Virtual inertia control and controller parameter setting method based on fractional derivative

Abstract

The invention discloses a fractional derivative-based virtual inertia control and controller parameter setting method, which comprises the following steps of: obtaining a characteristic equation of a system according to a transfer function from system interference to system frequency response; obtaining a real root boundary curve RRB and a multiple root boundary curve CRB according to a characteristic equation of the system; dividing the virtual inertia-virtual damping plane according to the boundary curve to obtain a stable region in which the system is kept stable in the parameter plane; obtaining an optimal control parameter of a fractional order virtual controller FOVIC according to the stable region; the controller output reference power is obtained according to the optimal control parameters; and according to the controller output reference power, the active power adjustment amount of the primary frequency modulation of the battery energy storage system is obtained through inner loop control of a GFL power converter. According to the method, the frequency deviation can be better suppressed, the robustness of the system is enhanced, and a quick frequency regulation service is provided for a low-inertia power grid integrated with renewable energy sources.

Description

Technical Field

[0001] The present invention belongs to the field of electrical engineering, and more specifically, relates to a virtual inertia control and controller parameter setting method based on fractional-order derivatives. Background Art

[0002] With the large-scale access of new energy represented by wind power and photovoltaic power generation to the power grid, power electronic equipment has penetrated rapidly in the power system, gradually replacing traditional synchronous generators, which has greatly reduced the rotational inertia and frequency support of the power system. However, the battery energy storage system (BESS) can provide fast response and is suitable for short-term frequency regulation services, that is, it can provide inertial response and primary frequency regulation, so the battery energy storage system has received great attention from academia and the academic community. An effective control scheme to provide inertia and frequency support for BESS is to implement virtual inertia control (VIC) on the converter and adopt a grid-following (GFL) converter. For GFL converters, most use differential as the VIC algorithm to simulate mechanical inertia.

[0003] The order of differentiation or integration in the fractional-order controller (FOC) is a decimal. Compared with the traditional integer-order differential equation, the fractional-order system model can more accurately describe the dynamic response of the actual engineering system, improve the ability to describe, analyze and control the dynamic performance of the system, and has excellent performance in frequency regulation performance and robustness compared to the traditional integer-order control (IOC). However, for the sake of tuning efficiency, a suitable parameter tuning method has not yet been explored for the fractional-order controller. Most of the existing tuning methods for fractional-order calculus controllers are to determine the order of the calculus link in advance according to experience and then perform parameter tuning according to the integer-order controller tuning method. Although the control effect can be better than that of the integer-order controller, the robustness of the existing methods is not high. Summary of the invention

[0004] The present invention provides a virtual inertia control and controller parameter tuning method based on fractional-order derivatives, which is used to solve the technical problems that the existing low-inertia power system lacks inertia and frequency support and the fractional-order controller (FOC) parameter tuning method has low robustness.

[0005] The technical solution of the present invention to solve the above technical problem is as follows: a virtual inertia control and controller parameter setting method based on fractional-order derivatives, comprising the following steps:

[0006] Step 1: Obtain the characteristic equation of the system according to the transfer function from system interference to system frequency response;

[0007] Step 2: According to the characteristic equation of the system, transform and set to zero to obtain the real root boundary curve RRB and the complex root boundary curve CRB;

[0008] Step 3: dividing the virtual inertia-virtual damping plane according to the real root boundary curve RRB and the complex root boundary curve CRB, and obtaining a stable region where the system remains stable in the parameter plane through a stability criterion of the fractional order system;

[0009] Step 4: According to the stable region, the optimal control parameters of the fractional order virtual controller FOVIC are obtained through a parameter optimization model and a particle swarm solving algorithm;

[0010] Step 5: According to the optimal control parameters, the controller output reference power is obtained through fractional-order virtual inertia control;

[0011] Step 6: According to the reference power output by the controller, the active power adjustment amount of the primary frequency modulation of the battery energy storage system is obtained through the inner loop control of the GFL power converter.

[0012] Furthermore, the characteristic equation of the system is the denominator part of the expression of the transfer function of the system interference to the system frequency response.

