Multi-module parallel DC microgrid system and droop control method thereof

Abstract

The invention discloses a multi-module parallel DC micro-grid system and a droop control method thereof, and relates to the field of DC micro-grid systems. According to the method, the parameter k and the quality factor Q are optimized and selected under the condition that the voltage gain is met, the optimized droop coefficient is obtained in combination with the balance coefficient and the speed regulation factor, and droop control is performed through the optimized droop coefficient. According to the invention, the bus voltage offset can be compensated, and the power balance is realized on the premise of not sacrificing the response speed.

Description

Technical Field

[0001] The present invention relates to the field of direct current microgrid systems, and in particular to a multi-module parallel direct current microgrid system and a droop control method thereof. Background Art

[0002] In distributed energy storage, the energy storage device needs to have the characteristics of good quality of stored power, high adjustability and fast response speed. Based on this, battery energy storage with the above characteristics is very suitable for new energy power generation systems. However, the voltage that the battery can withstand is often low, and directly connecting to the DC bus will damage the battery. In addition, the voltage across the battery fluctuates greatly with its SOC state, and the output voltage is unstable. Therefore, a bidirectional DC-DC converter is usually connected between the battery and the power grid to adjust the primary and secondary voltages and currents. However, distributed energy storage has problems such as difficult control, high energy loss, and complex and changeable working scenarios. Therefore, the bidirectional DC-DC converter working in this scenario must meet the characteristics of low energy loss and large input and output gain, so that it can adapt to complex working environments.

[0003] In view of the problem of bus voltage offset and power balance distribution accuracy, many studies have been conducted on droop control strategies at home and abroad. In the prior art, bus compensation control is designed, and the bus voltage fluctuates little when adding or reducing loads, but its charging and discharging calculation formulas are different, and the control system is complex; through cascade droop control, SOC can be quickly balanced. Although the impact of different battery capacities on the system is considered, there is a lack of research on overcharging and over-discharging of energy storage systems; by improving the power exponential droop coefficient, adjusting the speed factor can change the charging and discharging speed, but due to the protection interval limitation, the SOC adjustment effect is not obvious. The prior art also adopts variable speed factor droop control, which improves the SOC balancing speed but sacrifices the output power response speed; the prior art provides a variable speed factor droop control strategy based on time function, which can quickly distribute output power and shorten the power distribution response time, but its speed factor coefficient undergoes a sudden change, and the output power jitters, affecting the system stability; or dynamically adjust the droop control coefficient, the bus voltage offset value is low, but its SOC balancing speed does not change significantly. Summary of the invention

[0004] In view of the above-mentioned defects of the prior art, the present invention provides a multi-module parallel DC microgrid system, including a plurality of energy storage modules, each of which is connected to the power grid through a bidirectional CLLLC energy storage converter, wherein the bidirectional CLLLC energy storage converter includes a symmetrical primary side structure, a secondary side structure and a transformer, wherein the primary side structure and the secondary side structure both include a full-bridge circuit and a resonant circuit composed of four switch tubes;

[0005] When the maximum voltage gain of the converter is M ac_maxThe voltage gain M of the converter at the minimum frequency ac (ω n.min ) satisfies M ac_max <M ac (ω n.min ), the parameter k is the voltage gain M in the common area. ac_max The maximum point k on the boundary of max , otherwise, parameter k is the minimum point k min ;According to the parameter k, the quality factor Q is:

[0006]

[0007] When the selected parameter k value does not satisfy: When the parameter k is set to

[0008] The common area is the area where the parameter k and the quality factor Q are selected to achieve ZVS under the premise of satisfying the voltage gain of the converter; the parameter k is: k = L m / L r1 , L r1 and L m They are the resonant inductance and excitation inductance of the primary side structure respectively.

[0009] Furthermore, the maximum voltage gain of the converter is:

[0010]

[0011] Where n is the transformer ratio, V in.min is the minimum input voltage of the primary side structure, V out_max It is the maximum value of the output voltage of the secondary side structure.

[0012] Furthermore, the transformer ratio n is:

[0013]

[0014] Among them, V N-in is the rated voltage of the DC bus, V N-out It is the rated voltage output by the secondary side structure.

[0016] Furthermore, the excitation inductance L m for:

[0017]

[0018] Among them, R eq is the equivalent impedance of the converter; ω r is the resonant frequency of the converter.

