KR102272665B1 - MAS(Multi-Agent Systems)-based distributed control system and method in DC microgrid - Google Patents

Abstract

The present invention relates to a MAS-based distributed control system and method in a DC microgrid.
The MAS-based distributed control method in a DC microgrid according to the present invention comprises the steps of: a) acquiring information used for decision making for determining a control mode by a control unit through a communication network; b) determining whether a grid failure (failure) by the control unit based on the obtained information; c) determining whether the ESS or RES is faulty (failure) by the control unit based on the obtained information; d) If it is not a grid failure (failure) in the determination of step b), adjust the DCV and balance the system power according to the supply-demand power relationship by the control unit, and implement the grid recovery control algorithm when the grid recovers from the failure activating to determine a control mode; and e) if it is not a failure (failure) of the ESS or RES in the determination of step c), the control unit activates the DCV (DC-link) recovery algorithm or the grid recovery detection algorithm according to the DCV control capability of the grid and battery agent to determine a control mode.

Description

MAS (Multi-Agent Systems)-based distributed control system and method in DC microgrid}

The present invention relates to a MAS-based distributed control system and method in a DC microgrid, and more particularly, ensures the power balance and stability of the system under various conditions even if a communication failure occurs in the MAS-based distributed control. It relates to a MAS-based distributed control system and method in a DC microgrid that can

With the advantages of easy resource integration, flexible installation location and reliable operation today, distributed generation-based DC microgrid (DCMG) has become the future trend of power systems.

DCMG is a complex system consisting of several devices such as renewable energy sources, energy storage systems, grid-connected systems and loads. Because of their different characteristics and mode of operation, effective regulated power flow control strategies are needed to ensure stable and reliable operation of the overall system.

Communication networks play an important role in DCMG's distributed control. However, ubiquitous communication problems such as delays or breakdowns in the process of data transmission can cause system failure and instability. Communication failures prevent the entire DCMG system from functioning properly. In addition, this failure has a serious impact on DC grid voltage stabilization, especially in two situations: grid failure and grid recovery.

As an example of a grid failure, when a grid failure is detected immediately after a failure, the system operation quickly switches to an island mode (isolated mode) to control the DC grid voltage, allowing the Renewable Energy Resource (RES) or Energy Storage System (ESS) to control the system. It plays the role of power balance. Unfortunately, grid faults cannot be detected immediately due to unintentional large delays, including fault detection delays and communication delays. During the delay period, the power source does not help to balance the power and the system can hang.

Taking the example of grid recovery, in grid-connected mode, the grid agent is used as the main power source to balance supply-demand power. In the event of a grid fault (failure), the task of power balancing is taken over by the RES or ESS, depending on the implementation voltage control mode (VCM). When the grid is restored, the grid agent notifies the RES and ESS of the recovery status and consequently the RES or ESS terminates their VCM. Unfortunately, due to communication problems, RES and ESS are not aware of grid recovery. Accordingly, there is a problem in that the control of the DC grid voltage DCV collides with the two voltage control sources.

On the other hand, Korean Patent Application Laid-Open No. 10-2019-0118436 (Patent Document 1) discloses "a method of sharing power between microgrids in a multi-DC microgrid system". The method for sharing power between microgrids according to this is, (a) Each microgrid sets the reference value of the voltage operating point and the droop slope for the voltage source converter (VSC) and the DC/DC converter included in the microgrid according to long-term power scheduling, controlling the voltage through droop control; (b) In each microgrid, when the prediction error of the load and power generation occurs and the power value to be increased through the droop control is greater than the first threshold value for more than a threshold time, the short-term power scheduling is performed. recalculating the reference values of the voltage operating point and the droop slope; and (c) receiving power from another microgrid when the reference values of the voltage operating point and the droop slope that satisfy the constraint conditions are not calculated in the short-term power scheduling process in step (b). Step c) is characterized in that the power is shared from the microgrid in which the power value of the voltage-type converter or the DC/DC converter does not exceed a first threshold value.

In the case of Patent Document 1 as described above, there is an advantage that can be solved through power sharing when system instability occurs due to a change in the amount of renewable power generation or a change in the load in a multi-DC microgrid system, but MAS Since there are no countermeasures for communication failure in distributed control based on MAS, it is difficult to ensure the power balance and stability of the system when communication failure occurs in distributed control based on MAS.

Korean Patent Publication No. 10-2019-0118436 (2019.10.18.)

The present invention was created in consideration of the above, and provides an emergency operation algorithm and DCV (DC-grid voltage) stabilization solution for a DC microgrid, and provides a MAS (Multi-Agent Systems) based distributed control method. By implementing regulated control for DCMG (DC microgrid) using MAS-based distributed control, MAS-based distributed control in DC microgrid can ensure power balance and stability of the system under various conditions even if communication failure occurs in MAS-based distributed control. An object of the present invention is to provide a type control system and method.

In addition, it is another object of the present invention to quickly restore DCV to a nominal value in preparation for the effect of unintended grid fault detection delay even in the event of a communication failure, and to prevent a collision in DCV control when a communication problem occurs. An object of the present invention is to provide a distributed control system and method based on MAS in a grid.

