A Distributed Control Method for Low-Voltage Distribution Networks Based on Source-Load Matching

Abstract

This invention discloses a distributed control method for low-voltage distribution networks based on source-load matching, comprising the following steps: S1: acquiring power dispatch instructions from medium-voltage distribution networks and parameter information of each node in the low-voltage distribution network; S2: establishing a source-load matching optimization model for the low-voltage distribution network; S3: solving the source-load matching optimization model for the low-voltage distribution network in S2; S4: calculating the controllable power range of the low-voltage distribution area based on the output capacity of energy storage; S5: performing distributed control of the low-voltage distribution network; S6: determining whether the power dispatch instructions issued by the medium-voltage distribution network have been responded to. If yes, the response is implemented and the control ends; otherwise, step S1 is performed. This invention effectively solves the problems of severe node voltage exceeding limits and significantly increased network losses caused by the large number, quantity, and wide coverage of distributed photovoltaic (PV) points in low-voltage distribution networks. It improves the absorption capacity of existing distribution networks for distributed PV and ensures the safe, reliable, economical, and efficient operation of low-voltage distribution networks with a high proportion of residential PV.

Description

Technical Field

[0001] This invention relates to the field of distribution network operation analysis and control, specifically to a distributed control method for low-voltage distribution networks based on source-load matching. Background Technology

[0002] In recent years, with the construction of new power systems and the continuous advancement of the "dual carbon" target, the capacity and number of distributed photovoltaic grid-connected systems in low-voltage distribution networks have increased rapidly, showing a development trend of more points, larger quantities, and wider coverage, which has brought great challenges to the safe operation of distribution networks: (1) The voltage fluctuation at the grid connection point of low-voltage distribution networks is large. Photovoltaic output has anti-peak-shaving characteristics, and its net output with the load presents a duck curve shape, resulting in frequent occurrences of excessively high or low voltage at low-voltage distribution network nodes. (2) The network loss of distribution networks has increased. The high proportion of distributed photovoltaic grid connection has made the reverse heavy overload phenomenon of distribution networks significant, resulting in a significant increase in the network loss of distribution networks.

[0003] Solving the above problems involves two dimensions: distribution network planning and operation control. In terms of planning, by analyzing the current and future development trends of distributed photovoltaic (PV) power in low-voltage distribution networks, upgrades and renovations of transformer substations can be implemented, increasing line and transformer capacity to alleviate voltage overruns and reverse overloads. However, this method significantly increases investment costs. In terms of operation control, by coordinating distributed PV, energy storage, and load management, the power fed from low-voltage distributed PV to medium-voltage distribution networks can be effectively reduced, alleviating node voltage overruns and reverse overloads. This method has become a key research focus. Currently, there are two main methods for controlling distributed PV: centralized control and distributed control. The former requires high observability and controllability of distributed power sources, while the latter controls decentralized resources in low- and medium-voltage distribution networks where comprehensive observability and measurability are not possible. Chinese patent application number CN201711272249.8 discloses an optimized configuration and operation method for distributed energy storage in low-voltage distribution networks with photovoltaics. The patent mentions that the proposed optimization model can simultaneously optimize the installation location, rated power, rated power, and charging and discharging status and power of distributed energy storage at different times. It considers the voltage and reactive power control characteristics of distributed energy storage to support the voltage of low-voltage distribution networks, and also considers the limitation on the number of installation points of distributed energy storage in low-voltage distribution networks.

[0004] However, the above-mentioned method for optimizing the configuration and operation of distributed energy storage in low-voltage distribution networks with photovoltaics only considers the installation location, rated power, rated capacity, and charging and discharging status and power of distributed energy storage at different times. At the same time, this method limits the number of installation points of distributed energy storage. Therefore, the technical solution disclosed in Chinese patent application number CN201711272249.8 cannot solve the problem of severe over-limiting of node voltage and significant increase in network loss caused by the large number, quantity, and wide coverage of distributed photovoltaic points in low-voltage distribution networks. Summary of the Invention

[0005] This invention primarily addresses the problems of severe voltage exceedances at grid connection points and significantly increased network losses caused by the large number, wide distribution of distributed photovoltaic (PV) points in low-voltage distribution networks. It provides a source-load matching-based distributed control method for low-voltage distribution networks. By matching distributed PV, loads, and energy storage in the low-voltage distribution network, it enhances the local absorption capacity of distributed PV, reduces the backflow of power to the grid, and effectively solves the problems of severe voltage exceedances at grid connection points and increased network losses caused by the large number and wide distribution of distributed PV points in low-voltage distribution networks. This improves the absorption capacity of existing distribution networks for distributed PV, ensuring the safe, reliable, economical, and efficient operation of low-voltage distribution networks with a high proportion of residential PV.