[0019] The second aspect of the present invention provides a droop control method for a multi-module parallel DC microgrid system, wherein the droop coefficient z of the i-th energy storage module is i The output power of the energy storage module is multiplied and subjected to voltage and current double closed-loop control to obtain a PWM control signal, thereby realizing the charge and discharge control of the corresponding bidirectional CLLLC energy storage converter;

[0020] The droop coefficient z of the i-th energy storage module i for:

[0021]

[0022] Among them, k i is the SOC balance coefficient of the i-th energy storage module, z s is the adjustment coefficient; r is the speed control factor; S i is the available power of the i-th energy storage module.

[0023] Furthermore, the adjustment coefficient z s for:

[0024] z s =e β(0.2-s) ;

[0025] Wherein, β is the amplification factor, β>0, and S is the total available power of the energy storage module.

[0026] Furthermore, the SOC balancing coefficient k of the i-th energy storage module i The relationship is:

[0027]

[0028] Wherein, m is the number of energy storage modules.

[0029] Furthermore, the speed adjustment factor r is:

[0030]

[0031] Among them, r min is the minimum speed regulation factor; Δp is the output power difference between the two converters; Δp max is the maximum difference in output power, and α is a constant.

[0032] Compared with the prior art, the present invention has the following technical effects:

[0033] The present invention discloses a bidirectional CLLLC energy storage converter suitable for a DC microgrid system. The excitation inductance, parameters k and Q in the bidirectional CLLLC energy storage converter are designed to meet the voltage gain requirement. Based on the bidirectional CLLLC energy storage converter, a droop control method for a multi-module parallel DC microgrid system is provided. By introducing an adjustment coefficient, the output power of the energy storage system can change dynamically with the SOC, which solves the problem of overcharge and overdischarge of the battery and ensures the service life of the battery. The SOC balance of the energy storage system is achieved through the balance coefficient. By improving the speed regulation factor r, the speed regulation factor has a sine function characteristic. The simulation results show that the improved droop control can effectively shorten the SOC balance time of the energy storage system while taking into account the response speed and balance speed of the energy storage system.

[0034] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a structural principle diagram of a bidirectional CLLLC type distributed energy storage system according to a specific embodiment of the present invention;

[0036] Figure 2 It is a circuit modeling structure diagram of a CLLLC type energy storage converter according to a specific embodiment of the present invention;

[0037] Figure 3 It is a flow chart of the design of a resonant network of a CLLLC type energy storage converter according to a specific embodiment of the present invention;

[0038] Figure 4 In a specific embodiment of the present invention, M ac_max Relationship diagram with ZVS implementation area;

[0039] Figure 5 It is a block diagram of the overall strategy of improved droop control according to a specific embodiment of the present invention;

[0040] Figure 6 is a comparison diagram of different droop controls when sufficient power is available in a specific embodiment of the present invention. Figure 6 a is the droop coefficient transformation curve, Figure 6 b is the SOC conversion curve, Figure 6 c is the bus voltage transformation curve, Figure 6 d is the output power conversion curve;

[0041] Figure 7 In a specific embodiment of the present invention, SOC and output power conversion curves under power exponential droop control at different speed regulation factors are shown. Figure 7 a is the SOC conversion curve when r=2, Figure 7 b is the output power conversion curve when r=2, Figure 7 c is the SOC conversion curve when r=5, Figure 7 Output power conversion curve when d is r=5, Figure 7 e is the SOC conversion curve when r=7, Figure 7 Output power conversion curve when f is r=7;

[0042] Figure 8 In a specific embodiment of the present invention, the SOC and output power conversion curve under the time-based variable speed factor droop control is shown. Figure 8 a is the SOC conversion curve, Figure 8 b is the output power conversion curve;

[0043] Fig. 9 is a speed regulation factor r characteristic curve having a sinusoidal function characteristic in a specific embodiment of the present invention;

[0044] Fig.10 is a SOC and output power conversion curve under the speed regulation factor r droop control with a sinusoidal function characteristic in a specific embodiment of the present invention, Fig.10 a is the SOC conversion curve, Fig.10 b is the output power conversion curve;

[0045] Fig.11 In a specific embodiment of the present invention, SOC and output power conversion curves under power exponential droop control at different speed regulation factors are shown. Fig.11 a is the SOC conversion curve when r=2, Fig.11 b is the output power conversion curve when r=2, Fig.11 c is the SOC conversion curve when r=5, Fig.11 Output power conversion curve when d is r=5, Fig.11 e is the SOC conversion curve when r=7, Fig.11 Output power conversion curve when f is r=7;

