Active-reactive collaborative optimization scheduling method and system for power distribution network containing high-proportion distributed photovoltaic, and storage medium

Abstract

The invention discloses an active-reactive collaborative optimization scheduling method and system for a power distribution network containing high-proportion distributed photovoltaic, and a storage medium. The method comprises the following steps: S1, constructing a constraint model of a network power flow constraint condition of a power distribution network, a substation side reactive voltage regulation equipment operation constraint condition and a feeder side active and reactive regulation equipment operation constraint condition; s2, constructing an objective function of power distribution network optimization regulation and control of calculation formula weighted combination of a plurality of optimization objectives; s3, constructing an optimal scheduling model, wherein the optimal scheduling model takes the constraint model constructed in the step S1 as a constraint condition to solve a minimum value of the target function constructed in the step S2; s4, processing a non-convex constraint condition and a target function containing an absolute value in the optimal scheduling model, converting the optimal scheduling model into a mixed integer second-order cone programming model, and solving the mixed integer second-order cone programming model; according to the method, various active and reactive power regulation devices in the power distribution network can be accurately and effectively coordinated, optimized and dispatched according to a plurality of optimization targets.

Description

Technical Field

[0001] The present invention relates to active-reactive collaborative optimization dispatching of a distribution network, and in particular to an active-reactive collaborative optimization dispatching method, system and storage medium for a distribution network containing a high proportion of distributed photovoltaics. Background Art

[0002] In distribution networks with a high proportion of distributed photovoltaic penetration, the photovoltaic installed capacity penetration rate exceeds 100%, and the average annual photovoltaic power penetration rate reaches 30%~50%. There are rich types of equipment involved in active and reactive power regulation in such distribution networks, such as on-load tap changers (OLTCs) and capacitors (CBs) on the substation side, energy storage (ESS), static VAR compensators (SVCs) and photovoltaic (PV) grid-connected inverters on the feeder side, all of which can participate in the regulation and operation of distribution networks; the safe and efficient operation of distribution networks with a high proportion of distributed photovoltaics is to achieve the goals of reducing network losses, reducing voltage fluctuations, and improving photovoltaic absorption and operation economy. However, how to fully coordinate the regulation of various active and reactive power regulation equipment to achieve the various operation goals of distribution networks with a high proportion of distributed photovoltaics has become a major problem that plagues distribution network dispatching managers. Summary of the invention

[0003] Purpose of the invention: The purpose of the present invention is to provide a method, system and storage medium for collaborative optimization scheduling of active and reactive power in a distribution network containing a high proportion of distributed photovoltaics, which can accurately and effectively coordinate and optimize the scheduling of various active and reactive regulation devices in the distribution network according to multiple optimization objectives.

[0004] Technical solution: The active-reactive coordinated optimization dispatching method of the distribution network containing a high proportion of distributed photovoltaics described in the present invention comprises the following steps:

[0005] S1. Constructing a constraint model for the network flow constraints of the distribution network, the operation constraints of the reactive voltage regulation equipment on the substation side, and the operation constraints of the active and reactive regulation equipment on the feeder side;

[0006] S2. Constructing an objective function of distribution network optimization control by weighted combination of calculation formulas of multiple optimization objectives;

[0007] S3, constructing an optimization scheduling model, which uses all constraint models constructed in step S1 as constraint conditions to find the minimum value of the objective function constructed in step S2;

[0008] S4. Convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution to schedule the distribution network.

[0009] Based on the above scheme, constraint models are constructed for the network flow constraints of the distribution network, the operating constraints of the reactive voltage regulation equipment on the substation side, and the operating constraints of the active and reactive regulation equipment on the feeder side, which can fully take into account the various constraints of the coordinated operation of various active and reactive regulation equipment in the distribution network with a high proportion of distributed photovoltaics; then the calculation formulas of multiple optimization objectives are weightedly combined to construct the objective function of the distribution network optimization control, which can avoid the situation where some optimization objectives are not fully considered in the collaborative optimization scheduling process; on the basis of constructing the model and the objective function, the two are combined as a whole to construct the optimization scheduling model, and the optimization scheduling model is convexified and linearized, that is, after it is approximately simplified, it can be solved directly using a commercial solver, and according to the solution results, various active and reactive regulation equipment in the distribution network with a high proportion of distributed photovoltaics are coordinated and optimized, which can realize the distribution network according to different optimization objectives Under the premise of taking into account the constraints of each device, more accurate and effective collaborative optimization scheduling that is more in line with the actual scenario can be achieved.

