 and I1 are the voltage and current measured before the current change and V2 and I2 are the voltage and current measured after the current change.

[0067] The instantaneous OCV is determined by OCV=Vterm+Icell*Rcell, where Vterm is the measured instantaneous terminal voltage and cell is the measured instantaneous current.

[0068] Loop 2: [0069] IF charging (i.e. current, I<0) THEN [0070] Calculate OCV=Vterm+Icell*Rcell [0071] IF OCV<Vtop THEN [0072] Integrate cell current, Icell, into the DOD [0073] Convert DOD to SOC by way of SOC=100%−100*DOD/CapCell [0074] IF SOC>SOCtop THEN [0075] Set SOC=SOCtop [0076] Convert SOC to DOD by way of DOD=Capcell*(100%-SOC) [0077] ELSE [0078] Convert OCV to SOC using a straight line interpolation between SOCtop and SOCfull and between Vtop and Vfull [0079] Convert SOC to DOD [0080] IF discharging (current, I,>0) THEN [0081] IF OCV>Vbottom THEN [0082] Integrate cell current into DOD [0083] Convert DOD to SOC [0084] IF SOC<SOCbottom THEN [0085] Set SOC=SOCbottom [0086] Convert SOC to DOD [0087] ELSE [0088] Convert OCV to SOC using a straight line interpolation between SOCbottom and SOCempty and between Vbottom and Vempty [0089] Convert SOC to DOD

[0090] The above algorithm is given by way of example only. During use of the algorithm there is an accumulation of errors on account of integration of measured current to determine the DOD whereby the DOD measurement uncertainty normally increases over time. The algorithm therefore involves a reset of the DOD when the SOC goes through either SOCtop or SOCbottom which can result in a significant jump in the measured SOC and DOD.

[0091] A battery is modelled by way of algorithms that provide for the like of the summing of individual cell block conditions, derived electrical parameters and measurements and the determination of derived quantities such a mean or average based on summed quantities.

[0092] The cell group model described above is based on the State of Charge (SOC) versus open circuit voltage (OCV) curve for the cell block. The cell group model for each cell block is configured and calibrated by taking the cell block through at least one complete charge and discharge cycle. During the at least one complete cycle voltage, current, temperature and time are measured to high accuracy and the measured values are used to configure and calibrate the cell group model. Each cell group model is therefore configured specifically for a particular cell block, Normally the cell group model calibration is performed at the same time as calibration of the rest of the battery management system to thereby provide a simpler calibration procedure and otherwise avoid duplication of cell group model calibration operations and battery management system calibration operations.

[0093] A cycle of use of a battery cell is shown in FIG. 5. Upon manufacture the cell 80 has an SOH of 100% and is installed in an electric battery of a BEV 82. After a period of use in the BEV 82, the SOH of the cell as determined by the present invention drops to 80% which is a critical threshold for continued use in the BEV 82. Following identification of cell weakness by the present invention, the cell is removed from the BEV and installed in local electric battery storage 86 in an off-grid environment. The local electric battery storage 86 is configured in accordance with the present invention whereby monitoring of the cell is continued. When the SOH of the cell 88 as determined by the present invention drops to 50% which is a critical threshold for continued use in the local electric battery storage 86, the cell is removed from the local electric battery storage 86 and decommissioned 90.

[0094] It is to be noted that each cell group is operative of itself in respect of management functions such as: measurement of the like of voltage, current and temperature; determination of various parameters such as SOC, DOD, SOH and internal resistance; and the recordal of events such as charge/discharge cycles, NOA and SOA excursions. Each cell group is so operative irrespective of whether the battery has yet to be assembled from the cell group and other cell groups, the battery has been assembled or the battery has been disassembled such the cell group no longer forms part of the battery. Each cell group is therefore configured to perform the above described management functions. Furthermore each cell group is configured to perform the above described management functions in cooperation with at least one other cell group.

[0095] The present invention is also of application in Hybrid Electric Vehicles (HEVs), Plug-in Hybrid Electric Vehicles (PHEVs) and fuel cell electric vehicles (FCEVs). Where the present invention is applied to an FCEV, modifications are made to the cell/cell group model to take account of the different characteristics of fuel cells. Nevertheless models of the fuel cells comprised in the FCEV are determined in the same fashion as described above with reference to battery cells by way of s measurements at each fuel cell during an initial calibration procedure. Furthermore the nature of the condition data determined for the fuel cells is selected to cater for the different characteristics of fuel cells. Otherwise the invention is of a form and function as described above with reference to BEVs.