[0046] Fig.12 In a specific embodiment of the present invention, the SOC and output power conversion curve under the time-based variable speed factor droop control is shown. Fig.12 a is the SOC conversion curve, Fig.12 b is the output power conversion curve;

[0047] Fig.13 is a SOC and output power conversion curve under the speed regulation factor r droop control with a sinusoidal function characteristic in a specific embodiment of the present invention, Fig.13 a is the SOC conversion curve, Fig.13 b is the output power conversion curve;

[0048] Fig.14 In a specific embodiment of the present invention, the SOC and output power conversion curves under the improved droop control when the charge and discharge working modes are switched, Fig.14 a is the SOC conversion curve, Fig.14 b is the output power conversion curve. DETAILED DESCRIPTION

[0049] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0050] like Figure 1 As shown in the figure, the multi-module parallel DC microgrid system includes several energy storage modules, each of which is connected to the grid through a bidirectional CLLLC energy storage converter. Distributed energy such as light energy, hydropower, and geothermal energy are powered by a Boost converter directly connected to the bus. The battery pack as the energy storage module realizes energy interaction with the grid through the CLLLC energy storage converter. The CLLLC energy storage converter has a high voltage gain characteristic, which can adapt to the voltage fluctuations of most buses and battery packs and realize energy interaction between the battery pack and the bus through a certain SOC control algorithm.

[0051] like Figure 2 As shown, the bidirectional CLLLC energy storage converter in this embodiment includes a symmetrical primary side structure, a secondary side structure and a transformer. A resonant element L is added to the secondary side of the LLC converter. r2 , C r2 A CLLLC topology with bidirectional symmetry is formed, where S1-S8 are MOSFET tubes, S1-S4 form a full-bridge circuit in the primary side structure, S5-S8 form a full-bridge topology in the secondary side structure, D1-D8 are the anti-parallel parasitic body diodes of S1-S8, C1-C8 are the parasitic capacitances of the switch tubes, and the resonant network in the topology is located between the two full-bridge structures, including the primary resonant inductor L r1 , resonant capacitor C r1 , and its symmetrical inductor and capacitor L r2 , C r2 , Excitation inductance L m , the transformer T ratio is n:1, in the two full-bridge structures, C in , C out They are the input and output capacitors, which play a filtering role, V in , V outThe input and output voltages on both sides of the converter are connected to the primary side of the converter, which is generally connected to the DC bus, and the secondary side output is the battery or load voltage.

[0052] When energy flows from the primary side of the converter to the secondary side, MOSFET tubes S1, S4 and S2, S3 receive complementary drive signals. After the DC grid bus voltage is input into the primary inverter network, a voltage with an amplitude of V is formed at points A and B. in , -V in The square wave is converted back and forth; the current output from the inverter side is input into the secondary side rectifier network of the converter through the resonant network. The various components in the converter satisfy the following relationship:

[0053]

[0054] Where: r1 ,ω r2 is the resonant angular frequency of the nonlinear elements on the primary and secondary sides of the resonant network; Q is the quality factor.

[0055] The resonant components on the primary and secondary sides of the converter are combined and calculated, and the transfer function obtained is:

[0056]

[0057] Combining (1) and (2), we obtain:

[0058]

[0059] Assuming that the converter works in a quasi-resonant state, f s =f r , at this time, the imaginary part of formula (2-19) is 0, and the imaginary part relationship is:

[0060]

[0061] Define the relationship expression of parameters a, b, c as:

[0062]

[0063] Substituting equation (5) into equation (4), we can obtain the following relationship:

[0064]

[0065] In order to consider the actual engineering needs, L m The value should be much larger than L r , so r 2 / ω r1 2 After taking the square root, we should take the larger of the two solutions:

[0066]

[0067] The parameters in formula (2) are normalized, and the normalized parameters are:

[0068]

[0069] Among them, ω n is the normalized angular frequency; Q is the quality factor; Z1=jω s L r1 -j / ω s C r1 ; and then get:

[0070]

[0071] like Figure 3 As shown, the parameter design method of the CLLLC type energy storage converter resonant network in this embodiment includes:

[0072] Determine the transformer ratio n: When the converter operates in quasi-resonant mode, that is, f s =f r , at this time, the value of gain M is always 1, and the transformation ratio n can be determined according to the rated voltage of the converter:

[0073]

[0074] Among them, V N-in is the rated voltage of the DC bus, V N-out It is the rated voltage output by the secondary side structure.

