WO2019153305A1

WO2019153305A1 - Charging station and method and device for controlling charging station

Info

Publication number: WO2019153305A1
Application number: PCT/CN2018/076314
Authority: WO
Inventors: Xing Huang; Hailian XIE; Junjie GE
Original assignee: ABB Schweiz AG
Current assignee: ABB Schweiz AG
Priority date: 2018-02-11
Filing date: 2018-02-11
Publication date: 2019-08-15
Anticipated expiration: 2020-08-11

Abstract

A charging station, a method and device for controlling the charging station are provided. The charging station(100) includes a controller(118) configured to obtain load predictions of electric vehicles(120,122) and load predictions of a photovoltaic converter(110) for time slots over a time period; determine power values of an energy storage converter(114) for the time slots over the time period based on the load predictions and a cost model of the utility; and cause the energy storage converter(114) to supply power from an energy storage(112) for the first time slot based on a power value associated with the first time slot. The controller(118) may repeat the process with updated predictions.

Description

CHARGING STATION AND METHOD AND DEVICE FOR CONTROLLING CHARGING STATION

FIELD

Embodiments of the present disclosure generally relate to the field of electrical charging system, and in particular, to energy management in charging of electric vehicles (EVs) .

BACKGROUND

For an electric vehicle (EV) charging station, especially for medium or large scale EV charging station with DC fast chargers, the integration of photovoltaic (PV) power generation and energy storage (ES) can help save the operational cost for the station.

An existing solution is that the charging station uses utility power for EV charging only when PV power and ES power are smaller than some thresholds, and charges the extra PV power into utility when PV power and ES power are greater than some thresholds. This method can help to improve the PV utilization, but may also lead to frequent use of ES battery and thus significant battery degradation. Another existing solution improves the energy management method by designing EV charging curves based on the PV prediction results and utility peak valley electric price. However, the method may cause the problem that the EVs are not charged immediately when EV customer arrives. The characteristic is not suitable for fast charging station when the EV customers prefer to charge their EV and leave as soon as possible.

Therefore, there is a need for an improvement in energy management of an EV charging station.

SUMMARY

In general, example embodiments of the present disclosure provide a charging station and a method and device for controlling a charging station.

In a first aspect, there is provided a charging station for charging electric vehicles. The charging station comprises a utility converter coupled to a utility; a photovoltaic converter coupled to a photovoltaic cell; an energy storage converter coupled to an energy storage; and a controller. The controller is configured to: obtain a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of the photovoltaic converter for a first plurality of time slots over a first time period; determine a first plurality of power values of the energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility; obtain a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot; determine a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and cause the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.

In accordance with the embodiments of the present disclosure, based on the predicted PV and EV loads, only ES power is controlled for energy management, while EV customers can charge their EVs whenever they arrive. In addition, based on optimization window with certain time, so that the optimization accuracy can be improved with updated PV and EV load predictions.

In some embodiments, the first plurality of power values of the energy storage converter is determined by optimizing an objective function that considers the cost model of the utility. In this way, the cost of the utility may be minimized.

In some embodiments, the first plurality of power values of the energy storage converter is determined by optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell. In this way, the overall cost of the utility, energy storage, and PV cells may be minimized.

In some embodiments, the controller is further configured to: determine a power boundary of the energy storage converter for a first time in the first time slot; in response to determining the first power value is within the power boundary, supply the first output power with the first power value at the first time; in response to determining the first power value is not within the power boundary, supply an output power corresponding to the power boundary at the first time. In this way, real time operation of the charging station will be considered and inaccuracy of the predictions and optimization will be compensated.

In some embodiments, the power boundary is determined based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage. In this way, the power boundary may be accurately and comprehensively determined based on real-time feedback of the overall operation of the charging station. As a result, the optimization results will be made more accurate and the EV customers may charge their EVs whenever they arrive.