[0013] Furthermore, the step 2 comprises:

[0014] According to the characteristic equation of the system, after frequency domain transformation and Euler transformation, when the frequency ranges from 0 to positive infinity, a set of controller parameters satisfying the characteristic equation whose real part and imaginary part are both 0; wherein, the set of controller parameters when the frequency is equal to 0 constitutes the real root boundary curve RRB, and the set of controller parameters when the frequency is greater than 0 constitutes the complex root boundary curve CRB.

[0015] Furthermore, the step 3 comprises:

[0016] The virtual inertia-virtual damping parameter plane is divided into a number of closed areas according to the real root boundary curve RRB and the complex root boundary curve CRB. According to the stability criterion based on parameter sensitivity, it is judged whether each closed area meets the stability condition, so as to obtain the stable area where the system remains stable in the parameter plane.

[0017] Furthermore, the stability criterion based on parameter sensitivity is as follows:

[0018] If the following equation is satisfied, the system is unstable:

[0019] S X (w * ,k * )>0

[0020] Among them, k * is a point on the SBL curve, X∈{J,D}, S J and S D Respectively represent Jv and D v Root sensitivity.

[0021] Further, the step 4 comprises:

[0022] Taking the controller parameters in the stable region as variables, the frequency deviation, output power deviation and the distance of the parameter setting point from the center of the stable region are considered to build a parameter optimization model through weighted summation. The particle swarm algorithm PSO is used to solve the parameter optimization model and obtain the optimal control parameters of the fractional-order virtual controller FOVIC.

[0023] Furthermore, the controller parameters include virtual inertia, virtual damping, and fractional order.

[0024] Further, the step 5 comprises:

[0025] The tuning is based on the optimal control parameters and the frequency change of the system. The virtual damping parameter is used as the proportional coefficient and the virtual inertia parameter is used as the differential coefficient. The virtual inertia control is a proportional-differential controller, thereby obtaining the output reference power of the fractional-order virtual controller FOVIC.

[0026] Further, the step 6 comprises:

[0027] According to the reference power output by the controller, through the inner loop control of the GFL power converter, the GFL power converter is represented by an inertia link, and the output is limited to obtain the active power adjustment amount of the primary frequency modulation of the battery energy storage system.

[0028] The present invention adopts fractional-order virtual inertial control (FOVIC) and optimizes FOVIC control parameters to obtain optimal controller parameters by comprehensively considering frequency deviation, active power deviation and distance from the center of the stable area, which can improve the system frequency regulation performance and system robustness and provide inertia and frequency support for the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of a virtual inertia control and controller parameter tuning method based on fractional-order derivatives provided in an embodiment of the present invention;

[0030] Figure 2 A schematic diagram of a power system model provided by an embodiment of the present invention;

[0031] Figure 3 for Figure 2 Schematic diagram of the corresponding battery energy storage system model;

[0032] Figure 4 for Figure 3 Schematic diagram of the corresponding fractional-order virtual inertial control topology;

[0033] Figure 5 The SBL curve of the studied system when v = 0.1 and τ = 0.1s;

[0034] Figure 6 It is a schematic diagram of the symbol representation near the inflection points a and b of the SBL curve;

[0035] Figure 7 is the system frequency response curve affected by 0.01pu load disturbance;

[0036] Figure 8 When v changes, J v -D v Stable regions on a plane;

[0037] Fig. 9 is the frequency response of the power system under different optimized controllers under 0.01pu load disturbance;

[0038] Fig.10 (a) is the curve of domestic load, solar power generation and their sum, (b) is the frequency response of the power system under different optimization controllers;

[0039] Fig.11 The frequency response of the power system under different parameter disturbances. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0041] like Figure 1 As shown, the present invention provides a virtual inertia control and controller parameter tuning method based on fractional-order derivatives, the method comprising the following steps:

[0042] Step 1: Obtain the characteristic equation of the system according to the transfer function from system interference to system frequency response;

[0043] Step 2: According to the characteristic equation of the system, transform and set to zero to obtain the real root boundary curve RRB and the complex root boundary curve CRB;

