Pulse discharge system

Abstract

A power supply system can include an electrical battery for supplying electrical load using pulse discharge; and a control unit to control the electrical battery to pulse discharge by periodically switching between a discharge state of the electrical battery, when the electrical battery is connected to the electrical load, and a rest state of the electrical battery when the electrical battery is disconnected from the electrical load. During the pulse discharge, the control unit can control a power source for supplying an injection current to the electrical battery during a rest period when the electrical battery is in the rest state. The control unit can further determine the voltage of the electrical battery during the rest state, and when the voltage does not meet a threshold value increase the duration of the rest period, until said threshold value is reached. Other systems, methods and apparatuses are described.

Claims

1. A power supply system for supplying an electrical load using pulse discharge, the system comprising; a first electrical battery for supplying said electrical load using pulse discharge; and a control unit configured to control the first electrical battery to pulse discharge by periodically switching between a discharge state of the first electrical battery, when the first electrical battery is connected to said electrical load, and a rest state of the first electrical battery when the first electrical battery is disconnected from said electrical load; said control unit being further configured to, during the pulse discharge, control a power source for supplying a first injection current to the first electrical battery during a rest period when the first electrical battery is in the rest state; said control unit further configured to determine, during the rest state of the pulse discharge, the voltage of the first electrical battery and, during the rest state of the pulse discharge, when the voltage does not meet a threshold value, increase the duration of the rest period until said threshold value is reached, wherein the control unit is further configured to control the power source such that said rest period has a fixed minimum duration even if the threshold value is met.

2. The power supply system according to claim 1, wherein the control unit is configured such that the threshold value is a minimum threshold voltage.

3. The power supply system according to claim 1, wherein the control unit is further configured to determine the voltage at fixed time intervals during the rest state to determine when the threshold value is reached.

4. The power supply system according to claim 3, wherein the control unit is further configured to determine the difference between the voltage determined at two consecutive time intervals, and compare the difference to the threshold value to determine when the threshold value has been reached.

5. The power supply system according to claim 1, wherein the control unit is configured such that the increase in the duration of the rest period involves increasing the length of time the injection current is applied.

6. The power supply system according to claim 1, further comprising: a second electrical battery, wherein the control unit is further configured to control the second electrical battery for periodically switching between a discharge state of the second electrical battery and a rest state of the second electrical battery.

7. The power supply system according to claim 6, wherein the control unit is configured to periodically switch between the first electrical battery and the second electrical battery to alternate between the discharge state of the first battery and the discharge state of the second electrical battery.

8. The power supply system according to claim 7, wherein said control unit is further configured to control the power source to supply a second injection current to the second electrical battery during a rest period when the second electrical battery is in the rest state.

9. The power supply system according to claim 8, wherein the electrical output across the load from the power supply system is substantially continuous.

10. The power supply system according to claim 6, wherein the control unit is further configured to determine the voltage of the second electrical battery during the rest state of the second electrical battery, and when the voltage does not meet the threshold value increase the duration of the rest period of the second electrical battery until said threshold value is reached.

11. The power supply system according to claim 1, wherein the power source is any of an electric motor, the output of an inverter, a kinetic energy recovery system (KERS), a photovoltaic panel, or a battery.

12. A control unit comprising; a processor; and a switching generator configured to generate a switching signal for periodically switching between a discharge state of a first electrical battery and a rest period when the first electrical battery is in a rest state; wherein the control unit is further configured to generate a power signal for controlling a power source to supply an injection current to the first electrical battery when the first electrical battery is in the rest state; and wherein the control unit is further configured to determine, during the rest state, the voltage of the first electrical battery, and, during the rest state, when the voltage does not meet a threshold value, increase the duration of the rest period until said threshold value is reached, wherein control unit is further configured to control the duration of the rest period such that said rest period has a fixed minimum duration even if the threshold value is met.

13. A method for supplying an electrical load using pulse discharge, the method including: supplying, by a first electrical battery, said electrical load using pulse discharge; controlling, by a control unit, the first electrical battery to pulse discharge by periodically switching between a discharge state of the first electrical battery, when the first electrical battery is connected to said electrical load, and a rest state of the first electrical battery when the first electrical battery is disconnected from said electrical load; during the pulse discharge controlling, by the control unit, a power source for supplying a first injection current to the first electrical battery during a rest period when the first electrical battery is in the rest state; and determining, by the control unit during the rest state of the pulse discharge, the voltage of the first electrical battery and, during the rest state of the pulse discharge, when the voltage does not meet a threshold value, increasing the duration of the rest period until said threshold value is reached, and controlling, by the control unit, the power source such that said rest period has a fixed minimum duration even if the threshold value is met.

