Adaptive virtual inertia control method for DC microgrid with flywheel energy storage

By combining the sagging curve intercept adjustment amount and voltage change rate in the DC microgrid, the adaptive control of hyperbolic tangent function and power function is introduced, the problem of insufficient inertia of the flywheel energy storage system in the DC microgrid is solved, and the problem of inertia in the DC microgrid is realized. Fast response and voltage stability increase when power changes suddenly.

CN115425636BActive Publication Date: 2024-11-29SHANXI UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202211139071.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-09-19
Publication Date
2024-11-29
Estimated Expiration
2042-09-19

AI Technical Summary

Technical Problem

The existing flywheel energy storage system cannot effectively improve the system's inertia in the DC microgrid, and the dynamic performance and stability of the bus voltage when the power changes suddenly, and the traditional control method is complex and has many factors to consider.

Method used

A method of adaptive control of virtual inertia of DC microgrids with flywheel energy storage is proposed. By combining the sagging curve intercept adjustment amount and voltage change rate, and introducing hyperbolic tangent function and power function, adaptive virtual inertia control is realized, and the flywheel is used. The energy storage system provides inertia support to fast charge and discharge, shorten transient response time and prevent converter output from exceeding limits.

Benefits of technology

It effectively improves the inertial support capacity of the DC microgrid, improves voltage transients, reduces power impact on the AC power grid, and simplifies control complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115425636B_ABST
    Figure CN115425636B_ABST
Patent Text Reader

Abstract

This invention relates to the field of virtual inertia control technology for DC microgrids, and discloses an adaptive virtual inertia control method for DC microgrids with flywheel energy storage. First, a linear relationship is established between the voltage change rate and the droop curve intercept adjustment to achieve virtual inertia control. Second, a hyperbolic tangent function with nested power functions is used to adaptively adjust the virtual inertia, ensuring the linear relationship meets practical flexibility requirements. This shortens the transient response time when the voltage change rate is small, and ensures a large virtual inertia support capacity when the voltage change rate is large, preventing the converter output from exceeding limits. Simultaneously, the flywheel rotor speed is rapidly adjusted to provide timely auxiliary power to the system, enhancing the inertia support of the DC microgrid. Finally, a photovoltaic-energy storage DC microgrid control system is constructed using Matlab / Simulink. Simulations are conducted under sudden power changes, introducing additional virtual inertia control to verify the feasibility and effectiveness of the proposed control method.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of virtual inertia control of a DC microgrid, and in particular to a virtual inertia adaptive control method of a DC microgrid containing flywheel energy storage. Background Art

[0002] The problem of reduced grid inertia caused by large-scale photovoltaic grid connection has limited the development of photovoltaics to a certain extent. Domestic and foreign scholars use the ability of energy storage to quickly charge and discharge to provide inertial support for the system. Flywheel energy storage system (FESS) is widely used due to its advantages such as high power density, fast response speed, easy detection of discharge depth and long service life.

[0003] At present, PI feedback control method is widely used in the discharge process of FESS. Dai Xingjian et al. introduced load power and current feedforward compensation values ​​into voltage closed-loop control to eliminate the influence of power mutation; Zhang Xiang used feedforward compensation to compensate the load current, which improved the stability of the system, but required the addition of load current sensors; Gong Lei et al. introduced the immersed invariant manifold algorithm, which can effectively suppress the load fluctuation at the DC end, but required the correction of the parameters of multiple controllers. The above controls only adjust the feedback amount on the basis of the original PI to suppress the influence of load mutation, and do not consider the use of the potential inertial support capacity of FESS. There are many factors to consider, and the overall control is relatively complex.

[0004] In order to stimulate potential inertia, it is proposed to apply improved droop control in the energy storage system, and introduce voltage differential components in the droop coefficient to provide inertial support for the system. Some studies have proposed a dynamic virtual inertia control strategy, introducing the inverse tangent function to achieve droop coefficient adaptation, but it is more sensitive to parameter changes and the transient response duration is too long. There are also studies that, based on the above control strategy, introduce the hyperbolic tangent function to achieve virtual inertia control adaptation. Compared with the inverse tangent function, it has a better convergence speed when the voltage change rate is large, but its judgment of the adjustment direction and the relationship between the adjustment amount and the virtual capacitance are more complicated. In response to the above problems, the present invention proposes a virtual inertia adaptive control method for a DC microgrid containing flywheel energy storage. Summary of the invention