[0044] Step 3: dividing the virtual inertia-virtual damping plane according to the real root boundary curve RRB and the complex root boundary curve CRB, and obtaining a stable region where the system remains stable in the parameter plane through a stability criterion of the fractional order system;

[0045] Step 4: According to the stable region, the optimal control parameters of the fractional order virtual controller FOVIC are obtained through a parameter optimization model and a particle swarm solving algorithm;

[0046] Step 5: According to the optimal control parameters, the controller output reference power is obtained through fractional-order virtual inertia control;

[0047] Step 6: According to the reference power output by the controller, the active power adjustment amount of the primary frequency modulation of the battery energy storage system is obtained through the inner loop control of the GFL power converter.

[0048] Embodiment 1

[0049] This example takes a low-inertia power system with battery energy storage as an example. Figure 2 As shown, the power system studied includes traditional power plants such as reheat thermal power plants and hydroelectric power plants, including 550 MW thermal power plants, 250 MW hydroelectric power plants, and 200 MW photovoltaic power plants. The battery energy storage system in the system is 100 MW.

[0050] Among them, LFC is a load frequency controller, which is used for the secondary frequency modulation of the system and outputs the frequency deviation after the secondary frequency modulation; G c (s) is the transfer function of a conventional generator set, including the governor, steam turbine, and water turbine; G is the active unbalance caused by the disturbance, including the fluctuation of photovoltaic power generation and load fluctuation; ESS (s) is the transfer function of the battery energy storage system, which injects the required active power of primary frequency regulation into the system; is the inertial damping response of the power system; Δf is the frequency deviation of the primary frequency modulation of the system.

[0051] For the above system structure and method structure, the control target of the virtual inertia control and controller parameter tuning method based on fractional-order derivatives provided in this embodiment is to use the fractional-order virtual inertia controller FOVIC and optimize its control parameters (J v , D v , v), providing active power for the primary frequency regulation of BESS makes FOVIC show excellent robustness in the frequency regulation of low-inertia systems with large uncertainty in system parameters compared with integer-order virtual inertial controller (IOVIC). At the same time, FOVIC has a smaller frequency deviation during load disturbance than IOVIC.

[0052] The battery energy storage system BESS architecture used in this example is as follows Figure 3As shown in the figure, BESS takes the system frequency change as input, and outputs the active power adjustment provided by the battery energy storage system BESS participating in the primary frequency regulation of the system through the communication delay link, phase-locked loop PLL, fractional-order virtual inertia controller and inner loop control.

[0053] where e -τs Represents the communication delay link, and the phase-locked loop PLL is represented by a second-order transfer function:

[0054]

[0055] Where:

[0056] The transfer function of the fractional-order virtual inertia controller is expressed as:

[0057] G VIC (s) = J V s v +D V .

[0058] Among them, J v is the virtual inertia parameter, D v is the virtual damping parameter and v is the fractional order.

[0059] The transfer function of the battery energy storage system BESS is expressed as:

[0060]

[0061] Among them, T ESS is the first-order filter time constant, e -τs is the delayed communication link, ζ is the system damping ratio, ω n is the natural frequency of the system, G VIC is the transfer function of the fractional-order virtual inertia controller.

[0062] The virtual inertia control based on fractional derivatives used in this embodiment is as follows: Figure 4 The specific control process is that the frequency change of the power system is obtained through a proportional differential PD link to obtain the reference active power of the BESS participating in the primary frequency regulation, where the proportional coefficient is the virtual damping and the differential coefficient is the virtual inertia.

[0063] A virtual inertia control and controller parameter tuning method based on fractional-order derivatives provided by the present invention is described in detail below.

[0064] Step 1 includes: obtaining a characteristic equation of a system disturbance to a system frequency response.

[0065] The transfer function of each element in the system under study is obtained, and the transfer function of the system interference to the system frequency response can be obtained as:

[0066]

[0067] in, is the system disturbance, Δf is the system frequency change, M is the system inertia time constant, D is the load damping coefficient, G ESS is the transfer function of the battery energy storage system BESS, G c is the transfer function of a conventional generator set.

