Disclosure of Invention
The invention aims to overcome the defects and provides a method for controlling current sharing of a distributed power supply of a direct-current microgrid and bus voltage recovery of the distributed power supply. According to the method, current sharing precision among distributed power supplies can be greatly improved only by transmitting current information, and bus voltage drop caused by droop coefficients and line impedance is accurately compensated, so that the inherent problems of the traditional droop control method are solved.
In order to achieve the above object, the method comprises the following steps:
s1, locally adopting a droop control strategy by the direct-current micro-grid distributed power supply to sample local output current information;
s2, each distributed power supply in the direct-current micro-grid distributed power supplies communicates with an adjacent distributed power supply, and local output current information is sent to the adjacent distributed power supplies;
s3, after each distributed power supply obtains the output current information of the adjacent distributed power supplies, calculating the ratio of the local output current to the output current of the adjacent distributed power supplies;
s4, adjusting droop control coefficients of the direct-current micro-grid distributed power supplies according to the ratio of the local output current to the output current of the adjacent distributed power supplies, repeating the steps S1-S3, and enabling each distributed power supply to obtain the ratio of the local output current after the droop control coefficients are changed to the output current of the adjacent distributed power supplies;
s5, calculating the line impedance between the local output end and the direct current bus by the local controller of each distributed power supply according to the ratio of the local output current before and after the droop control coefficient is changed to the output current of the adjacent distributed power supply;
s6, setting corresponding droop coefficients of the distributed power supplies to enable the total equivalent output impedance of each distributed power supply to be equal;
s7, obtaining the average value of the output current of all the distributed power supplies of the direct-current micro-grid;
and S8, sending the product of the average value of the output current and the total equivalent output impedance as the compensation quantity of the voltage reference value into a voltage control loop of droop control by each distributed power supply local controller.
In S3, the ratio of the local output current to the output current of the adjacent distributed power source is calculated by the local controller.
In S5, the method for measuring the line impedance between the local output terminal and the dc bus includes:
wherein R isliThe line impedance value between the distributed power supply i and the direct current bus, k is the increment of the droop coefficient in S4, and RdIs the initial droop coefficient, x, for each distributed power supply in S11,x2And respectively changing the ratio of the output current of the distributed power supply i before and after the change of the droop coefficient to the output current of the adjacent distributed power supply.
In S6, the droop coefficient set for each distributed power supply is such that:
Rdi+Rli=Rdj+Rlj=…=Rdn+Rln=R
wherein R isdi Rdj,…,RdnThe droop coefficients R are respectively set for the direct current micro-grid distributed power supplies i, j, …, nli,Rlj,…,RlnThe line impedance from the output end of the direct current microgrid distributed power supply i, j, …, n to the direct current bus is respectively, and R is the line impedance from each distributed power supplyThe total equivalent output impedance of the power supply.
In S7, a current observer dynamic expression based on a consistency algorithm is adopted to obtain the average value of the output currents of all the distributed power supplies of the direct-current micro-grid.
In S7, the method for calculating the dynamic expression of the current observer based on the consensus algorithm is as follows:
wherein ioiIs the output current value i of the ith distributed power supplyavgi,iavgjObtaining the average value of output currents of the full-system distributed power supplies a for the ith and the jth distributed power supplies respectively by using a local current observerijFor communication weight factors in a communication network topology, a when a distributed power source i is in communication association with a distributed power source j ij1, otherwise, aij=0。
And calculating the average value of the output current of all the distributed power supplies of the direct-current micro-grid by using a local controller.
In S8, the voltage loop voltage reference value calculating method in droop control is as follows:
Urefi=(U*-Rdiioi)+Riavgi
wherein, U*,UrefiRespectively, the voltage reference value Ri of the distributed power source i before and after voltage compensationavgiThe amount of voltage compensation for distributed power source i.