[0005] In view of the problem that the traditional control method of the flywheel energy storage system cannot improve the potential inertia of the DC microgrid, and the dynamic performance and stability of the bus voltage are poor when the power of the system changes suddenly, the present invention proposes a virtual inertia adaptive control method for the DC microgrid containing flywheel energy storage. The method combines the intercept adjustment of the droop curve with the rate of change of the voltage, and introduces the hyperbolic tangent function and the power function to realize adaptive virtual inertia control. Finally, in the case of power mutation, the virtual inertia control with additional improved droop is simulated and compared with the traditional control without virtual inertia, verifying the effectiveness of the proposed method.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The present invention provides a method for adaptively controlling virtual inertia of a DC microgrid containing flywheel energy storage, comprising the following steps:

[0008] Step 1, virtual inertia analysis of DC microgrid with flywheel energy storage;

[0009] Step 2, constructing a linear relationship between the voltage change rate and the droop curve intercept adjustment amount to achieve virtual inertia control;

[0010] Step 3: Adopt the hyperbolic tangent function of the nested power function to realize adaptive adjustment of the virtual inertia, so that the linear relationship meets the actual flexible needs, and shortens the transient response time when the voltage change rate is small; when the voltage change rate is large, ensure that there is a large virtual inertia support capacity and prevent the output of the converter from exceeding the limit;

[0011] Step 4: Control the rotor speed of the flywheel to adjust rapidly, provide auxiliary power to the system in time, and enhance the inertial support of the DC microgrid.

[0012] Further, the specific process of step 1 is:

[0013] The inertia time constant of the DC microgrid is:

[0014]

[0015] Where: W ei Represented as parallel capacitance C i Stored energy; S Nci C i The capacity base value; U is the DC side voltage value;

[0016] H dc_microgrid The physical meaning is: the time required for the capacitor to release the stored energy at the rated voltage, the DC side capacitance value C i Small, inertia time constant H dc_microgrid Small, insufficient system inertia;

[0017] The power relationship on the DC capacitor side of the DC microgrid is:

[0018]

[0019] Where P r and P s Represent the required total power and the output power of the energy storage unit respectively; P c and U dc are the power flowing into the capacitor and the DC voltage respectively;

[0020] Udc When P is kept constant, r =P s Established,

[0021] When the DC microgrid is disturbed, the flywheel energy storage system quickly charges and discharges under additional control to provide inertia for the system, and a larger capacitor, C, is virtualized on the DC side. vir , and the power source provides auxiliary power ΔP s To reduce the voltage change speed, ΔP s for:

[0022]

[0023] Where P' s is the output power of the flywheel energy storage under the additional control action;

[0024] Output power fluctuation ΔP at load r When:

[0025] P' s -(P r +ΔP r )=P c +ΔP s (4)

[0026] Combining equations (2) and (3), we can obtain:

[0027]

[0028] After adding virtual inertia control, the capacitance value of the DC side increases from C to C+Cvir. Combining formula (1), it can be seen that the inertia time constant of the system becomes:

[0029]

[0030] Further, the specific process of step 2 is:

[0031] When the flywheel energy storage system adopts droop control based on PU characteristics, there are:

[0032]

[0033] Where U ref It represents the vertical intercept of the droop curve; 1 / k is the droop coefficient.

[0034] Since the PU characteristic droop control cannot reflect the sensitivity of the flywheel energy storage system to voltage changes and cannot provide inertia, it is proposed to control the converter to quickly absorb and release power by adjusting the intercept of the droop curve to achieve virtual inertia control;

[0035] Let the sag intercept adjustment amount ΔU refCombined with the voltage change rate dU / dt, it changes with dU / dt; when dU / dt is positive, reduce U ref To prevent the voltage from increasing further, U should be increased. ref ; we can get:

[0036]

[0037] Where U ref0 is the initial sag curve intercept; k d is the adjustment parameter, and k d >0;

[0038] Substituting formula (8) into formula (7), we get:

[0039]

[0040] Combined with formula (7), let U ref0 The power required for the voltage change produced by U is P′ s , so we can get:

[0041]

[0042] Comparing equation (3) and equation (10), eliminating dU / dt, we get:

[0043]

[0044] U ref The reverse linear control with dU / dt is C vir Constant virtual inertia control, and k d The larger the C vir The bigger.