[0068] After commonizing the denominator of the transfer function and integrating, we can obtain the characteristic equation of the system as shown below:

[0069] Δ(s)=P(s)+Q(s)(J V s v +D V ) -sτ

[0070] P(s)=p0+p1s+p2s 2 +p3s 3 +p4s 4 +p5s 5 +

[0071] p6s 6 +p7s 7 +p8s 8 +p9s 9 +p 10 s 10

[0072] Q(s)=q0+q1s+q2s 2 +q3s 3 +q4s 4 +q5s 5 +

[0073] q6s 6 +q7s 7

[0074] Where Δ(s) represents the characteristic equation, P(s) and Q(s) represent the active power and reactive power, and J V is the virtual inertia, D V is the damping coefficient, v is the fractional order, e -sτ It is the communication delay link.

[0075] Step 2 includes: obtaining stable boundary curves, including: real root boundary curve (RRB) and complex root boundary curve (CRB).

[0076] The controller parameters include: virtual inertia parameter J v , virtual damping parameter D v , fractional order v.

[0077] By transforming the characteristic equation and setting it to 0, we can obtain:

[0078] Δ(ω,J V ,D V )=P(jω)+Q(jω)[J V (jω) v +D V ]e -jωτ =0

[0079] Among them, Δ(ω,J V ,D V ) represents the characteristic equation, P(jω) and Q(jω) represent the active power and reactive power, J V is the virtual inertia, D V is the damping coefficient, v is the fractional order, e -jωτ It is the communication delay link.

[0080] The SBL curve is actually the curve that satisfies the above equation (J v ,D v ) is a collection of v -D v The curve on the plane can be transformed by Euler's formula to obtain:

[0081] e -jωτ =cos(ωτ)-jsin(ωτ)

[0082] Δ(ω,J V ,D V ,v)=J V A1(ω,v)+D V B1(ω,v)+P1(ω)+

[0083] j(J V A2(ω,v)+D V B2(ω,v)+P2(ω))

[0084] Where:

[0085]

[0086] B1(ω,v)=Q1cos(ωτ)+Q2sin(ωτ)

[0087] B2(ω,v)=Q2cos(ωτ)-Q1sin(ωτ)

[0088] P1(ω)=p0-p2ω 2 +p4ω 4 -p6ω 6+p8ω 8 -p 10 ω 10

[0089] P2(ω)=p1ω-p3ω 3 +p5ω 5 -p7ω 7 +p9ω 9

[0090] Q1(ω)=q0-q2ω 2 +q4ω 4 -q6ω 6

[0091] Q2(ω)=q1ω-q3ω 3 +q5ω 5 -q7ω 7 .

[0092] Among them, A1(ω,v) and A2(ω,v) are coefficients before virtual inertia, B1(ω,v) and B2(ω,v) are coefficients before virtual inertia, and ω represents the angular frequency.

[0093] By reducing the real and imaginary parts of the characteristic equation to 0, we can obtain:

[0094]

[0095] Among them, J V is the virtual inertia, D V is the damping coefficient.

[0096] At a fixed fractional order v, the characteristic equation in J v -D v There are two stable boundaries on the plane, namely the real root boundary (RRB) and the complex root boundary (CRB). When ω ranges from 0 to positive infinity, the RRB and CRB can be plotted. Figure 4 Shown is the SBL curve of the studied system when v = 0.1 and τ = 0.1s.

[0097] Step 3 includes: obtaining a stable region where the system remains stable in the parameter plane.

[0098] like Figure 5 As shown, J v -D v The plane is divided into many parts by RRB and CRB, including N0, N1, N2, etc. In order to determine the stable region where the system remains stable in the parameter plane, the present invention introduces two fractional-order stability criteria.

[0099] Theorem 1: For a system described by a fractional-order quasipolynomial:

[0100] Δ(ω,JV ,D V )=P(jω)+Q(jω)[J V (jω) v +D V ]e -jωτ

[0101] If and only if the inversion equation: w=s v All zeros of w * It is stable only if the following equation is satisfied:

[0102]

[0103] Among them, arg(w * ) represents the angle corresponding to the zero point, and v is the order of the fractional order.

