Electric power system

Abstract

To provide an electric power system appropriately combining centralized control and decentralized control.SOLUTION: An electric power system 100 including a plurality of groups each combining a plurality of batteries comprises: a central control unit 102 that determines discharging or charging target electric power for each group; and a decentralized controller C that controls batteries so as to output the determined target electric power to each group.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a power system that combines centralized control and distributed control. [Background technology]

[0002] A technology has been disclosed for a stationary energy storage system that combines multiple types of batteries with different properties, such as lead batteries, lithium-ion batteries, and lithium-ion capacitors, in which the long-cycle components of the system's power demand from the power grid are borne by the lead batteries and the short-cycle components are borne by the lithium-ion batteries, thereby optimizing the cost of the entire system (Non-Patent Document 1). [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] K.Takeda, C.Takahashi, H.Arita, N.Kusumi, M.Amano, A.Emori, “Design of Hybrid Energy Storage System using Dual Batteries for Renewable Applications”, IEEE PES General Meeting Power & Energy Society (2014) Summary of the Invention [Problem to be solved by the invention]

[0004] In a stationary energy storage system using different types of batteries, it is possible to distribute the power required by the system according to the characteristics of the batteries. However, when multiple batteries suitable for long cycle components and multiple batteries suitable for short cycle components are installed, there is no mention of how to distribute the power of the long cycle components and the power of the short cycle components to each battery. For example, it is not clear whether the power should be simply distributed equally depending on the number of batteries.

[0005] Also, in order to distribute power, a method is considered in which the target power of each battery is determined according to the SOC of the battery. Specifically, when there is a discharge request from the power grid, the target power (discharge power) is set to be higher for a battery with a higher SOC, and conversely, when there is a charge request, the target power (charge power) is set to be higher for a battery with a lower SOC. However, in the conventional technology, in order to set the target power by applying this method, a central control server constantly grasps the SOC of all the batteries and calculates each target power based on it. Therefore, as the number of batteries increases, the calculation cost and calculation time of the central control server increase, resulting in a problem of an increase in the operating cost of the system. [Means for solving the problem]

[0006] One aspect of the present invention is a power system including multiple groups each combining multiple batteries, the power system comprising a central control unit that determines a target power for discharging or charging for each of the groups, and distributed controllers provided in the groups and that control the batteries to output the target power determined for each of the groups.

[0007] Here, it is preferable that the central control unit determines the target power for discharging or charging for each of the groups so as to minimize an objective function J for the characteristic amount of the battery while satisfying a constraint condition.

[0008] The objective function J and the constraints are expressed as the set i (i=1, 2,...N g ) Battery B ij (j=1,2,M i ) based on the feature quantity of the set i, and the target power r i It is preferable to determine

[0009] It is also preferable that the distributed controller determines the target power for each battery in the group in accordance with a characteristic amount of the battery belonging to the group so as to track the target power for that battery.

[0010] It is also preferable that the distributed controller determines the target power for the batteries belonging to the group based on an error between the target power for the group from the central control unit and the actual output power of the group, and the SOC of the batteries belonging to the group.

[0011] In addition, to determine the target power for each of the groups, it is preferable to use the maximum discharge power of the group, the minimum discharge power of the group, the maximum charge power of the group, and the minimum charge power of the group, as well as the target powers for each of the groups in the past.

[0012] It is also preferable to further use an average SOC of batteries belonging to the group to determine the target power for each of the groups.