[0076] Determine the maximum and minimum voltage gains of the converter as:

[0077]

[0078] Among them, V in.min is the minimum input voltage of the primary side structure, V out_max It is the maximum value of the output voltage of the secondary side structure.

[0079] Determine the parameters k and Q:

[0080] like Figure 4 As shown, establish I m [Z in / Z r ] and the maximum converter gain M ac_max The common area in the figure is the k and Q selection area to achieve ZVS under the premise of satisfying the converter voltage gain. It can be seen from the figure that the maximum value of k·Q is in the common area M ac_max The boundary of the public area is M. ac_maxThe maximum point on the boundary of ac_max =M ac (ω n.min ) into formula (9), the relationship between k and Q is obtained as follows:

[0081]

[0082] Let x = k·Q and we have:

[0083]

[0084] Let y = 0, and we get two solutions of equation (13):

[0085]

[0086] The k min , k max Substituting into x, we can see that when M ac_max <M ac (ω n.min ), k min , k max are the minimum and maximum stationary points respectively. When M ac_max ≥M ac (ω n.min ), then only k max , and the relationship between Q1 and Q2 corresponding to the two solutions is:

[0087]

[0088] In summary, when the maximum voltage gain of the converter M ac_max The voltage gain M of the converter at the minimum frequency ac (ω n.min ) satisfies M ac_max <M ac (ω n.min ), the parameter k is the voltage gain M in the common area. ac_max The maximum point k on the boundary of max , the quality factor Q at this time is Q2 in the above formula, otherwise, the parameter k is the minimum point k min , the quality factor Q is Q1 in the above formula.

[0089] Test the determined parameters k and Q:

[0090] Although the k and Q values ​​selected based on this can achieve ZVS, we should also consider the situation when the energy storage system is not connected to the load. At this time, Q = 0, and we only need to consider the value of k, and take ω n =ω n.max , if the converter operates at the maximum frequency M ac (ω n.max), its voltage gain needs to be less than the gain M required under no-load conditions. ac_min , that is: M ac_min ≥M ac (ω n.max ), so the parameter k needs to satisfy the following formula:

[0091]

[0092] Therefore, after determining the parameters k and Q, it is necessary to verify whether they satisfy equation (16). If they do, no further calculation is required. If they do not, let Recalculate Q again.

[0093] After determining the parameters k and Q, the resonant inductance L is determined by the following formula r1 , L r2 , resonant capacitor C r1 , C r2 and the magnetizing inductance L m for:

[0094]

[0095] Among them, R eq is the equivalent impedance of the converter, ω r is the resonant frequency of the converter.

[0096] like Figure 5 As shown, in a specific embodiment, based on the above-mentioned bidirectional CLLLC type energy storage converter, a droop control method of a multi-module parallel DC microgrid system is provided, comprising the following steps:

[0097] S1, according to the SOC of the i-th energy storage module i The value is used to obtain the current available power S of the energy storage module. i ;

[0098] In this embodiment, the current SOC of the i-th energy storage module i for:

[0099]

[0100] Among them, SOC i_0 is the initial charge state of the i-th energy storage module, C e is the rated capacity of the energy storage module, U dc is the converter output voltage, U Bati is the voltage output by the i-th energy storage module, i out i is the current output by the i-th energy storage module, the SOC i Represents the percentage of the remaining battery power, so the current available power S of the i-th energy storage module is obtained i.

[0101] S2, based on the current available power S of the i-th energy storage module i Calculate the adjustment factor z s And the SOC balance coefficient k of the i-th energy storage module i , and then get the droop coefficient z of the i-th energy storage module i ;

[0102] By adjusting the coefficient z s Compensation bus voltage, the adjustment coefficient z s for:

[0103]

[0104] Wherein, β is the amplification factor, β>0, in this embodiment, β=1.7; S is the total available power of the energy storage module m is the number of energy storage modules.