In some embodiments, the objective function is optimized under constraints comprising: an energy balance requirement; the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter; a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and a SOC of the energy storage being within a boundary of the energy storage. With the constraints, the optimization accuracy can be further improved.

In some embodiments, the objective function is optimized under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter. With the constraint, the optimization accuracy can be further improved.

In a second aspect, there is provided a method for controlling a charging station configured to charge electric vehicles. The method comprises obtaining a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of a photovoltaic converter for a first plurality of time slots over a first time period; determining a first plurality of power values of an energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility, the energy storage converter being coupled to an energy storage; obtaining a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot; determining a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and causing the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.

In some embodiments, determining the first plurality of power values of the energy storage converter comprises: optimizing an objective function that considers the cost model of the utility.

In some embodiments, determining the first plurality of power values of the energy storage converter comprises: optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell.

In some embodiments, the method further comprises determining a power boundary of the energy storage converter for a first time in the first time slot; in response to determining the first power value is within the power boundary, supplying the first output power with the first power value at the first time; in response to determining the first power value is not within the power boundary, supplying an output power corresponding to the power boundary at the first time.

In some embodiments, determining the power boundary comprising: determining the power boundary based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage.

In some embodiments, optimizing the objective function comprises optimizing the objective function under constraints comprising: an energy balance requirement; the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter; a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and a SOC of the energy storage being within a boundary of the energy storage.

In some embodiments, optimizing the objective function comprises optimizing the objective function under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter.

In a third aspect, there is provided a device for controlling a charging station configured to charge electric vehicles the device comprising a controller configured to: obtain a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of a photovoltaic converter for a first plurality of time slots over a first time period; determine a first plurality of power values of an energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility, the energy storage converter being coupled to an energy storage; obtain a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot; determine a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and cause the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.

In some embodiments, the first plurality of power values of the energy storage converter is determined by optimizing an objective function that considers the cost model of the utility.

In some embodiments, the first plurality of power values of the energy storage converter is determined by optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell.

In some embodiments, the controller is further configured to: determine a power boundary of the energy storage converter for a first time in the first time slot; in response to determining the first power value is within the power boundary, supply the first output power with the first power value at the first time; in response to determining the first power value is not within the power boundary, supply an output power corresponding to the power boundary at the first time.

In some embodiments, the power boundary is determined based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage.

In some embodiments, the objective function is optimized under constraints comprising: an energy balance requirement; the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter; a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and a SOC of the energy storage being within a boundary of the energy storage.

In some embodiments, the objective function is optimized under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter.

The method and device for controlling the charging station may achieve similar technical effects to the charging station as described above. For the sake of clarity, the technical effects and benefits are not elaborated here.

It is to be understood that the summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

Through the more detailed description of some embodiments of the present disclosure in the accompanying drawings, the above and other objects, features and advantages of the present disclosure will become more apparent, wherein:

FIG. 1 is a schematic diagram illustrating a charging station in accordance with embodiments of the present disclosure;

FIG. 2 is a schematic diagram illustrating a hierarchical control system suitable for the charging station as shown in FIG. 1 in accordance with embodiments of the present disclosure;

FIG. 3 is a plot illustrating a droop control characteristic for a power converter in accordance with embodiments of the present disclosure;

FIG. 4 is a plot illustrating a timing diagram for the control signal in accordance with embodiments of the present disclosure;

FIG. 5 is a plot illustrating droop control characteristics for converters in accordance with embodiments of the present disclosure;

FIG. 6 is a block diagram illustrating a control system for an energy storage (ES) converter in accordance with embodiments of the present disclosure;

FIG. 7 is a flowchart illustrating a method for controlling a charging station in accordance with embodiments of the present disclosure; and

FIG. 8 is a plot illustrating a timing diagram for the control system or method in accordance with embodiments of the present disclosure.

Throughout the drawings, the same or similar reference numerals represent the same or similar element.

DETAILED DESCRIPTION

Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitations as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones describe below.

