Island micro-grid decentralized secondary control method based on generalized scouring filter and adaptive virtual impedance

Abstract

The invention relates to an island micro-grid decentralized secondary control method based on a generalized scouring filter and adaptive virtual impedance, and solves the defects of poor power sharing capability, large frequency deviation and strong dependence on communication when a traditional control method is applied to an island micro-grid under the condition that line impedance is not matched compared with the prior art. The method comprises the following steps: establishing a microgrid droop control model based on a generalized scouring filter; obtaining a steady-state output power error of the distributed power supply of the micro-grid; self-adaptive virtual impedance is generated based on active mode switching; and carrying out distributed secondary control on the island micro-grid. According to the method, the active power error is obtained based on the dynamic phasor model, then the adaptive virtual impedance is generated, and the constraint condition that the virtual impedance increment is zero is set, so that the stability of the voltage of the common coupling point of the microgrid can be effectively ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of hierarchical control of islanded microgrids, and specifically to a decentralized secondary control method for islanded microgrids based on a generalized washout filter and adaptive virtual impedance. Background Art

[0002] A microgrid (MG) composed of distributed generation (DG), energy storage, and loads can make more reasonable and effective use of renewable energy and reduce environmental pollution. The microgrid has two operating modes: islanded and grid-connected, and has the characteristics of control flexibility and stability. In the islanded mode, the microgrid uses hierarchical control of primary control and secondary control to coordinate different types of DGs to provide active and reactive power to local loads, achieve power sharing, and maintain voltage and frequency stability. To ensure that each DG can achieve power sharing without communication, droop control is generally used in the primary control of hierarchical control, which adjusts the frequency and voltage amplitude through droop control of active power frequency modulation and droop control of reactive power voltage regulation. Since frequency is a global variable, droop control can achieve active power sharing, but the presence of line impedance differences results in poor reactive power sharing performance. To solve the problem of poor reactive power sharing caused by line impedance mismatch, the main measure is to introduce virtual impedance control.

[0003] The voltage amplitude and frequency deviations that occur in droop control at steady state can be solved by secondary control (SC). Traditional methods use improved secondary control to restore the voltage amplitude and frequency to the rated value. However, these methods use low-bandwidth communication lines, there is a delay in control signals, which affects the secondary control effect.

[0004] Therefore, how to achieve power sharing and frequency restoration of each DG in an islanded microgrid without communication is a key technical problem that urgently needs to be solved in the technical field of hierarchical control of islanded microgrids, and it is of great significance for improving power supply reliability and promoting the development of renewable energy. Summary of the Invention

[0005] The purpose of the present invention is to solve the defects of poor power sharing ability, large frequency deviation, and strong dependence on communication existing in the application of traditional control methods in an islanded microgrid under the condition of line impedance mismatch, and to provide a decentralized secondary control method for an islanded microgrid based on a generalized washout filter (GWF) and adaptive virtual impedance to solve the above problems.

[0006] To achieve the above purpose, the technical solution of the present invention is as follows:

[0007] An islanded microgrid decentralized secondary control method based on a generalized washout filter and adaptive virtual impedance, comprising the following steps:

[0008] Establish a droop control model for the microgrid based on a generalized washout filter;

[0009] Obtain the steady-state output power error of the distributed power sources in the microgrid;

[0010] Generate an adaptive virtual impedance based on active mode switching;

[0011] Perform decentralized secondary control of the islanded microgrid.

[0012] The establishment of the droop control model for the microgrid based on a generalized washout filter is as follows: Design a droop control model for the microgrid based on a generalized washout filter using the droop control equation; comprising the following steps:

[0013] Suppose two DGs are connected to the point of common coupling PCC and the line impedances are mismatched, then:

[0014] DG i The active power and reactive power output are expressed as:

[0015]

[0016] Where: P i is the active power output by the i-th DG; Q i is the reactive power output by the i-th DG; V i represents the amplitude of the output voltage of DG i ; V L is the amplitude of the PCC voltage; δ i is the phase angle difference between the output voltage V i of the i-th DG and the PCC voltage V L ; R i is the resistance of the i-th line impedance; X i is the inductance of the i-th line impedance;

[0017] Assume that the value of δ i is extremely small, and there is sinδ i ≈δ i , cosδ i ≈1. Simplify equations (1) and (2) to:

[0018]

[0019] Among them, the active power P i is proportional to the phase angle difference δ i , and the reactive power Q i is related to the output voltage amplitude V i ;

[0020] Calculate the DG through the droop control equation i The frequency and amplitude of the output voltage, which are expressed as:

[0021] ω i = ω 0 - m i (P i - P 0,i ) (5)

[0022] V i = V 0 - n i (Q i - Q 0,i ) (6)

[0023] Where: ω i represents the output angular frequency of the i-th DG; V i represents the output voltage of the i-th DG; ω 0 and V 0 are the nominal angular frequency and the nominal output voltage amplitude respectively; m i and n i are the droop coefficients of the frequency droop controller and the voltage droop controller of the i-th DG respectively; P i is the output power of the i-th DG; Q i is the output power of the i-th DG; P 0,i and Q 0,i are the nominal active power and the nominal reactive power of the i-th DG respectively;

[0024] Establish a microgrid droop control model based on the generalized washout filter, and its expression is as follows:

[0025] ω i = ω 0 - m i · G GWF (s)(P i - P 0,i ) (7)

[0026] V i = V 0 - n i · G GWF (s)(Q i - Q 0,i ) (8)

[0027] Where: G GWF (s) represents the transfer function of the generalized washout filter;

[0028]

[0029] Where: ωl is the cut-off frequency of the low-pass filter; ω h is the cut-off frequency of the washout filter, and s represents the complex frequency domain.