In order to achieve the above object, the MAS-based distributed control system in the DC microgrid according to the present invention,

a three-phase bidirectional grid-connected converter connected to the commercial power grid, receiving commercial AC voltage, converting it into DC voltage, and supplying it to a load or an Energy Storage System (ESS);

a three-phase unidirectional renewable energy source-linked converter connected to a distributed renewable energy source (RES), receiving an AC voltage from a renewable energy source, converting it into a DC voltage, and supplying it to a load or an energy storage system;

Each of the three-phase bidirectional grid-connected converter and the three-phase unidirectional renewable energy source linked converter are connected to each other, and receive the DC voltage supplied by the three-phase bidirectional grid linked converter or the three-phase unidirectional renewable energy source linked converter to the battery. an interleaved bidirectional energy storage system linkage converter for storing and supplying electrical energy stored in the battery to a load;

Controls the status check and operation of the three-phase bi-directional grid-linked converter, the three-phase unidirectional renewable energy source-linked converter, and the interleaved bi-directional energy storage system-linked converter, and restores the grid by activating a grid recovery control algorithm in the event of a grid or communication failure; a control unit that activates a DCV recovery algorithm when DC grid voltage (DCV) is out of a normal range to restore DCV to a nominal value; and

The three-phase bi-directional grid-linked converter, the three-phase unidirectional renewable energy source-linked converter, the interleaved bi-directional energy storage system linkage converter, and a communication network that electrically connects the control unit and the load to each other and enables communication with each other,

The control unit includes a grid agent including the three-phase bi-directional grid-connected converter, a renewable energy source agent including the three-phase unidirectional renewable energy source-linked converter, and the interleaved two-way energy storage system-linked converter. It is characterized in that the operation mode of each agent of the energy storage system agent including the load agent and the load agent including the load is determined in real time in consideration of both the local agent state and the neighboring agent state obtained through the communication network.

Here, by detecting an abnormal change in DC grid voltage (DCV), the control unit immediately activates the local emergency control mode of Energy Storage Systems (ESS) and Renewable Energy Sources (RES) to relate to the information obtained through the communication network. DCV can be restored to its nominal value without

In addition, the distributed renewable energy source may be configured as a wind power generation system (WPGS).

In addition, in order to achieve the above object, the MAS-based distributed control method in the DC microgrid according to the present invention,

Based on MAS-based distributed control system in DC microgrid including 3-phase bi-directional grid-connected converter, 3-phase unidirectional renewable energy source-linked converter, interleaved bi-directional energy storage system-linked converter, and control unit and communication network. A MAS-based distributed control method in a DC microgrid, comprising:

a) acquiring information used for decision making for determining a control mode by the control unit through the communication network;

b) determining whether a grid failure (failure) by the control unit based on the obtained information;

c) determining whether an ESS (Energy Storage Systems) or RES (Renewable Energy Sources) has a failure (failure) by the controller based on the obtained information;

d) If it is not a grid failure (failure) in the determination of step b), adjust DCV and maintain system power balance according to the supply-demand power relationship by the control unit, and control grid recovery when the grid recovers from failure activating an algorithm to determine a control mode; and

e) If it is not a failure (failure) of the ESS or RES in the determination of step c), a DCV (DC-link) recovery algorithm or a grid recovery detection algorithm according to the DCV control capability of the grid and battery agent by the control unit It is characterized in that it includes the step of activating to determine a control mode.

Here, in step a), the information may be obtained from at least one of a grid operator, a local measurement (device), and an adjacent agent through the communication network.

(그리드 복구를 위해 사용된 최소 DCV) 및

(그리드 복구를 위해 사용된 최대 DCV)에 의해 생성된 사전 정의된 범위를 벗어날 경우, SCCM을 DCVM-REC 또는 DCVM-INV로 전환하여 DCV를 조정하도록 구성될 수 있다.In addition, in step d), the grid recovery control algorithm determines that DCV is

(minimum DCV used for grid recovery) and

(Maximum DCV used for grid recovery) can be configured to adjust DCV by switching SCCM to DCVM-REC or DCVM-INV when outside the predefined range generated by it.

In addition, in determining the control mode by activating the DCV recovery algorithm (DC-link recovery algorithm) by the control unit in step e), when the DC grid voltage (DCV) is out of the normal range, the DCV The DCV can be restored to its nominal value by activating a recovery algorithm (DC-Link Recovery Algorithm).

According to the present invention, an emergency operation algorithm and DC-grid voltage (DCV) stabilization solution for DC microgrid are prepared, and DC microgrid (DCMG) using MAS (Multi-Agent Systems) based distributed control method By implementing the regulated control, there is an advantage in that the power balance and stability of the system can be guaranteed under various conditions even if a communication failure occurs in the MAS-based distributed control.

1 is a diagram schematically showing the configuration of an MAS-based distributed control system in a DC microgrid according to an embodiment of the present invention.
2 is a flowchart illustrating an execution process of an MAS-based distributed control method in a DC microgrid according to an embodiment of the present invention.
3 is a flowchart illustrating a control method designed for a local controller of a grid agent in an MAS-based distributed control method.
FIG. 4 is a flowchart illustrating a grid recovery control algorithm in the flowchart of FIG. 3 .
5 is a flowchart illustrating a control method designed for a battery agent in an MAS-based distributed control method.
6 is a flowchart illustrating a DC-link recovery algorithm and a grid recovery detection algorithm in the flowchart of FIG. 5, respectively.
7 is a flowchart illustrating a control method designed for a WPGS agent in an MAS-based distributed control method.
8 is a flowchart illustrating a DC-link recovery algorithm and a grid recovery detection algorithm in the flowchart of FIG. 7, respectively.
9 is a diagram showing simulation results of each DCV recovery algorithm by a battery agent and a WPGS agent.
10 is a diagram showing simulation results of grid recovery detection by a battery agent and a WPGS agent under a communication failure condition.

The terms or words used in the present specification and claims should not be construed as being limited to their ordinary or dictionary meanings, and the inventor may properly define the concept of the term in order to best describe his invention. Based on the principle, it should be interpreted as meaning and concept consistent with the technical idea of the present invention.

