An island micro-grid distributed cooperative control method based on local information and a storage medium

Abstract

The application discloses a kind of based on local information's island microgrid distributed cooperative control method and storage medium, belong to the technical field of microgrid, island microgrid system includes network type control and follow network type control's converter.Follow network type control according to local information, and using maximum power MPPT tracking algorithm controls power output.Construction network type control and parallel converter reactive power is exported and distributed according to fixed voltage regulation droop coefficient;Construction network type control and parallel converter active power is exported and distributed according to the frequency droop coefficient based on local information adaptive update.The application realizes that the power instruction distribution of island microgrid follow network type control and construction network type control is generated by local information, without the instruction distribution of upper centralized control, can realize island microgrid inverter's plug and play, flexible reconfiguration and decentralized control, with good practicability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of microgrids, and particularly relates to a local-information-based distributed cooperative control method for an island microgrid and a storage medium. BACKGROUND

[0002] With the rise of the concept of "zero carbon", the island microgrid, which is not directly connected to the large power grid, has a clear carbon accounting boundary that can avoid the doubt of washing green zero carbon with grid connection, and can also be economical in special scenarios due to the reduction of over-network fees and capacity fees. Therefore, the island microgrid becomes an important support carrier for coping with foreign "carbon barriers" restrictions and realizing green manufacturing and export. In terms of configuration, the island microgrid is a heterogeneous interconnected system composed of network-forming control and network-following control. The network-forming control is mainly applied to energy storage systems, and the network-following control is mainly applied to wind turbine systems, photovoltaic systems, hydrogen energy systems, etc. In terms of control, the current island microgrid adopts high, medium and low frequency three-layer cooperative control based on master-slave control to realize stable operation. The bottom layer control relies on the network-forming energy storage to drive other network-following controls, the middle layer adopts secondary compensation control to coordinate the voltage and frequency between the converters to maintain the rated value, and the upper layer adopts centralized control to collect and process the information of each unit and issue instructions.

[0003] However, without the support and bottom of the large power grid, the delay, error and fault of the power instructions of each subject of the island microgrid will cause power imbalance, voltage and frequency deviation from the rated value of the system, and further cause the load equipment to be unable to work normally. In addition, the island microgrid itself has a very high proportion of new energy and power electronic characteristics, resulting in very low physical inertia, limited regulation capacity and frequent source and load fluctuations of the island microgrid, which will make the delay, error and fault of the power instructions likely to cause a chain of harm to the stability of the island microgrid. Therefore, in order to realize the efficient and reliable cooperation of multiple elements such as "wind, light, hydrogen and storage vehicles" of the zero-carbon island microgrid and improve the robust stability of the island microgrid, the power instructions of each converter need to be able to realize adaptive distributed allocation and adjustment according to local information, rather than relying on the communication transmission allocation of the central controller instructions.

[0004] The conventional island microgrid control does not realize adaptive allocation and adjustment of power instructions. In view of this problem, although there are currently some adaptive control strategies to optimize, these optimizations still take the power instructions P ref , Q ref as the optimization target to realize high-precision implementation, fast regulation speed, and as small overshoot fluctuation as possible in the process of executing the instructions issued by the central controller, and this optimization method still cannot get rid of the dependence of the power instructions P ref , Q refThe communication dependence and poor fault tolerance of the island micro-grid system have not been solved, and therefore, there is an urgent need for a scheme for adaptive and weak communication power coordination control and distribution based on local information of the island micro-grid system. SUMMARY

[0005] The application aims to provide an island micro-grid distributed collaborative control method and storage medium based on local information, which adaptively generates and adjusts power instructions through local information, and realizes plug-and-play, flexible reconstruction and decentralized control of the island micro-grid.