[0030] The obtained steady-state output power error of the DG in the microgrid is as follows: Based on the dynamic phasor method, the output power transfer function of the DG based on the generalized washout filter droop control is derived. Using the final value theorem, the steady-state output power of the DG based on the generalized washout filter droop control is obtained to obtain the steady-state output power error of the DG in the microgrid; it includes the following steps:

[0031] Derive the distributed generator DG based on the dynamic phasor model i , i = 1, 2, the active power and reactive power output:

[0032]

[0033] where: P i is the active power output by the i-th DG; Q i is the reactive power output by the i-th DG; V i represents the amplitude of the output voltage of the i-th DG; V L is the amplitude of the PCC voltage; δ i is the output voltage V of the i-th DG i and the PCC voltage V L phase difference between; R i is the resistance of the impedance of the i-th line; X i is the inductance of the impedance of the i-th line, and s represents the complex frequency domain;

[0034] Ignoring line losses, there is:

[0035]

[0036] where: P L and Q L are the active power and reactive power of the load;

[0037] Linearize equations (10) and (11) and use the final value theorem to obtain the steady-state output power of the DG based on the generalized washout filter droop control as:

[0038]

[0039] where: P ss i is the steady-state active power output by the i-th DG; Q ss i is the steady-state reactive power output by the i-th DG; X 1 and X 2 are the line inductances of the two transmission lines respectively; P LWith Q L are respectively the active power and reactive power; m 1 and m 2 are respectively the droop coefficients of the frequency droop controllers of two DGs; V L is the PCC voltage amplitude; ω h is the cut-off frequency of the wash filter; ω l is the cut-off frequency of the low-pass filter;

[0040] To achieve accurate power sharing, the relationship of the droop coefficients of multiple parallel DGs is:

[0041]

[0042] Where: P i * , are the expected active and reactive powers of the i-th DG, m i and n i are respectively the droop coefficients of the frequency droop controller and voltage droop controller of the i-th DG;

[0043] Obtain the steady-state output power error of the microgrid DG:

[0044] Define the active power sharing error as P erri , and the reactive power sharing error as Q erri , and there is:

[0045]

[0046]

[0047] Where: P erri is the active power sharing error of the i-th DG, Q erri is the reactive power sharing error of the i-th DG; P i is the output power of the i-th DG; Q i is the output power of the i-th DG;

[0048] Generalize to an islanded microgrid with multiple DGs operating in parallel. The active power sharing error ΔP of the DG based on the generalized wash filter droop control erri is:

[0049]

[0050] Where: ΔP erri is the active power sharing error of the i-th DG: m ∑ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of the line impedances; V L is the PCC line voltage amplitude.

[0051] The adaptive virtual impedance generated based on the active mode switching is as follows: The controller operates in the droop control mode based on the generalized washout filter, samples the power value, and then switches to the droop control mode to obtain the power sharing error. The virtual impedance generation formula is derived to adaptively generate the virtual impedance, including the following steps:

[0052] Design the virtual impedance as:

[0053]

[0054] Where: X vi The virtual impedance designed for the i-th DG; Represents the expected value of the line impedance of the i-th DG; X i Is the actual line impedance of the i-th DG;

[0055] Propose a constraint condition that the total increment of the virtual impedance is 0 to keep the total line impedance unchanged, then it satisfies:

[0056]

[0057] From Equation (19) and Equation (20), we get:

[0058]

[0059] Where: Is the sum of the expected values of the line impedance; X ∑ Is the sum of the actual line impedance;

[0060] The sufficient condition for k DGs with different capacities to be connected in parallel to achieve reasonable power sharing is:

[0061]

[0062] Where: Represents the expected value of the line impedance of the i-th DG; m i Is the droop coefficient of the frequency droop controller of the i-th DG;

[0063] Substitute Equation (19) and Equation (22) into Equation (21) to get:

[0064]

[0065] Where: X vi Is the generated virtual impedance; m ∑ Represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ Represents the sum of the line impedances; m i Is the droop coefficient of the frequency droop controller of the i-th DG; X iis the actual line impedance of the i-th DG;

[0066] Substitute Equation (18) into Equation (23) to generate the virtual impedance:

[0067]

[0068] where: X vi is the generated virtual impedance; m ∑ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of each line impedance; ΔP erri is the active power sharing error of the i-th DG; V L is the voltage amplitude of the PCC line;

[0069] For the droop control based on the generalized washout filter, after the active mode switching, it is switched to the droop control to generate ΔP erri , that is:

[0070]

[0071] where: P i is the output power of the i-th DG; P i * is the expected value of the active power of the i-th DG; P G.ave and P D.ave are the steady-state values of the active power under the droop control mode based on the generalized washout filter and the traditional droop mode respectively; P L is the load active power;

[0072] Taking into account the influence of the line resistance, introduce the compensation coefficient K to generate the adaptive virtual impedance X vi , and there is:

[0073]

[0074] where: K is the compensation coefficient; X vi is the generated virtual impedance; m ∑ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of each line impedance; ΔP erri is the active power sharing error of the i-th DG; V L is the voltage amplitude of the PCC line.