Throughout the specification, when a part "includes" a certain element, it means that other elements may be further included, rather than excluding other elements, unless otherwise stated. In addition, terms such as “…unit”, “…group”, “module”, and “device” described in the specification mean a unit that processes at least one function or operation, which is hardware or software or a combination of hardware and software. can be implemented as

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram schematically showing the configuration of an MAS-based distributed control system in a DC microgrid according to an embodiment of the present invention.

1, the MAS-based distributed control system 100 in the DC microgrid according to the present invention includes a three-phase bi-directional grid-linked converter 110, a three-phase unidirectional renewable energy source-linked converter 120, It is configured to include an interleaved bidirectional energy storage system linkage converter 130 , a control unit 140 , and a communication network 150 .

The three-phase bidirectional grid-connected converter 110 is connected to a commercial power grid, receives a commercial AC voltage and converts it into a DC voltage to convert it to a load 160 or an Energy Storage Systems (ESS) (as an ESS in this embodiment). A case in which the battery 170 is employed is exemplified).

The three-phase unidirectional renewable energy source linked converter 120 is connected to distributed Renewable Energy Sources (RES) (in this embodiment, a case in which a wind power generation system is employed as the RES), and from the renewable energy source The AC voltage is received, converted into a DC voltage, and supplied to the load 160 or the energy storage system (ie, the battery 170 ).

The interleaved bidirectional energy storage system linked converter 130 is respectively connected to the three-phase bidirectional grid linked converter 110 and the three-phase unidirectional renewable energy source linked converter 120, and the three-phase bidirectional grid linked converter 110 or It receives the DC voltage supplied by the three-phase one-way renewable energy source-linked converter 120 and stores it in the battery 170 , and supplies the electric energy stored in the battery 170 to the load 160 .

The control unit 140 controls the status check and operation of the three-phase bi-directional grid-linked converter 110, the three-phase unidirectional renewable energy source linkage converter 120, and the interleaved bidirectional energy storage system linkage converter 130, and controls the grid or communication In case of failure, the grid is restored by activating the grid recovery control algorithm, and when DCV (DC grid voltage) is out of the normal range, the DCV recovery algorithm is activated to restore DCV to the nominal value.

The communication network 150 is between the three-phase bi-directional grid-connected converter 110, the three-phase unidirectional renewable energy source linked converter 120, the interleaved bi-directional energy storage system linked converter 130, the control unit 140 and the load 160. are electrically connected to each other and enable mutual communication.

In the MAS-based distributed control system 100 in the DC microgrid according to the present invention having the above configuration, the control unit 140 is a grid agent ( agent), and a renewable energy source agent (ie, a wind power generation system (WPGS) agent) including the three-phase unidirectional renewable energy source linked converter 120, and the interleaved bidirectional energy storage system linked converter 130. The operation mode of each agent of the energy storage system agent (ie, battery agent) and the load agent including the load 160 is real-time considering both the local agent state and the neighboring agent state obtained through the communication network 150 can be configured to determine

Here, also, the control unit 140 detects an abnormal change in DC grid voltage (DCV), and immediately activates the local emergency control mode of Energy Storage Systems (ESS) and Renewable Energy Sources (RES) to the communication network 150 . It can be configured to restore DCV to its nominal value regardless of the information obtained through

In addition, the distributed renewable energy source may be configured as a wind power generation system (WPGS).

2 is a flowchart illustrating an execution process of an MAS-based distributed control method in a DC microgrid according to an embodiment of the present invention.

Referring to FIG. 2 , the MAS-based distributed control method in the DC microgrid according to the present invention includes the three-phase bi-directional grid-connected converter 110 and the three-phase unidirectional renewable energy source-linked converter 120 as described above. And, in the DC microgrid based on the MAS-based distributed control system 100 in the DC microgrid including the interleaved bidirectional energy storage system linked converter 130 and the control unit 140 and the communication network 150 As an MAS-based distributed control method of , first, information used for decision making for determining a control mode by the controller 140 is acquired through the communication network 150 (step S201). Here, the information may be obtained from at least one of a grid operator, a local measurement (device), and an adjacent agent through the communication network 150 .

When the information is acquired in this way, it is determined whether or not the grid is faulty (failure) by the controller 140 based on the acquired information (step S202).

In addition, based on the obtained information, it is determined whether a failure (failure) of an ESS (Energy Storage Systems) (here, a battery agent) or a RES (Renewable Energy Sources) (here, a WPGS agent) is performed by the controller 140 (step S203).

(그리드 복구를 위해 사용된 최소 DCV) 및

(그리드 복구를 위해 사용된 최대 DCV)에 의해 생성된 사전 정의된 범위를 벗어날 경우, SCCM(special current control mode)을 DCVM-REC(그리드 에이전트의 컨버터 모드에 의한 DC-링크 전압 제어 모드) 또는 DCVM-INV(그리드 에이전트의 인버터 모드에 의한 DC-링크 전압 제어 모드)로 전환하여 DCV를 조정하도록 구성될 수 있다.If it is not a grid failure (failure) in the determination of step S202, DCV is adjusted and system power balance is maintained (guaranteed) according to the supply-demand power relationship by the control unit 140, and the grid is restored from failure The grid recovery control algorithm is activated to determine the control mode (step S204). Here, the grid recovery control algorithm is DCV

(minimum DCV used for grid recovery) and

When outside the predefined range generated by (maximum DCV used for grid recovery), SCCM (special current control mode) is set to DCVM-REC (DC-Link voltage control mode by converter mode of grid agent) or DCVM Can be configured to adjust DCV by switching to INV (DC-Link Voltage Control Mode by Inverter Mode of Grid Agent).