Compared with the prior art, the distributed power supplies are communicated pairwise by adopting a distributed sparse communication means, the system reliability is higher, the line impedance between the output end of each distributed power supply and a direct current bus can be measured only by transmitting the output current information of the distributed power supplies, the total output impedance of each distributed power supply is equal by compensating the corresponding droop coefficient, the current equalizing purpose is further realized, and the current equalizing precision is higher. The invention not only can compensate the bus voltage drop caused by the droop coefficient, but also can compensate the bus voltage drop caused by the line impedance voltage drop, and the bus voltage compensation precision is higher. The line impedance measurement method adopted by the invention is not influenced by the size of the line impedance, is not influenced by the load property of the direct-current microgrid system, has small measurement error, and solves the problems that the current sharing precision is influenced by the line impedance and the bus voltage drops in the traditional droop control method.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
The conventional droop control method is shown in fig. 1, and it can be seen that in the voltage loop control of the distributed power supply, through adjustmentSag factor RdCan adjust the output voltage instruction U of the distributed power supplyrefTo adjust the output voltage UoTo adjust the output current i of the distributed power supplyoTo achieve the purpose of controlling the output power of the distributed power supply.
FIG. 2 is an equivalent circuit for droop control for parallel operation of two distributed power sources i, j, where U is*For a given value of the output voltage of the distributed power supply, UbusIs a DC bus voltage value, Uoi,UojThe actual output voltage values i, i of the distributed power supplies i, j respectivelyoi,iojAnd outputting current values for the distributed power supplies i and j respectively. RdFor a virtual resistance at the output of the distributed power supply, i.e. the set droop coefficient, Rli,RljRespectively, the line impedance R between the output end of the distributed power supply and the DC busLThe equivalent load resistance of the direct current microgrid system is obtained. According to kirchhoff's equivalent current-voltage law, the following can be obtained:
Uoi=U*-Rdioi,Uoj=U*-Rdioj (1)
Ubus=Uoi-Rliioi=Uoj-Rljioj (2)
the equations (1) and (2) are solved simultaneously to obtain:
wherein R isi、RjThe equivalent total output impedance of each distributed power source i, j, as can be seen from equation (3), the output current of each distributed power source cannot be proportionally distributed according to a given droop coefficient due to the existence of the line impedance, and meanwhile, as can be seen from equations (1) and (2), the droop coefficient and the existence of the line impedance cause the drop of the bus voltage.
In order to solve the problems of droop control, the invention introduces a method for controlling current sharing of a distributed power supply of a direct-current microgrid and bus voltage recovery of the distributed power supply, and fig. 3 shows a communication topology of the distributed power supply of the direct-current microgrid when the strategy is adopted for control. It can be seen that the control adopts a means of pairwise communication among the distributed power supplies, each distributed power supply only communicates with the adjacent distributed power supplies, and each distributed power supply local controller adopts a consistency algorithm to calculate the average value of the measurement information of the whole system by obtaining the local measurement information and the neighbor node measurement information. The dynamic expression of the current observer based on the consistency algorithm is shown as the following formula:
in the formula i
oiIs output current information of the ith distributed power supply i
avgi,i
avgjAnd respectively obtaining the average value of the output current of the full-system distributed power supply by the ith and the jth distributed power supplies by using a local current observer. a is
ijFor communication weight factors in a communication network topology, a when a distributed power source i is in communication association with a distributed power source j
ij>0, take a
ij1, otherwise, a
ij0. It can be proved that if only one directed spanning tree is included in the communication network topology, when t → ∞
And converge at i
oiThe arithmetic mean of (a), i.e.:
it can be seen that, when the consistency algorithm is adopted for control, when any two distributed power supplies have communication faults or a certain distributed power supply quits operation due to the faults, a cluster of directed spanning trees in the communication network can still be ensured. Therefore, the control method has high reliability, the problem that the whole system cannot normally operate due to the fault of a certain distributed power supply or a certain communication line does not exist, and meanwhile, all the distributed power supplies have the same priority and meet the plug and play requirement of the direct-current micro-grid on the distributed power supplies.
