Hybrid energy storage system charge state self-recovery method based on double-layer fuzzy control and high and low frequency compensation link

Abstract

The invention relates to a hybrid energy storage system charge state self-recovery method based on double-layer fuzzy control and a high and low frequency compensation link. The method comprises the following steps: obtaining the output current of each converter through a droop control relational expression; taking a hybrid energy storage system SOC and a DC bus voltage normalization value as a first-layer fuzzy recovery control input quantity and an output recovery quantity; taking the SOC of the hybrid energy storage system as a second-layer fuzzy constraint control input quantity, and outputting a constraint coefficient; performing derivative value judgment on the first-order derivative and the second-order derivative of the power of the super capacitor to obtain high-frequency and low-frequency band signals; the high-low frequency band signal, the derivative of the current of the supercapacitor and the difference value between the output voltage of the supercapacitor converter and the reference value of the supercapacitor converter are subjected to recovery compensation to output a recovery compensation amount; and the high and low frequency band signals, the recovery quantity and the recovery compensation quantity output by the recovery compensation link are subjected to decoupling compensation to output the recovery quantity with decoupling compensation. According to the invention, the SOC recovery effect of the energy storage system can be improved, and reliable operation of the optical storage DC micro-grid is ensured.

Description

Technical Field

[0001] The present invention relates to a method for self - recovery of the state of charge (SOC) of a hybrid energy storage system based on double - layer fuzzy control and high - and low - frequency compensation links, which is applicable to the fields of micro - grid technology and energy storage technology. Background Technique

[0002] In recent years, in order to solve problems such as environmental pollution and energy shortage, the penetration rate of renewable energy in micro - grids has been increasing. Due to the uncertainty, randomness, and volatility of the output of renewable energy, an energy storage system is required in the micro - grid to regulate power, smooth the output fluctuations of renewable energy, and improve the operation stability of the micro - grid.

[0003] To avoid over - charging and over - discharging of the energy storage system, it is also necessary to perform recovery control on the state of charge of the system to ensure that the energy storage system always operates within a safe range. Because the capacity of the battery is large, currently, the SOC of the supercapacitor is mostly restored by modifying the virtual resistance - capacitance value and using adjustment coefficients, without considering the SOC of the battery. When the system operates for a long time, there is a risk of SOC exceeding the limit. The reasonable distribution of power and the coordinated recovery of the SOC of both can be achieved through intelligent optimization algorithms such as the particle swarm algorithm or optimal control. At the same time, power over - limit constraints can be added to avoid power over - limit. However, there is still a problem that the supercapacitor and the battery hinder each other's recovery effect during the SOC recovery process. At the same time, frequency - division control and SOC recovery control are essentially power distribution, and the two controls will affect each other through coupling during operation. There is an urgent need to propose a method for self - recovery of the state of charge of a hybrid energy storage system based on double - layer fuzzy control and high - and low - frequency compensation links to improve the SOC recovery effect of the energy storage system and ensure the reliable operation of the photovoltaic - energy - storage DC micro - grid. Summary of the Invention

[0004] In order to achieve the above object, the present invention provides a method for self - recovery of the state of charge of a hybrid energy storage system based on double - layer fuzzy control and high - and low - frequency compensation links. The method includes the following numerical simulation steps:

[0005] Obtain the output current of each converter through the droop - control relationship:

[0006]

[0007] Where: I 1 , I 2 are the output currents of the supercapacitor converter and the battery converter respectively; R is the virtual resistance; C is the virtual capacitance; I is the output current of the hybrid energy storage system; G sc (s), G bat (s) are the transfer functions of the output currents of the supercapacitor and the battery converters respectively.

[0008] Normalize the DC bus voltage:

[0009]

[0010] Where: U dcn is the normalized value of the DC bus voltage; U dcmax , U dcmin are the maximum and minimum allowable voltage values of the DC bus voltage respectively, as shown in the following formula:

[0011] U dcmax =(1 + 5%)U dc_ref ;

[0012] U dcmin =(1 - 5%)U dc_ref ;

[0013] Where U dc_ref is the reference voltage of the DC bus.