As used herein, the term “includes” and its variants are to be read as open terms that mean “includes, but is not limited to. ” The term “based on” is to be read as “based at least in part on. ” The term “one embodiment” and “an embodiment” are to be read as “at least one embodiment. ” The term “another embodiment” is to be read as “at least one other embodiment. ” Other definitions, explicit and implicit, may be included below.

FIG. 1 is a schematic diagram illustrating a charging station 100 in accordance with embodiments of the present disclosure. The charging station 100 may be installed in medium scale city parking lots or in front of buildings, for example. It is to be understood that the charging station 100 is shown only for illustrative purpose without suggesting any limitation as to the scope of the present disclosure. The charging station 100 includes a utility 102 and a utility converter 106 coupled to the utility 102 and configured to convert the utility 102 to a DC bus voltage on the DC bus 116. Optionally, a transformer 104 may be coupled between the utility 102 and the utility converter 106. The transformer 104 may convert the AC voltage of the utility to a voltage range suitable for the utility converter 106.

The charging station 100 includes at least one photovoltaic (PV) cell 108 that generates electrical power from solar energy. A PV converter 110 is coupled to the PV cell 108 and may be configured to convert the PV voltage generated by the PV cell 108 to the DC bus voltage on the DC bus 116.

The charging station 100 includes energy storage (ES) 112 and an ES converter 114 coupled to the ES 112 and configured to convert the power discharged by the ES 112 to the DC bus voltage and/or convert the power on the DC bus 116 to electrical power so as to charge the ES 112.

The electric vehicles (EVs) 120-1, 120-2, and 120-N, which may be referred to collectively as EVs 120, are coupled to the DC bus 116 via the EV converters 130-1, 130-2, and 130-N, respectively. The EV converters 130-1, 130-2, and 130-N may be referred to collectively as the EV converters 130. The EVs 122-1, 122-2, and 122-N, which may be referred to collectively as EVs 130, are coupled to the DC bus 116 via the EV converters 132-1, 132-2, and 132-N, respectively. The EV converters 132-1, 132-2, and 132-N may be referred to collectively as the EV converters 132. The

EV converters

130 and 132 are configured to convert the DC bus voltage on the DC bus 116 to respective charging voltages of the

EVs

120 and 122. The term "electric vehicle" herein refers to not only pure electric vehicles, but also to hybrid vehicles that can be charged and actuated by electrical power, and the like.

The central controller 118 is communicatively coupled to the utility converter 106, the PV converter 110, the ES converter 114, and the

EV converters

130 and 132. As a result, the central controller 118 may be configured to receive feedback signals from and/or send control references to the utility converter 106, the PV converter 110, the ES converter 114, and the

EV converters

130 and 132. Each of the

converters

106, 110, and 114 may include a controller configured to receive feedback signals from the converter and send control signals to the converter. For the sake of clarity, the controllers are not shown in FIG. 1.

Although the charging station 110 has been described above with reference to FIG. 1, it is to be understood that the number and arrangement of the components such as the converters and the EVs can vary depending on various implementations and applications. In addition, the DC bus 116 may be replaced by an AC bus and the other components may be adapted accordingly.

FIG. 2 is a schematic diagram illustrating a hierarchical control system suitable for the charging station 100 in accordance with embodiments of the present disclosure. The control system may include three levels of control including a tertiary control, a secondary control, and a basic droop control. In some embodiments, the energy management control described herein may be implemented in the top level of control, for example, the tertiary control as shown in FIG. 2.

In some embodiments, the basic droop control may be implemented in a distributed way, while the secondary and tertiary controls may be implemented in a centralized way. For example, the basic droop control may be implemented in the controllers of the

converters

106, 110, and 114, and the secondary and tertiary controls may be implemented in the central controller 118.

In some embodiments, the basic droop control may be configured for automatic power distribution between the

converters

106, 110, and 114. The secondary control may be configured to regulate the DC bus voltage, and the tertiary control may be configured for energy management of the charging station 100.