[0045] Further, the specific process of step 3 is:

[0046] The linear function droop curve intercept adjustment algorithm in formula (8) adjusts the parameter k d Constant, flexible application in actual situations cannot be achieved; therefore, the design ΔU ref Regarding the nonlinear function of dU / dt passing through the origin, a power function is embedded in the hyperbolic tangent function. When |dU / dt| is small, the duration of the transient state is shortened; when |dU / dt| is large, a larger virtual inertia support is ensured to prevent the output of the converter from exceeding the limit.

[0047] The improved intercept adjustment algorithm is:

[0048]

[0049] Where U ref.max and U ref.min Uref The maximum and minimum values ​​of , k1 and k2 are virtual inertia adjustment parameters;

[0050] From formula (12), we can get that the intercept of the droop curve is adjusted by using nested functions, which has a certain adaptive ability. The slope k of the secant line of the curve passing through the origin is d Can represent C vir When k2 = 1, as |dU / dt| increases, the slope of the secant line gradually decreases. It can be seen that, using the hyperbolic tangent function, in the transient process after the load mutation, as |dU / dt| increases, C vir When k2>1, when |dU / dt| is small, k d The change trend is the same as that of |dU / dt|; when |dU / dt| continues to increase to the right of point e, k d It decreases with the increase of |dU / dt|; the coordinates of point e are expressed as the voltage change of ||dU / dt| max |, the vertical intercept of the converter droop curve should be close to its limit value |ΔU ref ,limit|;

[0051] Therefore, the intercept adjustment algorithm with k2>1 can provide a smaller virtual inertia and reduce the transient response time when |dU / dt| is small. When |dU / dt| is large, the converter outputs a larger auxiliary power to enhance the virtual inertia support capability of the system. When |dU / dt| is large, the limiting capability of the hyperbolic function is used to reduce the output power and effectively avoid converter over-limit.

[0052] Further, the specific process of step 4 is:

[0053] Since the flywheel has a speed limit when it is running, the maximum power released by the flywheel is limited to:

[0054]

[0055] In the formula, J F is the moment of inertia of the rotor; ω f is the rotor rotation angular velocity;

[0056] Similarly, U ref The adjustment range is also limited. The flywheel energy storage system provides auxiliary power during disturbances. Limited by its speed, the droop curve can rise to L high , L high It is expressed as the maximum vertical intercept of the droop curve under the condition that the slope of the original PU droop characteristic curve remains unchanged, and the flywheel energy storage can provide the system with the maximum power under the additional effect. Since 1 / k is constant, it can be known that U ref.max for:

[0057]

[0058] Similarly, the droop curve can be shifted to L low , L low It is expressed as the minimum vertical intercept of the droop curve under the condition that the slope of the original PU droop characteristic curve remains unchanged and the flywheel energy storage can provide the system with the minimum power under the additional effect. Therefore, U ref.min for:

[0059]

[0060] Compared with the prior art, the present invention has the following advantages:

[0061] (1) The present invention proposes a virtual inertia adaptive control method for a DC microgrid with flywheel energy storage, which combines the droop curve intercept adjustment with the voltage change rate, and introduces a hyperbolic tangent function and a power function to achieve adaptive virtual inertia control. Finally, in the case of power mutation, a simulation comparison is carried out between the virtual inertia control with improved droop and the traditional control without virtual inertia, verifying the effectiveness of the proposed method.

[0062] (2) The virtual inertia adaptive control method proposed in the present invention can use F-VSC to provide inertial support at the moment of load mutation and shorten the transient response time according to the dynamic performance requirements of the system.

[0063] (3) The control method of the present invention can not only improve the voltage transient phenomenon, but also reduce the power impact caused to the AC power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is the topological structure diagram of the DC microgrid.

[0065] Figure 2 This is the power relationship diagram on the DC capacitor side.

[0066] Figure 3 Adjust the graph for the sag curve intercept.

[0067] Figure 4 is the characteristic curve of the droop curve intercept change, where (a) is the secant slope graph and (b) is the adjustment parameter k d Change curve graph.

[0068] Figure 5 Adjust the range plot for the sag curve intercept.

[0069] Figure 6 This is the F-VSC control block diagram.

[0070] Figure 7 The transient response time test result diagram of voltage and power when power suddenly changes; where: (a) is the DC bus voltage Udc , (b) is the intercept of the sag curve U ref , (c) is the F-VSC output power P F , (d) is the G-VSC output power P G , (e) is P PV Output power, (f) is P Load Output power.