[0104] The meaning of Theorem 1 is as follows Figure 6 As shown, the system is stable when all zeros of the transformed characteristic equation are located in the white area.

[0105] Theorem 2: If the following equation is satisfied, the system is unstable.

[0106] S X (w * ,k * )>0

[0107] Among them, k * is a point on the SBL curve, X∈{J,D}, S J and S D Respectively represent J v and D v Root sensitivity.

[0108] S X (w * , k * ) is defined as:

[0109]

[0110] Where:

[0111]

[0112] Theorem 2 determines the stability of the system by the sensitivity to the controller parameters. If the root crosses the imaginary axis from the left half plane to the right half plane of the complex plane, or if Theorem 2 is satisfied, the system will be unstable.

[0113] By using Theorem 1 and Theorem 2, we can determine J v -D v Stable regions in each region of the plane divided by the SBL curve.

[0114] Step 4 includes: obtaining optimal control parameters of the fractional order virtual controller (FOVIC).

[0115] The parameters of the fractional-order controller have a significant impact on the effect of virtual inertia and damping. v It can reduce the median frequency and deviation during transients, thereby improving dynamic performance without excessively increasing J v The settling performance will be reduced. Virtual damping D v It can improve the settling performance and slightly reduce the transient frequency deviation, but too large D v The value of V will lead to large quadratic overshoot. In addition, the fractional order v will also affect the performance of FOVIC. Therefore, a comprehensive optimization model is needed to fine-tune the controller parameters (J v , D v 、v).

[0116] Since the purpose of VIC is to suppress frequency deviation, the objective function of the optimization model should include the frequency deviation multiplied by a high weight coefficient. A widely used performance indicator is the integrated squared error (ISE).

[0117]

[0118] Wherein, Δf represents the system frequency deviation, and T represents the integration time.

[0119] The time scale of inertia and main frequency response is usually within 30 seconds, so the integration time T is set to 30 seconds.

[0120] Generally speaking, the output power of the BESS is also considered in the objective function in the form of the integrated square of the control output deviation (ISDCO), which is defined as:

[0121]

[0122] Among them, ΔP BESS represents the output active power of the battery energy storage system, and T represents the integration time.

[0123] Usually, in order to obtain a lower ISE, the controller parameter (J v , D v ) is high and is located near the boundary of the stable region or SBL curve. When the system parameters change significantly, such (J v , D v ) value is likely to fall outside the stable region, thus posing a serious threat to the robustness of the system. From a geometric point of view, when the system parameters change, the point that is most likely to remain in the stable region is the geometric center of the stable region. Therefore, from the perspective of improving the robustness of the system, it is best to set (Jv , D v ) is set at the center of the stable region. To avoid a decrease in robustness, the parameter setting point (J v , D v ) to the center of the stable region should also be included in the objective function, and d is given by the following formula:

[0124]

[0125] Among them, (J v , D v ) is the controller parameter setting point, (C J , C D ) is the geometric center point of the stable region, and d is the distance from the parameter setting point to the center of the stable region.

[0126] Figure 5 The SBL curve shown is actually composed of n vertices (J Vj , D Vj ) between the line segments, j = 0 to n-1. Therefore, the center point ( Figure 5 The yellow star in the figure can be calculated using the polygon center point formula:

[0127]

[0128] in, (C J , C D ) is the geometric center point of the stable region.

[0129] In summary, the controller parameter optimization model of FOVIC can be obtained as follows:

[0130]

[0131] 0<v<1.5

[0132] Among them, ω1, ω2, ω3 are weight coefficients, for (J Vj , D Vj ) in the stable region under fractional order v. The optimization model is solved by particle swarm optimization algorithm (PSO).

[0133] Step 5 includes: obtaining a reference power output by the controller.

[0134] like Figure 4 As shown, the system frequency is controlled by a proportional differential PD composed of control parameters to obtain an active power reference value provided by the battery energy storage system participating in the primary frequency regulation.