[0105] Taking the continuous charging of the energy storage module as an example, the adjustment coefficient z s The change process is divided into two stages: When the SOC remaining power is insufficient, the converter output power needs to be small to extend the battery working time and avoid battery damage. s >1, improved droop coefficient z i As S decreases and continues to increase, the discharge speed slows down, and the bus voltage deviation is reduced by selecting the amplification factor. When the power is relatively sufficient, z i As S decreases, it continues to increase, but z s is always less than 1, the coefficient z will be adjusted s Controlled within a smaller range, the converter output power is large, the battery discharge speed is fast, and the small adjustment coefficient is conducive to reducing voltage fluctuations and maintaining system stability, while avoiding the risk of battery overcharge and over-discharge. Compared with traditional droop control, under the action of the adjustment coefficient, the output power of the energy storage system decreases as the discharge proceeds. The droop control with the adjustment coefficient is more flexible and suitable for more complex charging and discharging occasions. At the same time, it solves the problem of sudden changes in the segmented droop control coefficient affecting system stability.

[0106] When the DC bus is connected to multiple energy storage modules in parallel, when multiple energy storage modules are connected in parallel, for modules with higher battery SOC, it is required to output high power during discharge and absorb low power during charging. For modules with lower battery SOC, it is required to output low power during discharge and absorb high power during charging. In order to overcome the problem of insufficient power distribution accuracy of traditional droop control, the SOC balancing coefficient k is introduced. i , which is used to further improve the accuracy of droop control power distribution and achieve SOC balance of each module battery.

[0107] The balancing coefficient k of the i-th energy storage module i for:

[0108]

[0109] Then the droop coefficient z of the i-th energy storage module is obtained i for:

[0110]

[0111] Where r is the speed factor. The power balance response speed of the energy storage system decreases as the speed factor r increases, but the system power convergence speed is proportional to the speed factor r. Therefore, if the speed factor n remains unchanged, the response speed or convergence speed of the system balance power will inevitably be sacrificed. In order to solve this problem, the speed factor r in this embodiment is:

[0112]

[0113] Among them, r min is the minimum speed regulation factor; Δp is the output power difference between the two converters; Δp max is the maximum difference in output power; α is a constant. A variable speed factor r with sinusoidal function characteristics is used. The speed factor r is gradually increased in the initial operation stage of the energy storage system. On the basis of ensuring the power balance response characteristics of the system, r reaches the maximum value in the system balance power convergence stage, ensuring the convergence speed of the system. Compared with the traditional power exponential droop control, the improved droop control shortens the system balancing time, can achieve multi-module power board balancing faster, and can take into account the response speed and convergence speed of the system. Compared with the segmented power exponential droop control improved according to time, the improved droop control speed factor r does not change suddenly, and has adaptive adjustability with power fluctuations, and the power output curve is smoother, which shortens the system balancing time while ensuring the stability of the system.

[0114] Taking the continuous charging of the energy storage module as an example, the adjustment coefficient z s The change process is divided into two stages: <S<S min When the SOC remaining power is insufficient, the converter output power needs to be small to extend the battery working time and avoid damaging the battery. s >1, droop coefficient z i As S decreases and continues to increase, the discharge speed slows down, and the bus voltage deviation is reduced by selecting the amplification factor. min <S<S max When the battery is relatively sufficient, i As S decreases, it continues to increase, but z s is always less than 1, the coefficient z will be adjusted sControlled within a smaller range, the converter output power is large, the battery discharge speed is fast, and the small adjustment coefficient is conducive to reducing voltage fluctuations and maintaining system stability, while avoiding the risk of battery overcharge and over-discharge. Compared with traditional droop control, droop control with adjustment coefficient is more flexible and adaptable to more complex charging and discharging occasions. It also solves the problem of sudden changes in segmented droop control coefficient affecting system stability.

[0115] S3, the droop coefficient z i The PWM control signal is obtained by multiplying the output power of the energy storage module and undergoing voltage and current dual closed-loop control to realize the charge and discharge control of the bidirectional CLLLC energy storage converter.

[0116] In summary, the droop control expression in this embodiment is:

[0117]

[0118] Where: U dci is the output voltage of the i-th converter port; U dc is the DC bus voltage; ΔU dci is the bus voltage offset value caused by the droop control of the i-th converter.

[0119] In order to further verify the above embodiment, Matlab / Simulink simulation software is used to simulate and analyze the designed converter parameters to determine the CLLLC type energy storage converter prototype parameters as shown in Table 1 below.