[0071] Figure 8 Figure 2 is the transient response time test result diagram for different values ​​of k1 and k2 when the load changes suddenly; (a) is the DC bus voltage U when k1 changes dc , (b) is the F-VSC output power P when k1 changes F , (c) is the DC bus voltage U when k2 changes dc , (d) is the F-VSC output power P when k2 changes F . DETAILED DESCRIPTION

[0072] The following is a specific and detailed description of the technical solutions in the embodiments of the present invention in combination with the embodiments of the present invention and the accompanying drawings. It should be noted that for ordinary technicians in this field, several modifications and improvements can be made without departing from the principles of the present invention, which should also be considered to belong to the protection scope of the present invention.

[0073] Adaptive control method of virtual inertia of DC microgrid with flywheel energy storage:

[0074] The DC microgrid topology is as follows: Figure 1 As shown, it mainly includes the following parts: photovoltaic module, energy storage module, converter, AC power grid and load module.

[0075] The photovoltaic power generation unit is connected to the DC distribution network through the DC converter PV-DC, and the maximum power point tracking (MPPT) control is adopted during operation; the FESS and the AC grid are connected to the DC distribution network through the bidirectional AC / DC converters F-VSC and G-VSC respectively, and the converters both use droop control to adjust the DC voltage; the AC load is connected to the DC bus by the converter L-VSC.

[0076] 1. Virtual inertia analysis of DC microgrid with flywheel energy storage

[0077] (1) Definition of DC microgrid inertia

[0078] Inertia is the resistance of any material to changes in its speed. In AC systems, inertia can prevent sudden changes in system frequency. In traditional synchronous generator sets, the inertia time constant H s It can be described as:

[0079]

[0080] Where W k and S N are the kinetic energy stored in the rotor and the rated capacity of the generator respectively, J refers to the moment of inertia of the rotor, and ω is the rotor speed.

[0081] Similar to the definition of inertia of AC system, the inertia of DC microgrid can be defined as the ability to suppress sudden voltage changes in the system. Extending equation (1) to DC microgrid, the inertia time constant can be expressed as:

[0082]

[0083] Where W ei Represented as parallel capacitance C i Stored energy; S Nci C i The capacity base value; U is the DC side voltage value.

[0084] H dc_microgrid The physical meaning is: the time required for the capacitor to release the stored electrical energy at the rated voltage. i Small, inertia time constant H dc_microgrid Small, insufficient system inertia.

[0085] (2) Inertia analysis of flywheel energy storage system

[0086] FESS uses a high-speed rotating flywheel mass to store energy. Its kinetic energy E F for:

[0087]

[0088] In the formula, J F Expressed as the moment of inertia of the rotor; ω f is the rotor rotation angular velocity.

[0089] The output power change caused by the change of FESS rotor speed is:

[0090]

[0091] When power imbalance occurs on the DC side of the system, the charging and discharging power of the DC side capacitor is:

[0092]

[0093] Where P r and P s are the inlet and outlet powers of the DC capacitor respectively.

[0094] At the moment of load fluctuation, the charging and discharging power of the DC side capacitor supplied by the FESS, ignoring the loss, the DC bus voltage change is converted into speed fluctuation. Combining equations (4) and (5), we can get:

[0095]

[0096] Integrating both sides of equation (6) we get:

[0097]

[0098] In the formula, ω f1 ,ω f0 , U dc1 , U dc0 Mechanical angular velocity and DC bus voltage after FESS rotor changes and at the initial moment respectively.

[0099] Take the speed ω of FESS at the rated voltage value after the transformation f1 Normalizing the rated speed to get:

[0100]

[0101] Then ω f1_pu ,ω f0_pu are the per unit values ​​of the rotor angular velocity after FESS transformation and at the initial stage respectively; E C For U N The electrical energy stored in the capacitor. C <<E F Therefore, when the voltage changes, the flywheel speed does not change significantly. Therefore, additional inertia control is applied to the FESS, which can increase the speed change on the original basis, control the flywheel charging and discharging power, and provide inertial support for the system.

[0102] (3) Analysis of virtual inertia of DC microgrid

[0103] like Figure 2 As shown in the figure, the equivalent circuit of the converter on the DC side is outside the dotted box. r and P s Represent the required total power and the output power of the energy storage unit respectively; P c and U dc are the power flowing into the capacitor and the DC voltage respectively.