[0006] The above-mentioned technical problems of the present invention are mainly solved by the following technical solution: a distributed control method for low-voltage distribution networks based on source-load matching, comprising the following steps:

[0007] S1: Obtain power dispatch instructions from the medium-voltage distribution network and parameter information of each node in the low-voltage distribution network;

[0008] S2: Establish a source-load matching optimization model for low-voltage distribution networks;

[0009] S3: Solve the source-load matching optimization model of the low-voltage distribution network in S2;

[0010] S4: Calculate the adjustable power range of the low-voltage distribution area based on the energy storage output capacity;

[0011] S5: Perform distributed control of low-voltage distribution networks;

[0012] S6: Determine whether the power dispatch command issued by the medium-voltage distribution network has been responded to. If yes, respond and end the control; otherwise, proceed to step S1.

[0013] Preferably, step S2 specifically includes the following steps:

[0014] S21: Establish the objective function. The objective of distributed control in low-voltage distribution networks is to maximize the source-load matching degree index. Therefore, the objective function is:

[0015]

[0016] Where θ is the source-load matching degree, δ M δ is the correlation coefficient between the source and the load. N Let be the variance function between the source and load.

[0017] S22: Determine the constraints. The constraints for low-voltage distributed control are distributed energy storage constraints.

[0018] Preferably, in step S21, the specific correlation coefficient and variance function between the source and load are as follows:

[0019]

[0020]

[0021] in, This represents the initial power of each low-voltage distribution network control time window. This indicates the power at the end of the control time window; Let L represent load, PV represent photovoltaic, and ess represent energy storage; P η.αv η represents the average value of the starting power and the ending power, where L represents the load, PV represents photovoltaic power, and ess represents energy storage; α1 and α2 are the weights of the two parameters mentioned above, and α1+α2=1.

[0022] As a preferred option, the specific distributed energy storage constraints in step S21 are as follows:

[0023] 1) Energy storage charge / discharge state constraints:

[0024]

[0025] in, and These represent the discharge or charging state of energy storage unit k, respectively. Since energy storage cannot be in both charging and discharging states at any given time, the sum of the two states is less than or equal to 1.

[0026] 2) Energy storage unit power constraints:

[0027] -P ess.dischmax ≤P ess.k ≤P ess.chmax

[0028] Among them, P ess.k P represents the instantaneous power of energy storage unit k. ess.dischmax P represents the maximum discharge power of the energy storage unit. ess.chmax This represents the maximum charging power of the energy storage unit.

[0029] 3) Energy constraints of energy storage units:

[0030]

[0031] Among them, E ess.t E represents the energy stored in the energy storage unit at any given time. ess.max To store the maximum energy, E ess.min The minimum energy required for energy storage.

[0032] Preferably, step S4 specifically includes the following steps:

[0033] S41: First, when calculating source-load matching, the total net output of distributed photovoltaic power in the low-voltage distribution area and household load is calculated using the following formula:

[0034] P LAP =P M -P ESS

[0035] Among them, P LAP P represents the total net output of distributed photovoltaic power in the low-voltage distribution area and residential loads. M For the total output of source-load energy storage in the low-voltage distribution area, P ESS This refers to the total output of distributed energy storage in the low-voltage distribution area when the source and load are matched.

[0036] S42: Calculate the energy storage's output range. The energy storage's output range is determined by the capacity of the distributed energy storage and its SOCi. This range is also calculated when the energy storage is charging.

[0037]

[0038] Among them, P LL.ess E is the lower limit of energy storage output. ess.i For the capacity of energy storage unit i, SOC i.max η is the upper limit of the SOC of this energy storage unit. ch.i P represents the charging efficiency of the energy storage unit, Δt represents the control time interval, and P represents the charging efficiency of the energy storage unit. ess.i This is the rated power of the energy storage unit. When the stored energy discharges,

[0039]

[0040] Among them, P UL.ess E is the upper limit of energy storage output. ess.i For the capacity of energy storage unit i, SOC I.min η is the lower limit of the SOC of this energy storage unit. disch.i P represents the discharge efficiency of the energy storage unit, Δt represents the control time interval, and P represents the discharge efficiency of the energy storage unit. ess.i This is the rated power of the energy storage unit. S43: Calculate the adjustable power range of the low-voltage distribution area using the following formula:

[0041]

[0042] S44: Based on the calculation formula, the adjustable power range of the low-voltage distribution area is calculated to be [P]. LL P UL ].