[0120] Table 1 Parameters of CLLLC energy storage converter prototype

[0121]

[0122] Then, a simulation model of two batteries connected to the CLLLC energy storage converter in parallel to the DC bus was built in Matlab / Simulink. The simulation results of the traditional droop control and the improved droop control working in the charge and discharge mode and the load mutation were compared. The proposed improved droop control was simulated and verified. Taking the system discharge operation as an example, the DC bus voltage U dc =420V, its rated output power is P bus =10kW, battery output voltage U Bat * =48V, the load is purely resistive, and its rated power is P Load = 10kW When the system is running for 4s, a sudden load causes the load to consume power P Load Increased to 13kW, improved droop control and segmented droop control and traditional droop control simulation waveforms such as Figure 6 shown.

[0123] like Figure 6 As shown in a, the improved droop coefficient increases with the decrease of SOC. At 4s, the power required by the load suddenly increases. At this time, the SOC of the two energy storage modules is the initial value of 0.9. At this time, the adjustment coefficient z s With the improved droop coefficient taking the minimum value, under the improved dynamic droop control, its bus voltage is 419.3V, which is better than 418.6V of dynamic droop control and 417.5V of traditional droop control. The output power of its energy storage system is about 2.97kW, which is better than 2.9kW of dynamic droop control and 2.82kW of traditional droop control. Its transmission efficiency is improved by about 5% compared with traditional droop control. As the energy storage system continues to discharge, the battery SOC gradually decreases. At 80s, the SOC drops to 0.3, and the output power is 2.82kW, which is the same as the droop control. Under the improved dynamic droop control, when the energy storage system has sufficient available power, its output power and discharge amount are significantly better than traditional droop control. Compared with the segmented droop control, the bus voltage does not change suddenly, the compensation effect is good, and the voltage offset is small.

[0124] In order to verify the feasibility of the speed regulation factor with sinusoidal function characteristics, in the charging mode, this section uses the parameters shown in Table 2 to simulate and compare the improved droop control, time-based variable speed droop control and power exponential droop control.

[0125] Table 2 Simulation parameters based on droop control strategy in charging mode

[0126]

[0127] Under the power exponential droop control strategy, the speed regulation factor r is taken as 2, 5, and 7 for simulation verification. The simulation results are as follows: Figure 7 As shown. Figure 7 It can be seen that when the energy storage system works in charging mode, the control strategy can achieve SOC balance, and the balance speed is related to the speed factor n. The response speed and convergence speed of the energy storage system output power are affected by the value of the speed factor n. In the output power response stage, the smaller the value of n, the faster the response speed. When r=7, the energy storage system reaches the critical value of the power response stage in the 8th second. In the output power convergence stage, the larger the value of n, the faster the response speed, and the faster the energy storage system charges. Under the control of the three speed factors, the output power of the two energy storage modules eventually approaches the same, and the SOC state also tends to be balanced. In the charging mode, the larger the value of n, the smaller the system output power balance time. The simulation results verify the influence of the speed factor n on the balance speed of the energy storage system in the charging mode.

[0128] Under the time-based variable speed factor r droop control, r min Take 1, Δt=Δr=1, n max=7, ensuring that the speed regulation factor r increases by 1 every 1s, ensuring that it reaches the maximum when the energy storage system is at the critical value of the power response stage. The relationship expression is:

[0129]

[0130] The converter output power and SOC equilibrium speed under the time-based variable speed factor r droop control are simulated respectively, as shown in Figure 8 As shown. Figure 8 b. Under the time-based variable speed factor droop control, the energy storage system reaches the output power response critical point at 7s, and the output power reaches equilibrium at 43s, shortening the power balancing response time of the energy storage system. When the system runs to the output power response critical point, its speed factor reaches the maximum value, ensuring the convergence speed of the system. The simulation verifies that the variable speed factor can shorten the system power balancing response time and speed up the overall SOC balancing speed while taking into account the system response speed and convergence speed. Figure 4-14 b) It can be seen that although this control method improves the response speed of the output power, the sudden change of the droop coefficient will cause the output power of the energy storage system to jitter in the initial operation, and the power curve is not smooth enough, affecting the stability of the system. Based on this, this embodiment adopts a speed regulation factor droop control with a sinusoidal function characteristic.

