MMGS distributed critical bus voltage control method based on inclusion control

Abstract

The invention provides an MMGS distributed critical bus voltage control method based on inclusion control, and relates to the technical field of microgrids. The invention provides a distributed controller based on a container to balance two conflicting targets of voltage regulation and tide. The controller can control the critical bus voltage within a reasonable range under the condition that the voltage difference exists, so that power flow is realized; converting a critical bus voltage regulation problem of the alternating current automatic control system into a distributed output feedback tracking problem of a linear automatic control system with nonlinear dynamic characteristics; different from the current event-based secondary control strategy, an appropriate timer is designed to ensure the existence of the positive MIET of each DG unit. Further, the auxiliary controller depends only on information related to the observer, thereby preventing the use of global information related to the communication topology.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrids, and in particular to a distributed critical bus voltage control method for a multi-microgrid system (MMGS) based on containment control. Background Art

[0002] A microgrid is an efficient and autonomous small-scale power generation and distribution system, aiming to solve the problem of grid connection operation of a wide variety of, numerous, and relatively dispersed distributed generation units, thereby improving the grid's utilization ability of renewable energy. To cope with its highly complex dynamic characteristics, distributed control based on multi-agent consensus algorithms has been widely applied to the control technology of microgrids, and this scheme uses the communication between distributed power sources to achieve global information sharing. To avoid waste of communication and computing resources and achieve the convergence of the consensus algorithm at a low communication frequency, in recent years, many scholars have begun to study the event-triggered communication mechanism, that is, each distributed power source only transmits local state variables to adjacent nodes when the predefined event-triggered conditions are met.

[0003] For a power system, it is a more important requirement that the triggering mechanism has a strictly positive minimum event interval time than just excluding the Zeno phenomenon. A uniformly strictly positive minimum event interval time is sufficient to exclude Zeno behavior and is more practical than traditional event triggering because hardware constraints always impose a positive minimum time constraint between two consecutive event times, avoiding Zeno behavior while ensuring the existence of a positive MIET. However, in the secondary control problem of AC microgrids, almost no researchers have considered the above problems.

[0004] With the development of computer technology and artificial intelligence, distributed control based on multi-agent consensus algorithms provides a feasible solution for the reliable cooperative autonomy of MGs, and this scheme uses the communication between distributed power sources to achieve global information sharing. Compared with traditional centralized and decentralized control, distributed control eliminates the single-point failure of the central controller. The communication and computing burdens of the central controller are allocated to the local controllers of each distributed power source, taking into account the advantages of the high reliability of decentralized control and the global cooperative management of centralized control. However, the continuous communication mechanism in this scheme leads to waste of communication resources and computing resources, especially when the communication bandwidth is limited, and the convergence of the consensus algorithm needs to be achieved at a low communication frequency. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a distributed critical bus voltage control method for an MMGS based on containment control in view of the above-mentioned deficiencies of the prior art. The proposed protocol is fully distributed and scalable and does not rely on the global information of the network diagram.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0007] A distributed critical bus voltage control method based on inclusion control for MMGS, comprising the following steps:

[0008] Step 1: Establish a microgrid non-linear dynamic model, transform the model into a linear multi-agent system model through feedback linearization, and simplify it to obtain a simplified control model;

[0009] Step 2: Design a distributed secondary voltage regulation strategy based on inclusion control, which allows the critical bus voltage to be restricted within a defined range, while allowing a voltage difference to exist between sub-microgrids, thereby allowing power to flow between the respective sub-microgrids;

[0010] Step 3: Propose an event-triggering mechanism based on multiple measurement errors. Under this event-triggering mechanism, each distributed generator DG has a uniform and strictly positive minimum event interval time, and can avoid continuous updates of the secondary controller; and does not require the use of any global information in the communication graph.

[0011] Furthermore, the specific method of Step 1 is as follows:

[0012] Step 1.1: Establish a microgrid non-linear dynamic model:

[0013]

[0014] Wherein, is the state vector, represents the first derivative of the state vector x i δ i represents the angle of the i-th DG reference frame relative to the common reference frame, P i and Q i represent the active power and reactive power at the i-th inverter terminal, is the auxiliary state variable in the voltage controller, is the auxiliary state variable in the current controller, i ldi 、i lqi are filter current elements, v odi 、v oqi respectively represent the direct component and quadrature component of the output voltage v oi i < odi 、i oqi respectively represent the direct component and quadrature component of the output current i oi ; D i =[ω com ,v bdi ,v bqi ; ω iDenote the rotational speed of the reference frame of the $i$-th DG. Take the reference frame of one of the DGs as the common reference frame, $\omega$ com Denote the rotational frequency of the common reference frame; $v$ bdi 、$v$ bqi Denote the $d$-axis and $q$-axis components of the critical bus voltage; $u$ i Denote the control input; $f$ i $(x$ i )、$g$ i $(x$ i )、$k$ i $(x$ i )、$h$ i $(x$ i ) are functions related to the system characteristics, used to describe the nonlinear dynamic behavior of the distributed generator; Let it be $v$ bi , denote the critical bus voltage;

