Patent Application
20140159658
This invention relates to electric charging systems, and more particularly to electric vehicle chargers drawing power from an electrical grid.
The introduction of electric vehicles to the nation's highways is forecast to place additional electrical demands on the local power grids as the electric vehicles are connected to recharge. A need exists to reduce the instantaneous load on the electric grid from individual and groups of electric vehicle chargers during times of increased electrical grid demand.
Embodiments include methods and devices for restarting one or more electric vehicle chargers based on a delayed startup time, where the delayed startup time of each charger unit is based on a randomly generated number. One embodiment of an electric vehicle (EV) charger restart method includes determining a respective restart delay time (Tdel) for each of one or more electric vehicle chargers, each respective restart delay time (Tdel) comprising a respective delay time increment based on a generated random number and a group time interval for reset (Tint), determining if the determined respective restart delay time (Tdel) is met, and initiating a restart of at least one of the one or more electric vehicle chargers, if the determined restart delay time (Tdel) is met. In such an embodiment, the method may also include determining a difference between a measured source voltage and a normal source voltage at at least one of the one or more electric vehicle chargers and, if the determined difference between the measured source voltage and the normal source voltage is within a threshold, then setting an established time line start time (TPOK) for the at least one electric vehicle charger. The step of initiating a restart may begin at or after a time equal to TPOK plus Tdel, and the respective restart delay time (Tdel) may further comprise a predetermined time grid stabilization delay (Tint2). If the determined difference between the measured source voltage and the normal source voltage is not within a threshold, the method may include performing the determining a difference between a measured source voltage and a normal source voltage step again after a predetermined delay (DT1).
In embodiments where the respective restart delay time (Tdel) further comprises a predetermined time grid stabilization delay (Tint2) Tint2 may be between 20-60 seconds. Tdel may also be between 25-60 seconds. The group time interval for reset (Tint) may be between 5 and 20 seconds, and the generated random number may be between zero and one. The generated random number function may include an exponent (m) greater than 1.
Embodiments of the method may also include monitoring a difference between a reference phasor value and at least one PMU feedback signal indicative of a power grid phasor value, and if the difference is within a threshold (f), then setting an established time line start time (TPOK) for the at least one electric vehicle charger.
Alternative embodiments include receiving a power grid health signal from a power grid control processing unit in the at least one electric vehicle charger, setting an established time line start time (TPOK) for the at least one electric vehicle charger in response to receipt of the power grid health signal, and wherein the initiating a restart begins at or after a time equal to TPOK plus Tdel. In such configurations, the respective restart delay time (Tdel) may further include a predetermined time grid stabilization delay (Tint2), and Tint2 may be between 20-40 seconds. Tdel may be between 25-60 seconds. The generated random number may be between zero and one, and the generated random number may include an exponent (m) greater than 1.
The method may also include providing a pulse width modulated EVSE pilot signal to an electric vehicle, the pulse width modulated EVSE pilot signal over-riding an on-board charger current loading ramp function to extend the electric vehicle's power ramp time.
In one embodiment of a device including a processing module for restarting one or more electric vehicle chargers, wherein the processing module comprises a processor having addressable memory, the processor may configured to:
(i) determine if grid power quality is acceptable;
(ii) set an established time line start time (TPOK) for the at least one electric vehicle charger in response to determining grid power quality is acceptable;
(iii) determine a respective restart delay time (Tdel) for each of one or more electric vehicle chargers, the respective restart delay time (Tdel) comprising a delay time increment based on a respective generated random number function and a group time interval for reset (Tint); and
(iv) initiate a restart of at least one of the one or more electric vehicle chargers, if a current time (Tnow) is greater than Tdel plus TPOK. In such an embodiment, the respective restart delay time (Tdel) may include a predetermined time grid stabilization delay (Tint2). The generated random number function may further include an exponent (m) greater than 1.
In an alternative embodiment of a method, the method may include receiving in an electric vehicle (EV) charger a PMU power quality signal, determining a difference between the PMU power quality signal and a reference PMU power quality signal, and if the determined difference between the PMU power quality signal and a reference PMU power quality signal is within a threshold, then setting an established time line start time (TPOK) for the at least one electric vehicle charger, determining a respective restart delay time (Tdel) for the EV charger, the restart delay time (Tdel) comprising a respective delay time increment based on a generated random number and a group time interval for reset (Tint), and initiating a restart of the EV charger if an existing time (Tnow) is greater than TPOK plus Tdel The PMU power quality signal may be indicative of a measured power grid phasor. The restart delay time (Tdel) may include a predetermined time grid stabilization delay (Tint2), and
Tint2 may be between 20-60 seconds. In such an embodiment, Tdel may be between 25-60 seconds. The generated random number function may include an exponent (m) greater than 1.
Embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawing, and in which:
Embodiments are described of an electric vehicle charging system that has a restart processor configured to mitigate the effects of individual and groups of electric vehicle (EV) chargers restarting, such as occurs subsequent to interruption in power service. The electric vehicle charging system may determine a respective restart delay time (Tdel) for each of one or more EV chargers, the respective restart delay time (Tdel) comprising a respective delay time increment based on a generated random number and a group time interval for reset (Tint). The system may establish a time line start time (TPOK) based on comparison of reference and received PMU power quality signals, such as phasor current or phasor voltage values, and may initiate a restart of one or more of the EV chargers if an existing time (Tnow) is greater than TPOK plus Tdel.
More particularly, EV chargers, alternatively called “charging stations,” frequently have storage capacitors that result in a high power demand when each device is powering up. When connected in multiples of ten or one hundred, EV chargers may collectively present a sufficiently large load on the power grid to make reestablishing power service difficult. If the power outage is locally confined, e.g., due to a circuit breaker interruption, then the grid should have ample reserve to handle the startups. However, if the start-up requirements of the EV chargers on the re-powered circuit are extreme, for instance if all the chargers restart at once, then the local circuit breaker may blow, or open, due to high current demand.
Notably the restart power surge problem may be divided into two categories: restart after over-demand grid failure, and local restart after circuit interruption. To prevent overloading of the grid during the simultaneous start-up of multiple EV chargers, each EV charger may be configured to restart at a random time within a time interval, e.g., after a power outage, or after a power grid malfunction, and in so doing lower the probability that multiple EV chargers will startup simultaneously, or substantially simultaneously, and overload the grid. This random restart capacity is especially advantageous during periods of peak power demand when unpredictable loads serve to increase the instability of the power distribution network. Accordingly, to attenuate the injection of transient voltage drops into the grid, embodiments cause the effected EV chargers to power on at a restart delay time that has a delay time increment based on a generated random number.
The restart processor 104 is shown in communication with elements 112, 102, 114, 116, 118, interface circuitry (122, 124) for managing charge of the battery store, obtaining feedback from the battery store, controlling switching of AC/DC charge paths, managing conversion of DC to DC voltages, and managing conversion of DC to AC, respectively, and for managing interface circuitry for AC and DC electric vehicle charging. The restart processor 104 is depicted as in communication with a user interface 126 that may include a display 128, such as a touch-screen display, to enable the user to interact with the restart processor 104. The restart processor 104 is also shown as optionally in communication with a transmitter or a transmitter/receiver element 130, i.e., a transceiver or XCVR that may transmit and receive data via an antenna element 132, such as PMU feedback signals indicative of power grid health (i.e., signals indicative of current and/or voltage phasor values). The exemplary restart processor 104 includes a central processing unit (CPU) and addressable memory 120 where the CPU may be configured via computer-readable instructions to monitor current levels and charge levels within the EV charger and report portions of the monitored values to one or more external communication nodes via the XCVR 130 and antenna 132. The restart processor 104 may be further configured to read data stored in the data store 120, and output the read data to the XCVR 130 for transmitting to a remote site via the antenna 132. The interface circuitry 124 may be an EVSE and may be interposed between the inverter 118 and the electric vehicle 106, and may be detachably connected to the inverter 118 via a connector 134, and the interface circuitry 124, the restart processor 104, memory store 120, user interface 126 and display 128 may comprise a detachable module 136, e.g., a charger, that: (a) may be removed from the device 110 and fixedly attached to a support structure, such as a wall; and (b) wired to an AC power source such as a 220-240 VAC power line. The restart processor 104, user interface 126, display 128, and optional transceiver 130 may be powered via a power supply (not shown) that may receive as input 120 VAC and/or 104 VAC, or may be powered via the direct current charger 112, or other rectifying circuits, and a voltage regulator (not shown). Although described in terms of a “processor,” the restart processor is intended to encompass any manner of logic circuitry or firmware that processes or responds to basic instructions.