[0014] Take the SOC of the battery, the SOC of the supercapacitor, and the normalized value of the DC bus voltage as the input variables of the first-layer fuzzy restoration control, and output the restoration quantity. The first-layer fuzzy restoration control rule table is:

[0015]

[0016] Where: S SOC.sc is the SOC value of the supercapacitor; H sc is the restoration quantity of the supercapacitor before constraint.

[0017] Take the SOC of the hybrid energy storage system as the input variable of the second-layer fuzzy constraint control, and output the constraint coefficient. The second-layer fuzzy constraint control rule table is:

[0018]

[0019] Where: S SOC.bat is the SOC value of the battery; K sc is the constraint coefficient of the supercapacitor.

[0020] Divide the interval according to the magnitudes of the first derivative and the second derivative of the supercapacitor power, and add a delay link to obtain high and low frequency signals.

[0021] According to the derivative of the supercapacitor current, the difference between the output voltage of the supercapacitor converter and its reference value, and the high and low frequency signals, add a compensation quantity to the difference I' 1 between the output voltage U sc of the supercapacitor converter and its reference value. Add a compensation quantity to the reference value of the supercapacitor current, and add a compensation quantity to the supercapacitor current through the subsequent PI controller and arithmetic symbols.

[0022] The restored compensation amount output by the high and low frequency signals, the restoration amount, and the restoration compensation link is decoupled and compensated to output a restored amount with decoupling compensation. Description of the Drawings

[0023] Figure 1 is the structural framework of a DC microgrid with a hybrid energy storage system;

[0024] Figure 2 is the double-layer fuzzy SOC self-restoration control block diagram;

[0025] Figure 3 is the restoration compensation control flow chart;

[0026] Figure 4 is the decoupling compensation control flow chart;

[0027] Figure 5 is the fuzzy PI control block diagram;

[0028] Figure 6 is the SOC restoration curve of the battery under different control strategies;

[0029] Figure 7 is the power comparison diagram under the restoration control of the supercapacitor;

[0030] Figure 8 is the power of the supercapacitor without decoupling compensation;

[0031] Figure 9 is the power of the supercapacitor with decoupling compensation;

[0032] Figure 10 is the power comparison diagram of the supercapacitor under the restoration frequency division decoupling control;

[0033] Figure 11 is the SOC restoration curve of the supercapacitor under different control strategies. Detailed Implementation Manner

[0034] The present invention is based on a Figure 1 shown structural framework of a DC microgrid with a hybrid energy storage system, which consists of distributed photovoltaic power generation, a hybrid energy storage system (composed of a supercapacitor and a battery), DC loads, AC loads, and interface converters. The supercapacitor and the battery in the hybrid energy storage system respectively adopt droop control based on virtual capacitance and virtual resistance, and the droop control relationship formula is:

[0035]

[0036] Where: U dc_ref is the DC bus reference voltage; U 1 , I 1are the output voltage and output current of the supercapacitor converter respectively; U 2 , I 2 are the output voltage and output current of the battery converter respectively; R is the virtual resistance; C is the virtual capacitor; I is the output current of the hybrid energy storage system.

[0037] The output currents of each converter are obtained through the droop control relationship:

[0038]

[0039] where: G sc (s), G bat (s) are the transfer functions of the output currents of the supercapacitor and the battery converter respectively.