In addition, if it is not a failure (failure) of the ESS or RES in the determination of step S203, a DCV (DC-link) recovery algorithm or grid recovery detection is performed by the controller 140 according to the DCV control capability of the grid and the battery agent. The algorithm is activated to determine the control mode (step S205). Here, in determining the control mode by activating the DCV recovery algorithm (DC-link recovery algorithm) by the controller 140 , when the DC grid voltage (DCV) is out of the normal range, DCV can be restored to its nominal value by activating the DCV recovery algorithm (DC-Link Recovery Algorithm).

Hereinafter, in relation to the MAS-based distributed control method in the DC microgrid according to the present invention as described above, an amplification will be described.

3 is a flowchart illustrating a control method designed for a local controller of a grid agent in an MAS-based distributed control method.

As shown in Fig. 3, the control algorithm is divided into three layers, namely, information collection, control mode determination, and communication determination. In the information gathering layer, information can be obtained from three sources: grid operator (GO), local measurement, and neighbor agent. GO provides information about the maximum exchange power and state of the grid. Local measurement and neighbor agents provide information about voltage, current, and supply-demand power relationships. After receiving the necessary information, a control strategy is implemented to release the appropriate control mode. When the grid is faulty or maintained, the grid agent goes into IDLE mode, and at the same time G ^ctrl is set to zero. This means that the grid agent (see Fig. 1) cannot perform DCV control.

는 그리드로부터 DCMG로 공급된 요구 전력,

는 그리드로부터 DCMG로 공급된 최대 전력,

는 DCMG로부터 그리드로 주입된 요구 전력,

는 DCMG로부터 그리드로 주입된 최대 전력, G^ctrl은 그리드 에이전트에 의한 DCV 제어 능력을 나타내는 것으로, G^ctrl=1은 그리드 에이전트가 DCV를 제어할 수 있는 것을 의미하고, G^ctrl=0은 그리드 에이전트가 DCV를 제어할 수 없는 것을 각각 의미한다.The grid-free state is further divided into two sub-states. The grid is normal and the grid is restored. If the grid is normal, depending on the supply-demand power relationship, the grid agent will either "DCVM-REC" (DC-Link Voltage Control Mode by Grid Agent's Converter Mode) or "DCVM-INV" (DC by Grid Agent's Inverter Mode) - Link voltage control mode) to regulate DCV and ensure system power balance. In particular, DCVM-REC is used to compensate for the power shortage by injecting more power from the grid to the DCMG. On the other hand, DCVM-INV is implemented to absorb the power surplus of the DC-link. When the required power used for system power balancing becomes greater than the maximum level obtained in the GO, the grid agent switches operation to constant power control mode (CPCM). When CPCM is implemented, the power exchanged between the grid and DCMG is at the highest possible level and as a result, the grid agent cannot perform DCV control. When the grid recovers from a failure (failure), the grid recovery control algorithm is triggered. The function of this algorithm and the flags F _G1 , F _G2 and F _G3 will be described later. After the determination of the local control mode is released, the DCV control capability of the grid agent is notified to other agents through the communication line. In Fig. 3 P _L is the total load power, P _W is the output power of the WPGS agent,

is the required power supplied from the grid to the DCMG,

is the maximum power supplied from the grid to the DCMG,

is the required power injected from the DCMG into the grid,

is the maximum power injected from DCMG to the grid, G ^ctrl is the DCV control capability by the grid agent, G ^ctrl =1 means that the grid agent can control DCV, G ^ctrl =0 means that the grid agent can control DCV Each means that DCV cannot be controlled.

FIG. 4 is a flowchart illustrating a grid recovery control algorithm in the flowchart of FIG. 3 .

)에 유사한 파형이 나타난다. 배터리 및 WPGS 에이전트는 그리드 복구를 인식하기 위해 I_B 또는

의 분석에 의존할 수 있다. 그리드 복구를 인식한 후, 배터리 또는 WPGS 에이전트는 DCVM을 중지한다. 그 결과, 어떤 소스도 DCV를 제어하지 않기 때문에 DCV가 달라진다. DCV가

(그리드 복구 사례를 위해 사용된 최소 DCV) 및

(그리드 복구 사례를 위해 사용된 최대 DCV)에 의해 생성된 사전 정의된 범위를 벗어나자마자, 그리드 에이전트는 도 4에 도시된 바와 같이, SCCM을 DCVM-REC 또는 DCVM-INV로 전환하여 DCV를 조정한다. 플래그 F_G2 또는 F_G3은 해당 작동 모드를 나타내도록 설정된다. 알고리즘의 다음 시퀀스에서 F_G2 또는 F_G3가 1이므로 그리드는 해당 DCVM에서 계속 작동한다.Referring to Fig. 4, it shows the control strategy algorithm of the grid agent for the grid recovery case in the communication network problem. As soon as this algorithm is activated, the grid agent operates in a special current control mode (SCCM) to inject a special current pattern into the DC link. In this embodiment, the current pattern is a square wave with a high frequency _{of f G .} In DCV control mode (DCVM) of the battery or WPGS agent, any power injected into the DC-link is absorbed to ensure system power balance. Consequently, the battery current (I _B ) or the q-axis current of WPGS (

), a similar waveform appears. Batteries and WPGS agents to recognize grid recovery I _B or

can rely on the analysis of After recognizing grid recovery, the battery or WPGS agent stops the DCVM. As a result, the DCV varies because no source controls the DCV. DCV

(minimum DCV used for grid recovery case) and

As soon as it leaves the predefined range generated by (maximum DCV used for grid recovery case), the grid agent adjusts the DCV by switching SCCM to DCVM-REC or DCVM-INV, as shown in Fig. . A flag F _G2 or F _G3 is set to indicate the corresponding operating mode. In the next sequence of the algorithm, F _G2 or F _G3 is 1, so the grid continues to work on that DCVM.