During operation, the EV charger restart process may include a restart delay time (Tdel) that includes a delay time increment based on a generated random number, and may include a predetermined time grid stabilization delay (Tint2) to await stabilization or reestablishment of the power grid. Each EV charger may have a processor configured to execute the restart process by generating a random number. The EV charger may base the determination of a time delay between the time when the power from the power grid is reestablished and when the charger initiates restarts based entirely or in part on a locally generated random number. By taking into account the average start up duration of an EV charger, and approximating the toleration of downtime by the user, it is initially estimated that a total time interval of 15 seconds may be sufficient for a small group of EV chargers to restart. Using a conventional RAND( ) function, which, when executed by a digital processors, generates a random number with a uniform probability distribution between zero and one, the restart delay time (Tdel) for an individual charging station (in seconds) could be thus be determined by:
T
del=RAND( )*15 (Eq. 1)
Tdel would, in this example, have a uniform distribution over 15 seconds, with a mean of 7.5. Accordingly, the entire set of EV chargers collectively executing the EV restart process of the chargers would restart within 15 seconds. Generally, to accommodate various size groups of EV chargers and different power grid supply conditions, the estimated 15 second interval is replaced with the variable Tint, the desired time interval for a group restart of EV chargers. A value for Tint can be computed by multiplying the startup interval of an individual charger Tstrt by the number of chargers and by a redundancy factor of ten or so. For example 15 charger units with 0.1 second startup intervals would then result in a 15 second group interval, this bears a close relationship to equation 3 below.
The program step is then:
T
del=RAND( )*Tint (Eq.2)
Thus all chargers in the group start within the bounds of Tint, with the restarts distributed with even probability throughout the time interval.
The probability that just one of the chargers in the group is restarting at a particular instant during the restart interval is:
PRB
1
=N*T
strt
/T
int (Eq. 3)
where Tstrt is the actual duration it takes for an EV charger to power up.
The probability that two are simultaneously restarting is:
PRB
2
=N*(N−1)*(Tstrt/Tint)2 (Eq.4)
The probability that three are simultaneously restarting is:
PRB
3
=N*(N−1)*(N−2)*(Tstrt/Tint)3, (Eq. 5)
and so on for an additional number of restarts up to the probability that all N chargers are restarting at once:
PRB
N
=N!*(Tstrt/Tint)N (Eq. 6)
As is evident by the relationships of values expressed by the above equations, the probabilities are governed by the factor (N*Tstrt/Tint). For values of this factor significantly less than one, the probability of simultaneous restarting of multiple units drops rapidly with respect to increasing number of chargers simultaneously restarting. For example, if the maximum tolerability of simultaneous restarts for two chargers is 1% of the restart interval time period, the factor (N*Tstrt/Tint) should be less than one-tenth, i.e., less than 0.1. This is because the probability that two chargers are simultaneously restarting is roughly equal to the square of the probability that one unit alone is restarting (as long as N is fairly large). For small values of N, equation 4 gives the precise result for the probability of simultaneous restart of two charger units. The probability of three chargers being in restart at once would be approximately 1/10% or 0.001 when continuing this example. It is also important to consider that the chargers often go into a standby state after restart, whereby being in a standby state may include: waiting to be used to charge an electric vehicle, or hooked to an electric vehicle that is fully charged. In this standby state mode, after the initial rush of power associated with startup, the charger has very little power demand. When in the active “in use” mode a charger has a significant power demand after startup, which affects the local power surplus. The implication then is that the load on the grid from a series of restarting chargers in active use mode is higher at the end of the startup interval, Tint, (when more chargers are actively charging) than toward the beginning of the startup interval (when many of the chargers are still turned off). In a situation where most of the chargers are in an active mode the algorithm may be optimized by shifting the distribution of startup delay times to concentrate them near the beginning of interval Tint when the charging demand is relatively low. Accordingly, the determination of the time delay may be based on:
T
del=(RAND( ))m*Tint (Eq. 7)
where the exponent, m, is a number>1. For example when m=2 the median value of the restart delay intervals, Tdel, would be 0.25*Tint.
Selection of a value for the exponential factor (m) can be based on the ratio of the power necessary to start the charger, Pstart, to the power required to operate it during normal active use Pcharge, e.g., (Pstart/Pcharge), and on the minimum steady reserve capacity of the system, i.e., a steady reserve capacity, Pres, may be defined as the difference between the available power supply, Psupply, and the power consumed by the actively in use charging stations, Pcharge, e.g.,
P
res
=P
supply
−Nc*P
charge (Eq. 8)
Where Nc is the number of units actively charging.