[0013] It is also preferred to further use an average SOC of batteries belonging to all of said groups to determine said target power for each of said groups. Effect of the Invention

[0014] According to the present invention, it is possible to provide a power system in which a plurality of batteries are divided into a plurality of groups, and in which centralized control and distributed control are appropriately combined. [Brief description of the drawings]

[0015] [Figure 1] 1 is a diagram showing a power system according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a diagram illustrating definitions of physical quantities according to an embodiment of the present invention. [Diagram 3] FIG. 4 is a diagram showing a method of searching for a target power in the embodiment of the present invention. [Figure 4]FIG. 4 is a diagram showing setting values ​​of parameters used in a simulation of power control of a power system in an embodiment of the present invention. [Diagram 5] FIG. 11 is a diagram showing a result of a simulation of power control of the power system in the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] As shown in FIG. 1, a power system 100 according to an embodiment of the present invention includes a central control unit 102, monitoring units 104 (104-1 to 104-N g ), Battery B (B 11 ~B 1M1 ,···,B Ng1 ~B NgMNg ), and distributed controller C(C 11 ~C 1M1 ,···C Ng1 ~C NgMNg )

[0017] The power system 100 can be applied to a stationary power storage system used as a power buffer for a renewable energy system such as solar power generation or wind power generation, or to a vehicle Virtual Power Plant system (vehicle VPP). The power system 100 controls the discharge power supplied from battery B and the charge power supplied to battery B (hereinafter, collectively referred to as output power) so as to satisfy the power required from the power grid (system required power).

[0018] In the power system 100, the battery B is divided into a plurality of groups (group 1 to group N g In the example of the power system 100 shown in FIG. 1, the battery B 11 ~B 1M1 Battery B belongs to 11 ~B 1M1 Power y from each of 11 ~y 1M1 is output. Battery B in Group 1 11 ~B 1M1 The power output from each of 11~y 1M1 are added together and output as the total power y1 of group 1. The same applies to the other groups. Group N g Battery B Ng1 ~B NgMNg Battery B belongs to Ng1 ~B NgMNg Power y from each of Ng1 ~y NgMNg is output. Group N g Battery B Ng1 ~B NgMNg The power output from each of Ng1 ~y NgMNg are added together to form the group N g The total power of y Ng will be output as:

[0019] In the power system 100, battery B belonging to group 1 11 ~B 1M1 For each of 11 ~C 1M1 The other groups are similar, for example, group N g Battery B belonging to Ng1 ~B NgMNg For each of Ng1 ~C NgMNg will be established.

[0020] The distributed controller C is provided for each battery B and controls the power y output from the battery B. The distributed controller C sets the target power r to be output from each battery B based on the difference e between the target power r for each group and the total power y of the group. For example, for battery B belonging to group 1, 11 The corresponding distributed controller C 11 is calculated based on the difference e1 between the target power r1 in group 1 and the total power y1 in group 1. 11 The target power r to be output from 11 The other distributed controllers C are set in the same way. For example, g Battery B belonging to Ng1 The corresponding distributed controller C Ng1 is the group N g Target power at r Ng and the group N g Total power in yNg Difference of e Ng Based on battery B Ng1 The target power r to be output from Ng1 The method of setting the target power r in the distributed controller C will be described later.

[0021] A monitoring unit 104 is provided for each group of batteries B. The monitoring unit 104 monitors and detects the state of the batteries B belonging to each group. The total power y output from the batteries B belonging to each group is input to the monitoring unit 104. The monitoring unit 104 also monitors and detects the SOC, maximum discharge power, minimum discharge power, maximum charge power, and minimum charge power for each of the batteries B belonging to each group. The monitoring unit 104 then outputs the average SOC, maximum discharge power, minimum discharge power, maximum charge power, and minimum charge power of each group to the central control unit 102.