[0131] In this embodiment, a speed regulation factor droop control with a sinusoidal function characteristic is adopted, and the speed regulation factor simulation waveform is as follows: Fig. 9 As shown. Fig. 9 It can be seen that the speed regulation factor r with sinusoidal function characteristics continues to increase before the power imbalance reaches the maximum. The curve of this control method changes smoothly, and the coefficient does not change suddenly. It has the characteristics of a sinusoidal function with a steeper slope at the beginning of the response and a gentler slope at the end, and it conforms to the curve characteristics of the output power imbalance in the response stage.

[0132] The converter output power and SOC equilibrium speed under the speed regulation factor r droop control with sinusoidal function characteristics are simulated respectively, as shown in Fig.10 shown.

[0133] contrast Fig.10 , Fig. 9 It can be seen that under the droop control of the speed regulation factor r with sinusoidal function characteristics, the SOC balancing speed is faster. In the response stage of the energy storage system, the power output curve is smoother, and the output curve has a more obvious upward trend. The system response critical point is reached at 5s, and the output power reaches equilibrium at 35s. The improved droop control has a more stable output power and a faster output power response speed, which effectively shortens the system output power balancing time. The improved droop control introduces the adjustment coefficient z s, when the SOC is small, the output power is large, and when the SOC is large, the output power is small.

[0134] In the discharge mode, in order to verify the limitation of the improved droop control on over-discharge, the initial state of SOC should not be too high. The parameters shown in Table 3 are used to simulate and compare the improved droop control, time-based variable speed droop control and power exponential droop control.

[0135] Table 3 Simulation parameters based on droop control strategy in discharge mode

[0136]

[0137] Under the power exponential droop control strategy, the speed regulation factor r is taken as 2, 5, and 7 for simulation verification. The simulation results are as follows: Fig.11 As shown. Fig.11 It can be seen that when the energy storage system works in charging mode, the control strategy can achieve SOC balance. The larger the value of the speed regulation factor r, the slower the SOC decrease rate, and the balancing speed is related to the speed regulation factor r. The response speed and convergence speed of the energy storage system output power are affected by the value of the speed regulation factor r. The simulation analysis is similar to that of the charging mode and will not be repeated here. The simulation results verify the influence of the speed regulation factor r on the balancing speed of the energy storage system in the discharge mode.

[0138] Under the time-based variable speed factor r droop control, the system works in the discharge operation state, and the converter output power and SOC equilibrium speed under the time-based variable speed factor r droop control are simulated respectively, as shown in FIG. Fig.12 As shown. Fig.12 It can be seen that when the energy storage system works in the discharge mode, the time-based variable speed factor droop control can shorten the power balance response time and speed up the SOC balance speed. The simulation results are consistent with the variable speed factor theory, but the output power curve jitter problem caused by the sudden change of the speed factor still exists. The converter output power and SOC balance speed under the speed factor r droop control with sinusoidal function characteristics are simulated respectively. Fig.13 As shown. Fig.12 , Fig.13 It can be seen that in the discharge mode, the working time of the energy storage system is significantly extended, the output power response critical point is reached at 5.5s, and the output power reaches equilibrium at 35s. The improved droop control output power response speed is faster, which effectively shortens the system output power equilibrium time. s Under the improved droop control, the output power decreases as the SOC decreases, effectively extending the working time of the energy storage system. In order to verify whether the improved droop control can still achieve fast SOC balance and power sharing under the charge and discharge mode conversion, this section uses the parameters shown in Table 4 to simulate and analyze the improved droop control.

[0139] Table 4 Simulation Parameter Table Based on Droop Control Strategy under Charge and Discharge Modes

[0140]

[0141] In the initial stage of the simulation, the energy storage system operates in the charging mode, and the output power of the DC bus stabilizes at 15 kW. When the system operates at the 20th second, the output power of the DC bus is insufficient to keep the load running stably, and the energy storage system switches to the discharging mode. When the system operates at the 60th second, the output power of the DC bus is sufficient, and the energy storage system switches back to the charging mode again. The simulation waveforms of its SOC balance and output power are as Fig.14 shown.

[0142] Fig.14 In the figure, the improved droop control can still achieve SOC balance control when operating in the environment of charge and discharge mode switching. From 0 s to 20 s, the energy storage system operates in the charging mode, SOC1 > SOC2, and the output power of energy storage module 1 is lower than that of energy storage module 2. From 20 s to 48 s, the system operates in the discharging mode, SOC1 < SOC2, and the output power of energy storage module 1 is higher than that of energy storage module 2. The output powers of the two modules tend to be the same at the 58th second. The simulation results show that this control method can still achieve output power balance when the energy storage system encounters a sudden change in rated power and has good stability.