[0006] The application mainly realizes the technical scheme as follows:

[0007] An island micro-grid distributed collaborative control method based on local information, the island micro-grid system comprising network type control and network type control of the converter; comprising the following steps:

[0008] Step S1: the overall power of the island micro-grid satisfies the law of conservation of energy;

[0009] Step S2: the network type control controls the power output according to the local information and the maximum power MPPT tracking algorithm;

[0010] Step S3: the reactive power of the network type control parallel converter is output and distributed according to the fixed voltage regulation droop coefficient; the active power of the network type control parallel converter is output and distributed according to the frequency regulation droop coefficient adaptively updated based on the local information; the calculation formula of the frequency regulation droop coefficient adaptively updated based on the local information is:

[0011] k f = k f0 ×α SOC ×β T ×γ;

[0012] Wherein, k f0 =( ω n - ω min ) / P max ;

[0013] Wherein: k f is the frequency regulation droop coefficient;

[0014] k f0 is the initial value of the frequency regulation droop coefficient;

[0015] α SOC is the SOC adjustment factor;

[0016] β T is a temperature adjustment factor;

[0017] γ is a frequency change rate adjustment factor;

[0018] ω n is a rated angular frequency of the island microgrid system;

[0019] ω min is a minimum rated angular frequency allowing fluctuation;

[0020] P max is a maximum active power output of the grid-forming converter.

[0021] To better implement the present application, further, in the default state, when the state of charge SOC of the energy storage is in the interval [0, 20%], the charging α SOC is 0.5, the discharging α SOC is 2; when the state of charge SOC of the energy storage is in the interval [20, 40%], the charging α SOC is 0.8, the discharging α SOC is 1.5; when the state of charge SOC of the energy storage is in the interval [40, 60%], the charging α SOC is 1, the discharging α SOC is 1; when the state of charge SOC of the energy storage is in the interval [60, 80%], the charging α SOC is 1.2, the discharging α SOC is 0.8; when the state of charge SOC of the energy storage is in the interval [80, 100%], the charging α SOC is 1.5, the discharging α SOC is 0.5. Meanwhile, the SOC adjustment factor can be further updated with values in different intervals according to the type and state of the grid-forming energy storage to improve the adjustment performance.

[0022] To better implement the present application, further, when the microgrid frequency change rate |df / dt|≤threshold A1, γ=1; when threshold A1<|df / dt|≤threshold A2, γ=1-(|df / dt|-A1); when |df / dt|>threshold A2, γ=0.

[0023] To better implement the present application, further, when the temperature T of the energy storage is in the optimal working interval, β T =1; when the temperature T of the energy storage is lower than the lower limit of the optimal working interval, β T increases step by step with the decrease of temperature, and the attenuation rate of lithium precipitation in the low-temperature charging process is much higher than that in the discharging process; when the temperature T of the energy storage is higher than the upper limit of the optimal working interval, βT The step increases with the temperature until it is triggered to the high-temperature protection threshold, and then it is reduced to zero.

[0024] To better realize the present application, further, when the energy storage temperature T≤0, the charging β T is 3, the discharging β T is 2.5; when the energy storage temperature 0<T≤10, the charging β T is 2, the discharging β T is 1.5; when the energy storage temperature 10<T≤45, the charging β T is 1, the discharging β T is 1; when the energy storage temperature 45<T≤55, the charging β T is 1.5, the discharging β T is 1.2; when the energy storage temperature T>55, the charging β T is 0, the discharging β T is 0.

[0025] To better realize the present application, further, in the step S3, the local information includes the state of charge SOC of the energy storage, the energy storage temperature and the micro-grid frequency change rate df / dt; wherein the state of charge SOC of the energy storage and the energy storage temperature are collected by the battery management system and the thermal management system of the energy storage respectively; the micro-grid frequency change rate df / dt is obtained based on the current sampling analysis of the LC filter.

[0026] To better realize the present application, further, in the step S3, the grid-forming control further includes voltage secondary compensation and frequency secondary compensation, which need to keep weak communication between the grid-forming control and update to the droop control, and carry out PI correction through the distributed deviation of voltage and frequency; the voltage and current of the LC filter are collected, the real active power and reactive power are obtained through power calculation and filtering, the frequency droop coefficient, the secondary compensation frequency value and the secondary compensation voltage value are updated in the active ring and the reactive ring, and the voltage vector is generated; finally, the voltage and current loop based on the virtual impedance generates the switch driving signal SVPWM wave of the inverter.

[0027] To better realize the present application, further, in the step S2, when the grid-forming control enters the stable state, the angular frequency ω g and the voltage E g generated by the current through the local public point are detected, and if |ω pref -ω g |≤△ω and | E mpref - E g |≤△ E, the follow network type control accesses the public point, realizes the follow network type control and constructs the network control collaborative operation;Wherein: delta omega and delta E Respectively, the angular frequency access allowed threshold and the voltage access allowed threshold.