[0075] The decentralized secondary control of the islanded microgrid is as follows: Each DG unit samples and records the active power based on the droop control of the generalized washout filter; then switch the controller to the droop control mode, and record the active power value at this time as the expected value P D,ave; Enable the proposed adaptive virtual impedance control to enter a new steady state; then switch the active power control mode to the droop control mode of the generalized washout filter to achieve bias-free frequency control, that is, finally achieve the sharing of DG power and the restoration of frequency; including the following steps:

[0076] Based on the active mode switching, obtain the active power sharing error ΔP erri , and then generate the adaptive virtual impedance X v . The d-axis and q-axis reference voltages of the DG controlled by the adaptive virtual impedance are:

[0077]

[0078] In the formula: and are the d-axis reference voltage and the q-axis reference voltage respectively; V d and V q are the d-axis actual voltage and the q-axis actual voltage respectively; X v is the generated virtual impedance; I od and I oq are the dq-axis output currents of the DG;

[0079] and serve as the voltage outer loop reference values in the voltage-current double closed-loop system. Among them, both the voltage outer loop and the current inner loop adopt PI control to track the frequency and voltage reference values;

[0080] Define the controller operation mode and basic events;

[0081] Based on the droop control of the generalized washout filter and the adaptive virtual impedance, realize the decentralized secondary control of the islanded microgrid.

[0082] The above-mentioned definition of the controller operation mode and basic events includes the following steps:

[0083] Set F v = 1 indicates enabling the adaptive virtual impedance control; F G = 1 indicates that the DG operates in the droop control mode based on the generalized washout filter, and F G = 0 indicates that the DG operates in the traditional droop control mode; F t = 1 indicates the transient operation of the islanded microgrid, and F t = 0 indicates the steady-state operation of the islanded microgrid;

[0084] Input the active power change rate and the reactive power change rate into the low-pass filter with a cut-off frequency of 10, or Output after low-pass filtering;

[0085] If it exceeds the defined threshold ξ PQ , it is determined that the islanded microgrid is in transient operation, i.e., F t = 1; and The outputs after low-pass filtering are both less than the threshold, and it is determined that the islanded microgrid is in steady-state operation, i.e., F t = 0;

[0086] Set event A to record the current local output active power P G,ave , which is recorded by the sample-and-hold module, and switch the active power control mode to the traditional droop control mode, F G = 0;

[0087] Event B is defined as recording and holding the current active power P D,ave , and enabling the proposed adaptive virtual impedance control;

[0088] Event C is defined as switching the active power control mode to the droop control mode based on the generalized washout filter, F G = 1.

[0089] The droop control based on the generalized washout filter and the adaptive virtual impedance to achieve decentralized secondary control of the islanded microgrid includes the following steps:

[0090] When the load of the islanded microgrid changes and the system enters the transient state, i.e., F t = 1, each distributed generator operates in the droop control mode based on the washout filter to achieve decentralized secondary control;

[0091] After monitoring that the system reaches the steady state, i.e., F t = 0, trigger event A, the sample-and-hold module records the active power P G,ave , and switch the active power control mode to the droop control mode, i.e., F G = 0, and the reactive power control mode continues to use the mode based on the generalized washout filter;

[0092] After monitoring that the active power is stable, record the active power value at this time as the expected value P D,ave , trigger event B, and enable the proposed adaptive virtual impedance control, i.e., F v = 1, and wait for the system to enter the steady state again;

[0093] After monitoring that the system re-enters the steady state, trigger event C, and switch the active power control mode to the droop control mode based on the generalized washout filter, i.e., F G = 1, to achieve bias-free power and frequency control, and finally realize the decentralized secondary control of the islanded microgrid based on the droop control of the generalized washout filter and the adaptive virtual impedance.