According to the research, the average values of the output voltage and the current of all the distributed power supplies are solved in the local controllers of all the distributed power supplies by using the dynamic expression of the consistency algorithm, and are respectively compared with the instruction voltage and the local output current information, so that the current output voltage and current values of the distributed power supplies are corrected, the current equalizing precision is improved, and the voltage drop of a direct current bus caused by a droop coefficient is compensated. However, no current research considers the influence of the voltage drop caused by the line impedance on the bus voltage drop, and the bus voltage drop caused by the line impedance voltage drop cannot be ignored in the low-voltage dc microgrid or the dc microgrid system with large line impedance. Therefore, the invention provides a line impedance measuring method, which sets corresponding droop coefficients after obtaining the impedance information of each distributed power supply line, so that the total equivalent output impedance of each distributed power supply is equal, and the purpose of current sharing is further achieved. Meanwhile, each distributed power local controller obtains the average value of the output current of each distributed power by using a current observer based on a consistency algorithm, and the product of the average value and the total equivalent output impedance is used as the local voltage compensation quantity, so that the voltage drop of the direct current bus caused by a droop coefficient and line impedance is compensated, and the aim of accurately compensating the bus voltage is fulfilled. The method has the following specific principle:
taking two distributed power supplies i, j as an example, measuring the line impedance, and locally adopting droop control:
Ubus=U*-Riioi=U*-Rjioj (7)
Ri=Rd+Rli,Rj=Rd+Rlj (8)
wherein, U*Is a command power of a distributed power supply i, jAnd pressing the reference value. U shapebusIs a DC bus voltage value, Urefi,UrefjRespectively and actually outputting voltage reference values i, i by the distributed power supplies i, joi,iojRespectively outputting current values, R, for distributed power supplies i, jdFor distributed power droop coefficient, Rli,RljLine impedance R between the output end of the distributed power supply i, j and the DC busi,RjRespectively, the total equivalent output impedance of the distributed power sources i, j. Is easily obtained from the formula (7):
each distributed power supply increases k on the original droop coefficient, and the k is as follows:
wherein ioi′,ioj' the output current values of the distributed power sources i, j after the droop coefficient is increased, respectively. x is the number of1,x2And increasing the output current ratio of the front distributed power supply i and the rear distributed power supply j for the droop coefficient. The following equations (9) and (10) can be obtained:
Rli=Ri-Rd (12)
namely, the distributed power supply i obtains the current value i before and after the droop coefficient is changed through the communication with the distributed power supply joj,ioj', according to (11) and (12), the line impedance R between the distributed power source i and the DC bus is calculated by the local controllerli。
And carrying out error analysis on the line impedance measurement method:
as shown in formulas (13) and (14), x ″)1,x2"respectively the output currents ioi,iojRatio and output current ioi′,ioj' measured error value of ratio.
From the formula (13):
the line impedance R is known from the formulas (12) and (15)liMeasured error of (2) is represented by RiDenominator term x of1″-x2"and molecular terms kx2"composition, wherein the denominator term x1″-x2"cannot be eliminated, belongs to an uncontrollable factor, and kx is the molecular term2", should have k<1, to avoid amplification errors.
In order to enable the current to be evenly divided under the steady state and the dynamic condition, the droop coefficient of each distributed power supply should be as follows:
Rdi+Rli=Rdj+Rlj=…=Rdn+Rln=R (16)
so that
ioi=ioj=…=ion (17)
Wherein R isdi Rdj,…,RdnThe droop coefficients i, i are respectively set for the direct current micro-grid distributed power sources i, j, …, noi,ioj,…,ionThe output current values are i, j and …, n, and R is the total equivalent output impedance of the distributed power supply, which should satisfy
Wherein, Delta UmaxMaximum voltage drop allowed for bus voltage, INThe rated current of the bus at full load. The droop coefficient R should be made to account for line impedance measurement errorsdiRdj,…,RdnAnd if the voltage is as large as possible, R is also as large as possible, so that the current sharing precision is ensured, and meanwhile, the bus voltage compensation is carried out on the basis. The dynamic expression of the current observer based on the consistency algorithm is shown in a formula (4), namely, each distributed power supply only communicates with the adjacent distributed power supply, the local controller calculates the average value of the output currents of all the distributed power supplies of the direct-current microgrid system according to the formula (4), and at the moment, the voltage compensation quantity of each distributed power supply is
ΔU=Riavgi=Riavgj=…=Riavgn (19)
The command voltage reference value of the distributed power source i is
Urefi=(U*-Rdiioi)+Riavgi (20)
Therefore, the method can measure the line impedance between the output end of each distributed power supply and the direct current bus only by transmitting adjacent current information, realizes current sharing, compensates the bus voltage drop caused by the droop coefficient and the line impedance, and is a more accurate bus voltage compensation mode.