14. The method according to claim 13, wherein the threshold value is a minimum threshold voltage.

15. The method according to claim 13, further including: determining the voltage at fixed time intervals during the rest state to determine when the threshold value is reached; and determining the difference between the voltage determined at two consecutive time intervals, and compare the difference to the threshold value to determine when the threshold value has been reached.

16. The method according to claim 13, wherein the increase in the duration of the rest period involves increasing the length of time the injection current is applied.

17. The method according to claim 13, further comprising: controlling a second electrical battery for periodically switching between a discharge state of the second electrical battery and a rest state of the second electrical battery.

18. The method according to claim 17, further comprising: periodically switching between the first electrical battery and the second electrical battery to alternate between the discharge state of the first battery and the discharge state of the second electrical battery; and controlling the power source to supply a second injection current to the second electrical battery during a rest period when the second electrical battery is in the rest state.

Description

DESCRIPTION OF DRAWINGS

(1) FIGS. 1A and 1B show simplified circuit diagrams of an example power supply system in an embodiment;

(2) FIG. 2A is a graph showing the voltage across a battery terminal during one switching cycle for the example power supply system as shown in FIGS. 1A and 1B;

(3) FIG. 2B is a graph showing the current with respect to the battery terminal during one switching cycle for the power supply system as shown in FIGS. 1A and 1B;

(4) FIGS. 3A and 3B show simplified circuit diagrams of a second example of a power supply system in another embodiment;

(5) FIG. 4 is a graph showing a section of a voltage versus time discharge curve for the system shown in FIGS. 3A and 3B without a variable rest period, and for a continuous discharge system;

(6) FIG. 5 is a graph showing a complete discharge profile of a voltage versus time discharge curve for the system shown in FIGS. 3A and 3B without variable rest period, and for a continuous discharge system;

(7) FIG. 6 shows a further example power supply system in an embodiment;

(8) FIG. 7A shows a graph showing a section of a voltage versus time discharge curve for the system shown in FIG. 6 with a variable rest period showing the maximum and minimum voltage envelope;

(9) FIG. 7B shows a zoomed in section of the graph of a portion of FIG. 7A showing the voltage profile;

(10) FIG. 8 shows a graph showing variation of the length of the rest period over time for the discharge curve shown in FIG. 7A;

(11) FIG. 9 shows a graph of variation of the length of the rest period over time for a further example discharge curve;

(12) FIG. 10 shows a graph of the energy exchanged per cycle for the test shown in FIG. 9;

(13) FIG. 11 shows an example photovoltaic system showing one example of the type of system that can be used in conjunction with an example power supply system of embodiments;

(14) FIG. 12 shows a further example power supply system in an embodiment.

DETAILED DESCRIPTION

(15) FIGS. 1A and 1B are simplified circuit diagrams of an example power supply system 2. The power supply system 2 includes a battery 4 which is connected in series to a load 6 by a two way switch 8. The power supply 2 system drives the load by supplying an electrical current. A power source 10 is connected to the other terminal of the two way switch 8. FIG. 1A shows the circuit in the configuration where the two-way switch 8 is connected to the load 6 so that the battery 4 supplies power to the load 6. FIG. 1B shows the alternative arrangement where the two way switch 8 is connected to the power source 10 so that the battery 4 can be recharged. The control unit 12 is configured to generate a switching signal SW1. The switching signal SW1 activates the switch 8 causing it to switch periodically between its two positions.

(16) In FIG. 1A when the switch 8 is closed between the battery 4 and the load 6, current flows from the battery 4 to the load 6. This results in a discharge of the battery 4 during the time period that it is connected to the load 6. The switch 8 is open between the battery 4 and the power source 10, and therefore the power source 10 does not supply any electrical current to the battery 4.