[0028] A computer readable storage medium, which stores a computer program, the program is executed by the processor to realize the above-mentioned one kind of based on local information's island microgrid distributed collaborative control method.

[0029] The beneficial effects of the present application are as follows:

[0030] The present application is an adaptive, weak communication collaborative control scheme for island microgrid system, based on SOC, temperature, frequency rate of change and other local information, and parallel network energy storage realizes the adaptive update of droop coefficient, which helps to improve the life of island microgrid system energy storage battery, system safety and power grid performance. The present application realizes that the power instruction distribution of island microgrid follow network type control and network type control is generated by local information, without the instruction distribution of upper centralized control, which can realize the plug and play, flexible reconstruction and decentralized control of island microgrid inverter, and has good practicability. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 It is a principle diagram of circuit topology for island microgrid system;

[0032] Figure 2 It is a principle diagram of network type control;

[0033] Figure 3 It is a principle diagram of follow network type control;

[0034] Figure 4 It is a principle diagram of centralized, distributed and decentralized communication control;

[0035] Figure 5 It is a voltage and current double closed loop control principle diagram of network control based on virtual impedance. DETAILED DESCRIPTION

[0036] Embodiment 1:

[0037] A kind of based on local information's island microgrid distributed collaborative control method, aims at realizing the plug and play, flexible reconstruction and decentralized control of island microgrid.

[0038] As Figure 1As shown, the circuit topology of the island microgrid system is mainly composed of grid-following (GFL) converters and grid-forming (GFM) converters, power sources, LC filters, inductive current sensors, LC filter voltage and current sensors, line impedance, point of common coupling (PCC), PCC voltage and current sensors, and loads. Among them, the voltage source of the grid-forming control parallel converter controlled by the battery energy storage and other subjects is E 1、 E 2,The current source of the grid-following converter controlled by the photovoltaic, wind turbine, hydrogen fuel cell and other subjects is I 1、 I 2.

[0039] The present application specifically includes the following steps:

[0040] 1、The overall power balance of the island microgrid satisfies P PV + P wind = P battery1 + P battery2 + P load + P line +∑ P i ;

[0041] Among them: P PV Ppv is the active power of photovoltaic power generation;

[0042] P wind Pw is the active power of wind power generation;

[0043] P battery1 P1 is the active power of grid-forming energy storage 1;

[0044] P battery2 P2 is the active power of grid-forming energy storage 2;

[0045] P load Pload is the active load;

[0046] P line Ploss is the line loss power;

[0047] P i Pout is the output power at the first i place.

[0048] 2. Grid-connected converters, based on local information such as irradiance, wind speed, and electrode voltage, etc. Figure 3 As shown, power output can be achieved through methods such as maximum power per minute (MPPT) based on photovoltaic voltage and current without requiring additional control modifications. Specifically: V 1PV Photovoltaic voltage; I 1PV Photovoltaic current; P pvref1 This is a reference value for the maximum power of photovoltaic power. I pvref1 This is the reference value for the output current of the photovoltaic inverter.

[0049] like Figure 3 As shown, grid-following control GFLs often control power output based on local information (illuminance, wind speed, electrode voltage, etc.) using methods such as maximum power point tracking (MPPT) algorithms. The microgrid's reference voltage is generated and maintained by the grid-following control. Once the grid-following control reaches a stable state, the following control will detect the angular frequency and voltage generated by the current passing through the local common point. If both the angular frequency and voltage are within the allowable range of the access conditions, i.e., |ω... pref -ω g |≤△ω and| E mpref - E g |≤△ E At the same time, it connects to the common point of the network control to achieve coordinated operation with the network control.

[0050] 3. Collect local information, including the energy storage's state of charge (SOC), energy storage temperature, and microgrid frequency change rate. The energy storage's SOC and temperature can be obtained from the energy storage's battery management system (BMS) and thermal management system (TMS), respectively, while the microgrid frequency change rate can be derived from the current sampling of the LC filter.

[0051] Based on the aforementioned local information, the grid-controlled parallel converter achieves adaptive updates of the droop coefficient. Although the local converter does not receive commands from the upper-level central controller at this time, i.e., the local reference power command P of the grid-controlled parallel converter... refThe power output and distribution of the grid-forming converter of the microgrid system is updated according to the adaptive adjustment of the droop coefficient, so as to balance the fluctuations and mutations of the grid-following converter and the load on the basis of the overall power balance of the microgrid system. The above analysis of the grid-following control and the grid-forming control in the island microgrid system is the basic principle of not depending on the instruction of the central controller.