Example (b):
two distributed power sources are taken as examples, and the specific implementation mode is explained as follows:
FIG. 4 is a schematic diagram of the control of the parallel operation of two distributed power sources, C1、C2Respectively two distributed power supply output end capacitors, Rl1、Rl2Line impedance i between the output ends of the two distributed power supplies and the DC buso1、io2Respectively output currents, R, for two distributed power suppliesLIs the load equivalent resistance. Each distributed power supply locally adopts droop control, and the two-layer control adopts the control strategy and improves the bus voltage regulation rate, as shown in fig. 5. In MATLAB according to the circuit diagram shown in FIG. 4 and the control method shown in FIG. 5The strategy of the invention is verified by a SIMULINK platform, wherein the rated voltage value of a bus is 80V, and the line resistance R of two distributed power suppliesl1=2Ω,Rl22.5 Ω, load R L80 Ω. The specific implementation steps are as follows:
1. initially setting the droop coefficient of each distributed power supply to be 2.5 omega, respectively sampling the output current of the local controllers of the two distributed power supplies in a steady state, respectively sampling 5 times of output current values by adopting an average filtering method, simultaneously filtering high-frequency components in the output current by adopting a low-pass filter, and then obtaining the average value of the local output current after 5 times of sampling.
2. And each distributed power supply is communicated with the adjacent distributed power supply, and the local output current average value of each distributed power supply is sent to the adjacent distributed power supply. At this time, each distributed power supply local controller calculates the ratio of the local output current average value to the adjacent distributed power supply output current average value through formula (9), and the ratio is recorded as x1。
3. Adding 0.5 omega to the droop coefficient of each distributed power supply, repeating the step 1 to obtain the output current average value of each distributed power supply after the droop coefficient is changed, repeating the step 2 to obtain the ratio of the local output current average value of each distributed power supply after the droop coefficient is changed to the output current average value of the adjacent distributed power supply, and recording the ratio as x2。
4. The local controller of each distributed power supply can calculate the line impedance from the output end of the distributed power supply to the direct current bus according to the formulas (11) and (12). Experimental calculation to obtain Rl1=2.0098Ω,Rl2The measurement error from the actual resistance value was 0.49%, 0.43%, respectively, when it was 2.5107 Ω.
5. Uniformly setting the total output impedance R of each distributed power supply to be 5 omega, and calculating the droop coefficients R of the distributed power supplies 1 and 2 by the formula (12)d1=2.9902Ω,Rd22.4893 Ω. At the moment, the total output impedance of the two distributed power supplies is basically consistent, and the purpose of current sharing is achieved.
6. When the local controller of each distributed power supply communicates with the adjacent distributed power supplies, the two-layer control adopts a current observer based on the formula (4) to obtain the average value of the output current of all the distributed power supplies, and the product of the average value and the total output impedance R is used as the compensation quantity of the voltage reference instruction of the local controller to be sent into a voltage control loop, as shown in the formulas (19) and (20), the purpose of accurately compensating the voltage drop of the bus is achieved.
Fig. 6 to 11 show simulation results in experiments using the above inventive procedure. Fig. 6 shows the simulation results of the dc bus voltage when the inventive control strategy is applied, wherein the load is weighted when t is 1 s. Fig. 7 and 8 are respectively the steady-state values of the dc bus voltage before and after the load is increased, and it can be seen that the steady-state values of the dc bus voltage before and after the load is changed are both rated voltage 80V, so that it can be seen that the method has a relatively accurate dc bus voltage compensation effect.
Fig. 9 shows the output current values of two parallel distributed power supplies when the inventive strategy is employed. Fig. 10 and fig. 11 are steady-state values of output currents of two distributed power supplies before and after load change, respectively, and it can be seen that the output currents of the two distributed power supplies before and after load change almost tend to be consistent, so that it can be seen that the output current sharing accuracy of the distributed power supplies is high when the method is used for controlling.
Fig. 12 and 13 show the dc bus voltage and the output current values of the two distributed power sources when droop control is performed only by using the initial droop coefficient, respectively. It can be seen that when only droop control is adopted, due to the influence of inconsistent line impedance, the current sharing precision is poor, meanwhile, the direct current bus voltage also drops to a certain extent, and meanwhile, when the load is heavier, the current sharing precision is worse, and the direct current bus voltage drop is also larger.
In summary, by comparing the simulation result of the control strategy with the simulation result of only using droop control, it can be seen that the method has higher current sharing accuracy and bus voltage compensation accuracy.