[0015] Step 1.2: Perform feedback linearization transformation on the nonlinear dynamics model of the AC microgrid:

[0016]

[0017] Among them, is 's second derivative; $F$ i $(x$ i ) = $f$ i $(x$ i ) + $k$ i $(x$ i )$D$ i , Denote the Lie derivative of $h$ i , defined as: Defined as:

[0018] $V$ ni Is equivalent to the control input $u$ i ; $v$ li Denote the line voltage between the critical bus and the inverter output, Is the second derivative of $v$ li ; This step transforms the nonlinear model into a form convenient for control design;

[0019] Step 1.3: Simplify the control model:

[0020]

[0021] Among them, $r$ i Denote the $d$-axis component $v$ bdi of the critical bus voltage and its derivative The two-dimensional vector composed of is used to build the control algorithm model. The A, B, and C matrices determine the dynamic characteristics and input-output relationship of the system. B=[0,1] T , C = [1,0]; For r i The first-order derivative of ; N represents the number of sub-microgrids.

[0022] Furthermore, the specific method of step 2 is:

[0023] Step 2.1: Design the state observer:

[0024]

[0025] Among them, the estimated parameters The first-order derivative of is γ θi 、k θi ,π1,R are design parameters; is r i (t) is estimated, yes The first derivative of ; is the observer output; is the error parameter between the observer output and the critical bus voltage; is the observer gain, represents the field of real numbers; are estimated parameters; Represents a mathematical mapping that transforms state estimation information Converted into quantities that can be used for control algorithms and parameter estimation operations;

[0026] Step 2.2: Propose a distributed inclusion protocol:

[0027]

[0028] Among them, c i is a positive control gain; ρ i is the adaptive dynamic parameter, is ρ i The first derivative of ; is the kth triggering moment of the i-th DG; P is a positive definite matrix; is a positive constant; are parameters related to neighbor nodes and reference values, and are defined as follows:

[0029]

[0030] in, a ijis an element of the weighted adjacency matrix, reflecting the communication relationship between nodes; r g (t) is the state related to the virtual leader.

[0031] Furthermore, the specific method of step 3 is as follows:

[0032] Step 3.1: Propose a triggering condition: o i represents a dynamic variable related to the triggering condition;

[0033] Step 3.2: Design a multi-measurement error mechanism according to the error parameter given in step 2.1 and define a dynamic variable which is determined by the following formula:

[0034]

[0035] where, σ i is a design parameter; is the combined measurement error, which determines the update time of the controller, and e i represents the error between the observer output and the critical bus voltage.

[0036] The beneficial effects of adopting the above technical solution are as follows: The MMGS distributed critical bus voltage control method based on containment control provided by the present invention proposes a container-based distributed controller to balance two conflicting objectives of voltage regulation and tide. The controller can control the critical bus voltage within a reasonable range in the presence of a voltage difference to achieve power flow; transform the critical bus voltage regulation problem of the AC automatic control system into a distributed output feedback tracking problem of a linear automatic control system with nonlinear dynamic characteristics; different from the current event-based secondary control strategy, a suitable timer is designed to ensure the existence of positive MIET for each DG unit. In addition, the auxiliary controller only depends on the information related to the observer, thus preventing the use of global information about the communication topology. Compared with the mas-based algorithm, the present invention innovatively depends on multiple measurement errors to avoid continuous updates of the controller. Through simulation examples, it is verified that the method of the present invention is effective, and it is also applicable to the cases of load switching and plugging and unplugging capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a block diagram of MMGS provided by an embodiment of the present invention and a schematic diagram of a proposed secondary voltage regulation strategy based on containment control;

[0038] Figure 2 is a block diagram of an inverter-based DG provided by an embodiment of the present invention;

[0039] Figure 3 Schematic diagram of the MGs test system and the considered communication structure provided by an embodiment of the present invention;

[0040] Figure 4 Critical bus voltage v provided by an embodiment of the present invention bi Curve graph;

[0041] Figure 5 Error y' between the observer output and the critical bus voltage provided by an embodiment of the present invention i Curve graph;