[0135] Step 6 includes: obtaining the active power adjustment amount of the battery energy storage system participating in primary frequency regulation.

[0136] like Figure 3 As shown in FIG. 1 , the active power reference value output by the FOVIC controller is obtained after the inner loop control to obtain the active power provided by the battery energy storage system participating in the primary frequency modulation.

[0137] This example uses a virtual inertia control and controller parameter tuning method based on fractional-order derivatives. By adding a fractional-order virtual controller to the battery energy storage system and optimizing the controller parameters, the low-inertia power system can reduce the frequency drop in the case of load disturbance, improve the system's frequency regulation capability, and enhance the robustness of frequency regulation of low-inertia systems with large uncertainties in system parameters.

[0138] The virtual inertia controller parameter tuning method based on fractional-order derivatives provided in the embodiment is simulated, analyzed and verified in a low-inertia power system. Figure 5 As shown in the figure, under the condition of v = 0.1, four points are selected on the SBL curve of the characteristic equation, which are denoted as k (1) , k (2) , k (3) and k (4) By using the above-mentioned Theorem 2 of the stability criterion, we can obtain the J corresponding to each point as shown in Table 1. v , D v ,ω,S J , S D The value of S J , S D The positive or negative value of J v -D v The stability of each area divided in the plane. As can be seen from Table 1, k (1) S J and S D are all positive, indicating that one root of the transformed characteristic equation will enter the unstable region, so k (1) The upper right region (N2) is unstable. Similarly, we can conclude that regions N0, N1, and N4 are unstable, while the shaded region N3 is stable because k (2) S J and S D They are all negative values.

[0139] like Figure 6 Shown is S J and S D The symbol expression near the inflection points a and b of the SBL curve. Figure 6 As can be seen in (b), as J v The increase of S J The sign at the inflection point a changes from negative to positive. This indicates that the area above the lower part and below the upper part of the SBL curve is stable. Figure 6 As can be seen in (c), the area above the inflection point b is stable. Therefore, it is mathematically verified that the N3 region is a stable region.

[0140] like Figure 7 The figure shows the frequency response curve of the system under a load disturbance of 0.01pu at 5s when the controller parameters corresponding to the points located inside, outside and at the boundary of N3 are used respectively. It can be seen from the figure that the frequency response corresponding to the points located inside is constantly converging, the frequency response corresponding to the points located outside is constantly diverging, and the frequency response corresponding to the points located at the boundary is a constant amplitude oscillation. Therefore, from the perspective of simulation, it is verified again that the N3 region is a stable region.

[0141] like Figure 8 The figure shows the J of the low inertia system under any fractional order v∈(0,1.5). v -D v The stable region R(v) on the plane. The space enclosed by the three-dimensional surface and the RRB boundary line in the figure is the stable region under different fractional orders.

[0142] The virtual inertia control strategy based on fractional order derivatives provided in the embodiment is compared with the traditional virtual inertia control method based on integer order derivatives. Fig. 9 As shown in the figure, there is a load step change of 0.01pu at 5s, and the system frequency response diagram is without virtual inertia controller, with fractional order virtual inertia controller and with integer order virtual inertia controller. The simulation results show that the low inertia power system without virtual inertia controller VIC will have a considerable frequency deviation (soil 0.1Hz). When the BESS with VIC controller is connected to the grid, the frequency deviation will be significantly reduced. Compared with integer order controller IOVIC (soil 0.028Hz), FOVIC controller increases the lowest point of frequency drop (soil 0.023Hz), thereby reducing the rate of change of frequency (ROCOF) and providing better frequency support.

[0143] like Fig.10 (a) shows the fluctuations of photovoltaic power generation and load and the external disturbance of the system (i.e. the sum of photovoltaic power generation fluctuations and load changes). Fig.10 (b) shows the frequency response curve of the system under FOVIC and IOVIC when external disturbance is injected. It can be seen from the figure that the maximum frequency deviation of the system under IOVIC and FOVIC is ±0.150Hz and ±0.103Hz respectively. Obviously, under the action of FOVIC, the frequency variation is significantly reduced and the frequency deviation is also well alleviated.