[0143] In summary, the present invention improves based on the traditional droop control, introduces an adjustment coefficient, while preventing the system from overcharging and over-discharging, extends the working time of the system. In order to realize the reasonable distribution of the output power of multiple energy storage modules, an equalization coefficient with the available electricity as the independent variable is added to improve the power distribution accuracy. To improve the power distribution response speed and solve the problem of output power jitter caused by the time-based variable speed regulation factor droop control, a variable speed regulation factor with sinusoidal function characteristics is constructed. On the premise of being able to take into account both the output power response speed and the convergence speed of the system, the output power curve of the energy storage system is smoother, ensuring the stability of the system. Finally, a simulation model of two energy storage modules connected in parallel to the DC bus is built in the Matlab / Simulink software, and the simulation results are consistent with the theory, verifying the feasibility of this control strategy.

[0144] Based on the above analysis, the dynamic droop coefficient adjustment is smoother, the compensation effect is better, and the adjustment accuracy is higher. It can output a large power compensation to the bus voltage to keep the system stable when the electricity is sufficient, extend the working time when the electricity is insufficient, prevent overcharging and over-discharging to extend the service life of the battery, and can adapt to more complex energy storage working scenarios.

[0145] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A multi-module parallel DC microgrid system, comprising a plurality of energy storage modules, each of which is connected to a power grid through a bidirectional CLLLC energy storage converter, wherein the bidirectional CLLLC energy storage converter comprises a symmetrical primary side structure, a secondary side structure and a transformer, wherein the primary side structure and the secondary side structure both comprise a full-bridge circuit and a resonant circuit consisting of four switch tubes, characterized in that: When the maximum voltage gain of the converter is M ac_max The voltage gain M of the converter at the minimum frequency ac (ω n.min ) satisfies M ac_max <M ac (ω n.min ), the parameter k is the voltage gain M in the common area. ac_max The maximum point k on the boundary of max , otherwise, parameter k is the minimum point k min ;According to the parameter k, the quality factor Q is: When the selected parameter k value does not satisfy: When the parameter k is set to The common area is the area where the parameter k and the quality factor Q are selected to achieve ZVS under the premise of satisfying the voltage gain of the converter; the parameter k is: k = L m / L r1 , L r1 and L m They are the resonant inductance and excitation inductance of the primary side structure respectively.

2. The multi-module parallel DC microgrid system according to claim 1, characterized in that: The maximum voltage gain of the converter is: Where n is the transformer ratio, V in.min is the minimum input voltage of the primary side structure, V out_max It is the maximum value of the output voltage of the secondary side structure.

3. The multi-module parallel DC microgrid system according to claim 2, characterized in that: The transformer ratio n is: Among them, V N_in is the rated voltage of the DC bus, V N_out It is the rated voltage output by the secondary side structure.

4. The multi-module parallel DC microgrid system according to claim 1, characterized in that: Excitation inductance L m for: Among them, R eq is the equivalent impedance of the converter; ω r is the resonant frequency of the converter.

5. A droop control method for a multi-module parallel DC microgrid system, characterized in that: The droop coefficient z of the i-th energy storage module i The output power of the energy storage module is multiplied and subjected to voltage and current double closed-loop control to obtain a PWM control signal, thereby realizing the charge and discharge control of the corresponding bidirectional CLLLC energy storage converter; The droop coefficient z of the i-th energy storage module i for: Among them, k i is the SOC balance coefficient of the i-th energy storage module, z s is the adjustment coefficient; r is the speed control factor; S i is the available power of the i-th energy storage module.

6. The droop control method of the multi-module parallel DC microgrid system according to claim 5, characterized in that: The adjustment factor z s for: with s =e β(0.2-s) ; Wherein, β is the amplification factor, β>0, and S is the total available power of the energy storage module.

7. The droop control method of the multi-module parallel DC microgrid system according to claim 6, characterized in that: The SOC balancing coefficient k of the i-th energy storage module i The relationship is: Wherein, m is the number of energy storage modules.

8. The droop control method of the multi-module parallel DC microgrid system according to claim 6, characterized in that: The speed regulation factor r is: Among them, r min is the minimum speed regulation factor; Δp is the output power difference between the two converters; Δp max is the maximum difference in output power, and α is a constant.

Images