[0042] Figure 6 Reactive power Q of the DG corresponding to the critical bus provided by an embodiment of the present invention i Curve graph;

[0043] Figure 7 Active power P of the DG corresponding to the critical bus provided by an embodiment of the present invention i Curve graph;

[0044] Figure 8 Interval time provided by an embodiment of the present invention Schematic diagram;

[0045] Figure 9 Interval time provided by an embodiment of the present invention Schematic diagram;

[0046] Figure 10 Interval time provided by an embodiment of the present invention Schematic diagram;

[0047] Figure 11 Interval time provided by an embodiment of the present invention Schematic diagram. Specific embodiments

[0048] The following combines the accompanying drawings and embodiments to further describe in detail the specific embodiments of the present invention. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0049] As Figure 1As shown in the figure, this embodiment proposes a container-based distributed critical bus voltage control strategy. First, a nonlinear dynamic model of the microgrid is established. Through feedback linearization, the model is transformed into a linear multi-agent system (MAS) model and simplified to obtain a simplified control model. Secondly, a distributed secondary voltage regulation strategy based on inclusion control is designed, which allows the critical bus voltage to be limited within a defined range while allowing voltage differences between submicrogrids (SMGs), thereby allowing power to flow between individual submicrogrids. In addition, an event-triggering mechanism based on multiple measurement errors is proposed. Under this event-triggering mechanism, each distributed generator (DG) has a uniform and strictly positive minimum interevent time (MIET), and continuous updates of the secondary controller are avoided. Moreover, this voltage control strategy does not require any global information in the communication graph. Finally, the results are verified by simulation. The specific method is described as follows.

[0050] Step 1: Establish a nonlinear dynamic model of the microgrid. Through feedback linearization, the model is transformed into a linear multi-agent system (Multi-Microgrid System, MAS) model and simplified to obtain a simplified control model:

[0051] S1.1: As Figure 1 shown, an MMGS is composed of N SMGs connected in series and can generally be regarded as a MAS, where the N SMGs are its agents and the SMGs can communicate through a communication network for communication. Among them is the node set, is the edge set. The weighted adjacency matrix is defined as follows: when a ij = 1; otherwise, a ij = 0. The Laplacian matrix satisfies and when i ≠ j,

[0052] In the SMG, as Figure 2 shown, each DG includes an inner-loop voltage-current power source, an LC filter, and a line model. According to the large-signal dynamic model of the i-th DG, the nonlinear dynamic model of the AC microgrid is written as follows:

[0053]

[0054] Among them, is the state vector, Denote the state vector x i The first derivative of, δ i Denote the angle of the i-th DG reference frame relative to the common reference frame, P i And Q i Denote the active power and reactive power at the i-th inverter terminal, Is the auxiliary state variable in the voltage controller, Is the auxiliary state variable in the current controller, i ldi 、i lqi Are filter current elements, v odi 、v oqi Respectively denote the direct component and quadrature component of the output voltage v oi The direct component and quadrature component of the output current i odi 、i oqi Respectively denote the direct component and quadrature component of the output current i oi ; D i =[ω com ,v bdi ,v bqi ; ω i Denote the rotational speed of the reference frame of the i-th DG, taking the reference frame of one of the DGs as the common reference frame, ω com Denote the rotational frequency of the common reference frame; v bdi 、v bqi Denote the d-axis and q-axis components of the critical bus voltage; u i Denote the control input; f i (x i ), g i (x i ), k i (x i ), h i (x i ) Are functions related to the system characteristics, used to describe the nonlinear dynamic behavior of the distributed motor; Set as v bi , Denote the critical bus voltage.

[0055] S1.2: Perform feedback linearization transformation on the nonlinear dynamic model of the AC microgrid:

[0056]

[0057] Among them, Is The second derivative of; F i (x i ) = f i (x i ) + k i (x i )D i, denotes the Lie derivative of h i and is defined as: is defined as: (In differential geometry, the Lie derivative is an operator named after Sophus Lie that acts on tensor fields, vector fields, or functions on a manifold, taking the directional derivative of the tensor along the flow of a vector field.) Similarly, V ni is equivalent to the control input u i . This step transforms the nonlinear model into a form convenient for control design.

[0058] Feedback linearization is a method of transforming a nonlinear system into a linear system through coordinate transformation and nonlinear feedback. In this process, complex mathematical processing and transformation are performed on the nonlinear part of the system. Specifically: Define a new control input u i such that This step recombines and defines the nonlinear terms related to control in the original system, aiming to design the controller as if it were for a linear system later. After such transformation, the original system can be simplified to in the form, which is convenient for designing control strategies based on linear system theory to effectively control key system quantities such as the critical bus voltage.