[0022] For example, the monitoring unit 104-1 belonging to group 1 is connected to the battery B 11 ~B 1M1 The monitoring unit 104-1 monitors and detects the total power y1 output from the battery B belonging to group 1. 11 ~B 1M1 For each of the SOCs 1j , maximum discharge power u D1j , minimum discharge power l D1j , maximum charging power u C1j , minimum charging power l C1j Then, the monitoring unit 104-1 monitors and detects the average SOCs1 and maximum discharge power U D1 , minimum discharge power L D1 , maximum charging power U C1 , minimum charging power L C1 The other monitoring units 104 are similarly calculated and output to the central control unit 102. For example, g Monitoring unit 104-N belonging to g is the group N g Battery B belonging to Ng1 ~B NgMNg The total power output from y NgIn addition, the monitoring unit 104-N g is the group N g Battery B belonging to Ng1 ~B NgMNg For each of the SOCs Ngj , maximum discharge power u DNgj , minimum discharge power l DNgj , maximum charging power u CNgj , minimum charging power l CNgj (j=1,2,M Ng ) and monitoring unit 104-N g Based on this information, group N g About average SOCs Ng , maximum discharge power U DNg , minimum discharge power L DNg , maximum charging power U CNg , minimum charging power L CNg is calculated and output to the central control unit 102.

[0023] For each group i (i = 1, 2, , N g ) maximum discharge power U Di , minimum discharge power L Di , maximum charging power U Ci , minimum charging power L Ci can be calculated by the following formulas (1) to (4).

number

number

number

number

[0024] The maximum discharge power of each group can be set as the sum of the maximum discharge powers of the batteries B belonging to the group, as shown in formula (1). However, from the viewpoint of battery protection, if it is desired to avoid operating all of the batteries B at the maximum discharge power as much as possible, the maximum discharge power of each group may be set to be smaller than the sum of the maximum discharge powers of the batteries B belonging to the group. Furthermore, the maximum charge power of each group can be set as the sum of the maximum charge powers of the batteries B belonging to the group, as shown in formula (3). However, from the viewpoint of battery protection, if it is desired to avoid operating all of the batteries B at the maximum charge power as much as possible, the maximum charge power of each group may be set to be smaller than the sum of the maximum charge powers of the batteries B belonging to the group.

[0025] Moreover, the minimum discharge power of each group can be set as the sum of the minimum discharge power of the battery B belonging to the group, as shown in formula (2). However, in order to operate the power converter connected to the battery B in a region where the power conversion efficiency is high, it is preferable that the target power of each battery B is larger than the minimum discharge power of each battery B. Therefore, the minimum discharge power of each group may be set to be larger than the sum of the minimum discharge power of the battery B belonging to the group. Moreover, the minimum charge power of each group can be set as the sum of the minimum charge power of the battery B belonging to the group, as shown in formula (4). However, in order to operate the power converter connected to the battery B in a region where the power conversion efficiency is high, it is preferable that the target power of each battery B is larger than the minimum charge power of each battery B. Therefore, the minimum charge power of each group may be set to be larger than the sum of the minimum charge power of the battery B belonging to the group.

[0026] Also, the monitoring unit 104 may not be installed, and the central control unit 102 may have the functions of the monitoring unit 104. However, in order to reduce the calculation cost and calculation time of the central control unit 102, it is preferable to install the monitoring unit 104.

[0027] The central control unit 102 receives a required power r for the system from the power grid to which the power system 100 is connected, and calculates target power r (r1 to r Ng At this time, the central control unit 102 uses the average SOC, maximum discharge power, minimum discharge power, maximum charge power, and minimum charge power of each group received from the monitoring unit 104 to solve an optimization problem with constraint conditions to set the target power r (r1 to r Ng ) is set for each group in the central control unit 102. Ng The setting method for this will be described later.

[0028] Although the system required power and the target power for each battery B depend on time, in this embodiment, the time-dependent physical quantity A(t) is simply referred to as the physical quantity A.

[0029] [Power system control method] The following describes a control method in the power system 100. In the power system 100, the discharge power output from each battery B or the charge power supplied to each battery B is controlled by combining centralized control by the central control unit 102 and distributed control by distributed controllers C corresponding to each battery B belonging to each group. Note that Fig. 2 shows physical quantities used in the following control method.