[0043] Preferably, in step S5, energy storage is specifically used as the distributed control object of the low-voltage distribution network.

[0044] Preferably, in step S6, determining whether a power dispatch command issued by the medium-voltage distribution network has been responded to specifically involves comparing the power dispatch command P. ref With response power P res The difference ΔP between k If the value is less than the set value, then respond and end the control; otherwise, proceed to S1.

[0045] As a preferred option, in step S3, the low-voltage distribution network source-load matching optimization model constructed in step S2 is solved using the second-order cone programming method.

[0046] The beneficial effects of this invention are: taking into account the communication characteristics, observability and measurability of low-voltage distribution networks and the local absorption factors of distributed power sources, a distributed control method for low-voltage distribution networks based on source-load matching is proposed. (1) It effectively solves the problem of serious over-limit of node voltage and significant increase in network loss caused by the large number, large quantity and wide coverage of distributed photovoltaic points in low-voltage distribution networks; (2) It improves the absorption capacity of existing distribution networks for distributed photovoltaics; (3) It ensures the safe, reliable, economical and efficient operation of low-voltage distribution networks with a high proportion of household photovoltaics. Attached Figure Description

[0047] Figure 1 This is a flowchart of a low-voltage distribution network distributed control method based on source-load matching according to an embodiment of the present invention. Detailed Implementation

[0048] The process is as follows: Figure 1 As shown. Considering the communication characteristics, observability and measurability of low-voltage distribution networks, and the local absorption of distributed power sources, distributed control is implemented for low-voltage distribution networks to achieve source-load matching in low-voltage distribution areas while responding to power commands from medium-voltage distribution networks.

[0049] Specific Implementation: A distributed control method for low-voltage distribution networks based on source-load matching, such as... Figure 1 As shown, it includes:

[0050] S1: Obtain power dispatch instructions from the medium-voltage distribution network and parameter information of each node in the low-voltage distribution network; specifically: obtain the power dispatch instruction P from the medium-voltage distribution network. ref Information such as photovoltaic output and load curves of each user in the low-voltage distribution network, as well as the status of distributed energy storage in the low-voltage distribution area.

[0051] S2: Establish a source-load matching optimization model for low-voltage distribution networks; specifically: S21: Establish the objective function. The distributed control objective of low-voltage distribution networks is to maximize the source-load matching degree index, so the objective function is:

[0052]

[0053] Where θ is the source-load matching degree, δM δ is the correlation coefficient between the source and the load. N Let V be the variance between the source and source loads.

[0054] The specific correlation coefficients and variance functions between the source and load are as follows:

[0055]

[0056]

[0057] in, This represents the initial power of each low-voltage distribution network control time window. This indicates the power at the end of the control time window; Let L represent load, PV represent photovoltaic, and ess represent energy storage; P η.αv This represents the average of the initial power and the final power. η represents the load (L), photovoltaic (PV) power (PV), and energy storage (ess). α1 and α2 are the weights of these two parameters, and α1 + α2 = 1. S22: Determine the constraints. The constraints for low-voltage distributed control are distributed energy storage constraints. Specific distributed energy storage constraints: 1) Energy storage charge / discharge state constraints:

[0058]

[0059] in, and These represent the discharge or charging state of energy storage unit k, respectively. Since energy storage cannot be in both charging and discharging states at any given time, the sum of the two states is less than or equal to 1.

[0060] 2) Energy storage unit power constraints:

[0061] -P ess.dischmax ≤P ess.k ≤P ess.chmax

[0062] Among them, P ess.k P represents the instantaneous power of energy storage unit k. ess.dischmax P represents the maximum discharge power of the energy storage unit. ess.chmax This represents the maximum charging power of the energy storage unit.

[0063] 3) Energy constraints of energy storage units:

[0064]

[0065] Among them, E ess.t E represents the energy stored in the energy storage unit at any given time. ess.max To store the maximum energy, E ess.min The minimum energy required for energy storage.