(17) In FIG. 1B the switching signal SW1 generated by the control unit 12 has caused the switch 8 to change orientation compared to in FIG. 1A. The switch 8 is now open between the battery 4 and the load 6. In this case the battery 4 is no longer connected to the load 6 and no current flows from the battery 4 to the load 6. The battery 4 is no longer discharging and is in a rest state. The switch 8 is closed between the battery 4 and power source 10; the power source 10 can then supply current to the battery 4 for a portion of the time that it is connected.

(18) The control unit 12 further controls the power source 10 to supply an electric current for a certain period during the rest state by generating a power control signal PC1 to control the power source between an ON and an OFF state. The power source 10 is connected to the terminals of the battery 4 such that the supply of current to the battery 4 from the power source 10 causes a charging effect to occur.

(19) The control unit 12 causes periodic switching of the switch 8 during discharge of the battery 4. This creates a pulsed discharge of the battery 4 with charging during a portion of the rest state.

(20) The power supply system 2 further includes a voltmeter 5 that is in connection with the control unit 12. The voltmeter 5 measures the voltage of the battery 4 and sends signal VS1 indicating the voltage to control unit 12. This enables the control unit 12 to control the periodic switching of switch 8 when the voltage of the battery drops below a threshold minimum level, resulting in an increase in the length of time that the battery is in the rest state.

(21) FIG. 2A is a graph showing the voltage across a battery 4 terminal during one switching cycle for the power supply system 2. At t=0 the battery 4 is connected to the load 6 and starts discharging. Between t=0 and t=30 s the battery 4 is connected to the load 6 and is in the discharge state for a discharge period, T.sub.on. During the discharge period the voltage across the battery 4 decreases. At t=30 s the battery 4 is disconnected from the load 6 when the two-way switch 8 is switched to its alternate position. The battery 4 remains disconnected from the load 6 in this rest state for a rest period between t=30 s and t=60 s. At the beginning of the rest period the voltage across the battery 4 increases rapidly, since the load 6 has been disconnected. The voltage across the battery 4 then increases gradually over the remainder of the rest period. Between t=50 and t=60 a current is injected into the battery 4 by the power source 10 under control of the control unit 12. This is termed a current injection period, and it results in an additional voltage increase across the battery 4. The rest period therefore comprises a recovery period during which the battery 4 is disconnected from the load 6 but is not supplied with a current, and a current injection period where the battery 4 is disconnected from the load 6 and a current is supplied by the power source 10.

(22) The switching cycle has a switching period T.sub.c that is defined as the sum of the discharge period T.sub.on, and the rest period T.sub.t. As explained above, the rest period comprises the recovery period T.sub.off and the current injection period T.sub.ON. Thus:
T.sub.t=T.sub.off+T.sub.ON
and
T.sub.c=T.sub.t+T.sub.on
FIG. 2B is a graph showing the current supplied at one of the terminals of the battery 4 for one switching cycle. Between t=0 and t=30 during the discharge period the battery 4 discharges at constant current (I.sub.on). Between t=30 and t=50 (the recovery period) the battery 4 is disconnected from the load 6, and no current is supplied to the battery 4 from the power source 10; therefore, during the recovery period the current is zero (I.sub.off=0). Between t=50 and t=60 (the current injection period) a current is supplied to the battery 4 by the power source 10; the current measured at the terminal of the battery 4 during the current injection period I.sub.ON.

(23) FIGS. 3A and 3B show simplified circuit diagrams of another example of a power supply system 2 for supplying an electrical current to a load 6. The power supply system 2 includes a second battery 13, in addition to the first battery 4 as shown in the example power supply system 2 in FIGS. 1A and 1B.

(24) As is apparent from FIG. 3A, the first battery 4 can be connected in series with the load 6 by selected operation of two-way switches 14, 16. The second battery 13 is connected in series with the power source 10 through selected operation of two-way switches 18, 19. It is noted that, with the two-way switches 14, 16, 18, 19 in the configuration shown in FIG. 3A, the second battery 13 is disconnected from the load 6, and the first battery 4 is disconnected from the power source 10.

(25) Voltmeters 5a and 5b in communication with the control unit 12 are connected to the first battery 4 and the second battery 13, respectively.

(26) FIG. 3B shows an alternative switching configuration. In this arrangement the first battery 4 is connected in series to the power source 10 through selected control of the two-way switches 14, 18. The second battery 13 is connected in series to the load 6 through selected control of the two-way switches 16, 19. In this switching configuration the first battery 4 is not connected to the load 6, and the second battery 13 is not connected to the power source 10.