[0030] The central control unit 102 receives the system required power r and each group i (i=1, 2, . . . , N g ) average SOCsi, maximum discharge power U Di , minimum discharge power L Di , maximum charging power U Ci , minimum charging power L Ci Based on this, we solve the constrained optimization problem (minimization problem) to obtain the target power r for each group i. i Specifically, the central control unit 102 sets the target power r i Determine.

number

[0031] Here, the estimated mean SOC of group i one step ahead, s ^ i (t) can be expressed by equation (6).

number

[0032] Moreover, the average SOCs(t) of all batteries B included in the power system 100 can be expressed by equation (7).

number

[0033] As constraints for solving the formula (5), the system required power r(t) for the power system 100, the power condition r in the group i, i (t) and the SOC conditions in group i are set as formulas (8) to (10), respectively.

number

number

number

[0034] Here, g included in the objective function J in formula (5) i ,q i (i=1,2,...,N g ) is an arbitrary design parameter. The time step T is the time step for updating the target power of each group, and when entering a market where the amount of power traded is in 30-minute intervals, such as the Japanese wholesale electricity market, it is preferable to set the time step T to a value of 30 minutes, for example.

[0035] Each term of the objective function J was set with the following intention.

[0036] <First term of objective function J> Since the capacity of the battery B tends to deteriorate when the SOC is extremely high or low, it is preferable to make the SOC of all the batteries B constituting the power system 100 uniform so that the capacity deterioration of the batteries B is as uniform as possible. i Average SOCs one step ahead from (t) ^ i (t) is estimated using equation (6) and is made to match the average SOCs(t) of all batteries B at the current time, which is expressed by equation (7).

[0037] In addition, the constraint condition in equation (9) is set so that when a discharge request is made from the power grid (r(t)>0), none of the groups will charge (discharge or output power will be 0), and when a charge request is made (r(t)<0), none of the groups will discharge (charge or output power will be 0).

[0038] This allows control such that when a discharge request is made, the discharge amount is increased for the group with a higher SOC, and when a charge request is made, the charge amount is increased for the group with a lower SOC. As a result, the SOCs of the batteries B included in the power system 100 become uniform.

[0039] <Second term of objective function J> In order to prevent the total power from becoming oscillatory, it is preferable to operate the battery B at a constant power as much as possible. i The second term was introduced to reduce the fluctuation of (t).

[0040] A group i (i=1, 2, . . . N) is selected so that the capacity deterioration of all batteries B included in the power system 100 is as uniform as possible and the objective function J is minimized under the constraints of formulas (8) to (10) in order to operate the batteries at a constant power. g ) target power r i Explore (t).

[0041] Figure 3 shows the target power r iA specific method for searching for (t) is shown below. The target power r i (t) (i = 1, 2, . . . N g Once the target power r i (t) and the output power y i Difference e of (t) i Based on (t), the distributed controller C ij Battery B ij Target power of r ij Determine (t). Battery B ij is the target power r ij (t) depending on power y ij Charging and discharging are performed by (t).

[0042] Battery B ij Output power of y ij (t) is expressed by formulas (11) to (13).

number

number

number

[0043] Battery B ij Target power of r ij When (t) is positive, i.e., discharge is requested, battery B ij The power output from y ij (t) is the target power r ij (t) depending on the minimum discharge power l Dij Maximum discharge power u Dij The battery B is controlled as follows. ij Target power of r ij When (t) is negative, i.e., charging is requested, battery B ij The power supplied to y ij (t) is the target power r ij (t) depending on the minimum charging power l Cij Maximum charging power over Cij The battery B is controlled as follows.ij Target power of r ij When (t) is 0, battery B ij Charging and discharging power of y ij (t) is set to 0. Under such constraints, the entire power system 100 is controlled so that the objective function J is minimized.

[0044] In this way, by setting a lower limit on the discharge power and the charge power, when the required power of battery B is close to 0, ij This prevents the sudden switching of charging and discharging of battery B from causing vibration. ij This makes it possible to suppress sudden charging and discharging of the battery.

[0045] The validity of the power control in the power system 100 in this embodiment is verified by simulation. The system required power for the power system 100 is the target power r i (t) was set in the same pattern as the example searching.