[0040] Normalize the DC bus voltage:

[0041]

[0042] where: U dcn is the normalized value of the DC bus voltage; U dcmax , U dcmin are the maximum and minimum allowable voltage values of the DC bus voltage respectively, as follows:

[0043] U dcmax =(1 + 5%)U dc_ref ;

[0044] U dcmin =(1 - 5%)U dc_ref ;

[0045] In order to make the SOC of the hybrid energy storage system self-recover, a recovery amount is introduced into the voltage outer loop of the virtual capacitor and virtual resistor control branches respectively. The output power of the supercapacitor and the battery is redistributed through the recovery amount to achieve the purpose of recovering the SOC. Taking the battery SOC, supercapacitor SOC, and normalized value of the DC bus voltage as the first-layer fuzzy recovery control input quantities and outputting the recovery amount, the first-layer fuzzy recovery control rule table is:

[0046]

[0047] where: S SOC.sc is the SOC value of the supercapacitor; H sc is the recovery amount of the supercapacitor before constraint.

[0048] Add the second - layer fuzzy constraint control on the first - layer fuzzy recovery control. The recovery amount output by the first layer is constrained by the output constraint coefficient to ensure the SOC recovery effect. Taking the SOC of the hybrid energy storage system as the input of the second - layer fuzzy constraint control and outputting the constraint coefficient, the second - layer fuzzy constraint control rule table is as follows:

[0049]

[0050] Where: S SOC.bat is the SOC value of the battery; K sc is the constraint coefficient of the supercapacitor.

[0051] Finally, the block diagram of the double - layer fuzzy SOC self - recovery control is Figure 2 .

[0052] Divide the interval according to the magnitudes of the first - order derivative and the second - order derivative of the supercapacitor power, and add a delay link to obtain high - and low - frequency signals.

[0053] According to the derivative of the supercapacitor current, the difference between the output voltage of the supercapacitor converter and its reference value, and the high - and low - frequency signals, a compensation amount is added to the difference I' 1 between the output voltage U sc of the supercapacitor converter and its reference value. After passing through a PI controller, a compensation amount is added to the reference value of the supercapacitor current A compensation amount is added, and after subsequent PI controllers and arithmetic symbols, a compensation amount is added to the supercapacitor current. The flowchart of the recovery compensation control is Figure 3 .

[0054] The high - and low - frequency signals, the recovery amount, and the recovery compensation amount output by the recovery compensation link pass through decoupling compensation to output a recovery amount with decoupling compensation. The flowchart of the decoupling compensation control is Figure 4 .

[0055] Combine fuzzy control with traditional PI control, and use fuzzy control to perform real - time adaptive tuning and adjustment of PI parameters to improve the system robustness and reduce the workload of parameter tuning. The block diagram of the fuzzy PI control is Figure 5 , and the fuzzy control rule tables for the fuzzy control output quantities ΔP and ΔI are respectively:

[0056]

[0057] For easy understanding, in this embodiment, a photovoltaic - energy - storage DC micro - grid simulation model containing a hybrid energy storage system as shown in Figure 1 is built. The simulation waveforms of the SOC and power of the hybrid energy storage system are as shown in Figures 6 - 11 .

[0058] As shown in Figure 6As shown, the single-layer restoration control only improves the battery SOC by 0.013% compared to the traditional control, with a poor restoration effect and a great hidden danger of SOC exceeding the limit. While the double-layer restoration control improves the SOC by 0.2% compared to the traditional control and by 0.187% compared to the single-layer restoration control. The restoration effect is significantly enhanced, reducing the risk of the energy storage SOC exceeding the limit and increasing the SOC margin during the system operation.

[0059] As Figure 7 shown, when only the double-layer restoration control is added to restore the SOC, as in the curve of Strategy 2, the part of the restoration amount that plays a role in restoration will gradually decrease due to the characteristics of the capacitor in the RC filter, resulting in a gradual decrease in the restoration power and finally fluctuating around 0, no longer performing SOC restoration, and the restoration effect is very poor. When the restoration compensation is added at 0.73 s, as in the curve of Strategy 3, the SOC restoration power will not gradually decrease, but can maintain the maximum power that can be achieved under this working condition for constant-power SOC restoration, achieving the maximum restoration effect.