[0052] Based on the above theory, although the grid-forming control can realize power distribution not depending on the communication of the central controller by using a fixed droop coefficient, the fixed droop coefficient will lead to poor flexibility of the microgrid system, and therefore the droop coefficient needs to be adaptively changed according to local information to realize the flexibility adjustment of the island microgrid system. This is because the design of the fixed droop coefficient is determined by the frequency / voltage range allowed to fluctuate and the maximum power release capacity of the microgrid system, as follows:

[0053] k f = (ω n - ω min ) / ω P max ;

[0054] wherein, k f is the frequency adjustment droop coefficient, ω n is the rated angular frequency of the microgrid system, ω min is the minimum rated angular frequency allowed to fluctuate, (ω n - ω min ) represents the maximum angular frequency fluctuation range allowed to fluctuate of the microgrid system, P max represents the maximum active power output of the microgrid system.

[0055] However, the maximum power output capacity of the grid-forming energy storage varies with the change of the state of the energy storage (i.e. the above-mentioned battery SOC, temperature, etc.). For the SOC, the charge and discharge capacity of the battery energy storage at different SOCs is not the same and has similar characteristics, for example, the charge capacity of the battery decreases with the increase of the battery SOC, and if the charge and discharge is performed beyond the power that the battery can withstand, it will cause lithium current to be generated at the negative electrode of the battery, thereby being detrimental to the service life and safety of the energy storage. In addition, if the initial SOCs of the parallel grid-forming energy storage batteries are different, the power distribution of the fixed droop coefficient will also cause the energy gap of the parallel energy storage to become larger and larger, which is not conducive to the energy proportioning and the stability support to the island microgrid.

[0056] For temperature, as the most important factor of battery safety, temperature also directly affects the charging and discharging power of the battery. On the one hand, the capacity of the battery decreases with the decrease of temperature, and the lithium precipitation phenomenon of overcurrent is also intensified at low temperature; on the other hand, too high temperature may cause damage to the integrated switching device, and may also be affected by the faults such as thermal runaway and electric arc, at this time, the temperature should be reduced to avoid further expansion of the harm.

[0057] For the frequency change rate df / dt, since the island microgrid is prone to frequency fluctuation, when the small fluctuation of the frequency change rate is within the allowable range, the fixed droop coefficient is still adopted for adjustment; when the frequency change rate is too large, in order to suppress or slow down the impact of the sharp change of the frequency on the system, since the frequency is a global variable, all network-forming controls need to reduce the droop coefficient, so that the same frequency deviation can trigger a larger power response, thereby suppressing the sharp change of the frequency; at the same time, the power proportional distribution between the parallel network-forming controls does not change.

[0058] In summary, SOC, temperature and frequency change rate are the key elements for the island microgrid system to realize adaptive power collaborative control and distribution based on local information and weak communication. In actual operation, more local information can be coupled to update the droop coefficient adaptively, such as the battery health state (SOH) provided by EMS.

[0059] The principle of adaptive adjustment of network-forming droop control for active power distribution of island microgrid is that since the frequency as a global variable is not affected by the difference of line impedance, the frequency droop coefficient and active power of the parallel converter of network-forming control are respectively set as k f1 and P 1、 k f2 and P 2、… k fi and P i When the island microgrid system no longer uses the central controller to issue instructions, i.e. P 1= P 2= P 3=…= P i =0, at this time, the frequency droop coefficient and active power satisfy k f1 · P 1= k f2 · P 2= k f3 · P 3…= k fi ·P i ; similarly, let the voltage droop coefficient and reactive power of the voltage source converter be k v1 and , k v2 and ,… k vi and , while the voltage as a local variable will be affected by the difference in line impedance, let the corresponding line impedance of the grid-forming controlled parallel converter be X l1 , X l2 ,… X li When the product of line impedance and reactive capacity is equal, that is X l1 · = X l2 · = X l3 · …= X li · , at this time the voltage droop coefficient and reactive power satisfy k v1 · = k v2 · = k v3 · …= k vi · . Therefore, for the grid-forming controlled parallel converter, when the power dispatch reference instruction is 0, its active power is inversely proportional to the frequency modulation coefficient (P P ∝1 / k f ), and its reactive power is also inversely proportional to the voltage droop coefficient under ideal line impedance conditions (Q ∝1 / k v ).