As shown in FIG. 2, the tertiary control may be configured to output signal Deltal with minute scale regulating speed, for example. The secondary control may be configured to output signal Delta2 with second scale regulating speed, for example. The basic droop control may have the output (droop) characteristics as shown in FIG. 3, which is a plot illustrating a droop control characteristic for a power converter. The graph 300 shows the relationship between the nominal power of the converter and the nominal DC bus voltage, where V _{dc_n} denotes the nominal DC bus voltage when the nominal power of the converter is zero, and Delta denotes a shift from the nominal DC bus voltage.

The control signal Delta may be obtained by adding Deltal and Delta2. The control signal Delta may be sent from the central controller 118 to the controllers of the

converters

106, 110, and 114, so that the output characteristics of the

converters

106, 110, and 114 may be changed, as shown in FIG. 4. FIG. 4 is a plot illustrating a timing diagram of the control signal in accordance with embodiments of the present disclosure. In particular, the regulating speed is set to be a minute, which is illustrative without suggest any limitations to the scope of the present disclosure.

FIG. 5 is a plot illustrating droop control characteristics for a PV converter (PVC) , an ES converter (ESC) , and a utility converter (UC) . As shown in FIG. 5, with the same DC bus voltage Vdc, the power distribution among the converters can be controlled by Delta, and the DC bus voltage can be modified accordingly.

As described above, the energy management described herein may be arranged in the top level controller -tertiary control. The tertiary controller will output a control signal Delta1 for energy management, and Delta1 will be sent to ES controller with certain regulating speed (for example 60 seconds) .

FIG. 6 is a block diagram illustrating a control system 600 for an ES converter 114 in accordance with embodiments of the present disclosure. The control system 600 includes a tertiary control block 620, a secondary control block 610, a communication delay block 630, and an ES converter control block 640. The tertiary control block 620 includes an energy management system (EMS) block 622 configured to output a power reference for the ES converter P _{ESC_ref}. The tertiary control block 620 receives a feedback power for the ES converter P _{ESC_fb}, which is subtracted from the P _{ESC_ref} at the subtractor 624. The output of the subtractor 624 is provided to a proportional integral (PI) block 626. The PI block 626 provides its output to a zero-order hold block 636 to obtain Deltal. The zero-order hold block 636 may achieve a regulating speed of one minute (60 seconds) , for example. It is noted that the energy management described herein may be implemented in the block 650.

The secondary control block 610 includes a subtractor 612 that subtracts a DC bus feedback voltage V _{bus_fb} from a DC bus reference voltage V _{bus_ref}. The result is then provided to a PI block 614. The output of the PI block 614 is delayed by a zero-order hold block 632, which outputs Delta2. The zero-order hold block 632 may achieve a second scale regulating speed, for example, 50ms. The adder 636 calculates the sum of Deltal and Delta2 as Delta = Delta1 + Delta2. Delta, as an output of the communication delay block 630, is then provided to an adder 642, which adds Delta to DC bus offset voltage Vdc_n and outputs the result to a subtractor 644. The real-time feedback power P of the ES converter 114 is multiplied at block 646 by a droop coefficient k_droop to be fed into the subtractor 644. The subtractor 644 obtains a difference of its inputs and then output the difference to the subtractor 648. The subtractor 648 subtracts a real-time feedback voltage of the DC bus from the difference. The PI block 649 outputs a DC bus voltage reference as the output of the ES converter control block 640.

Although the control system 600 has been described above with reference to FIG. 6, it is to be understood that the control system 600 can vary depending on various implementations and applications. For example, although the ES converter control block 640 utilizes a loop control characteristic, any other suitable control mechanism may be employed instead.

FIG. 7 is a flowchart illustrating a method 700 for controlling a charging station 100 in accordance with embodiments of the present disclosure, and FIG. 8 is a plot illustrating a timing diagram for the control system or method in accordance with embodiments of the present disclosure. The method 700 may be implemented in the central controller 118 as shown in FIG. 1 or the block 650 as shown in FIG. 6.