5 is a flowchart illustrating a control method designed for a battery agent in an MAS-based distributed control method.

Referring to FIG. 5 , similarly to FIG. 3 described above, the control algorithm of the battery agent is also divided into three layers. At the information gathering layer, information can be obtained from local measurements and neighboring agents. By using local measurements, information about battery failure, state of charge (SOC), voltage and current can be obtained locally. Information from adjacent agents, including supply-demand power relationships and the grid agent's DCV control capabilities, can be obtained via communication lines. DCV recovery and current identification algorithms will be described later. Also, _{the functions of the flags F B1} and F _B2 will be described later.

의 필수 충전 전력 및

의 필수 방전 전력이

및

의 최대 레벨을 초과하는 경우, CPCM(constant power control mode)은 최대 용량으로 배터리를 충전/방전하도록 실현되었다. 결과적으로, 배터리 작동 중 과열 또는 손상이 제거될 수 있다. 마찬가지로, 정전압 제어 모드(CVCM)는 배터리를 최대 전압 레벨로 충전하도록 실현되었으며, 그것은 배터리가 과충전되는 것을 방지한다. 로컬 제어 모드의 결정이 해제된 후, 배터리 에이전트의 DCV에 대한 능력(B^ctrl)은 통신 라인을 통해 다른 에이전트에 동시에 알려진다.If the battery is faulty, the battery agent goes into IDLE mode and at the same time B ^ctrl (indicating the ability to control DCV by the battery agent) is set to 0, which means that the battery agent cannot perform DCV control. indicates When the battery is normal, according to the supply-demand power relationship, the battery SOC, voltage and current, an appropriate control mode is determined to ensure the system power balance of DCMG in various conditions. If the battery parameters of SOC, voltage and required power are within the predefined ranges, the battery agent will cause DCVM-DIS (DC-Link Voltage Control Mode by Battery Discharge) to inject power into the DC-Link or DCVM-CHA ( DC-link voltage control mode by battery charging) realizes to absorb power from DC-link, as a result, DCV is adjusted to the nominal value. Battery SOC is SOC ^min and SOC ^max , the battery agent switches to IDLE mode to avoid overcharging or overdischarging. Also,

of the required charging power and

the required discharge power of

and

When the maximum level of CPCM (constant power control mode) is exceeded, it is realized to charge/discharge the battery to its maximum capacity. As a result, overheating or damage during battery operation can be eliminated. Likewise, the constant voltage control mode (CVCM) is realized to charge the battery to the maximum voltage level, which prevents the battery from being overcharged. After the determination of the local control mode is released, the capability of the battery agent to DCV (B ^ctrl ) is simultaneously announced to other agents via the communication line.

6 is a flowchart illustrating a DC-link recovery algorithm and a grid recovery detection algorithm in the flowchart of FIG. 5, respectively.

)보다 큰 경우, 배터리 에이전트는 유휴(IDLE) 모드이다. DCV가

보다 낮아지자마자, CPCM은 최대 방전 전력으로 배터리를 방전하도록 활성화된다. 동시에, 플래그 F_B1은 1로 설정된다. 알 수 있는 바와 같이, DCV가 미리 정의된 제 2 최소 레벨(

)보다 낮은 한, 배터리의 동작 모드는 CPCM으로 유지된다. 이 구현의 목표는 가능한 한 빨리 DCV를 복구하는 것이다. DCV가

보다 높아지면, 배터리가 자동으로 DCVM-DIS로 전환되어 DCV를 공칭 레벨로 조절하고 플래그 F_B2가 1로 설정된다. 다음 시퀀스에서 F_B1과 F_B2가 1이기 때문에 배터리는 DCVM-DIS에서 계속 작동한다. 도 6의 (A)의 알고리즘을 사용하여 통신 문제가 있는 배터리 에이전트의 작동 모드를 결정한 후, 통신 결정은 출력(1 또는 2)에 따라 도 5에 도시된 바와 같이 선택된다.Referring to Fig. 6(A), this shows the details of the DCV recovery algorithm for the battery agent, where DCV is still at a predefined first minimum level (

), the battery agent is in IDLE mode. DCV

As soon as it goes lower, the CPCM is activated to discharge the battery to its maximum discharge power. At the same time, the flag F _B1 is set to 1. As can be seen, DCV is at a predefined second minimum level (

), the battery's operating mode remains CPCM. The goal of this implementation is to restore DCV as quickly as possible. DCV

When higher, the battery automatically switches to DCVM-DIS to adjust DCV to the nominal level and flag F _B2 is set to 1. In the following sequence, since F _B1 and F _B2 are 1, the battery continues to operate in DCVM-DIS. After determining the operating mode of the battery agent having communication problem using the algorithm of Fig. 6(A), the communication decision is selected as shown in Fig. 5 according to the output (1 or 2).

)는 제로 검출에 의존하여 계산될 수 있다. 검출된 주파수

가 그리드 에이전트에 의해 생성된 주파수

와 동일하면, 카운터는 단계 1로 카운트 업 한다. 카운터가 사전 정의된 N의 수에 도달하면, 배터리 또는 WPGS 에이전트는 그리드 복구를 검출한다. 다음 동작을 위해 카운터가 재설정된다. 출력(1 또는 2)을 결정한 후, 배터리 에이전트의 제어 알고리즘은 도 5와 같이 지속적으로 실현된다.Referring to FIG. 6B , it shows a current identification algorithm used to detect a special current pattern generated by a grid agent. A high-pass filter removes low-frequency components. The frequency of that output (

) can be calculated depending on the zero detection. detected frequency

frequency generated by the grid agent

equal to, the counter counts up to step 1. When the counter reaches a predefined number of N, the battery or WPGS agent detects grid recovery. The counter is reset for the next operation. After determining the output 1 or 2, the control algorithm of the battery agent is continuously realized as shown in FIG.