[0104] Depend on Figure 2 We can get:

[0105]

[0106] U dc When P is kept constant, r =P s Established.

[0107] When the DC microgrid is disturbed, the FESS quickly charges and discharges under the additional control to provide inertia for the system, and a large capacitor, C vir , and the power source provides auxiliary power ΔP s To reduce the voltage change speed. ΔP s for:

[0108]

[0109] Where P' s is the output power of the flywheel energy storage under the additional control action;

[0110] Output power fluctuation ΔP at load r When:

[0111] P' s -(P r +ΔP r )=P c +ΔP s (11)

[0112] Combining equations (9) and (10), we can obtain:

[0113]

[0114] After adding virtual inertia control, the capacitance value of the DC side increases from C to C+Cvir. Combining formula (2), it can be seen that the inertia time constant of the system becomes:

[0115]

[0116] In the formula, C vir_i is the virtual capacitance value of the i-th converter involved in the control.

[0117] 2. Construct a linear relationship between the voltage change rate and the droop curve intercept adjustment to achieve virtual inertia control

[0118] When FESS adopts droop control based on PU characteristics, there are:

[0119]

[0120] Where U ref It represents the vertical intercept of the droop curve; 1 / k is the droop coefficient.

[0121] PU characteristic droop control cannot reflect the sensitivity of FESS to voltage changes, nor can it provide inertia. Therefore, it is proposed to control the converter to quickly absorb and release power by adjusting the intercept of the droop curve to achieve virtual inertia control. The principle diagram is as follows:

[0122] like Figure 3 As shown, a is the initial working point. When the load increases to P b , detected U dc Lower than expected, increase the intercept U of the sag curve ref , so that the converter outputs auxiliary power and slows down the speed of DC voltage reduction; vice versa. It can be seen that there is a linear correlation between the adjustment amount of the droop curve intercept and the change in converter output power.

[0123] Let the sag intercept adjustment amount ΔU ref Combined with the voltage change rate dU / dt, it changes with dU / dt; when dU / dt is positive, reduce U ref To prevent the voltage from increasing further, U should be increased. ref ; we can get:

[0124]

[0125] Where U ref0 is the initial sag curve intercept; k d is the adjustment parameter, and k d >0;

[0126] Substituting formula (15) into formula (14), we get:

[0127]

[0128] Combined with equation (14), let U ref0 The power required for the voltage change produced by U is P′ s , so we can get:

[0129]

[0130] Comparing equation (10) and equation (17), eliminating dU / dt, we get:

[0131]

[0132] U ref The reverse linear control with dU / dt is C vir Constant virtual inertia control, and k d The larger the C vir The bigger.

[0133] 3. The hyperbolic tangent function of the nested power function is used to realize adaptive adjustment of the virtual inertia, so that the linear relationship can meet the actual flexible needs, and shorten the transient response time when the voltage change rate is small; when the voltage change rate is large, it ensures a large virtual inertia support capacity and prevents the output of the converter from exceeding the limit;

[0134] The intercept adjustment algorithm of the linear function droop curve in formula (15) is to adjust the parameter k d Constant, flexible application in actual situations cannot be achieved; design ΔU ref Regarding the nonlinear function of dU / dt passing through the origin, a power function is embedded in the hyperbolic tangent function. When |dU / dt| is small, the duration of the transient state is shortened; when |dU / dt| is large, a larger virtual inertia support is ensured to prevent the output of the converter from exceeding the limit.

[0135] The improved intercept adjustment algorithm is:

[0136]

[0137] Where U ref.max and U ref.min U ref The maximum and minimum values ​​of , k1 and k2 are virtual inertia adjustment parameters;

[0138] From formula (19), we can see that the intercept of the droop curve is adjusted by using nested functions, which has a certain adaptive ability. When the k value changes, the intercept of the droop curve changes as follows: Figure 4 shown.

[0139] Figure 4 (a) Given two function curves, draw a secant line through the origin. The slope of the secant line is k. d Can represent C vir size.

[0140] When k2=1, as |dU / dt| increases, k d3 <k d2 <k d1 , the slope of the secant line of the curve gradually decreases. It can be seen that, using the hyperbolic tangent function, in the transient process after the load mutation, as |dU / dt| increases, C vir When k2>1, when |dU / dt| is small, k d4 <k d5 , k d The change trend is the same as that of |dU / dt|; when |dU / dt| continues to increase to the right of point e, there is k d6 >k d7 , k d It decreases with the increase of |dU / dt|.