[0066] S3: Solve the low-voltage distribution network source-load matching optimization model in S2; specifically: for the constructed low-voltage distribution network source-load matching optimization model, the second-order cone programming method is used for solution, and its expression is:

[0067] ||Ax+b||2≤c T x+d

[0068] S4: Calculate the adjustable power range of the low-voltage distribution area based on the energy storage output capacity; specifically: S41: First, calculate the total net output of distributed photovoltaic power and household load in the low-voltage distribution area when matching source and load, using the following formula:

[0069] P LAP =P M -P ESS

[0070] Among them, P LAP P represents the total net output of distributed photovoltaic power in the low-voltage distribution area and residential loads. M For the total output of source-load energy storage in the low-voltage distribution area, P ESS This refers to the total output of distributed energy storage in the low-voltage distribution area when the source and load are matched.

[0071] S42: Calculate the energy storage's output range. The energy storage's output range is determined by the capacity of the distributed energy storage and its SOCi. This range is also calculated when the energy storage is charging.

[0072]

[0073] Among them, P LL.ess E is the lower limit of energy storage output. ess.i For the capacity of energy storage unit i, SOC i.max η is the upper limit of the SOC of this energy storage unit. ch.i P represents the charging efficiency of the energy storage unit, Δt represents the control time interval, and P represents the charging efficiency of the energy storage unit. ess.i This is the rated power of the energy storage unit. When the stored energy discharges,

[0074]

[0075] Among them, P UL.ess E is the upper limit of energy storage output. ess.i For the capacity of energy storage unit i, SOC I.min η is the lower limit of the SOC of this energy storage unit. disch.i P represents the discharge efficiency of the energy storage unit, Δt represents the control time interval, and P represents the discharge efficiency of the energy storage unit. ess.i This is the rated power of the energy storage unit. S43: Calculate the adjustable power range of the low-voltage distribution area using the following formula:

[0076]

[0077] S44: Based on the calculation formula, the adjustable power range of the low-voltage distribution area is calculated to be [P]. LL P UL ].

[0078] S5: Perform distributed control of low-voltage distribution networks; specifically: since household photovoltaic and loads in low-voltage distribution networks are not yet controllable, the distributed control object of low-voltage distribution networks is energy storage.

[0079] Suppose that at a certain moment, the power scheduling instruction P... ref With response power P res The difference between them is △P k Taking the SOC change of distributed energy storage as the consistency variable, the low-voltage distribution network can respond to power commands within the control time interval T. Therefore:

[0080] ΔSOC k =λ k ΔP k

[0081] Where k is the number of control iterations for the low-voltage distribution network, and λ k This is the conversion factor.

[0082] The iterative process of energy storage control at any node can be expressed as: ΔSOC k =∑w ij ΔSOC k-1

[0083] Among them, w ij The weight of the node that has a communication connection with this node.

[0084] Based on ΔSOC k =λ k ΔP k Sum of ΔSOC k =∑w ij ΔSOC k-1 It enables distributed control of energy storage.

[0085] S6: Determine whether a power dispatch command issued by the medium-voltage distribution network has been responded to; specifically: compare the power dispatch command P. ref With response power P res The difference between △P k If the value is less than the set value, then respond and end the current cycle of regulation; otherwise, proceed to S1.

[0086] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A low-voltage power distribution network distributed control method based on source-load matching, characterized in that, The method comprises the following steps: S1: obtaining a medium-voltage power distribution network power scheduling instruction and low-voltage power distribution network node parameter information; S2: establishing a low-voltage power distribution network source-load matching optimization model; a target function is established by a correlation coefficient and a variance function between the source and the load, the source-load correlation coefficient is the product of the load power and the photovoltaic energy storage net power at the beginning and the end of the time window, and the source-load variance is the square root of the sum of squares of the difference between the photovoltaic energy storage net power and the load power; the low-voltage power distribution network distributed control target is to maximize the source-load matching degree index, and the constraint condition is determined, and the constraint condition of the low-voltage distributed control is the distributed energy storage constraint; S3: solving the low-voltage power distribution network source-load matching optimization model in S2; S4: calculating the low-voltage transformer area controllable power range according to the output capacity of the energy storage; calculating the total net output of the distributed photovoltaic and household load in the low-voltage transformer area during source-load matching, calculating the energy storage output range, and the energy storage output range is determined by the capacity and SOCi of the distributed energy storage, calculating the low-voltage transformer area controllable power range, and obtaining the low-voltage transformer area controllable power range; S5: performing low-voltage power distribution network distributed control; S6: determining whether the power scheduling instruction issued by the medium-voltage power distribution network is responded, if yes, the response is realized, and the control is ended; if not, the step S1 is performed.