7 is a flowchart illustrating a control method designed for a WPGS agent in an MAS-based distributed control method.

Referring to FIG. 7 , like the grid agent and battery agent described above, the control algorithm of the WPGS agent is also divided into three layers. At the information gathering layer, information can be obtained from local measurements and neighboring agents. Local measurements can be used to obtain the power extracted from _{the wind turbine (P W ).} Information from the adjacent agent provides the supply-demand power relationship of the battery and grid agent and DCV control capabilities. DCV (DC-Link) recovery algorithm and current identification algorithm (grid recovery detection algorithm) will be described later.

If the WPGS is faulty, the WPGS agent switches to IDLE mode and at the same time W ^ctrl (indicating DCV control capability by the WPGS agent) is set to 0, indicating that the WPGS agent cannot perform DCV control. According to the supply-demand power relationship, an appropriate control mode is determined to balance the system power in various conditions. In particular, the maximum power point tracking (MPPT) mode is used to extract maximum power from the wind turbine to the DC-link. On the other hand, DCV control by limiting the output power of the wind turbine (DCVM-LIM) is implemented to adjust the output power of the wind turbine to the load demand. After the determination of the local control mode is released, the DCV control capability (W ^ctrl ) of the WPGS agent is simultaneously announced to other agents through the communication line.

8 is a flowchart illustrating a DC-link recovery algorithm and a grid recovery detection algorithm in the flowchart of FIG. 7, respectively.

)보다 낮은 경우 WPGS 에이전트는 MPPT 모드에서 작동한다. DCV가

보다 높아지면 DCVM-LIM이 트리거되어 WPGS의 출력 전력을 총 부하 요구량과 같은 수준으로 제한한다. 따라서, 공급-수요 전력 관계가 보장되므로 DCV는 공칭 값으로 안정적으로 유지된다. 동시에, 플래그 F_W는 1로 설정된다. 알고리즘의 다음 시퀀스에서, F_W는 1과 같기 때문에 WPGS는 DCVM-LIM에서 계속 작동한다. 통신 문제가 나타나는 WPGS 에이전트의 작동 모드를 결정한 후, 통신 결정은 도 7과 같이 출력(1 또는 2)에 따라 해제된다.Referring to (A) of Figure 8, this shows the details of the DCV (DC-Link) recovery algorithm for the WPGS agent, where the DCV is set to a predefined maximum level (

), the WPGS agent operates in MPPT mode. DCV

When it goes higher, the DCVM-LIM is triggered, limiting the output power of the WPGS to a level equal to the total load demand. Thus, the supply-demand power relationship is ensured so that the DCV remains stable at its nominal value. At the same time, the flag F _W is set to 1. In the next sequence of the algorithm, WPGS continues to work in DCVM-LIM because _{F W is equal to 1.} After determining the operating mode of the WPGS agent in which the communication problem appears, the communication decision is released according to the output (1 or 2) as shown in FIG.

)는 제로 검출에 의존하여 계산될 수 있다. 검출된 주파수

가 그리드 에이전트에 의해 생성된 주파수

와 동일하면, 카운터는 단계 1로 카운트 업 한다. 카운터가 사전 정의된 N의 수에 도달하면, 배터리 또는 WPGS 에이전트는 그리드 복구를 검출한다. 다음 동작을 위해 카운터가 재설정된다. 출력(1 또는 2)을 결정한 후, 배터리 에이전트의 제어 알고리즘은 도 5와 같이 지속적으로 실현된다.Referring to Fig. 8B, it shows the current identification algorithm used to detect the special current pattern generated by the grid agent. A high-pass filter removes low-frequency components. The frequency of that output (

) can be calculated depending on the zero detection. detected frequency

frequency generated by the grid agent

equal to, the counter counts up to step 1. When the counter reaches a predefined number of N, the battery or WPGS agent detects grid recovery. The counter is reset for the next operation. After determining the output 1 or 2, the control algorithm of the battery agent is continuously realized as shown in FIG.

9 is a diagram showing simulation results of each DCV recovery algorithm by a battery agent and a WPGS agent.

으로 떨어지자 마자, 배터리 에이전트는 CPCM으로 전환하여 2kW의 최대 방전 전력으로 배터리를 방전한다. DCV가 390V의

에 도달할 때, 배터리의 제어 모드는 DCVM-DIS로 전환되어 공칭 값 400V에서 DCV를 점진적으로 조정한다. 그리드 오류가 t = 0.5초에서 감지된 경우, 그리드 에이전트의 작동이 유휴(IDLE)로 전환되고 배터리 에이전트는 DCVM-DIS에 의해 DCV를 지속적으로 제어한다. 도 9의 (A)에서 (a)는 a상 그리드 전압, (b)는 에이전트들의 출력 전력, (c)는 DCV를 각각 나타낸다.Referring to (A) of FIG. 9 , this shows the simulation results of the DCV recovery algorithm implemented by the battery agent without SHED in the presence of grid fault detection and communication delay. Before the grid fault occurs, the grid agent performs DCV It takes over control and system power balance. Meanwhile, the battery agent is in idle mode and the WPGS agent operates on the MPPT to extract maximum power from the wind turbine. At t = 0.3 s, the grid shuts down, but the GO and all system agents cannot immediately recognize it due to grid fault detection and communication delays. As a result, the grid agent still works in DCVM-REC while the battery is still in idle mode. As can be seen in (A) of Figure 9, DCV (refer to (c)) rapidly decreases due to power imbalance. DCV is 370V

As soon as it drops to , the battery agent switches to CPCM to discharge the battery with a maximum discharge power of 2 kW. DCV of 390V

When , the control mode of the battery switches to DCVM-DIS, which gradually adjusts the DCV from the nominal value of 400V. When a grid error is detected at t = 0.5 seconds, the operation of the grid agent switches to IDLE and the battery agent continuously controls the DCV by DCVM-DIS. In (A) of FIG. 9, (a) is a-phase grid voltage, (b) is the output power of the agents, (c) is DCV, respectively.