[0010] The active-reactive coordinated optimization dispatching system of the distribution network containing a high proportion of distributed photovoltaics according to the present invention comprises:

[0011] Constraint model building module: a constraint model used to build the network flow constraints of the distribution network, the operating constraints of the reactive voltage regulation equipment on the substation side, and the operating constraints of the active and reactive regulation equipment on the feeder side;

[0012] Objective function construction module: used to construct the objective function of distribution network optimization control with multi-objective weighted combination;

[0013] Optimization scheduling model construction module: used to construct an optimization scheduling model, which uses all constraint models constructed by the constraint model construction module as constraint conditions to find the minimum value of the objective function constructed by the objective function construction module;

[0014] Solution scheduling module: It is used to convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution results to schedule the distribution network.

[0015] The computer-readable storage medium storing one or more programs described in the present invention includes one or more programs including instructions, and when the instructions are executed by a computing device, the computing device executes any one of the above methods.

[0016] Beneficial effects: Compared with the prior art, the present invention has the following significant effects: by constructing a constraint model for the network flow constraints of the distribution network, the operating constraints of the reactive voltage regulation equipment on the substation side, and the operating constraints of the active and reactive regulation equipment on the feeder side, various constraints in the operation of the distribution network are fully considered, and then the calculation formulas of multiple optimization objectives are weighted combined to obtain the objective function of the optimized regulation, avoiding the problem that some optimization objectives are not fully considered, and then combining the two to construct an optimized scheduling model and make an approximate simplified solution to it, thereby achieving more accurate and effective optimal scheduling of various active and reactive regulation equipment in the distribution network according to different optimization objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0018] Figure 2 The topology diagram of the IEEE 33-node test system;

[0019] Figure 3 Power prediction curves for photovoltaic and load;

[0020] Figure 4a This is a schematic diagram of node voltage in different time periods for Example 1;

[0021] Figure 4b This is the schematic diagram of the network active power loss in Example 1;

[0022] Figure 5a This is the schematic diagram of node voltage in different time periods of Example 2;

[0023] Figure 5b This is the schematic diagram of the network active power loss in Example 2;

[0024] Figure 6a This is the schematic diagram of node voltage in time periods of Example 3;

[0025] Figure 6b This is a schematic diagram of the network active power loss in Example 3. DETAILED DESCRIPTION

[0026] As shown in the figure, the active-reactive coordinated optimization dispatching method of the distribution network containing a high proportion of distributed photovoltaics according to the present invention comprises the following steps:

[0027] S1. Constructing a constraint model for the network flow constraints of the distribution network, the operation constraints of the reactive voltage regulation equipment on the substation side, and the operation constraints of the active and reactive regulation equipment on the feeder side;

[0028] Among them, network flows include node voltage, line current and power; reactive voltage regulation equipment on the substation side includes on-load tap-changing transformers and capacitor banks; active and reactive regulation equipment on the feeder side includes energy storage systems, photovoltaic grid-connected inverters and static VAR compensators.

[0029] The model of the network flow constraint condition of the distribution network is: (1) (2) (3) (4) (5) (6) in, and They represent the square of the voltage amplitude at the head node i and the terminal node j of line ij in time period t, and Respectively represent the resistance and reactance values ​​of line ij, and They represent the active power and reactive power of line ij in time period t, respectively. represents the square of the current value of line ij in time period t, and They represent the active power and reactive power of line jk in time period t respectively, line jk represents the line with the first end node j and the last end node k, and set J represents the set of all lines with the first end node j; and They represent the net active load and net reactive load of node j in time period t, respectively. and They represent the actual active load and reactive load of node j in time period t, and They represent the active power and reactive power generated by the photovoltaic grid-connected inverter connected to node j in time period t, and They represent the charging and discharging active powers of the energy storage grid-connected point connected to node j in time period t, represents the reactive compensation power of the static VAR compensator connected to node j in time period t, represents the reactive compensation amount of the capacitor bank connected at node j in time period t, and They represent the minimum and maximum values ​​of the square of the node voltage, and They represent the minimum and maximum values ​​of the square of the line current respectively.