[0046] The power system 100 includes 15 batteries B (#1 to #15), and the initial SOC, maximum discharge power, and the like are set as shown in FIG. g =3), with batteries #1 to #5 in group 1, batteries #6 to #10 in group 2, and batteries #11 to #15 in group 3.

[0047] The capacity of each battery B was 1.2 kWh, and was the same for all 15 batteries B. In other words, the total capacity of the batteries in each group was P1 = P2 = P3 = 1.2 kWh × 5 = 6 kWh. In addition, the maximum discharge power U Di ,Minimum discharge power L Di are set to 4 kW and 1 kW, respectively. That is, the maximum discharge power U Dij is set to 0.8 kW, and the maximum discharge power U Di The minimum discharge power of each battery B is 0.8kW × 5 = 4kW. Dijis set to 0.1 kW, and the minimum discharge power L of the five batteries B in each group is Dij The total is 0.1kW × 5 = 0.5kW, but the minimum discharge power L Di was set to a larger value of 1 kW. Similarly, the maximum charging power U Ci ,Minimum charging power L Ci were set to -4kW and -1kW, respectively.

[0048] In addition, any design parameter g included in the objective function J i ,q i (i=1,2,3) are 2×10 11 and 1, and the time step T was 60 minutes.

[0049] Fig. 5 shows the results of simulating power control in the power system 100. Fig. 5 shows the system required power for the power system 100 (Fig. 5(a)), the target power setting values ​​for each group (Fig. 5(b)), and the power (Fig. 5(c)) and SOC (Fig. 5(d)) of battery B included in each group over time.

[0050] As shown in FIG. 5(c), the power of each battery B in each group is in the unusable region, i.e., the minimum discharge power L Dij (=0.1kW) or less, and when charging is requested, the minimum charging power L Cij The discharge power or charge power of each battery B was controlled so as not to enter the region of (=-0.1 kW) or more. In addition, the target power was determined so that the SOC of each battery B was as uniform as possible, and some groups were rested without being charged or discharged.

[0051] In this embodiment, control is performed to make the average SOC of each group as uniform as possible, but if constraints on SOC are set for each group and it is acceptable to vary the SOC based on those conditions, it is not necessary to control to make the average SOC of each group uniform.

[0052] As described above, according to the power system 100 of this embodiment, the power converter can be operated in a region where the power conversion efficiency is high. In other words, the power loss of the power converter can be reduced. In a stationary energy storage system and a vehicle VPP, in order to reduce the power loss (Joule loss) that occurs when transmitting power from a battery connected to these systems to a power grid, the voltage is increased by the power converter before transmission, and the greater the power, the greater the power conversion efficiency of the power converter can be.

[0053] That is, in power system 100 in this embodiment, the target power of a certain group is set to a large value, and the batteries belonging to that group are operated at a relatively large power, while the target power of another group is set to 0, and the batteries belonging to that group are allowed to rest without being charged or discharged. In this way, by allowing the batteries of some groups to rest, the central control unit determines the target power of each group so that the target power per operating battery does not become small, and therefore the power converter can be operated in a range where the power conversion efficiency is high.

[0054] Furthermore, when attempting to operate the batteries and power converters in a region with high power conversion efficiency, the computation cost and computation time of the central control unit increases as the number of batteries increases. Furthermore, since information on all batteries is required for calculation, the system is vulnerable to delays associated with information exchange between the batteries and the central control unit. In other words, when a delay occurs, the central control unit calculates the target power of each battery based on the delayed information, and as a result, a time lag occurs between the system required power and the target power of each battery, deteriorating tracking. In contrast, the power system 100 of this embodiment can reduce the computation cost and computation time of the central control unit.

[0055] That is, in the power system 100 according to the present embodiment, the physical quantity to be calculated is the target power of the battery group, and the number of the battery groups is equal to the number of groups. Therefore, the calculation cost and calculation time of the central control unit can be reduced compared to simple centralized control.