에 도달하자마자, WPGS 에이전트는 DCVM-LIM으로 전환하여 WPGS의 출력 전력을 제한한다. 그리드 오류가 t = 0.5초에서 감지되면, 그리드 에이전트의 작동이 유휴(IDLE)로 전환되고 WPGS 에이전트는 DCVM-LIM에 의해 DCV를 지속적으로 제어한다. 도 9의 (B)에서 (a)는 a상 그리드 전압, (b)는 에이전트들의 출력 전력, (c)는 DCV를 각각 나타낸다.Referring to FIG. 9B , this shows the simulation result of the DCV recovery algorithm implemented by the WPGS agent in the presence of grid fault detection and communication delay, in which the battery is idle (IDLE) before the grid fault occurs. The WPGS agent operates in MPPT mode while in operation. DCV is regulated to 400V by DCVM-REC of the grid agent. A grid fault occurs at t = 0.3 s, but the GO and all agents are unaware of it due to grid fault detection and communication delays. This results in a rapid increase in DCV due to supply-demand power imbalance. DCV

Upon reaching , the WPGS agent switches to DCVM-LIM to limit the output power of the WPGS. When a grid error is detected at t = 0.5 seconds, the operation of the grid agent switches to IDLE and the WPGS agent continuously controls the DCV by DCVM-LIM. In FIG. 9(B), (a) is a-phase grid voltage, (b) is the output power of the agents, (c) is DCV, respectively.

10 is a diagram showing simulation results of grid recovery detection by a battery agent and a WPGS agent under a communication failure condition.

에 도달하자마자, 그리드 에이전트는 SCCM을 DCVM-INV로 전환하여 DCV를 조정한다. 도 10의 (A)에서 (a)는 DCV, (b)는 그리드 에이전트의 출력 전력, (c)는 부하 전력, (d)는 WPGS의 출력 전력, (e)는 배터리 전류를 각각 나타낸다.Referring to FIG. 10A , this shows the simulation result of the grid recovery detection method using the battery agent, and DCV is adjusted by implementing DCVM-CHA of the battery agent before the grid is restored. At t = 0.2 s, the grid recovers, but the battery agent is not aware of it due to a communication failure. At the same time, the grid agent activates the special current control mode (SCCM) to inject a special current pattern into the DC-link. The current pattern is a square wave with a frequency of 20 Hz. Due to the influence of the special current pattern generated by the grid agent, a similar waveform appeared in the _{battery current I B , as can be seen in (e) of Fig. 10(A).} Then, the battery current waveform is analyzed by the current identification algorithm described in FIGS. 6 (B) and 8 (B). Variations in wind power and load power are considered disturbances during the identification process. After the counter reaches a predefined N level, grid recovery is detected, and the battery agent stops DCVM-CHA. DCV is 410V

Upon reaching , the grid agent adjusts the DCV by switching the SCCM to DCVM-INV. In (A) of FIG. 10, (a) is DCV, (b) is the output power of the grid agent, (c) is the load power, (d) is the output power of WPGS, (e) is the battery current, respectively.

에 도달하자마자, 그리드 에이전트는 SCCM을 DCVM-INV로 전환하여 DCV를 제어한다. 본 발명의 방법을 사용함으로써, 풍력 및 부하 전력의 변화로 인한 통신 장애 하에서도 그리드 복구가 감지될 수 있음을 확인하였다. 또한, DCV는 전압 제어의 충돌없이 잘 조정된다. 도 10의 (B)에서 (a)는 DCV, (b)는 그리드 에이전트의 출력 전력, (c)는 부하 전력, (d)는 WPGS 전류를 각각 나타낸다.Referring to (B) of FIG. 10 , this shows the simulation result of grid recovery detection by the WPGS agent under the communication failure condition. Before the grid is restored, the DCV is adjusted by implementing DCVM-LIM of the WPGS agent. At t = 0.2 seconds the grid is restored, but the WPGS agent is not aware of it and it still works in DCVM-LIM. When the counter reaches N level, grid recovery is detected and the WPGS agent converts DCVM-LIM to MPPT. DCV is 410V

Upon reaching , the grid agent switches SCCM to DCVM-INV to control DCV. By using the method of the present invention, it was confirmed that grid restoration can be detected even under communication disturbances due to changes in wind power and load power. Also, the DCV is well regulated without conflicting voltage control. In (B) of FIG. 10, (a) is DCV, (b) is the output power of the grid agent, (c) is the load power, (d) is the WPGS current, respectively.

As described above, the MAS-based distributed control system and method in the DC microgrid according to the present invention provides an emergency operation algorithm and DCV (DC-grid voltage) stabilization solution for the DC microgrid, and MAS (Multi- Agent Systems)-based distributed control method is used to implement regulated control for DCMG (DC microgrid), so even if communication failure occurs in MAS-based distributed control, the power balance and stability of the system can be guaranteed under various conditions. there is this

In addition, even in the event of a communication failure, DCV can be quickly restored to a nominal value in preparation for the influence of unintended grid fault detection delay, and there is an effect of preventing a collision in DCV control when a communication problem occurs.