[0141] Therefore, the intercept adjustment algorithm with k2>1 can provide a smaller virtual inertia and reduce the transient response time when |dU / dt| is small. When |dU / dt| is large, the converter outputs a larger auxiliary power to enhance the virtual inertia support capability of the system. When |dU / dt| is large, the limiting capability of the hyperbolic function is used to reduce the output power and effectively avoid converter over-limit.

[0142] Figure 4 (b) Three curves with k2>1 are given, and the k2 values ​​are different. It can be seen from the curve graph that when |dU / dt| is the same, k d3 <k d2 <k d1 That is, the control gradually reduces the virtual inertia provided to the system as the value of the virtual inertia adjustment parameter k2 increases. It can be seen that selecting appropriate adjustment parameters can provide appropriate inertial support for the system.

[0143] 4. Control the rotor speed of the flywheel to quickly adjust and provide auxiliary power to the system in time to enhance the inertial support of the DC microgrid

[0144] From the above analysis, we can see that the greater the power released or absorbed by the FESS, the stronger the voltage stabilization effect of the DC bus. However, there is a speed limit when the flywheel is running, so the maximum power released by the flywheel is limited to:

[0145]

[0146] Similarly, U ref There are also limited ranges for adjustment, such as Figure 5 shown.

[0147] FESS provides auxiliary power during disturbances. Limited by its speed, the droop curve can rise to L high , since 1 / k is constant, we know that U ref.max for:

[0148]

[0149] Similarly, the droop curve can be shifted to L low , so U ref.min for:

[0150]

[0151] 5. Implementation of flywheel energy storage control system

[0152] The additional control link acts on the outer loop of FESS, and in coordination with the traditional PI control of FESS, the F-VSC exhibits the inertia characteristics of a synchronous generator. In practical applications, since the measurement of the voltage change rate is easily affected by noise and harmonics, this study proposes to first low-pass filter the voltage information, and then replace dU / dt with the voltage variable δU output by high-pass filtering.

[0153] When the system is disturbed, δU≠0, under the additional control ΔU ref Adjust in the opposite direction of δU, and calculate the new droop curve intercept U by formula (19): ref , and then the output power reference value P of the converter at this moment is obtained by the PU droop characteristic ref-F , then perform Laplace transformation on equation (4) to obtain the angular velocity reference value ω of the flywheel energy storage ref-F , and then use PI control to generate the current inner loop reference value, which uses i sd = 0 vector control. Since the original PI control of the flywheel energy storage system has a large current inner loop bandwidth, the FESS can track the reference value of the output power in time and provide real-time inertial support for the system. Figure 6 shown.

[0154] 6. Simulation analysis:

[0155] In order to verify the feasibility of the virtual inertia adaptive control method of DC microgrid with flywheel energy storage mentioned above, according to Figure 1 The model is built in the Matlab / Simulink simulation platform, and the proposed control algorithm is applied to the simulation model. The main parameters of the simulation model are shown in Table 1.

[0156] Table 1 Main parameters of DC microgrid simulation

[0157]

[0158] In order to verify the effectiveness of the control method, the transient response time of the virtual inertia adjustment parameters (k1, k2) under different power mutations and load surges was tested.

[0159] (1) Power mutation test

[0160] Initially, P PV =10kW, AC load P Load 13kW, U dc The F-VSC and G-VSC jointly regulate the voltage at 397.56V. At t = 1s, P PV Increase to 15kW; t = 2s, P Load Sudden increase of 10kW, working conditions such as Figure 7(e) and (f) are shown. The F-VSC control simulation comparison is based on non-virtual inertia control and the adaptive virtual inertia control of the present invention. The waveform is shown in Figure 7 As shown in the figure: P G and P F are the output powers of G-VSC and F-VSC respectively.

[0161] Figure 7 (a) and (b) show the power sudden change U dc Transient response test results show that when adaptive virtual inertia control is not added, F-VSC and G-VSC jointly bear the power disturbance according to the droop characteristics, and the droop curve intercept U ref Remaining unchanged, the system lacks inertia, U dc After the introduction of adaptive virtual inertia control, U under F-VSC control ref As dU / dt changes, the converter can quickly absorb or release auxiliary power to provide virtual inertia support and enable U dc Slowly stabilizing.