(27) In this example the control unit 12 of the power supply system 2 supplies a switching signal SW2 to the two-way switches 14, 16, 18, 19 in order to switch between the first battery 4 being connected to the load 6 and the second battery 13 being connected to the load 6. In this way, only one of the batteries 4, 13 is arranged to discharge by supplying a current to the load 6 at any one time. The first battery 4 is in the discharge state whilst the second battery 13 is in the rest state, and the first battery 4 is in the rest state whilst the second battery 13 is in the discharge state. The control unit 12 is also arranged to generate a power control signal PC2 to control the power source 10 in order to supply an electric current to each of the batteries 4, 13 during a portion of their respective rest states.

(28) Voltmeter 5a measures the voltage of the first battery 4, and voltmeter 5b measures the voltage of the second battery 13, during their respective rest states. These voltage values are provided to the control unit 12 in signals VS1 and VS2 such that the control unit 12 can determine if the voltage of each battery reaches a minimum threshold voltage during the rest phase of that battery.

(29) If the minimum voltage threshold is met during the rest state then the switching cycle for the power supply system 2 shown in FIGS. 3A and 3B is equal and symmetric between the discharge and rest phase; thus T.sub.on=T.sub.t. In this case the duty cycle of the switching signal is 50%.

(30) However, when the voltage threshold is not met for a particular battery 4, 13 during its rest state then the rest period of that battery 4, 13 is extended until the open voltage of that battery reaches the minimum voltage threshold. In the embodiment shown in FIG. 3B, an increase in the length of the rest period of one battery leads to an increase in the discharge phase of the other battery, in order for a continuous current output from the respective batteries 4, 13, with the load 6 alternately supplied with current from the first battery 4 and the second battery 13. Advantageously, this is useful for systems requiring a continuous supply of power.

(31) Initially the length of the rest period T.sub.t is equal to the length of the discharge period. However, once the voltage drops below the minimum threshold value the length of the rest period can be extended.

(32) In some embodiments, the length of the rest period T.sub.t. can be extended based on monitoring the recovery of the voltage of the battery 4, 13 when the battery is in the rest state.

(33) Taking t.sub.0 as the instant at which the rest period begins, V.sub.0 is the voltage value of the cell measured at the instant in which there is no current flow through the battery when the battery voltage then starts to rise, said voltage defined by the product of the current previously circulating and the battery internal resistance The voltage of the battery is then measured every 5 s to determine a series of voltage values at each of the points in time (V.sub.0, to =0; V.sub.1, t.sub.1=5 s; V.sub.2, t.sub.2=10 s, . . . ). Thus, at a given time period t.sub.x,x+1 there is a voltage variation V.sub.x,x+1. The value of the rest period T.sub.t is then extended until the point where the following condition is met:

(34) $\frac{V_{x,x+1}}{V_{0,1}}<0.2$

(35) Thus, when the ratio of 0.2 is reached the rest period is no longer extended and the battery reverts to the discharge state. This time period of when the ratio is reached is the optimum rest period for a battery for that particular discharge cycle. This is because once this condition is satisfied any increase in the recovery of the battery voltage becomes negligible.

(36) This process is then repeated for each pulse discharge cycle, with the value of V.sub.0,1 measured for the first discharge cycle taken to be the same for each of the subsequent pulse discharge cycles. In this way there is a variable rest period that is optimised for each pulse discharge cycles.

(37) Alternatively, the state of charge (SoC) and/or depth of discharge (DoD) may be measured and used as the threshold condition for increasing the length of the rest period, with the aim of maximizing recovery for the voltage of the battery during the rest state. Other threshold conditions may instead by used, such as directly determining if a minimum voltage at a certain point in time is met.

(38) In other embodiments, the ratio in the equation above need not necessarily be 0.2. For instance, if a different rest period is desired, such as longer or shorter rest periods, which may involve further improvements in efficiency.

(39) The configuration described in the embodiments can readily be scaled to integrate more than two batteries operating in a coordinated way so that at any given time, only a portion of batteries are in a discharge stage while the others are in a rest stage during which a current is supplied by the power source.