At block 702, the central controller 118 obtains load predictions of the

electric vehicles

120 and 122 and load predictions of the PV converter 110 for time slots over a time period. For example, the time period may be the first time period as shown in FIG. 8. The time period may include a number of time slots, for example, 72 hours with a time slot corresponding to an hour. For the sake of clarity, reference will be made to an example time period of 72 hours and an example time slot of 1 hour. However, it is to be understood that the duration of the time period or the time slot may vary depending on various implementations or application.

In some embodiments, the predictions may be provided by a third-party and may include average predicted value in our hour for the next 72 hours. For example, 72 load prediction values of the EVs may be obtained for the next 72 hours, and 72 load prediction values of the PV converter may be obtained for the next 72 hours. The predictions may be obtained up to date at time 0, such that the central controller 118 performs the operation at block 704 with currently available data.

At block 704, the central controller 118 determines power values of the ES converter 114 for the time slots over the first time period based on load predictions of the

EVs

120 and 122 and the PV converter 110 and a cost model of the utility. The power values may be the average power references of the ES converter 114 during respective time slots.

Based on the predicted PV and EV loads, only ES power is controlled for energy management, while EV customers can charge their EVs whenever they arrive and leave quickly. In addition, based on optimization window with certain time, so that the optimization accuracy can be improved with updated PV and EV load predictions.

In some embodiments, the central controller 118 may determine the power values of the ES converter 114 by optimizing an objective function that considers the cost model of the utility. For example, the cost model of the utility considers the peak/valley price. In some embodiments, the power values of the ES converters 114 may be obtained by minimizing the cost of the utility based on the load predictions of the

EVs

120 and 122 and the PV converter 110.

In some embodiments, the objective function may further consider a cost model of degradation of the ES and/or a cost model of degradation of the PV cell (s) . For example, the objective function may be the sum of the cost model of the utility, the cost model of degradation of the ES and/or the cost model of degradation of the PV cell (s) .

The objective function may calculate the operational cost of the charging station for a predefined duration, for example one day (24 hours) . The input array x (i) of the objective function is the average ES converter power in one hour, which includes 72 values for the following 72 hours, x (i) , i=1, 2...72. The output of the objective function is the operational cost in one day, for example. As described above, the operational cost may include an operational cost for electric charge, an operational cost for battery degradation, and/or an operational cost for PV degradation.

The average utility converter power in each hour may be calculated based on the ES converter power x (i) , predicted PV power generation PVC (i) and EV load requirement EVL (i) . The ES converter power x (i) is the variable to be optimized, while the predicted PV power generation PVC (i) and EV load requirement EVL (i) are obtained at block 702. The operational cost for electricity bill in each hour can be calculated considering the cost model of the utility, for example, peak/valley price.

In some embodiments, the objective function may consider degradation caused by shelf life and/or degradation caused by charge charge/discharge. For example, the shelf life may be defined as the time until the battery is degraded to a predefined value, for example 80%. If a battery has a shelf life of 20 years, the battery will experience 1%of degradation each year. Therefore, the shelf life of the battery may be calculated by:

where ΔC _shelf (t) denotes the degradation of battery capacity caused by shelf life in t days, C ₀ denotes the initial capacity of a new battery. In the example above, the shelf life is 20*365=7300 days and the unit oft is day.

In some embodiments, if the Depth of Discharge (DOD) is constant, the capacity loss in one charging/discharging cycle may be estimated as the reciprocal of the lifecycle number. However, the DOD of the ES battery is fluctuant in real application. To estimate the capacity loss of ES battery based on this fluctuant DOD, the SOC change of ES battery within each continuous charging or discharging period (even if the charging or discharging rate varies in this period) is calculated. The SOC change in the i ^th period may be denoted as ΔSOC _i.