[0094] Beneficial effects

[0095] Compared with the prior art, the islanded microgrid decentralized secondary control method based on the generalized washout filter and adaptive virtual impedance of the present invention obtains the active power error based on the dynamic phasor model, and then generates the adaptive virtual impedance, and sets the constraint condition that the virtual impedance increment is zero, which can effectively ensure the stability of the voltage at the point of common coupling of the microgrid. In addition, combining the advantages of traditional droop control and droop control based on the generalized washout filter, an active mode switching method is innovatively proposed to achieve power sharing and frequency recovery of the islanded microgrid without communication. The present invention has the advantages of not requiring communication support, high reliability, and strong stability. Description of the drawings

[0096] Figure 1 It is the sequence diagram of the method of the present invention;

[0097] Figure 2 It is the single-line diagram of the islanded microgrid with two DGs;

[0098] Figure 3 It is the flowchart of the decentralized secondary control without communication of the DG in the islanded microgrid based on the droop control of GWF and adaptive virtual impedance control;

[0099] Figure 4 It is the structure diagram of the islanded microgrid with three distributed DGs;

[0100] Figure 5 It is the simulation result diagram. Specific implementation manners

[0101] To have a further understanding and recognition of the structural features and achieved effects of the present invention, the following is a detailed description with the preferred embodiments and drawings:

[0102] As Figure 1 shown, the islanded microgrid decentralized secondary control method based on the generalized washout filter and adaptive virtual impedance of the present invention breaks through the limitation of traditional secondary control relying on communication conditions, and innovatively proposes a decentralized secondary control method for the islanded microgrid by an active mode switching method, realizing power sharing and frequency recovery of the islanded microgrid without communication, and effectively solving the problems of poor power sharing ability, large frequency deviation, and strong dependence on communication when applying traditional control methods in the islanded microgrid with mismatched line impedances. The specific steps are as follows:

[0103] First step, establish a microgrid droop control model based on the generalized washout filter. Derive the active power and reactive power output of the DG, and simplify the results in combination with the actual line impedance to obtain the traditional droop control equation. Then establish a microgrid droop control model based on the generalized washout filter. The specific steps are as follows:

[0104] (1) Derivation of the traditional droop control equation: As Figure 2 shown, take the example of two DG units connected to a common PCC point with mismatched line impedances for analysis. The active power and reactive power output by DG i (i = 1, 2) can be expressed as:

[0105]

[0106] where: P i is the output power of the i-th DG; Q i is the output power of the i-th DG; V i represents the amplitude of the output voltage of DG i ; V L is the amplitude of the PCC line voltage; δ i is the phase angle difference between the output voltage V i of the i-th DG and the PCC line voltage V L ; R i is the resistance of the i-th line impedance; X i is the inductance of the i-th line impedance;

[0107] Assume that the value of δ i is extremely small, with sinδ i ≈δ i , cosδ i ≈1. Simplify equations (1) and (2) to:

[0108]

[0109] where the active power is proportional to the phase angle difference δ i , and the reactive power is related to the amplitude of the output voltage V i ;

[0110] Calculate the frequency and amplitude of the output voltage through the droop control equation, which is expressed as:

[0111] ω i = ω 0 - m i (P i - P 0,i ) (5)

[0112] V i = V 0 - n i (Qi -Q 0,i ) (6)

[0113] where: ω i represents the output angular frequency of the i-th DG; V i represents the output voltage of the i-th DG; ω 0 and V 0 are the nominal angular frequency and the nominal output voltage amplitude respectively; m i and n i are the droop coefficients of the frequency droop controller and the voltage droop controller of the i-th DG respectively; P i is the output power of the i-th DG; Q i is the output power of the i-th DG; P 0,i and Q 0,i are the nominal active power and reactive power of the i-th DG respectively.

[0114] (2) Establish a microgrid droop control model based on the generalized washout filter, and its expression is as follows:

[0115] ω i = ω 0 - m i ·G GWF (s)(P i - P 0,i ) (7)

[0116] V i = V 0 - n i ·G GWF (s)(Q i - Q 0,i ) (8)

[0117] where: G GWF (s) represents the transfer function of the generalized washout filter;

[0118]

[0119] where: ω l is the cut-off frequency of the low-pass filter; ω h is the cut-off frequency of the washout filter, and s represents the complex frequency domain.

[0120] Second step, obtain the steady-state output power error of the microgrid DG: Derive the active power and reactive power output of the distributed generator based on the dynamic phasor model, then perform linearization processing and use the final value theorem to obtain its steady-state output power. This method cleverly utilizes the characteristic of the natural equal sharing of the droop control active power to obtain the expected output power of each DG, and finally obtains the steady-state output power error of each DG. This method can be extended to multiple DGs and has strong universality.

[0121] Obtain the steady-state output power error of the microgrid DG, and the specific steps are as follows:

[0122] (1) Derive the active and reactive power outputs of the distributed generator DG i , i = 1, 2:

[0123]

[0124] Where: P i is the output power of the i-th DG; Q i is the output power of the i-th DG; V i represents the amplitude of the output voltage of the i-th DG; V L is the amplitude of the PCC line voltage; δ i is the phase angle difference between the output voltage V i of the i-th DG and the PCC line voltage V L ; R i is the resistance of the impedance of the i-th line; X i is the inductance of the impedance of the i-th line, and s represents the complex frequency domain;

[0125] Ignoring line losses, we have:

[0126]

[0127] Where: P L and Q L are the active and reactive powers of the load;

[0128] (2) Linearize equations (10) and (11) and use the final value theorem to obtain the steady-state output power of the DG based on the droop control of the generalized washout filter as:

[0129]