(40) The configuration of batteries can also be adjusted to supply any specific current signal to the load. For instance, where a constant supply to the load is not required, the length of the discharge phase of the batteries may be kept constant, even when the length of the rest period is increased due to the threshold voltage not being met.

(41) FIG. 4 is a graph showing the voltage across the first and second batteries 4, 13 versus time in a device using pulse discharge where the rest period is not varied; these are shown in trace 41 and trace 43 respectively. A trace 40 is also shown for the voltage across a conventional battery versus time, where there is no pulsed discharge. In a conventional battery discharge the voltage across the battery gradually decreases over time. The traces 41, 43 for the first and second batteries 4, 13 follow the shape of FIG. 2A for a single switching cycle. However, the traces 41, 43 are out of phase such that the first battery 4 is in its rest state while the second battery 13 is in its charging state, and vice-versa. In each successive switching cycle the respective voltages across the first and second batteries 4, 13 are lower than in the previous cycle. Thus, there is a downward trend in the voltage values for successive switching cycles in the traces 41, 43. However, the downward trend in voltages is less for traces 41, 43 than for trace 40 which represents a conventional battery discharge. This indicates an improved efficiency of operation for the first and second batteries 4, 13 using the described charge/discharge cycle with fixed rest period.

(42) FIG. 5 is another graph showing the voltage across the first and second batteries 4, 13 over a longer time period. A plot is also shown for the voltage across a conventional battery versus time, where there is no pulsed discharge. The trace 23 represents the voltage across the first and second batteries 4, 13; this corresponds to the traces 41 and 43 shown in FIG. 4, but over an extended time period, i.e. having a fixed rest period. The trace 21 shows the voltage across a conventional battery with no pulsed discharge. As can be seen from FIG. 5 the trace 23, representing the voltages of the first and second batteries 4, 13, has a much shallower gradient.

(43) For the first example power system shown in FIGS. 1A and 1B, and the second example power system shown in FIGS. 3A and 3B, the power source 10 can be configured to supply current to the battery 4 at any point during the rest period of a battery. More specifically, the intermediate injection charging period T.sub.ON can be set to occur at any point during the rest period T.sub.t of that particular battery. In alternatives, the control unit 12 may control the power source 10 to supply electricity to the respective battery 4, 12 during the entire rest period, at the end of the rest period, at the beginning of the rest period, only during a portion of the rest period. There could also be intermittent charging during the rest period; this may include at the end, at the beginning, or at various times throughout the rest period. In the case of the current being intermittently supplied during the rest period the current may be supplied for various time periods during one rest period.

(44) A number of alternative timing schedules may be applied by the control unit 12. In one arrangement the discharge period may last from 5 to 100 seconds. The rest period may have a duration of 1 to 180 s. The current injection period may have a duration of 1 to 100 s. A further rest period may be provided at the end of the current injection period with a duration of 0-5 s. The current supplied to the battery 4 by the power source 10 during the current injection period may be less than 70% of the current provided by the battery 4 to the load 6 during the discharge period.

(45) The efficiency of the power supply system can be determined in a number of different ways: The net energy delivered by the battery during discharge may be calculated as follows:
E.sub.net=E.sub.onE.sub.ON Where E.sub.on is the total energy delivered to the load, measured starting from when the battery has its nominal voltage down to its minimum value, and E.sub.ON is the energy supplied by the partial charges of the battery. Where E.sub.ON=0 in the case of continuous discharge. Starting from the net energy, the discharge efficiency may be defined as the ratio between the net energy discharge of the switched battery power supply system (E.sub.net,M) and the net energy of the continuous discharge battery (E.sub.net,C).
.sub.discharge=E.sub.net,M/E.sub.net,C Discharge/charge cycle efficiency:
.sub.cycle=E.sub.on/(E.sub.ON+E.sub.charge) Where E.sub.charge is the energy required to charge the battery up to the initial voltage (arbitrarily set). As can be seen above the energy provided by the power source during the intermediate charging phase is taken into account in these calculations. Amperometric efficiency:

(46) ${}_a=\frac{i_{\mathrm{on}}}{i_{\mathrm{ch}}}=\frac{Q_{\mathrm{on}}}{Q_{\mathrm{ON}}+Q_{\mathrm{charge}}}$ Where i.sub.on is the discharge current, and i.sub.ch is the battery charge current. Q.sub.on is the charge supplied by the battery to the load during the discharge state, Q.sub.ON is the total partial charge supplied to the battery during the current injection periods, and Q.sub.charge is the total final charge required to bring the battery to its nominal voltage value. We note that when a current is passed through a battery gas is generated; this reduces the amperometic efficiency by limiting the electrical current flowing outside the battery.