[0059] S1.3: Simplify the control model:

[0060]

[0061] where r i denotes the two-dimensional vector composed of the d-axis component v bdi of the critical bus voltage and its derivative and is used to construct the control algorithm model, The A, B, C matrices determine the dynamic characteristics and input-output relationship of the system, B = [0,1] T , C = [1,0]; is the first derivative of r i ; N represents the number of sub-microgrids.

[0062] Step 2: Design a distributed secondary voltage regulation strategy based on containment control, which allows the critical bus voltage to be restricted within a defined range while allowing a voltage difference to exist between sub-microgrids (SMGs), thus allowing power to flow between individual sub-microgrids:

[0063] S2.1: Design a state observer

[0064]

[0065] Among them, the estimated parameter The first derivative of γ θi , k θi , π1, and R are design parameters; is the estimate of r i (t), is The first derivative of is the observer output; is the error between the observer output and the critical bus voltage; is the observer gain, represents the real number field; is the estimated parameter; represents a mathematical mapping that transforms the state estimation information into a quantity that can be used for operations such as control algorithms and parameter estimation.

[0066] S2.2: Propose a distributed inclusion protocol:

[0067]

[0068] Among them, c i is a positive control gain; ρ i is an adaptive dynamic parameter, is the first derivative of ρ i ; is the k-th triggering time of the i-th DG (distributed generator); P is a positive definite matrix; is a positive constant; is a parameter related to neighbor nodes and reference values.

[0069] Definition of related parameters:

[0070]

[0071] Among them, a ij is an element of the weighted adjacency matrix, reflecting the communication relationship between nodes; r g (t) is the state related to the virtual leader.

[0072] Step 3: Propose an event-triggering mechanism based on multiple measurement errors. Under this event-triggering mechanism, each distributed generator (DG) has a uniform and strictly positive minimum interevent time (MIET), and it can avoid continuous updates of the secondary controller; moreover, it does not require any global information in the communication graph:

[0073] S3.1: Propose the triggering condition: o i Denote the dynamic variables related to the triggering condition;

[0074] S3.2: According to the error parameters given in Step 2.1 Design a multiple measurement error mechanism and define a dynamic variable which is determined by the following formula:

[0075]

[0076] where, σ i is the design parameter; is the combined measurement error, and this condition determines the update moment of the controller. e i denotes the error between the observer output and the critical bus voltage.

[0077] To verify the feasibility and effectiveness of the event-triggered secondary control strategy based on control and output feedback proposed in this embodiment in the critical bus voltage regulation, use matlab / simulink to simulate the MG scenario, as Figure 3 shown. The MMGS is tested with a combination of four SMGs, where v b1 , v b2 , v b3 , v b4 are the critical bus voltages of MG#1, MG#2, MG#3, and MG#4 respectively. Solve PA T +AP - PBB T P + I n = 0 to obtain:

[0078] Define v ref1 = 230, v ref2 = 210, and it can be obtained:

[0079] Let ρ i (t), sh The initial values are ρ i (0.4) = 1, where, i = 1, 2, 3, 4. In addition, c i = 1.5 * 107 , σ i = 0.5, v i = 40.

[0080] As Figure 4 shown, the effectiveness of the critical bus voltage algorithm based on inclusion control is verified. Initially relying only on primary control, at t = 0.4 s, the secondary control based on the critical bus voltage regulation of inclusion control is switched in, and the critical bus voltage returns to the convex hull spanned by all reference values. At t = 0.75 s, Load #3 is disconnected, at t = 1.1 s, MG#4 is disconnected, and is restored at t = 1.7 s. The simulation results are as Figure 5 , Figure 6 and Figure 7 shown. The control method proposed in this embodiment can regulate the critical bus voltage, active power, and reactive power after large-signal disturbances (i.e., load changes and MMGS structure reconfiguration). For clear representation, the triggering times of DG#1, DG#2, DG#3, DG#4 each time they are triggered are respectively as Figure 8 , Figure 9 , Figure 10 and Figure 11 shown.

[0081] Therefore, the critical bus voltage regulation strategy based on inclusion control proposed in this embodiment is proven to be effective, and it is also applicable to the cases of load switching and plugging and unplugging capabilities, demonstrating the feasibility of the monitoring scheme of the present invention.

[0082] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope defined by the claims of the present invention.