2. The low voltage distribution network distributed control method based on source load matching according to claim 1, characterized in that, Step S2 specifically comprises the following steps: S21: establishing a target function, and the low-voltage power distribution network distributed control target is to maximize the source-load matching degree index, so the target function is: Wherein, θ is the source load matching degree, δ M is the correlation coefficient between the source loads, δ N is the variance between the source loads; α1 is the weight coefficient of the source load correlation coefficient, and α2 is the weight coefficient of the source load variance. S22: determining the constraint condition, and the constraint condition of the low-voltage distributed control is the distributed energy storage constraint.

3. The low voltage distribution network distributed control method based on source load matching according to claim 2, characterized in that, In step S21, the specific correlation coefficient and variance function between the source and the load are: wherein, Pstart represents the start power of each low-voltage distribution grid control time window, Pend represents the end power of the control time window; L represents the load, PV represents photovoltaic, and ess represents energy storage; P η.αv Pstart and Pend represent the average value of the start power and the end power, L represents the load, PV represents photovoltaic, and ess represents energy storage; α1 and α2 are the weights of the two parameters, and α1 + α2 = 1.

4. The low voltage distribution network distributed control method based on source load matching according to claim 3, characterized in that, In step S21, the specific distributed energy storage constraint is: 1) energy storage charging and discharging state constraint: wherein, and respectively represent the discharge or charge state of the energy storage unit k, since the energy storage cannot be in both a charging and discharging state at any time, the sum of both is less than or equal to 1; 2) energy storage unit power constraint: - P ess.dischmax ≤ P ess.k ≤ P ess.chmax where P ess.k represents the instantaneous power of the energy storage unit k, P ess.dischmax is the maximum discharge power of the energy storage unit, P ess.chmax is the maximum charge power of the energy storage unit; 3) energy storage unit energy constraint: where E ess.t is the energy stored by the energy storage unit at any time, E ess.max is the maximum energy of the energy storage, E ess.min is the minimum energy of the energy storage.

5. The low voltage distribution network distributed control method based on source load matching according to claim 1, characterized in that, Step S4 specifically comprises the following steps: S41: first, calculating the total net output of the distributed photovoltaic and household load in the low-voltage transformer area during source-load matching, and the calculation formula is as follows: P LAP = P M - P ESS wherein P LAP is the total net output of the distributed photovoltaic and household load in the low-voltage transformer area, P M is the total output of the source-load-storage energy in the low-voltage transformer area, P ESS is the total output of the distributed storage energy in the low-voltage transformer area when the source-load is matched. S42: calculating the energy storage output range, which is determined by the capacity and SOCi of the distributed energy storage, when the energy storage is charging, where P LL.ess is the lower limit of the energy storage output, E ess.i is the capacity of the energy storage unit i, SOC i.max is the upper limit of the SOC of the energy storage unit, η ch.i is the charging efficiency of the energy storage unit, Δt is the control time interval, P ess.i is the rated power of the energy storage unit; when the energy storage is discharging, where P UL.ess is the upper limit of the energy storage output, E ess.i is the capacity of the energy storage unit i, SOC I.min is the lower limit of the SOC of the energy storage unit, η disch.i is the discharge efficiency of the energy storage unit, Δt is the control time interval, P ess.i is the rated power of the energy storage unit; S43: calculating the low-voltage transformer area controllable power range, and the calculation formula is as follows: S44: According to the calculation formula, the adjustable power range of the low-voltage transformer area is calculated as [P LL , P UL ] 6. The low voltage distribution network distributed control method based on source load matching according to claim 1, characterized in that, In step S5, the energy storage is taken as the low-voltage power distribution network distributed control object.

7. The low voltage distribution network distributed control method based on source load matching according to claim 1, characterized in that, The determination in step S6 whether to respond to the power dispatching instruction issued by the medium-voltage distribution network is specifically to compare the difference ΔP between the power dispatching instruction P ref and the response power P res , whether the difference is less than a set value, if yes, the response is realized, and the regulation and control is ended; if no, S1 is performed. k ​ 8. The low voltage distribution network distributed control method based on source load matching according to claim 1, characterized in that, In step S3, the low-voltage power distribution network source-load matching optimization model constructed in step S2 is solved by using a second-order cone programming method.

Images