(47) Example test results achieved using embodiments will now be demonstrated. FIG. 6 shows an example power supply system 60 according to an example embodiment that has been used for said tests. It includes control unit 62 which is a central computerised system, running a proprietary dedicated software developed in environment Labview. The control unit 62 is configured to supervise all the connected devices. DC power source 70 is a programmable DC power supply model DC EA-PSI 8080-510. This is configured to apply the injection current during the rest phase of battery 64, and also to provide the full recharge of the battery 64.

(48) Diode 76 has its anode connected to the positive output of the power source 70 and its cathode connected to the positive pole of the battery 64. This is to provide decoupling the output of the power source 70 from the battery 64.

(49) The load of this circuit comprises two contributions. The first is active electronic load 66, (EA EL9080-600). The second component of the load is a power resistor array 72 capable to dissipate a maximum power of 9.4 kW. The resistors enable the active load to dissipate for a longer duration. The active electronic load 66 and the resistor array 72 are in series with each other. Control unit 62 supplies the loads with current from battery 64. This set up enables discharge currents of up to 250 A to be achieved.

(50) Transducers and converters 68 and 74 enable current and voltage measurements to be taken. These include dedicated transducers and analogue to digital converters with 16 bit of resolution, at concurrent sampling (model DAQ NI9215). These measurements are provided to the control unit 62.

(51) Battery 64 is of the lead acid type (model 10OPzS1000 produced by TAB), being one cell of 2V nominal voltage having 10 plates of 100 Ah. The nominal capacity is 1000 Ah, with a discharge current of 100 A, a full discharge is achieved in 10 hours, and at 250 A a full discharge is achieved in 3 hours at continuous discharge. In the present experiments 20 cells are connected in series.

(52) For all of the results, multistep potentiometric analysis has been selected. The current is set constant in the time interval selected and the battery voltage evolution is recorded (data sampling time=1 s). It has to be noted that the voltage measured has two contributions (V.sub.m=V+): the battery voltage V and the ohmic contribution =I*(R.sub.int+R.sub.c) due to battery internal resistance (R.sub.int) and contact resistance (R.sub.c). In order to have higher accuracy in the results obtained the contacts have been optimized to reduce this contribution. Energy calculation is performed from the current and voltage data recorded from the instrument. The Rectangle method has been employed for the area calculation under the voltage curve evolution over time interval (V-t graph); the quantity obtained is subsequently multiplied by the charge/discharge current. By definition, the energy corresponding to a single sampling interval is equal to E.sub.s=V.sub.m*A*T.sub.C [Joule].

(53) TABLE-US-00001 TABLE 1 Minimum Partial charge threshold Test Test current Voltage Duration T.sub.1 .sub.cycle .sub.a 1 N/A N/A 8 hrs N/A 0.832 N/A Discharge at continuous current 2 N/A N/A 17 hrs 92 s 0.845 N/A Pulse discharge - no injection current during rest phase 3 115 A N/A 8 hrs 92 s 0.879 0.959 Pulsed discharge with injection current during rest phase 4 200 A N/A 5 hrs 92 s 0.873 0.957 Pulsed discharge with injection current during rest phase 5 200 A N/A 5 hrs 92 s 0.871 0.954 Pulsed discharge with injection current during rest phase 6 70.6 A 43 V 16 hrs Variable 0.900 0.980 Variable T.sub.t (min 92 s) 7 70.6 A 44 V 24 hrs Variable 0.926 0.996 Variable T.sub.t 8 70.6 A 44 V 96 hrs Variable 0.933 0.997 Variable T.sub.t 9 70.6 A 44.2 V 24 hrs Variable 0.927 0.999 Variable T.sub.t

(54) The experimental results shown in Table 1 indicate an increase in terms of efficiency of the charge/discharge cycle of the power supply system using a variable T period according to embodiments (experiments 6 to 9), with respect to the continuous cycle (experiment 1), pulse discharge (experiment 2) and pulse discharge with injection current during the rest phase (experiment 3 to 5), as indicated by a larger n.sub.cycle. The power supply system with variable T.sub.t can also provide higher amperometric efficiencies, indicated by the larger n.sub.a when using a variable rest period. This is an unexpected and advantageous effect.