As an example, the lifecycle of a Li-ion battery is about 5000 cycles when its Depth of Discharge (DOD) is 100%, while the lifecycle of a Li-ion battery is about 400000 cycles when its DOD is 3%. The lifecycle was tested with constant discharging rate in each charging and discharging cycle during the whole life time.

If the ES battery is charged or discharged with a constant DOD of ΔSOC _i, the lifecycle (Cycle _i) of the battery may be calculated by:

where DOD ₁ = 3%, DOD ₂ = 80%, Cycle ₁ = 400000, Cycle ₂ = 5000.

The battery capacity loss (ΔC _i) caused by the i ^th charging or discharging period may be estimated by:

where C ₀ denotes initial capacity of a fresh new battery. The factor of 0.5 is applied considering that the i ^th period only covers either a charging or a discharging process, not both.

In some embodiments, the battery degradation in t days may be represented by the sum of ΔC _shelf (t) and Σ ΔC _i. For example, if the battery price is 200USD/kWh, the ES battery operation cost caused by battery degradation may be represented by:

Cost_battery (t) = (ΔC _shelf (t) + Σ ΔC _i) × 200 (USD/kWh) (7)

In some embodiments, the central controller 118 optimizes the objective function under a number of constraints. The constraints may include at least one of an energy balance requirement; the power values of the ES converter 114 being within nominal power capacity of the ES converter 114; the power values of the utility converter 106 being within nominal power capacity of the utility converter 116; and a SOC of the ES 112 being within a boundary of the ES 112. In some embodiments, the constraints further comprise the power values of the PV converter 110 being within nominal power capacity of the PV converter 110.

For example, the constraints may be represented by

1. Energy balance requirement:

ESC_power (i) +PVC_power (i) +UC_power (i) +EVL_power (i) =0,

where ESC_power (i) represents the power reference of the ES converter in the i ^th time slot, PVC_power (i) represents the power reference of the PV converter in the i ^th time slot, UC_power (i) represents the power reference of the utility converter in the i ^th time slot, and EVL_power (i) represents the power reference of the EV loads in the i ^th time slot;

2. The power references of the ES converter being within a nominal power capacity of the ES converter:

ESC_power_min ≤ ESC_power (i) ≤ ESC_power_max,

where ESC_power_min and ESC_power_max represent minimum and maximum power capacities of the ES converter, respectively;

3. The power references of the ES converter being within a nominal power capacity of the PV converter:

0 ≤ PVC_power (i) ≤ PV_power_max (i) ,

where PV_power_max represents a maximum power capacity of the PV converter;

4. The power references of the utility converter being within a nominal power capacity of the utility converter:

0 _≤ UC_power (i) ≤ UC_power _max,

where UC_power_max represents a maximum power capacity of the utility converter;

5. The SOC of the energy storage being within a boundary of the energy storage:

SOC_min _≤ SOC (i) ≤ SOC_max,

where SOC_min and SOC_max are minimum and maximum power capacities of the SOC, and SOC (i) may be calculated by SOC (i) =SOC0+Σ [x (i) *1hour÷Battery_Capacity] .

The parameters of the objective function may include the PV power prediction values PVC (i) and EV load prediction values EVL (i) . The optimization results of the objective function may be the average ES converter power in a time slot, for example, one hour, totally 72 output values for the following 72 hours, x (i) , i=1, 2...72. For example, the objective function may be optimized by an optimization function in Matlab/optimization toolbox, FMINCON.

The optimization results by minimizing the objective function are an array of average ESC power references in a time slot (for example, one hour) for the time period. If the time period is 72 hours, the results include 72 output values for the next 72 hours, x (i) , i=1, 2, ..., 72.

At block 706, the central controller 118 causes the ES converter to supply output power based on a power value from the ES for a time slot. The power value is one of the power values associated with the time slot. The time slot is the first time slot in the time slots over the time period. For example, the time slot may be the time from 0 to Ti, as shown in FIG. 8.