[0060] Due to the requirement of microgrid power factor and the actual impedance characteristics mainly resistive and inductive, the reactive power of the grid-forming controlled parallel converter still adopts a fixed voltage droop coefficient for distribution.

[0061] As shown in Figure 2 , the active power of the grid-forming controlled parallel converter adopts an adaptive frequency droop coefficient based on local information such as SOC, temperature, and microgrid frequency change rate for distribution, and its relationship is as follows:

[0062] k f = k f0 ×α SOC ×β T ×γ;

[0063] wherein: k f denotes a frequency modulation coefficient, k f0 denotes an initial value of the frequency modulation coefficient, a SOC denotes an SOC adjustment factor, b T denotes a temperature adjustment factor, g denotes a frequency variation rate adjustment factor.

[0064] SOC adjustment factor a SOC can enhance the energy storage life, the optimal charge and discharge current rate (power) of energy storage at different SOC gradients is different, and adaptive adjustment of power distribution according to different energy storage SOC can improve the stability and life of network energy storage, thereby being beneficial to long-term operation and maintenance economy. Temperature adjustment factor b T can improve the safety of energy storage and avoid large power charge and discharge at low temperature and accelerated side reactions at high temperature. Frequency variation rate adjustment factor g can improve the stability performance of island microgrid.

[0065] Initial value of frequency modulation coefficient k f0 is obtained by the product of the allowable fluctuation range of angular frequency and the maximum value of the converter power, and the specific expression is: k f0 = (ω n - ω min ) / P max .

[0066] Preferably, first, the SOC adjustment factor a SOC is analyzed, at this time k f = k f0 · a SOC ;

[0067] Assuming that the capacity / power ratio of battery1 and battery2 is 1:n, and the corresponding droop coefficients of network type control are k f1 and k f2 , respectively. Then the fixed droop coefficient under the conventional control is k f10 : k f20 = n:1, and the variable droop coefficient is definedk f1 = k f10 ·α SOC1 and k f2 = k f20 ·α SOC2 .

[0068] When the energy storage is charged, the smaller the SOC is, the more the charging power should be satisfied, that is, the larger the charging power allocation is, and the smaller the droop coefficient is; the larger the SOC is, the more the charging power should be satisfied, that is, the smaller the charging power allocation is, and the larger the droop coefficient is; at the same time, the power capacity of the energy storage itself should be considered, especially in the high SOC area to prevent lithium precipitation. Therefore, the network construction energy storage adjustment factor a of the charging state SOC increases with the increase of SOC.

[0069] When the energy storage is discharged, the smaller the SOC is, the more the discharging power should be satisfied, that is, the smaller the discharging power allocation is, and the larger the droop coefficient is; the larger the SOC is, the more the discharging power should be satisfied, that is, the larger the discharging power allocation is, and the smaller the droop coefficient is; therefore, the network construction energy storage adjustment factor a of the discharging state SOC decreases with the increase of SOC.

[0070] In order to avoid unnecessary overshoot and shock of the system caused by frequent changes of the droop coefficient, the adjustment dead zone is set to 20% in combination with the state interval characteristics of the battery SOC. The charging and discharging current boundaries of the energy storage battery under different SOC states are calibrated, and the corresponding adjustment factor a SOC .

[0071] Here, according to the experience value of the charging and discharging rate of the conventional lithium battery under different SOC, the adjustment factor a of the battery energy storage system under different SOC intervals can be obtained SOC roughly corresponding to the change table, as shown in Table 1. In the specific implementation process, a SOC can be calibrated and updated according to the specifications of the energy storage battery.

[0072] Table 1 a SOC value information under different SOC states

[0073]

[0074] Since the state of the battery energy storage in the charging and discharging SOC interval is not completely corresponding, the adjustment factor a SOCThe charging [0, 20%], [20%, 40%], [40%, 60%] and discharging [80%, 100%], [60%, 80%], [40%, 60%] are in symmetrical relationship. The energy storage in the charging [60%, 80%], [80%, 100%] and discharging [0, 20%], [20%, 40%] have different maximum safe charging and discharging current capacity, so they are not completely symmetrical.