(55) For any given battery and battery configuration, the parameters that affect the efficiency of operation may be: the length of switching cycle, the discharge current, the charging current, the combination of the three periods (T.sub.on, T.sub.ON and T.sub.OFF) considered in terms of duration and sequence, and the value of the minimum threshold voltage. By varying these parameters it may be possible to optimise efficiency in different operational conditions.

(56) FIG. 7A shows a graph of the voltage across the battery over time for Test 6 shown in Table 1, having a switching cycle with variable rest phase. FIG. 7A shows the voltage envelope 79, showing the maximum voltage 81 i.e. the voltage reached during the rest period, and the minimum voltage 83 i.e. the voltage reached during the discharge period. Only the voltage envelope is shown in FIG. 7A, however the voltage will vary between each of the maximum voltage 81 and the minimum voltage 83 throughout the discharge cycles. FIG. 7B shows a zoomed in section of voltage verses time graph between 400 minutes and 420 minutes corresponding to that shown in FIG. 7A. FIG. 7B unlike FIG. 7A shows the voltage at each point throughout the discharge cycle, similar to FIG. 4. FIG. 7B shows the minimum voltage 85 and the maximum voltage 87, and the change in the voltage between successive pulse discharge cycles.

(57) The experiment in FIG. 7A employs a pulsed discharge current of 115 A, a constant value of T.sub.on of 92 s, a minimum rest period (T.sub.t)=92 s, of which the current injection period T.sub.ON=32 s. The minimum threshold voltage during the rest period was set to 43 V (rather than using the inequality previously shown). Once the battery voltage reaches a maximum value of 44 V during the application of the injection current, said injection current is stopped. As can be seen the experiment was carried out for a duration of 16 hours using repeated pulse discharge cycles. At the end of the test a full charge has been applied to the battery to get it back to its nominal voltage.

(58) It is also noted that for experiments 4 and 5, with a fixed rest period, that value the current of the injection current (200 A) is higher than that for the discharge cycle (115 A). However, in these experiments as the injection current is only supplied for a short time period, the overall charge supplied by the injection current during each cycle is less that the charge supplied by the discharge of the battery during each cycle.

(59) As can be seen, the injection current for experiments 6 to 9 have a value of 70.6 A, whereas the discharge current is larger at 115 A. Again the injection current is only supplied for a portion of the rest period. To account for the variable rest period the injection current is lower than in experiments 4 and 5. However, the charge supplied by the injection current during each cycle is less that the charge supplied by the discharge of the battery during each cycle.

(60) As can be seen, unlike the continuous charge and the pulse discharge with fixed rest period as shown in FIG. 5, the voltage of the battery when using variable rest period enables the charge of the battery to be maintained at a much higher level. Thus, after the 16 hour experiment the battery still has a high state of charge, which is not possible for a continuous discharge or a pulse discharge (with or without injection current) without a variable rest period.

(61) FIG. 8 shows a graph of how the value of the rest period varies for the Test shown in FIG. 7A. It can be seen that between t=0 and t=750 minutes the rest period is constant at 92 s. During this time the voltage of the battery during the rest period exceeds the minimum threshold voltage of 43 V. However, after t=750 minutes the length of the rest period progressively increases for successive cycles, due to the fact that the voltage of the battery measured during the rest period does not reach the minimum threshold voltage of 43V before expiry of the minimum rest period of 92 s.

(62) It has been found that if the experiment is run for long enough, after so many cycles the variable rest period will eventually plateau and reach a steady state. This is shown in FIG. 9. As can be seen the rest period starts to increase at around t=400 minutes and reaches a steady state of 150 s at t=600 minutes.

(63) FIG. 10 shows the energy exchanged per cycle for the experiment shown in FIG. 9 over the length of the experiment. Plot 110 shows the energy delivered by the battery for each discharge cycle, and plot 112 shows the energy applied by the injection current for each cycle. It can be seen that the energy applied by the injection current for each cycle increases over time until it plateaus at t=10 hours, in line with the plateau of the variable rest period shown in FIG. 9.