In some embodiments, the PV and EV loads fluctuate in real time, and even if the predictions of PV power and EV load are correct, deviation between the real time value and the predicted average value in one hour still exists. To keep power balance of the station in real time, the power reference of each converters for energy management should be further calculated based on not only the optimized average power reference ESC (i) , but also the real time feedback of the power of the converters and/or the SOC of the ES battery.

In some embodiments, the central controller 118 may determine a power boundary of the ES converter for a first time in the first time slot. For example, the first time slot is from 0 to T1 and the first time may be any time between 0 and T1. If it is determined that the power value obtained at block 704 (as referred to as average power reference) is within the power boundary of the ES converter 114, the ES converter 114 will be controlled to supply the output power with the power value at the first time. If it is determined that the power value is not within the power boundary, the ES converter 114 will be controlled to supply an output power corresponding to the power boundary at the first time.

For example, the central controller 118 may calculate the power boundary of the ES converter 114, for example, the upper and lower power limits Limit_max and Limit_min of the ES converter 114. If it is determined that Limit_min≤x (1) ≤Limit_max, the final ES converter power reference y (1) =x (1) , if it is determined that Limit_min＞x (1) , y (1) =Limit_min, and if it is determined that x (1) ＞Limit_max, y (1) =Limit_max. The final ES converter power reference y (1) is sent from central controller 118 to the ES converter 114 during 0＜t＜T1.

In some embodiments, the central controller 118 may receive real-time feedback including at least one of an output power of the PV converter 110, a load requirement of the

EVs

120 and 122, upper and lower power limits of the utility converter 106, upper and lower power limits of the ES converter 114, and a State of Charge (SOC) of the ES 112. The central controller 118 may calculate the power boundary (for example, the upper and lower power limits Limit_max and Limit_min) of the ES converter 114 based on the real-time feedback.

Referring to FIG. 8 again, at the time Ti, the method 700 may proceed to block 702. The central controller 118 may obtain load predictions of the

EVs

120 and 122 and load predictions of the PV converter 110 for the time slots over the second time period, for example, 72 hours from T1 as shown in FIG. 8. The second time period is offset from the first time period by the time slot from 0 to T1. The load predictions may be obtained up to date for the time T1, such that the average power reference for the ES converter 114 will be determined based on updated data so as to improve accuracy of energy management. As a result, the

EVs

120 and 122 may be charged as they arrive at the charging station 100. The method 700 may proceed to block 704 and repeat the process.

In accordance with the embodiments of the present disclosure, based on the predicted PV and EV loads, only ES power is controlled for energy management, while EV customers can charge their EVs whenever they arrive. In addition, based on optimization window with certain time, so that the optimization accuracy can be improved with updated PV and EV load predictions and optionally ES battery SOC feedback value and/or the like.

Although the present disclosure has been described in language specific to structural features and/or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

A charging station for charging electric vehicles comprising:

a utility converter coupled to a utility;

a photovoltaic converter coupled to a photovoltaic cell;

an energy storage converter coupled to an energy storage; and

a controller configured to:

obtain a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of the photovoltaic converter for a first plurality of time slots over a first time period;

determine a first plurality of power values of the energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility;

obtain a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot;

determine a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and

cause the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.
The charging station of claim 1, wherein the first plurality of power values of the energy storage converter is determined by:

optimizing an objective function that considers the cost model of the utility.
The charging station of claim 2, wherein the first plurality of power values of the energy storage converter is determined by:

optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell.
The charging station of claim 1, wherein the controller is further configured to:

determine a power boundary of the energy storage converter for a first time in the first time slot;

in response to determining the first power value is within the power boundary, supply the first output power with the first power value at the first time;

in response to determining the first power value is not within the power boundary, supply an output power corresponding to the power boundary at the first time.
The charging station of claim 4, wherein the power boundary is determined based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage.
The charging station of claim 2 or 3, wherein the objective function is optimized under constraints comprising:

an energy balance requirement;

the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter;

a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and

a SOC of the energy storage being within a boundary of the energy storage.
The charging station of claim 6, wherein the objective function is optimized under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter.
A method for controlling a charging station configured to charge electric vehicles, the method comprising:

obtaining a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of a photovoltaic converter for a first plurality of time slots over a first time period;

determining a first plurality of power values of an energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility, the energy storage converter being coupled to an energy storage;

obtaining a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot;

determining a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and

causing the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.
The method of claim 8, wherein determining the first plurality of power values of the energy storage converter comprises:

optimizing an objective function that considers the cost model of the utility.
The method of claim 9, wherein determining the first plurality of power values of the energy storage converter comprises:

optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell.
The method of claim 8, further comprising:

determining a power boundary of the energy storage converter for a first time in the first time slot;

in response to determining the first power value is within the power boundary, supplying the first output power with the first power value at the first time;

in response to determining the first power value is not within the power boundary, supplying an output power corresponding to the power boundary at the first time.
The method of claim 11, wherein determining the power boundary comprising:

determining the power boundary based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage.
The method of claim 9 or 10, wherein optimizing the objective function comprises optimizing the objective function under constraints comprising:

an energy balance requirement;

the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter;

a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and

a SOC of the energy storage being within a boundary of the energy storage.
The method of claim 13, wherein optimizing the objective function comprises optimizing the objective function under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter.
A device for controlling a charging station configured to charge electric vehicles the device comprising a controller configured to:

obtain a first plurality of load predictions of the electric vehicles and a second plurality of load predictions of a photovoltaic converter for a first plurality of time slots over a first time period;

determine a first plurality of power values of an energy storage converter for the first plurality of time slots over the first time period based on the first and second plurality of load predictions and a cost model of the utility, the energy storage converter being coupled to an energy storage;

obtain a third plurality of load predictions of the electric vehicles and a fourth plurality of load predictions of the photovoltaic converter for a second plurality of time slots over a second time period, the second plurality of time slots being offset from the first plurality of time periods by a first time slot;

determine a second plurality of power values of the energy storage converter for the second plurality of time slots over the second time period based on the third and fourth plurality of load predictions and the cost model of the utility; and

cause the energy storage converter to supply first output power based on a first power value from the energy storage for the first time slot, and to supply second output power based on a second power value from the energy storage for a second time slot following the first time slot, the first power value being one of the first plurality of power values associated with the first time slot and the second power value being one of the second plurality of power values associated with the second time slot.
The device of claim 15, wherein the first plurality of power values of the energy storage converter is determined by:

optimizing an objective function that considers the cost model of the utility.
The device of claim 16, wherein the first plurality of power values of the energy storage converter is determined by:

optimizing the objective function that further considers a cost model of degradation of the energy storage and a cost model of degradation of the photovoltaic cell.
The device of claim 15, wherein the controller is further configured to:

determine a power boundary of the energy storage converter for a first time in the first time slot;

in response to determining the first power value is within the power boundary, supply the first output power with the first power value at the first time;

in response to determining the first power value is not within the power boundary, supply an output power corresponding to the power boundary at the first time.
The device of claim 18, wherein the power boundary is determined based on an output power of the photovoltaic converter, a load requirement of the electric vehicles, upper and lower power limits of the utility converter, upper and lower power limits of the energy storage converter, and a State of Charge (SOC) of the energy storage.
The device of claim 16 or 17, wherein the objective function is optimized under constraints comprising:

an energy balance requirement;

the first plurality of power values of the energy storage converter being within nominal power capacity of the energy storage converter;

a third plurality of power values of the utility converter being within nominal power capacity of the utility converter; and

a SOC of the energy storage being within a boundary of the energy storage.
The device of claim 20, wherein the objective function is optimized under constraints further comprising a fourth plurality of power values of the photovoltaic converter being within nominal power capacity of the photovoltaic converter.