[0075] The adjustment factor α SOC Substituting the adaptive frequency droop coefficient formula, the adaptive frequency droop coefficient ratio of the parallel grid-type energy storage battery charging and discharging can be obtained.

[0076] Taking two parallel grid-type energy storages as an example, when the SOC interval of the parallel grid-type energy storage is the same, the frequency droop coefficient remains the initial value calculated by the design capacity of the grid-type converter, at this time the frequency droop coefficient ratio is unchanged; with the change of the SOC interval of the grid-type energy storage, the adjustment factor α SOC The update causes the droop coefficient ratio to change, at this time the conventional lithium battery energy storage can obtain the frequency droop coefficient ratio during charging as shown in Table 2. Similarly, the frequency droop coefficient ratio of the conventional lithium battery energy storage during discharging is shown in Table 3.

[0077] Table 2: Frequency droop coefficient ratio information during charging

[0078]

[0079] Table 3: Frequency droop coefficient ratio information during discharging

[0080]

[0081] Preferably, the frequency change rate adjustment factor γ selection principle is: when |df / dt| is in the normal small disturbance range, γ=0 does not activate the intensive adjustment; when |df / dt| is in the medium or large disturbance, γ activates the adaptive to reduce the droop coefficient to have greater power adjustment capability, and reduce the frequency change rate. For example, according to the experience value of the current grid frequency change rate on stability, when |df / dt|≤1Hz / s, γ=1; when 1Hz / s<|df / dt|≤2 Hz / s, γ=1-(|df / dt|-1)=2-|df / dt|; when |df / dt|>2Hz / s, γ=0. In the specific implementation process, γ can be updated according to the requirements of the island microgrid for the frequency change rate.

[0082] Preferably, the temperature of the parallel grid type energy storage will also be different due to the unevenness of external environmental factors and the unevenness of internal operating state. Among them, the external environment includes sunshine difference, ventilation condition, installation position, etc., and the internal operating state includes aging degree, previous power operating state, inconsistency between clusters, unevenness of cooling system efficiency, etc. Based on the adaptive factor adjustment of local temperature, compared with the fixed droop coefficient, when the heterogeneity of the parallel grid type energy storage is small, the fine management of power distribution and life optimization can be realized; and when the heterogeneity is significant, the key safety protection function can be played. The battery temperature safety is established as the highest priority in the control logic. By adjusting the temperature regulation factor β T As a global coefficient acting on the entire droop coefficient calculation formula, it realizes the unified safety constraint of all output instructions including steady-state power and transient frequency response power, and fundamentally eliminates the risk of battery thermal runaway caused by pursuing SOC state charging and discharging potential and power grid performance under extreme temperature conditions.

[0083] The temperature regulation factor β T The selection principle is: when the temperature T is in the optimal working zone, β T = 1, at this time the temperature regulation factor does not affect the adaptive update of the droop coefficient; when the temperature T is lower than the lower limit of the optimal working zone, β T Stepwise increases with temperature decrease, and the lithium precipitation produced in the low-temperature charging process leads to a decay rate much higher than that in the discharging process; when the temperature T is higher than the upper limit of the optimal working zone, β T Stepwise increases with temperature increase, and is reduced to zero when the high-temperature protection threshold is triggered.

[0084] Here, according to the experience value of the charging and discharging rate of the conventional lithium battery at different temperatures considering the life and safety, as shown in Table 4, the battery energy storage system regulation factor β T corresponding to the rough change table. In the specific implementation process, β T can be updated according to the specification of the energy storage battery.

[0085] Table 4 β T value information of different temperature intervals

[0086]

[0087] In summary, the parameter adaptive update rules of the SOC regulation factor α SOC , the frequency change rate regulation factor γ, and the temperature regulation factor β T The grid type energy storage is adaptively adjusted according to the real-time state of local information, so as to realize the adaptive cooperation and distribution of power instructions of the island microgrid parallel grid type energy storage relying on local information without the instruction of the central controller.

[0088] Preferably, the overall implementation of the grid-connected energy storage droop control is: through sampling the voltage and current of the LC filter, the real active power and reactive power are obtained through power calculation and filtering, the active ring and the reactive ring update the droop coefficient, the secondary compensation frequency value and the secondary compensation voltage value on the basis of droop, and the generated voltage vector finally generates the SVPWM wave of the switch drive through the voltage and current ring.