The preceding relationships for determining a time delay for the EV restart process consider the group EV charger circuit in isolation from wider power demand issues. This presumption may be applicable in the case where the circuit breaker that serves only the group of EV chargers is cycled, i.e., where the circuit breaker is turned off then on again. Although occurring less frequently than a circuit interruption, the restart difficulties after a general power outage may be much more significant, as the power supply may be marginal for about the first twenty seconds or longer, after power is restored and while a wide variety of devices start up again throughout the grid network. Accordingly, the relationship for determining a time delay for the EV charger restart process may be further refined to accommodate an additional time delay interval Tint2 with the intent to ensure that the chargers may power on after the grid has had a sufficient time to stabilize: a satisfactory value for Tint2 is estimated to be around thirty seconds, but a more accurate value can be obtained from local grid parameters and experience. The charger may receive information regarding the nature of the power outage and adjust the value of Tint2 accordingly.
The relationship for determining a restart delay time Tdel for an individual EV charger restart may be expressed as:
T
del
=T
int2+(RAND( ))m*Tint. (Eq. 9)
In the preceding examples, the values for Tint and Tint2 are estimated, and may be specified before initiation of restart for any particular EV charger. Other embodiments may determine the time delay intervals based on the conditions at a particular EV charging station. For example, Tint may be determined from a function based on the difference between the normal supply voltage, Vnorm, and the measured supply voltage; Vmeas:
T
int=(Vnorm−Vmeas)e1*c1 (Eq. 10)
T
int2=(Vnorm−Vmeas)e2*c2+c3, (Eq. 11)
where c1, c2 and c3 are characteristic time calculating factors specific to the charger and e1 and e2 are exponents greater than one. If a desired threshold voltage must be reached before the charger is restarted, the process may include additional criteria, such as the critical cutoff voltage difference, k, where the startup procedure is only initiated if Vnorm−Vmeas is less than k, where Vnorm is the nominal power grid voltage and Vmeas is the measured power grid voltage. Accordingly, the limiting factor k helps to ensure that the charger is only restarted when the grid is ready to assume greater loads. It will be evident to those skilled in the art that more complex functions can be employed for determining suitable values of the delay intervals, and that computer-implemented methods using these more complex algorithms fall within the scope of this invention.
An exemplary process 200 is depicted in the flowchart of
T
now
>T
pok
+T
del (Eq. 12)
The EV restart process may comprise an exponential term, (RAND( ))m, which facilitates more immediate restarting of the chargers—to take advantage of the increased availability of reserve power as the individual charging systems initially start to come back on line. The EV restart process may include a hedge against grid instability and the marginal power supply at the end of a total outage, and more particularly may include an ancillary time delay Tint2. The EV restart process may be positioned to restart EV chargers so as to balance between when the grid is available and when the EV chargers require the most power transmittance, and so the EV restart process may be applied in the EV chargers of power networks governed by utilities, i.e., networks that must provide reliable service while minimizing load perturbations or spikes.
In one embodiment, the control processing unit 302 may be configured to provide an EV charger start-up signal to a restart processor 312 in each EV charger 314 subsequent to interruption in power grid service. The control processing unit 302 may be further configured to provide control signals to grid-level energy stores 316, 318 each being configured, responsive to a control signal from the control processing unit 302, to draw power from, and provide power to, the power grid.
The EVSE 400 may provide an EVSE pilot signal to the electric vehicle 412 to establish current draw from near zero current up to the pre-determined maximum current draw using a current ramp function to ease EV restart loading of the utility power source 404. However, instead of using an EV-defined current ramp profile, the restart processor 406 may enable the EVSE 400 to override and extend the ramp-up time for the EV current draw to mitigate the effects of individual and groups of EV chargers restarting, such as occurs subsequent to interruption in power service (See
Although illustrated as linear, the extended ramp line 502 may be non-linear, such as exponential, to weight current draw rate of change toward the end or beginning of the start-up process.
It is contemplated that various combinations and/or sub-combinations of the specific features and aspects of the above embodiments may be made and still fall within the scope of the invention. Accordingly, it should be understood that various features and aspects of the disclosed embodiments may be combined with or substituted for one another in order to form varying modes of the disclosed invention. Further it is intended that the scope of the present invention herein disclosed by way of examples should not be limited by the particular disclosed embodiments described above.
This application claims priority to and benefit of Provisional Patent Application No. 61/417,078 filed Nov. 24, 2010, the contents of which are hereby incorporated by reference herein for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 61417078 | Nov 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US11/61529 | Nov 2011 | US |
| Child | 13902557 | US |