In addition, when the control system and method of the present invention are applied, there is an effect that can effectively simultaneously solve concerns about the control and management of DCMG, including supply-demand power balance, system stability, grid failure, grid recovery, and communication failure. .

As described above, the present invention has been described in detail through preferred embodiments, but the present invention is not limited thereto, and it is common in the art that various changes and applications can be made without departing from the technical spirit of the present invention. self-explanatory to the technician. Therefore, the true protection scope of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

100: (Invention) MAS-based distributed control system in DC microgrid
110: 3-phase bidirectional grid-connected converter
120: 3-phase unidirectional renewable energy source coupled converter
130: interleaved bidirectional energy storage system linkage converter
140: control unit 150: communication network
160: load 170: battery

Claims (5)

a three-phase bidirectional grid-connected converter connected to the commercial power grid, receiving commercial AC voltage, converting it into DC voltage, and supplying it to a load or an Energy Storage System (ESS);
a three-phase unidirectional renewable energy source-linked converter connected to a distributed renewable energy source (RES), receiving an AC voltage from a renewable energy source, converting it into a DC voltage, and supplying it to a load or an energy storage system;
Each of the three-phase bidirectional grid-connected converter and the three-phase unidirectional renewable energy source linked converter are connected to each other, and receive the DC voltage supplied by the three-phase bidirectional grid linked converter or the three-phase unidirectional renewable energy source linked converter to the battery. an interleaved bidirectional energy storage system linkage converter for storing and supplying electrical energy stored in the battery to a load;
Controls the status check and operation of the three-phase bi-directional grid-linked converter, the three-phase unidirectional renewable energy source-linked converter, and the interleaved bi-directional energy storage system-linked converter, and restores the grid by activating a grid recovery control algorithm in the event of a grid or communication failure; a control unit that activates a DCV recovery algorithm when DC grid voltage (DCV) is out of a normal range to restore DCV to a nominal value; and
The three-phase bi-directional grid-linked converter, the three-phase unidirectional renewable energy source-linked converter, the interleaved bi-directional energy storage system linkage converter, and a communication network that electrically connects the control unit and the load to each other and enables communication with each other,
The control unit includes a grid agent including the three-phase bi-directional grid-connected converter, a renewable energy source agent including the three-phase unidirectional renewable energy source-linked converter, and the interleaved two-way energy storage system-linked converter. The operation mode of each agent of the energy storage system agent including the load agent and the load agent including the load is determined in real time in consideration of both the local agent state and the neighboring agent state obtained through the communication network,
By detecting an abnormal change in DC grid voltage (DCV), the control unit immediately activates the local emergency control mode of Energy Storage Systems (ESS) and Renewable Energy Sources (RES), regardless of the information obtained through the communication network. restores to its nominal value,
The DCV recovery algorithm includes a DCV recovery algorithm for the battery agent, wherein the DCV recovery algorithm for the battery agent includes a first minimum level (

), the battery agent remains in IDLE mode, and DCV is

As soon as it goes lower, the CPCM is activated to discharge the battery to the maximum discharge power, and the DCV is set to a second predefined minimum level (

), the battery's operating mode remains CPCM, and DCV is

MAS-based decentralized control system in DC microgrid, characterized in that when higher, the battery automatically switches to DCVM-DIS to regulate DCV to the nominal level.
delete Based on MAS-based distributed control system in DC microgrid including 3-phase bi-directional grid-connected converter, 3-phase unidirectional renewable energy source-linked converter, interleaved bi-directional energy storage system-linked converter, and control unit and communication network. A MAS-based distributed control method in a DC microgrid, comprising:
a) acquiring information used for decision making for determining a control mode by the control unit through the communication network;
b) determining whether a grid failure (failure) by the control unit based on the obtained information;
c) determining whether an ESS (Energy Storage Systems) or RES (Renewable Energy Sources) has a failure (failure) by the controller based on the obtained information;
d) If it is not a grid failure (failure) in the determination of step b), adjust DCV and maintain system power balance according to the supply-demand power relationship by the control unit, and control grid recovery when the grid recovers from failure activating an algorithm to determine a control mode; and
e) If it is not a failure (failure) of the ESS or RES in the determination of step c), a DCV (DC-link) recovery algorithm or a grid recovery detection algorithm according to the DCV control capability of the grid and battery agent by the control unit activating to determine a control mode;
In step e), in determining a control mode by activating a DCV recovery algorithm (DC-link recovery algorithm) by the control unit, when DC grid voltage (DCV) is out of a normal range, the DCV recovery algorithm is controlled by the control unit. (DC-Link Recovery Algorithm) to restore DCV to its nominal value,
The DCV recovery algorithm includes a DCV recovery algorithm for the battery agent, wherein the DCV recovery algorithm for the battery agent includes a first minimum level (

), the battery agent remains in IDLE mode, and DCV is

As soon as it goes lower, the CPCM is activated to discharge the battery to the maximum discharge power, and the DCV is set to a second predefined minimum level (

), the battery's operating mode remains CPCM, and DCV is

If higher, the battery is automatically switched to DCVM-DIS to regulate DCV to the nominal level, MAS-based decentralized control method in DC microgrids.
4. The method of claim 3,
In step d), the grid recovery control algorithm determines that DCV is

(minimum DCV used for grid recovery) and

Distributed control method based on MAS in DC microgrid to adjust DCV by switching SCCM to DCVM-REC or DCVM-INV when outside the predefined range generated by (maximum DCV used for grid recovery) .
delete

Images