(64) It can be seen from FIG. 10 that once the steady state is achieved the energy applied by the injection current for each cycle is slightly higher than the energy delivered during the discharge period for each cycle. Keeping the battery state of charge constant cycle after cycle implies that the energy applied by the injection current has to be higher than the discharge current. This is due to power losses generated by the current flow (both charge and discharge) and internal battery power losses due to the internal battery resistance, and any external power losses due to the cell interconnections. In the present case these differences are noticeable as current values can be in excess of 100 A. However, the present power supply system and methodology results in better efficiencies, as explained above.

(65) FIG. 11 shows an example photovoltaic system showing one example of the type of system that can be used in conjunction with an example power supply system 80 of embodiments. FIG. 11 shows a photovoltaic device 86 (such as a solar panel), connected to a maximum power point tracking circuit 88, which is connected to power supply system 80. The power supply system 80 includes control unit 82 connected to energy storage block 84. The power supply system 80 is connected through inverter 90 to the public mains which supply loads 92.

(66) The MPPT (Maximum Power Point tracking) circuit 88, is embedded into a dedicated DC-DC converter, and can maximise the instantaneous energetic efficiency that would otherwise be extremely variable due to the constantly changeable exposure to solar radiation of the photovoltaic device 86 and the required load conditions.

(67) The control unit 82 oversees both the charge of the energy storage block 84 and the management of the inverter 90. This enables the power supply system 80 to perform the pulse discharge method of embodiments.

(68) The inverter 90 delivers energy to the mains from the photovoltaic device 82 or from the storage block 84. A portion of the energy from the photovoltaic device 86 may also be used as the source of the injection current applied to the energy storage block 80 when it is in its rest phase. This for instance could be a public grid, or a private grid such as a power station dedicated to refill electrical energy for electrical vehicles. In some cases the DC-AC inverter 90 could be replaced with a simpler DC-DC converter as some electrical vehicles only accept DC charge.

(69) The loads 92 can be of various nature, domestic and industrial, resistive, capacitive and inductive. The energy storage unit 84 consists of batteries connected in series and/or parallel, depending on the capacity required and the voltage and current parameters needed.

(70) FIG. 12 shows an embodiment of an example power supply system 100 according to embodiments. FIG. 12 shows an energy storage block 103 having three battery units 106 connected in parallel, each battery unit comprising three cells connected in series. The battery management control logic 104 controls the switches 108 connecting the battery units 106 to implement the pulse discharge method of embodiments. The control unit 102 further controls the power supply system 100 such that the power is supplied to the load (not shown) when the power from the storage block 103 is required.

(71) The power supply system of embodiments can be used for both grid use (i.e. supplying power in excess of megawatts) or for micro grid use (smaller power networks).

(72) For instance, the power supply system of FIG. 12 could be used as a charging station, such as for electric vehicles. This is an example of a micro grid. Typically loads connected to micro grids require high peaks of power at certain times. Placing the power supply system of embodiments in close proximity to charging points can provide the advantage of allowing a better management of peak power locally delivered, ultimately avoiding overloads and power cuts which could affect the mains supply.

(73) The power source used to provide the battery with the injection current can be any type of device which is capable of providing an electric current to a battery. This could be any type of energy storage device that is capable of delivering a current over time, for instance a battery or a capacitor. The power source could also be any type of renewable energy source that is capable of supplying a current such as a photovoltaic system, or a wind turbine. The power source may provide DC or AC, depending on the requirements of the load 6. Where AC is required an inverter (not shown) may be provided.

(74) The power source could alternatively be the output from a kinetic energy recovery system (KERS) where the kinetic energy generated during the breaking process can be captured and turned into electrical energy. Alternatively, the output of a combustion engine can cause an electrical generator to generate a current. This power supply system is suitable for use in hybrid and electric vehicles.

(75) The control unit of embodiments can be a microcontroller, a microprocessor or a programmable logic device (for example, a FPGA=Field Programmable Gate Array). Preferably, the control unit and switch is implemented with a single component which is programmable logic device (for example FPGA) or an application-specific integrated circuit (ASIC).

(76) The control unit may generate a switching signal with a low logic signal indicating the selection of a discharge phase and a high logic value indicating the selection of the rest phase. Alternatively the low logic signal may indicate rest phase and high logic signal may indicate discharge phase.