[0130] Where: P ss i is the steady-state active power output by the i-th DG; Q ss i is the steady-state reactive power output by the i-th DG; X 1 and X 2 are the line inductances of the two transmission lines respectively; P L and Q L are the active and reactive powers of the load respectively; m 1 and m 2 are the droop coefficients of the frequency droop controller and the voltage droop controller of the two DGs respectively; V L is the amplitude of the PCC line voltage; ω h is the cut-off frequency of the washout filter; ωl is the cut-off frequency of the low-pass filter;

[0131] To achieve precise power sharing, the droop coefficient relationship of multiple parallel DGs is:

[0132]

[0133] where: P i * and are the expected active and reactive power of the i-th DG; m i and n i are the droop coefficients of the frequency droop controller and voltage droop controller of the i-th DG respectively;

[0134] (3) Obtain the steady-state output power error of the microgrid DG:

[0135] Define the active power sharing error as P erri and the reactive power sharing error as Q erri , and there is:

[0136]

[0137] where: P erri is the active power sharing error of the i-th DG, Q erri is the reactive power sharing error of the i-th DG; P i is the output power of the i-th DG; Q i is the output power of the i-th DG;

[0138] For an islanded microgrid with multiple DGs operating in parallel, the active power sharing error ΔP of the DG based on the generalized washout filter droop control is erri as follows:

[0139]

[0140] where: ΔP erri is the active power sharing error of the i-th DG: m ∑ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of the line impedances; V L is the voltage amplitude of the PCC line.

[0141] In the third step, based on the active mode switching to generate an adaptive virtual impedance, the present invention proposes to set a constraint condition that the virtual impedance increment is 0, so as to achieve only improving the line impedance distribution without increasing additional voltage drop and having no influence on the voltage at the point of common coupling; in addition, compared with the traditional virtual impedance generation method, this method has the advantages of not requiring communication conditions, higher reliability and stronger scalability.

[0142] Generate adaptive virtual impedance based on active mode switching, specifically including the following steps:

[0143] (1) Design the virtual impedance as:

[0144]

[0145] where: X vi is the virtual impedance designed for the i-th DG; represents the expected value of the line impedance of the i-th DG; X i is the actual line impedance of the i-th DG;

[0146] Propose a constraint condition that the total increment of the virtual impedance is 0, and keep the total line impedance unchanged, then it satisfies:

[0147]

[0148] From equations (19) and (20), we get:

[0149]

[0150] where: is the sum of the expected values of the line impedance; X ∑ is the sum of the actual line impedances;

[0151] (2) The sufficient condition for k DGs with different capacities to be connected in parallel to achieve reasonable power sharing is:

[0152]

[0153] where: represents the expected value of the line impedance of the i-th DG; m i is the droop coefficient of the frequency droop controller of the i-th DG;

[0154] Substitute equations (19) and (22) into equation (21):

[0155]

[0156] where: X vi is the generated virtual impedance; m ∑ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of the line impedances; m i is the droop coefficient of the frequency droop controller of the i-th DG; X i is the actual line impedance of the i-th DG;

[0157] Substitute equation (18) into equation (23) to generate the virtual impedance:

[0158]

[0159] Among them: X vi is the generated virtual impedance; m Σ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X Σ represents the sum of the line impedances; ΔP erri is the active power sharing error of the i-th DG; V L is the amplitude of the PCC line voltage;

[0160] (3) For the droop control based on the generalized washout filter, after active mode switching, it is switched to the droop control to generate ΔP erri , that is:

[0161]

[0162] Among them: P i is the output power of the i-th DG; P i * is the expected active power of the i-th DG; P G.ave and P D.ave are the steady-state values of the active power under the droop control mode based on the generalized washout filter and the traditional droop mode respectively; P L is the load active power;

[0163] Considering the influence of the line resistance, a compensation coefficient K is introduced to generate an adaptive virtual impedance X vi , and there is:

[0164]

[0165] Among them: K is the compensation coefficient; X vi is the generated virtual impedance; m Σ represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ represents the sum of the line impedances; ΔP erri is the active power sharing error of the i-th DG; V L is the amplitude of the PCC line voltage.

[0166] Fourthly, perform decentralized secondary control of the islanded microgrid, use the generated adaptive virtual impedance to compensate for the mismatch of the line impedance of the islanded microgrid, and adopt the method of active mode switching to achieve power sharing and voltage and frequency recovery. Specifically, it includes the following steps:

[0167] (1) Based on active mode switching, obtain the active power sharing error ΔP erri , and then generate the adaptive virtual impedance X v , and the d-axis and q-axis reference voltages of the DG controlled by the adaptive virtual impedance are:

[0168]

[0169] Wherein: and are the d-axis reference voltage and the q-axis reference voltage respectively; V d and V q are the d-axis actual voltage and the q-axis actual voltage respectively; X v is the generated virtual impedance; I od and I oq are the dq-axis output currents of the DG;

[0170] V d * and V q * are used as the voltage outer-loop reference values in the voltage-current double closed-loop system. Among them, both the voltage outer loop and the current inner loop adopt PI control to track the frequency and voltage reference values;

[0171] (2) Define the controller operation mode and basic events;