[0162] Figure 7 (c) and (d) power fluctuation curves show that, compared with non-virtual inertial control, FESS with additional control can automatically adjust its charging and discharging speed at the moment of power mutation, so that auxiliary power can be input into the system immediately and quickly, improving system inertia and system dynamic performance. After the inertial response ends, the power curve of FESS using virtual inertial control is parallel to the energy curve of G-VSC without virtual inertial control, that is, it can continuously provide the same steady-state power in the droop control stage to ensure the stable operation of the system.

[0163] By comparing the simulation experimental results of the two controls, it can be seen that the use of additional virtual inertia control can utilize the auxiliary power provided by the energy storage side to increase the inertial support capacity of the system, effectively slow down the power output of the AC power grid, and enhance the overall stability of the system.

[0164] (2) The impact of different k1 and k2 on transient response

[0165] At the initial moment, P PV =15kW, P Load =13kW, U dc F-VSC and G-VSC act at 400.04V; when t = 1s, P Load The power suddenly increases to 23kW. The working condition is as follows Figure 8 In order to compare the over-response time after a sudden load increase, k1 and k2 are adjusted to make the load increase moment U ref The waveform is as follows. Figure 8 shown.

[0166] Depend on Figure 8 From the bus voltage and FESS converter output power curves shown in the figure, it can be seen that as the k value changes, the virtual inertia provided by the virtual inertia control for the system also changes. The specific experimental data are shown in Tables 2 and 3.

[0167] Table 2 Transient response test results when k2 value is constant and k1 value changes

[0168]

[0169] Table 3 Transient response test results when k1 value is constant and k2 value changes

[0170]

[0171] At the moment of sudden load increase, F-VSC immediately outputs auxiliary power to mitigate the impact of power mutation to ensure the stability of the system. From the data in Table 2, it can be seen that when the k2 value is constant, as the k1 value increases, the greater the auxiliary power provided by F-VSC, the longer the transient response time, and the lower the bus voltage deviation rate.

[0172] It can be seen from the data in Table 3 that when the k1 value is constant, as the k2 value decreases, the inertia provided to the system by the virtual inertia control adopted increases, the bus voltage deviation is reduced, and the stability of the DC bus voltage is improved.

[0173] Therefore, by selecting appropriate k1 and k2, F-VSC can provide greater inertia support for the system at the moment of load mutation, and can also reduce ΔU according to the dynamic performance requirements of the system. ref Sensitivity to small voltage change rates, reducing transient response time.

Claims

1. A virtual inertia adaptive control method for a DC microgrid with flywheel energy storage, characterized in that: The following steps are involved: Step 1, virtual inertia analysis of DC microgrid with flywheel energy storage; Step 2, constructing a linear relationship between the voltage change rate and the droop curve intercept adjustment amount to realize virtual inertia control; Step 3: Adopt the hyperbolic tangent function of the nested power function to realize adaptive adjustment of the virtual inertia, so that the linear relationship meets the actual flexible needs, and shortens the transient response time when the voltage change rate is small; when the voltage change rate is large, ensure that there is a large virtual inertia support capacity and prevent the output of the converter from exceeding the limit; Step 4: Control the rotor speed of the flywheel to adjust quickly, provide auxiliary power to the system in time, and enhance the inertial support of the DC microgrid; The specific process of step 3 is as follows: Design ΔU ref Regarding the nonlinear function of dU / dt passing through the origin, a power function is embedded in the hyperbolic tangent function. When |dU / dt| is small, the duration of the transient state is shortened; when |dU / dt| is large, a larger virtual inertia support is ensured to prevent the output of the converter from exceeding the limit. The improved intercept adjustment algorithm is: Where U ref.max and U ref.min U ref The maximum and minimum values ​​of , k1 and k2 are virtual inertia adjustment parameters; From formula (12), we can get that the intercept of the droop curve is adjusted by using nested functions, which has a certain adaptive ability. The slope k of the secant line of the curve passing through the origin is d Can represent C vir When k2 = 1, as |dU / dt| increases, the slope of the secant line gradually decreases. It can be seen that, using the hyperbolic tangent function, in the transient process after the load mutation, as |dU / dt| increases, C vir When k2>1, when |dU / dt| is small, k d The change trend is the same as that of |dU / dt|; when |dU / dt| continues to increase to the right of point e, k d It decreases with the increase of |dU / dt|; the coordinates of point e are expressed as the voltage change of ||dU / dt| max |, the vertical intercept of the converter droop curve should be close to its limit value |ΔU ref ,limit|; Therefore, the intercept adjustment algorithm with k2>1 can provide a smaller virtual inertia and reduce the transient response time when |dU / dt| is small. When |dU / dt| is large, the converter outputs a larger auxiliary power to enhance the virtual inertia support capability of the system. When |dU / dt| is large, the limiting capability of the hyperbolic function is used to reduce the output power and effectively avoid converter over-limit.