[0060] As Figure 8 shown, when no decoupling compensation is performed on the decoupling process, a relatively large power fluctuation of about 1.786×10 5 W will occur at 1 s, affecting the system stability. While through Figure 9 it can be seen that when the decoupling compensation is adopted, the supercapacitor generates a power fluctuation of about 0.133×10 5 W at 1 s, reduced to about 7% of the power fluctuation without the decoupling compensation control, which better improves the system stability.

[0061] As Figure 10 shown, from 0 s to 0.55 s and from 1 s to 1.27 s are the frequency division stages. According to Strategy 2 in the figure, it can be seen that the frequency division control and the SOC restoration control will be coupled and affect each other, making it impossible to achieve the frequency division effect of the droop frequency division traditional control. In contrast, the power of Strategy 3 of the present invention coincides with the power of Strategy 1 during the frequency division stage, and the frequency division control achieves the due frequency division effect, improving the power response speed and prolonging the service life of the energy storage device. At the same time, from 0.55 s to 1 s and from 1.27 s to 2 s, the power of Strategy 3 of the present invention coincides with the power of Strategy 2, achieving the same restoration effect and performing constant-power restoration with the maximum restoration power. The SOC restoration curve of the supercapacitor under this working condition is as Figure 11 shown, which can verify again that the frequency division control of the control strategy proposed in this paper is not affected and can make the SOC perform constant-power restoration with the maximum restoration power.

Claims

1. A method for self-recovery of state of charge (SOC) of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation, characterized in that: The output current of each converter is obtained through the droop control relationship; the DC bus voltage is normalized; the battery SOC, supercapacitor SOC, and DC bus voltage normalization value are used as the first-layer fuzzy recovery control input, and the recovery value is output; the hybrid energy storage system SOC is used as the second-layer fuzzy constraint control input, and the constraint coefficient is output; the first-order derivative and second-order derivative of the supercapacitor power are judged by the derivative value to obtain the high and low frequency band signals; the high and low frequency band signals, the derivative of the supercapacitor current, and the difference between the supercapacitor converter output voltage and its reference value are restored and compensated to output the recovery compensation value; The high and low frequency band signals, the recovery amount, and the recovery compensation amount output by the recovery compensation link are decoupled and compensated to output a recovery amount with decoupling compensation.

2. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The output current of each converter is: Where: I1, I2 are the output currents of the supercapacitor converter and the battery converter respectively; R is the virtual resistance; C is the virtual capacitance; I is the output current of the hybrid energy storage system; G sc (s), G bat (s) are the transfer functions of the output current of the supercapacitor and battery converter respectively.

3. The method for self-recovery of charge state of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The DC bus voltage normalization process is: Among them: U dcn is the normalized value of DC bus voltage; U dcmax , U dcmin They are the maximum and minimum voltage values ​​allowed by the DC bus voltage, as shown in the following formula: IN dcmax =(1+5%)U dc_ref ; IN dcmin =(1-5%)U dc_ref ; Among them U dc_ref is the DC bus reference voltage.

4. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The first-layer fuzzy recovery control rule table is: Where: S SOC.sc is the SOC value of the supercapacitor; H sc is the recovery amount of the supercapacitor before constraint.

5. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The second-layer fuzzy constraint control rule table is: Where: S SOC.bat is the SOC value of the battery; K sc is the constraint coefficient of the supercapacitor.

6. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The high and low frequency band signals are obtained by adding a delay link to the intervals divided according to the size of the derivative value.

7. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The recovery compensation is the difference between the output voltage U1 of the supercapacitor converter and its reference value I s ' c Add compensation, and the supercapacitor current reference value is adjusted by PI controller Add compensation amount, and add compensation amount to the supercapacitor current through subsequent PI controller and operation symbol.

8. The method for self-recovery of the state of charge of a hybrid energy storage system based on double-layer fuzzy control and high and low frequency compensation according to claim 1 is characterized in that: The decoupling compensation is to add the decoupling compensation amount to the restoration amount according to the high and low frequency band signals and the restoration amount value.

Images