[0056] Also, the actual total power of all batteries B included in power system 100 can be made to follow the system required power without fluctuating over time. That is, in power system 100 of the present embodiment, the central control unit determines the target power of each group so that the target power per battery in operation does not become small, making it possible to suppress the oscillation of the total power.

[0057] [Configuration of the invention] [Configuration 1] In a power system including a plurality of groups each including a plurality of batteries, a central control unit that determines a target power for discharging or charging for each of the groups; A power system comprising: a distributed controller provided in each of the groups, the distributed controller controlling batteries to output the target power determined for each of the groups. [Configuration 2] 2. The power system according to claim 1, The power system according to claim 1, wherein the central control unit determines the target power for discharging or charging for each of the groups so as to minimize an objective function J for a characteristic quantity of the battery while satisfying a constraint condition. [Configuration 3] 3. The power system according to claim 2, The objective function J and the constraints are expressed as the set i (i=1, 2,...N g ) Battery B ij (j=1,2,M i ) based on the feature quantity of the set i, and the target power r i The power system according to claim 1, [Configuration 4] 3. The power system according to claim 2, The power system according to claim 1, wherein the distributed controller determines the target power for each battery in the group in accordance with a characteristic amount of the battery belonging to the group so as to track the target power for the battery. [Configuration 5] 5. The power system according to claim 4, The power system is characterized in that the distributed controller determines the target power for the battery belonging to the group based on an error between the target power for the group from the central control unit and an actual output power of the group, and a SOC of the battery belonging to the group. [Configuration 6] The power system according to any one of configurations 1 to 5, A power system comprising: a maximum discharge power of the group; a minimum discharge power of the group; a maximum charge power of the group; and a minimum charge power of the group; and the target powers for each of the groups in the past, for determining the target power for each of the groups. [Configuration 7] 7. The power system according to claim 6, The power system further comprising an average SOC of batteries in said groups to determine said target power for each of said groups. [Configuration 8] 8. The power system according to claim 7, The power system further comprising an average SOC of batteries in all of said groups to determine said target power for each of said groups. [Explanation of symbols]

[0058] 100 power system, 102 central control unit, 104 monitoring unit.

Claims

1. In a power system including a plurality of groups each including a plurality of batteries, a central control unit that determines a target power for discharging or charging for each of the groups; A power system comprising: a distributed controller provided in each of the groups, the distributed controller controlling batteries to output the target power determined for each of the groups.

2. 2. The power system of claim 1, The power system according to claim 1, wherein the central control unit determines the target power for discharging or charging for each of the groups so as to minimize an objective function J for a characteristic amount of the battery while satisfying a constraint condition.

3. 3. The power system of claim 2, The objective function J and the constraints are expressed as the set i (i=1, 2, . . . N ｇ ) Battery B ｉｊ (j=1, 2,...M ｉ ) based on the feature quantity of the group i, and the target power r ｉ The power system according to claim 1,

4. 4. The power system of claim 3, The power system according to claim 1, wherein the distributed controller determines the target power for each battery in the group in accordance with a characteristic amount of the battery belonging to the group so as to track the target power for the battery.

5. 5. The power system of claim 4, The power system is characterized in that the distributed controller determines the target power for the battery belonging to the group in accordance with an error between the target power for the group from the central control unit and an actual output power of the group, and an SOC of the battery belonging to the group.

6. The power system according to any one of claims 1 to 5, A power system comprising: a maximum discharge power of the group; a minimum discharge power of the group; a maximum charge power of the group; and a minimum charge power of the group; and the target powers for each of the groups in the past, for determining the target power for each of the groups.

7. 7. The power system of claim 6, 11. The power system of claim 10, further comprising: an average SOC of batteries in said groups for determining said target power for each of said groups.

8. 8. The power system of claim 7, 11. The power system of claim 10, further comprising: an average SOC of batteries in all of said groups to determine said target power for each of said groups.

Images