[0089] Specifically, as shown in Figure 2 , for the droop control, the LC filter voltage N u and the LC filter current N i are used for power calculation and low-pass filter (LPF) processing to obtain the actual active power of the grid-connected energy storage P g and the actual reactive power of the grid-connected energy storage ; further, the droop frequency coefficient K f and the droop voltage coefficient K v are obtained respectively.

[0090] For the grid-connected voltage source phase angle θ , the rated angular frequency ω n , the angular frequency adjustment amount Δω and the angular frequency secondary compensation amount ω nc are comprehensively considered; the grid-connected voltage source phase angle θ is obtained by integration. For the grid-connected voltage source voltage E mi , the rated voltage E 0, the voltage adjustment amount Δ E q and the voltage secondary compensation amount E nc are comprehensively considered; then, the grid-connected voltage source E abc output of the grid-connected energy storage is formed.

[0091] The d-axis component of the grid-connected voltage source of the grid-connected energy storage E dref ;

[0092] The q-axis component of the grid-connected voltage source of the grid-connected energy storage E qref .

[0093] Specifically, the grid-forming control adopts droop control based on the existing secondary compensation and adaptive droop coefficient. Among them, the voltage secondary compensation and frequency secondary compensation need to keep weak communication between the grid-forming control and update to the droop control, and the PI correction is carried out through the distributed deviation of voltage and frequency, so as to avoid the deviation of voltage and frequency from the rated value in the process of traditional droop control. Among them:

[0094] E mpref is the voltage reference value of the public PCC point;

[0095] E mp is the actual voltage value of the public PCC point;

[0096] ω pref is the frequency reference value of the public PCC point;

[0097] ω i is the actual frequency value of the public PCC point.

[0098] As shown in Figure 4 , wherein (a) is the principle diagram of centralized communication control; (b) is the principle diagram of distributed communication control; (c) is the principle diagram of decentralized communication control. The application designs adaptive droop coefficient, which changes the droop coefficient according to local information such as SOC of grid-forming energy storage, temperature of energy storage, frequency change rate of microgrid, so as to realize power instruction distribution adjustment without relying on upper centralized communication control, and improve the robustness and fault tolerance of the system.

[0099] As shown in Figure 2 and Figure 5 , the application is based on droop control of secondary compensation and adaptive droop coefficient, and then the switch driving signal of the inverter is produced through the voltage and current loop based on virtual impedance. The introduction of virtual impedance can cope with the influence of parallel grid-forming energy storage line impedance difference on reactive power distribution, and the virtual inductance and virtual resistance calculation formula of virtual impedance is:

[0100] ;

[0101] Among them: L V is the virtual inductance of the virtual impedance;

[0102] L V0 is the initial value of the virtual impedance;

[0103] k L is the adjustment coefficient of the virtual impedance;

[0104] is the difference of the reactive power of the distributed grid-forming converter.

[0105] R V is a virtual resistance;

[0106] R V0 is a virtual resistance initial value;

[0107] k R is a virtual resistance adjustment coefficient;

[0108] I Ld is an LC filter inductance current d-axis component;

[0109] I Lq is an LC filter inductance current q-axis component;

[0110] E dref , E qref are respectively a d-axis voltage component reference value and a q-axis component reference value of a network-internal potential reference value;

[0111] ω is an island microgrid angular frequency;

[0112] E d 、E q are respectively a d-axis voltage component and a q-axis component of an LC filter voltage;

[0113] C r is an LC filter capacitance value;

[0114] L r is an LC filter inductance value;

[0115] QPR is a quasi-proportional resonant controller.

[0116] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Any simple modification or equivalent change of the above embodiment according to the technical essence of the present application falls within the protection scope of the present application.