Claims

1. A distributed critical bus voltage control method based on inclusion control, characterized in that: Including the following steps: Step 1: Establish a microgrid non-linear dynamic model, transform the model into a linear multi-agent system model through feedback linearization, and simplify it to obtain a simplified control model; Step 2: Design a distributed secondary voltage regulation strategy based on inclusion control, which allows the critical bus voltage to be limited within a defined range, while allowing a voltage difference to exist between sub-microgrids, thus allowing power to flow between individual sub-microgrids; Step 3: Propose an event-triggering mechanism based on multiple measurement errors. Under this event-triggering mechanism, each distributed generator DG has a uniform and strictly positive minimum event interval time, and can avoid continuous updates of the secondary controller; and does not require any global information to be used in the communication graph.

2. The MMGS distributed critical bus voltage control method based on inclusion control according to claim 1, characterized in that: In the said Step 1, the microgrid non-linear dynamic model is established as: Among them, is the state vector, denotes the first derivative of the state vector x i , and δ i represents the angle of the i-th DG reference frame relative to the common reference frame, P i and Q i represent the active power and reactive power at the i-th inverter terminal, is the auxiliary state variable in the voltage controller, is the auxiliary state variable in the current controller, i ldi , i lqi are the filter current components, v odi , v oqi respectively represent the direct component and quadrature component of the output voltage v oi , i odi , i oqi respectively represent the direct component and quadrature component of the output current i oi ; D i = [ω com , v bdi , v bqi ; ω i represents the rotational speed of the reference frame of the i-th DG. Taking the reference frame of one of the DGs as the common reference frame, ω com represents the rotational frequency of the common reference frame; v bdi , v bqi represent the d-axis and q-axis components of the critical bus voltage; u i represents the control input; f i (x i ), g i (x i ), k i (x i ), h i (x i ) are functions related to the system characteristics and are used to describe the nonlinear dynamic behavior of the distributed motor; Let it be v bi , representing the critical bus voltage.

3. The MMGS distributed critical bus voltage control method based on inclusion control according to claim 2, characterized in that: In the said Step 1, the model is transformed by feedback linearization according to the following formula: Among them, is the second derivative; F i (x i ) = f i (x i ) + k i (x i )D i , L Fi h i represents the Lie derivative of h i , defined as: Defined as: V ni is equivalent to the control input u i ; v li represents the line voltage between the critical bus and the inverter output, is li the second derivative of v This step transforms the non - linear model into a form convenient for control design.

4. The MMGS distributed critical bus voltage control method based on inclusion control according to claim 3, characterized in that: In the said Step 1, the simplified control model is: where r i represents the two-dimensional vector composed of the d-axis component v bdi of the critical bus voltage and its derivative v bdi , which is used to construct the control algorithm model. Matrices A, B, and C determine the dynamic characteristics and input-output relationship of the system. B = [0, 1] T , and C = [1, 0]; is the first derivative of r i . N represents the number of sub-microgrids.

5. The MMGS distributed critical bus voltage control method based on inclusion control according to claim 4, wherein: The specific method of the said Step 2 is: Step 2.1: Design a state observer: Among them, the first-order derivative of the estimated parameter is γ θi 、k θi 、π1, R are design parameters; is the estimate of r i (t), is the first derivative of; is the observer output; is the error parameter between the observer output and the critical bus voltage; is the observer gain, denotes the real number field; is the estimated parameter; denotes a mathematical mapping that transforms the state estimation information into a quantity that can be used in control algorithms and parameter estimation operations; Step 2.2: Propose a distributed inclusion protocol: where c i is a positive control gain; ρ i is an adaptive dynamic parameter, is the first derivative of ρ i ; is the k-th triggering time of the i-th DG; P is a positive definite matrix; is a positive constant; is a parameter related to neighbor nodes and reference values, defined as follows: Among them, a ij is an element of the weighted adjacency matrix, reflecting the communication relationship between nodes; r g (t) is the state related to the virtual leader.

6. The MMGS distributed critical bus voltage control method based on inclusion control according to claim 5, characterized in that: The event trigger mechanism in step 3 has the following trigger conditions: o i represents the dynamic variables related to the trigger conditions.

7. The distributed critical bus voltage control method based on containment control according to claim 6, characterized in that: The event-triggering mechanism based on multiple measurement errors in step 3 is based on the error parameters given in step 2.1 Design a multiple measurement error mechanism and define a dynamic variable which is determined by the following formula: Among them, σ i is a design parameter; is the combined measurement error, and this condition determines the update moment of the controller. e i represents the error between the observer output and the critical bus voltage.

Images