[0172] Defining the controller operation mode and basic events includes the following steps:

[0173] A1) Set F v = 1 to indicate enabling adaptive virtual impedance control; F G = 1 indicates that the DG operates in the droop control mode based on the generalized washout filter, F G = 0 indicates that the DG operates in the traditional droop control mode; F t = 1 indicates the transient operation of the islanded microgrid, F t = 0 indicates the steady-state operation of the islanded microgrid;

[0174] The active power change rate and the reactive power change rate are input to a low-pass filter with a cut-off frequency of 10, or are output after low-pass filtering;

[0175] If it exceeds the defined threshold ξ PQ , it is determined that the islanded microgrid is in transient operation, that is, F t = 1; and The outputs after low-pass filtering are both less than the threshold, and it is determined that the islanded microgrid is in steady-state operation, that is, F t = 0;

[0176] A2) Set event A to record the current local output active power P G,ave, this value is recorded by the sample and hold module, and the active power control mode is switched to the traditional droop control mode, F G = 0;

[0177] Event B is defined as recording and holding the current active power P D,ave , and enabling the proposed adaptive virtual impedance control;

[0178] Event C is defined as switching the active power control mode to the droop control mode based on the generalized washout filter, F G = 1.

[0179] (3) The droop control based on the generalized washout filter and the adaptive virtual impedance are used to realize the decentralized secondary control of the islanded microgrid. The process of realizing the decentralized secondary control of the islanded microgrid by the droop control based on the generalized washout filter and the adaptive virtual impedance is as Figure 3 shown, and specifically includes the following steps:

[0180] B1) When the load of the islanded microgrid changes and the system enters the transient state, that is, when F t = 1, each distributed generator operates in the droop control mode based on the washout filter to achieve decentralized secondary control;

[0181] B2) After monitoring that the system reaches the steady state, that is, when F t = 0, event A is triggered, and the sample and hold module records the active power P G,ave , and the active power control mode is switched to the droop control mode, that is, when F G = 0, the reactive power control mode continues to adopt the mode based on the generalized washout filter;

[0182] B3) After monitoring that the active power is stable, the active power value at this time is recorded as the expected value P D,ave , event B is triggered, and the proposed adaptive virtual impedance control is enabled, that is, when F v = 1, and wait for the system to enter the steady state again;

[0183] B4) After monitoring that the system re-enters the steady state, event C is triggered, and the active power control mode is switched to the droop control mode based on the generalized washout filter, that is, when F G = 1, to achieve unbiased power and frequency control, and finally realize the decentralized secondary control of the islanded microgrid based on the generalized washout filter and the adaptive virtual impedance.

[0184] To verify the correctness of the proposed control, based on the MatLab / Simulink software, an islanded microgrid simulation model with three distributed DGs as shown in Figure 4 is established, and the main parameters of the islanded microgrid are shown in Table 1.

[0185] Table 1 Parameters of an islanded microgrid with two DGs in parallel supplying a common load

[0186]

[0187] The simulation results are as Figure 5 shown. When the rated powers of all DG units in the islanded microgrid are the same, the system first operates in the GWF droop control mode, and the power increases rapidly. The system enters the steady state at t = 1.1 s and actively switches to the droop control mode, achieving accurate sharing of active power. The proposed adaptive virtual impedance (AVI) control is triggered at t = 1.82 s, and the reactive power sharing ability is improved. The proposed control actively switches back to the GWF droop control mode at t = 2.75 s, achieving unbiased frequency control. Figure 5 (a) and (b) clearly reveal that, under the proposed control, the AVI generated by the active mode switching of the DG in the islanded microgrid compensates for the influence of line impedance mismatch on the power sharing performance, achieving a significant improvement in the active power and reactive power sharing performance; as Figure 5 (c) shows, when the load changes, the frequency fluctuation is small, and the frequency can be restored to the nominal value, achieving decentralized secondary control; Figure 5 (d) shows the output voltage of each DG. It can be seen that the control method proposed in the present invention can effectively restore the voltage.

[0188] From the simulation results, it can be obtained that a non - communication secondary control method for an islanded microgrid based on a generalized washout filter and adaptive virtual impedance proposed in the present invention effectively solves the problems of poor power sharing ability, large frequency deviation, and strong dependence on communication when applying traditional control methods in an islanded microgrid with mismatched line impedances.

[0189] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above - mentioned embodiments. What is described in the above - mentioned embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A decentralized secondary control method for an island microgrid based on a generalized washout filter and adaptive virtual impedance, characterized in that: The following steps are involved: 11) Establish a microgrid droop control model based on generalized flushing filter; 12) Obtaining the steady-state output power error of the distributed power source of the microgrid; 13) Generate adaptive virtual impedance based on active mode switching; 14) Conduct decentralized secondary control of isolated microgrid.

2. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 1 is characterized in that: The method of establishing a microgrid droop control model based on a generalized flushing filter is as follows: using a droop control equation to design a microgrid droop control model based on a generalized flushing filter; the method comprises the following steps: 21) Assume that two DGs are connected to the common coupling point PCC and the line impedance is not matched, then: DG i The output active power and reactive power are expressed as: Where: P i is the active power output by the i-th DG; Q i is the reactive power output by the i-th DG; V i Indicates DG i Output voltage amplitude; V L is the PCC voltage amplitude; δ i is the output voltage V of the i-th DG i and PCC voltage V L The phase angle difference between i is the resistance of the ith line impedance; X i is the inductance of the ith line impedance; Assume δ i The value is extremely small, and there is sinδ i ≈δ i ,cosδ i ≈1, simplifying equations (1) and (2) to: Among them, the active power P i and phase angle difference δ i Proportional to reactive power Q i With the output voltage amplitude V i related; Calculate DG by the droop control equation i The frequency and amplitude of the output voltage are expressed as: oh i =ω0-m i (P i -P 0,i ) (5) V i =V0-n i (Q i -Q 0,i ) (6) Where: i represents the output angular frequency of the i-th DG; V i represents the output voltage of the i-th DG; ω0 and V0 are the nominal angular frequency and nominal output voltage amplitude respectively; m i and n i are the droop coefficients of the frequency droop controller and voltage droop controller of the i-th DG respectively; P i is the output power of the i-th DG; Q i is the output power of the i-th DG; P 0,i and Q 0,i are the nominal active power and nominal reactive power of the i-th DG respectively; 22) A microgrid droop control model based on generalized flushing filter is established, and its expression is as follows: oh i =ω0-m i ·G GWF (s)(P i -P 0,i ) (7) V i =V0-n i ·G GWF (s)(Q i -Q 0,i ) (8) Where: G GWF (s) represents the transfer function of the generalized washout filter; Where: l is the low-pass filter cutoff frequency; ω h is the cutoff frequency of the flushing filter, and s represents the complex frequency domain.

3. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 1 is characterized in that: The method of obtaining the microgrid DG steady-state output power error is as follows: based on the dynamic vector method, a DG output power transfer function based on the generalized flushing filter droop control is derived, and the terminal value theorem is used to obtain the DG steady-state output power based on the generalized flushing filter droop control, so as to obtain the microgrid DG steady-state output power error; the method comprises the following steps: 31) Derivation of distributed generator DG based on dynamic phasor model i , i=1,2, output active power and reactive power: Where: P i is the active power output by the i-th DG; Q i is the reactive power output by the i-th DG; V i represents the output voltage amplitude of the i-th DG; V L is the PCC voltage amplitude; δ i is the output voltage V of the i-th DG i and PCC voltage V L The phase difference between i is the resistance of the ith line impedance; X i is the inductance of the ith line impedance, s represents the complex frequency domain; Ignoring line losses, we have: Where: P L With Q L is the load active power and reactive power; 32) Linearize equations (10) and (11) and use the final value theorem to obtain the DG steady-state output power based on the droop control of the generalized flushing filter: Where: P ss i is the steady-state active power output by the i-th DG; Q ss i is the steady-state reactive power output by the i-th DG; X1 and X2 are the line inductances of the two transmission lines respectively; P L With Q L are the load active power and reactive power respectively; m1 and m2 are the droop coefficients of the two DG frequency droop controllers respectively; V L is the PCC voltage amplitude; ω h is the cutoff frequency of the flush filter; ω l is the low-pass filter cutoff frequency; In order to achieve accurate power sharing, the droop coefficient relationship of multiple parallel DGs is: in: is the expected value of active and reactive power of the i-th DG, m i and n i are the droop coefficients of the frequency droop controller and voltage droop controller of the i-th DG respectively; 33) Obtain the steady-state output power error of the microgrid DG: Define the active power sharing error as P erri , the reactive power sharing error is Q erri , and there are: Where: P erri is the active power sharing error of the i-th DG, Q erri is the reactive power sharing error of the i-th DG; P i is the output power of the i-th DG; Q i is the output power of the i-th DG; Extended to the isolated microgrid with multiple DGs in parallel, the DG active power sharing error ΔP based on the generalized flushing filter droop control is adopted. erri for: Where: ΔP erri is the active power sharing error of the i-th DG: m ∑ It represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ Represents the sum of the impedances of each line; V L is the PCC line voltage amplitude.

4. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 1 is characterized in that: The method of generating an adaptive virtual impedance based on active mode switching is as follows: operating the controller in a droop control mode based on a generalized flushing filter, sampling a power value, and then switching to a droop control mode to obtain a power sharing error, and adaptively generating a virtual impedance using a derived virtual impedance generation formula; the method comprises the following steps: 41) Design the virtual impedance as: Where: X vi The virtual impedance designed for the i-th DG; represents the expected value of the line impedance of the i-th DG; X i is the actual line impedance of the i-th DG; The constraint condition of setting the total increment of virtual impedance to 0 is proposed to keep the total line impedance unchanged, which satisfies: From equation (19) and equation (20), we can get: in: is the sum of the expected values ​​of line impedance; X ∑ is the sum of actual line impedances; 42) When k DGs with different capacities are connected in parallel, the sufficient condition for achieving reasonable power distribution is: in: represents the expected value of the line impedance of the i-th DG; m i is the droop coefficient of the frequency droop controller of the i-th DG; Substituting equation (19) and equation (22) into equation (21), we obtain: Where: X vi is the generated virtual impedance; m ∑ It represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ Represents the sum of the impedances of each line; m i is the droop coefficient of the frequency droop controller of the i-th DG; X i is the actual line impedance of the i-th DG; Substituting equation (18) into equation (23) generates the virtual impedance: Where: X vi is the generated virtual impedance; m ∑ It represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ Represents the sum of the impedances of each line; ΔP erri is the active power sharing error of the i-th DG; V L is the PCC line voltage amplitude; 43) For the droop control based on the generalized flushing filter, after active mode switching, it is switched to the droop control to generate ΔP erri , that is: Where: P i is the output power of the i-th DG; is the expected value of active power of the ith DG; P G.ave and P D.ave are the steady-state values ​​of active power in the droop control mode based on generalized flushing filter and the traditional droop mode respectively; P L is the load active power; Taking into account the influence of line resistance, the compensation coefficient K is introduced to generate an adaptive virtual impedance X taking into account the influence of line resistance. vi , and there are: Where: K is the compensation coefficient; X vi is the generated virtual impedance; m ∑ It represents the sum of the droop coefficients of the frequency droop controllers of each DG; X ∑ Represents the sum of the impedances of each line; ΔP erri is the active power sharing error of the i-th DG; V L is the PCC line voltage amplitude.

5. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 1 is characterized in that: The decentralized secondary control of the isolated microgrid is as follows: each DG unit samples and records the active power based on the droop control of the generalized flushing filter; then switches the controller to the droop control mode, and after stabilization, the active power value at this time is recorded as the expected value P D,ave ; Enable the proposed adaptive virtual impedance control to enter a new steady state; then switch the active power control mode to the droop control mode of the generalized flushing filter to achieve unbiased frequency control, that is, ultimately achieve power sharing among DGs and frequency recovery; including the following steps: 51) Based on active mode switching, the active power sharing error ΔP is obtained erri , and then generate the adaptive virtual impedance X v , the d-axis and q-axis reference voltages of the DG controlled by adaptive virtual impedance are: Where: and are d-axis reference voltage and q-axis reference voltage respectively; V d With V q are the actual voltage of d-axis and q-axis respectively; X v is the generated virtual impedance; I od and I oq is the dq axis output current of DG; and As the voltage outer loop reference value in the voltage-current double closed loop system, where both the voltage outer loop and the current inner loop adopt PI control to track the frequency and voltage reference values; 52) Define controller operation modes and basic events; 53) Decentralized secondary control of isolated island microgrid based on droop control and adaptive virtual impedance of generalized flushing filter.

6. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 5 is characterized in that: Defining the controller operation mode and basic events includes the following steps: 61) Set F v =1 means enabling adaptive virtual impedance control; F G =1 means that DG operates in the droop control mode based on the generalized flushing filter, F G =0 means DG operates in the traditional droop control mode; F t =1 indicates the island microgrid is in transient operation, F t =0 indicates that the island microgrid operates in a steady state; The active power change rate and the rate of change of reactive power Input to a low-pass filter with a cutoff frequency of 10. or Output after low-pass filtering; If the defined threshold ξ is exceeded PQ , then the isolated microgrid is judged to be in transient operation, that is, F t =1; and The outputs after low-pass filtering are all less than the threshold, and the isolated microgrid is judged to be in steady-state operation, that is, F t =0; 62) Set event A to record the current local output active power P G,ave , the value is recorded by the sampling and holding module, and the active power control mode is switched to the traditional droop control mode, F G =0; Event B is defined as recording and maintaining the current active power P D,ave ,enabling the proposed adaptive virtual impedance control; Event C is defined as switching the active power control mode to the droop control mode based on the generalized flushing filter, F G =1.

7. The decentralized secondary control method for isolated island microgrid based on generalized flushing filter and adaptive virtual impedance according to claim 5 is characterized in that: The method for realizing decentralized secondary control of an island microgrid based on droop control and adaptive virtual impedance of a generalized flushing filter comprises the following steps: 71) When the load of the isolated microgrid changes, the system enters a transient state, i.e., F t =1, each distributed generator set operates in the droop control mode based on the flushing filter to achieve decentralized secondary control; 72) After the system reaches a steady state, F t =0, trigger event A, the sample and hold module records the active power P G,ave , and the active power control mode is switched to the droop control mode, that is, F G =0, the reactive power control mode continues to adopt the generalized flushing filter mode; 73) After the active power is monitored to be stable, the active power value at this time is recorded as the expected value P D,ave , triggering event B, enabling the proposed adaptive virtual impedance control, i.e., F v =1, waiting for the system to re-enter the steady state; 74) After the system is detected to have re-entered the steady state, event C is triggered, and the active power control mode is switched to the droop control mode based on the generalized flushing filter, that is, F G =1, achieving unbiased power and frequency control, and finally realizing decentralized secondary control of island microgrid based on droop control of generalized flushing filter and adaptive virtual impedance.

Images