[0144] like Fig.11The figure shows the power system parameters (M, D, T) under FOVIC control in BESS. ESS Table 2 is the objective function J value of the FOVIC controller parameter optimization model under different system parameter disturbances. Figure 8 It can be seen from Table 2 that in most cases, the J value and frequency response performance of FOVIC are better than those of IOVIC, that is, FOVIC has excellent robustness in the frequency regulation of low inertia systems with large uncertainties in system parameters. It is worth noting that when τ decreases by 50%, the J value of IOVIC exceeds that of FOVIC, which may be due to the fact that the component d in the objective function increases sharply with the expansion of the stable region, and its center point also moves away from the set parameters.

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A virtual inertia control and controller parameter tuning method based on fractional derivatives, characterized in that: The steps include: Step 1: Obtain the characteristic equation of the system according to the transfer function from system interference to system frequency response; Step 2: According to the characteristic equation of the system, transform and set to zero to obtain the real root boundary curve RRB and the complex root boundary curve CRB; Step 3: dividing the virtual inertia-virtual damping plane according to the real root boundary curve RRB and the complex root boundary curve CRB, and obtaining a stable region where the system remains stable in the parameter plane through a stability criterion of the fractional order system; Step 4: According to the stable region, the optimal control parameters of the fractional order virtual controller FOVIC are obtained through a parameter optimization model and a particle swarm solving algorithm; Step 5: According to the optimal control parameters, the controller output reference power is obtained through fractional-order virtual inertia control; Step 6: According to the reference power output by the controller, the active power adjustment amount of the primary frequency modulation of the battery energy storage system is obtained through the inner loop control of the GFL power converter.

2. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1 is characterized in that: The characteristic equation of the system is the denominator of the expression of the transfer function from the system disturbance to the system frequency response.

3. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1 is characterized in that: The step 2 comprises: According to the characteristic equation of the system, after frequency domain transformation and Euler transformation, when the frequency ranges from 0 to positive infinity, a set of controller parameters satisfying the characteristic equation whose real part and imaginary part are both 0; wherein, the set of controller parameters when the frequency is equal to 0 constitutes the real root boundary curve RRB, and the set of controller parameters when the frequency is greater than 0 constitutes the complex root boundary curve CRB.

4. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1 is characterized in that: The step 3 comprises: The virtual inertia-virtual damping parameter plane is divided into a number of closed areas according to the real root boundary curve RRB and the complex root boundary curve CRB. According to the stability criterion based on parameter sensitivity, it is judged whether each closed area meets the stability condition, so as to obtain the stable area where the system remains stable in the parameter plane.

5. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 4 is characterized in that: The stability criterion based on parameter sensitivity is as follows: If the following equation is satisfied, the system is unstable: S X (w * ,k * )>0 Among them, k * is a point on the SBL curve, X∈{J,D}, S J and S D Respectively represent J v and D v Root sensitivity.

6. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1, characterized in that: The step 4 comprises: Taking the controller parameters in the stable region as variables, the frequency deviation, output power deviation and the distance of the parameter setting point from the center of the stable region are considered to build a parameter optimization model through weighted summation. The particle swarm algorithm PSO is used to solve the parameter optimization model and obtain the optimal control parameters of the fractional-order virtual controller FOVIC.

7. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 6 is characterized in that: The controller parameters include virtual inertia, virtual damping, and fractional order.

8. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1, characterized in that: The step 5 comprises: The tuning is based on the optimal control parameters and the frequency change of the system. The virtual damping parameter is used as the proportional coefficient and the virtual inertia parameter is used as the differential coefficient. The virtual inertia control is a proportional-differential controller, thereby obtaining the output reference power of the fractional-order virtual controller FOVIC.

9. The method for virtual inertia control and controller parameter setting based on fractional derivatives according to claim 1, characterized in that: The step 6 comprises: According to the reference power output by the controller, through the inner loop control of the GFL power converter, the GFL power converter is represented by an inertia link, and the output is limited to obtain the active power adjustment amount of the primary frequency modulation of the battery energy storage system.

Images