Claims

1. A local information-based island microgrid distributed cooperative control method, characterized in that, The island micro-grid system comprises grid-forming control and grid-following control converters; and comprises the following steps: Step S1: the overall power of the island micro-grid satisfies energy conservation; Step S2: the grid-following control controls the power output according to local information and using a maximum power point tracking (MPPT) algorithm; Step S3: the grid-forming control outputs and distributes the reactive power of the parallel converters according to a fixed voltage regulation droop coefficient; the grid-forming control outputs and distributes the active power of the parallel converters according to a frequency regulation droop coefficient that is adaptively updated based on local information; the calculation formula of the frequency regulation droop coefficient adaptively updated based on local information is: k f = k f0 x a SOC x b T x g; wherein k f0 = ( ω n - ω min ) / P max ;​ wherein: k f is the frequency modulation droop coefficient; k f0 initial value for frequency modulation droop coefficient; a SOC SOC regulating factor; β T is the temperature regulation factor; γ is a frequency change rate adjustment factor; ω n is the nominal angular frequency of the islanded microgrid system; ω min Minimum rated angular frequency to allow fluctuations; P max For maximum active power output of grid-forming converters; When the state of charge SOC of the energy storage is in the interval [0, 20%], the charging a SOC is 0.5, the discharging a SOC is 2; when the state of charge SOC of the energy storage is in the interval [20, 40%], the charging a SOC is 0.8, the discharging a SOC is 1.5; when the state of charge SOC of the energy storage is in the interval [40, 60%], the charging a SOC is 1, the discharging a SOC is 1; when the state of charge SOC of the energy storage is in the interval [60, 80%], the charging a SOC is 1.2, the discharging a SOC is 0.8; when the state of charge SOC of the energy storage is in the interval [80, 100%], the charging a SOC is 1.5, the discharging a SOC is 0.5; When the micro-grid frequency change rate |df / dt| is less than or equal to a threshold A1, γ=1; when the threshold A1<|df / dt| is less than or equal to a threshold A2, γ=1-(|df / dt|-A1); and when |df / dt| is greater than the threshold A2, γ=0; When the energy storage temperature T is in the optimal working range, β T = 1; when the energy storage temperature T is lower than the lower limit of the optimal working range, β T increases stepwise with the decrease of temperature, and the lithium precipitation caused by the low-temperature charging process leads to a much higher attenuation rate than that of the discharging process; when the energy storage temperature T is higher than the upper limit of the optimal working range, β T increases stepwise with the increase of temperature, and decreases to zero when the high-temperature protection threshold is triggered.

2. The distributed cooperative control method of islanded microgrid based on local information according to claim 1, characterized in that, When the energy storage temperature T≤0, the β of the charge T The value is 3, and the discharge β is 3. T It is 2.5; when the energy storage temperature is 0 < T ≤ 10, the β of charging is... T The value is 2, and the discharge β T The value is 1.5; when the energy storage temperature is 10 < T ≤ 45, the β of charging is... T The value is 1, and the discharge β T The value is 1; when the energy storage temperature is 45 < T ≤ 55, the β of the charge is 1. T The value is 1.5, and the discharge β T The value is 1.2; when the energy storage temperature T > 55°C, the β value of charging is... T The discharge β is 0. T It is 0.

3. The distributed cooperative control method of islanded microgrid based on local information according to claim 1, characterized in that, In the step S3, the local information comprises a state of charge (SOC) of the energy storage, a temperature of the energy storage, and a micro-grid frequency change rate df / dt; wherein the SOC of the energy storage and the temperature of the energy storage are collected by a battery management system and a thermal management system of the energy storage, respectively; and the micro-grid frequency change rate df / dt is obtained based on current sampling analysis of an LC filter.

4. The distributed cooperative control method of islanded microgrid based on local information according to claim 1, characterized in that, In the step S3, the grid-forming control further comprises voltage secondary compensation and frequency secondary compensation, which need to keep weak communication between the grid-forming controls to update into the droop control, and are corrected by PI based on distributed deviation of voltage and frequency; the voltage and current of the LC filter are collected, and the real active power and reactive power are obtained through power calculation and filtering; the active ring and the reactive ring update the frequency regulation droop coefficient, the secondary compensation frequency value, and the secondary compensation voltage value, and generate a voltage vector; finally, a voltage and current loop based on virtual impedance generates a switch driving signal SVPWM wave of the inverter.

5. The distributed cooperative control method of islanded microgrid based on local information according to claim 1 or 4, characterized in that, The step S2, when the network type control enters the stability, detects the angular frequency ω generated by the current through the local public point g and voltage E g , if |ω pref -ω g |≤△ω and | E mpref - E g |≤△ E , the network type control accesses the public point, realizes the follow-up network control collaborative operation with the network type control; wherein:△ω and△ E respectively are the angular frequency access allowable threshold and the voltage access allowable threshold; E mpref is the public PCC point voltage reference value; ω pref is the public PCC point frequency reference value.

6. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the island micro-grid distributed collaborative control method based on local information in any one of claims 1-5.

Images