2. A method for adaptively controlling virtual inertia of a DC microgrid with flywheel energy storage according to claim 1, characterized in that: The specific process of step 1 is as follows: The inertia time constant of the DC microgrid is: Where: W ei Represented as parallel capacitance C i Stored energy; S Nci C i The capacity base value; U is the DC side voltage value; H dc_microgrid The physical meaning is: the time required for the capacitor to release the stored energy at the rated voltage, the DC side capacitance value C i Small, inertia time constant H dc_microgrid Small, insufficient system inertia; The power relationship on the DC capacitor side of the DC microgrid is: Where P r and P s Represent the required total power and the output power of the energy storage unit respectively; P c and U dc are the power flowing into the capacitor and the DC voltage respectively; U dc When P is kept constant, r =P s Established, When the DC microgrid is disturbed, the flywheel energy storage system quickly charges and discharges under additional control to provide inertia for the system, and a larger capacitor, C, is virtualized on the DC side. vir , and the power source provides auxiliary power ΔP s To reduce the voltage change speed, ΔP s for: Where P' s is the output power of the flywheel energy storage under the additional control action; Output power fluctuation ΔP at load r When: Combining equations (2) and (3), we can obtain: After adding virtual inertia control, the capacitance value of the DC side increases from C to C+Cvir. Combining formula (1), it can be seen that the inertia time constant of the system becomes:

3. The method for virtual inertia adaptive control of a DC microgrid with flywheel energy storage according to claim 2 is characterized in that: The specific process of step 2 is: When the flywheel energy storage system adopts droop control based on PU characteristics, there are: Where U ref It represents the vertical intercept of the droop curve; 1 / k is the droop coefficient; Since the PU characteristic droop control cannot reflect the sensitivity of the flywheel energy storage system to voltage changes and cannot provide inertia, it is proposed to control the converter to quickly absorb and release power by adjusting the intercept of the droop curve to achieve virtual inertia control; Let the sag intercept adjustment amount ΔU ref Combined with the voltage change rate dU / dt, it changes with dU / dt; when dU / dt is positive, reduce U ref To prevent the voltage from increasing further, U should be increased. ref ; we can get: Where U ref0 is the initial sag curve intercept; k d is the adjustment parameter, and k d >0; Substituting formula (8) into formula (7), we get: Combined with formula (7), let U ref0 The power required for the voltage change produced by U is P′ s , so we can get: Comparing equation (3) and equation (10), eliminating dU / dt, we get: U ref The reverse linear control with dU / dt is C vir Constant virtual inertia control, and k d The larger the C vir The bigger.

4. The method for virtual inertia adaptive control of a DC microgrid with flywheel energy storage according to claim 3 is characterized in that: The specific process of step 4 is as follows: Since the flywheel has a speed limit when it is running, the maximum power released by the flywheel is limited to: In the formula, J F is the moment of inertia of the rotor; ω f is the rotor rotation angular velocity; Similarly, U ref The adjustment range is also limited. The flywheel energy storage system provides auxiliary power during disturbances. Limited by its speed, the droop curve can rise to L high , L high It is expressed as the maximum vertical intercept of the droop curve under the condition that the slope of the original PU droop characteristic curve remains unchanged, and the flywheel energy storage can provide the system with the maximum power under the additional effect. Since 1 / k is constant, it can be known that U ref.max for: Similarly, the droop curve can be shifted to L low , L low It is expressed as the minimum vertical intercept of the droop curve under the condition that the slope of the original PU droop characteristic curve remains unchanged and the flywheel energy storage can provide the system with the minimum power under the additional effect. Therefore, U ref.min for:

Citation Information

Patent Citations

  • Microgrid voltage control method for fuzzy self-adaptive operation of high-permeability fan

    CN111064196A

  • Direct-current distribution network virtual inertia control method based on variable droop coefficient

    CN111478310A