[0030] Constraints (1) and (2) represent the power flow equations between the node voltages at both ends of the line in the distribution network and the current and power flowing through the line; constraints (3) and (4) represent the node power balance equations; constraints (5) and (6) represent the upper and lower limits of the node voltage and line current, respectively.

[0031] The load-controlled tap-changing transformer changes the node voltage of the substation by adjusting the transformer tap position. The model of its operation constraint condition is: (7) (8) (9) (10) (11) (12) Where Ntaps represents the total number of tap positions of the load-controlled tap-changing transformer. The state indicator variable indicating whether the tap position of the load-controlled tap-changing transformer in time period t is at the i-th tap, 1 means yes, 0 means no; Represents the square of the node voltage value of the substation bus node sub in time period t; Indicates the transformer ratio corresponding to the tap position of the load-regulating transformer at the ith tap, that is, the ratio of the primary side to the secondary side of the transformer. The per-unit value is generally used in power system flow calculations. The load-regulating transformer ratio is generally between 0.9 and 1.1; and The state indicator variables respectively indicate whether the tap position of the load-regulating transformer is adjusted up or down in the time period t compared with the previous time period, and the corresponding value is 1 for yes and 0 for no; It represents the maximum gear interval of the tap position of the load-controlled voltage-changing transformer for a single increase or decrease, T represents the set of all time periods within the entire scheduling time range, It indicates the total number of maximum up- and down-regulations of the load-controlled tap-changing transformer within the entire dispatching time range.

[0032] Constraint (7) indicates that the tap of the load-controlled tap-changing transformer is at a position between 1 and Ntaps in any time period; constraint (8) indicates the relationship between the voltage of the busbar node of the feeder outgoing line of the substation and the tap position of the load-controlled tap-changing transformer; constraint (9) indicates that the upward and downward adjustment actions of the load-controlled tap-changing transformer are mutually exclusive in any time period; constraints (10) and (11) indicate the maximum interval constraints for the single upward and downward adjustment of the tap position of the load-controlled tap-changing transformer; constraint (12) indicates the total number of upward and downward adjustment of the load-controlled tap-changing transformer within the entire scheduling time range.

[0033] The capacitor bank is usually installed in parallel at the busbar node of the feeder outgoing line of the substation. The model of its operation constraint condition is: (13) (14) (15) (16) in, The variable indicating the switch-on and switch-off status of the capacitor bank connected to node j in time period t, with a value of 1 indicating switch-on and a value of 0 indicating switch-off; represents the reactive capacity of the capacitor bank connected at node j, A state indicator variable indicating whether the capacitor bank connected to node j starts to be put into operation or withdrawn in time period t. A value of 1 indicates that the capacitor bank is put into operation, and a value of 0 indicates that the capacitor bank is withdrawn. It represents the maximum number of times the capacitor bank connected to node j is allowed to be switched on and off within the entire scheduling time range.

[0034] The energy storage system can regulate both active power and reactive power. Its operation constraints include power constraints, energy constraints, and charge and discharge state constraints. The specific model is: (17) (18) (19) (20) (twenty one) (twenty two) (twenty three) (twenty four) (25) (26) (27) in, and They represent the charging and discharging power of the energy storage system connected to node j in time period t, and are state indicator variables that respectively indicate whether the energy storage system connected to node j is charging or discharging in time period t, with a value of 1 indicating yes and a value of 0 indicating no; represents the maximum charging and discharging power of the energy storage system connected to node j, represents the capacity of the energy storage system connected to node j, and They represent the minimum and maximum ratios of the actual power and capacity of the energy storage system connected to node j, represents the remaining power of the ESS connected to node j after the end of time period t, Represents the last time period of the energy storage system connected to node j in the entire scheduling time range The remaining power after the end, represents the initial remaining power of the ESS connected to node j at the beginning of the first time period, represents the ratio of the initial actual power and capacity of the energy storage system connected to node j, and They represent the charging efficiency and discharging efficiency of the energy storage system connected at node j, Indicates the length of each time period, usually 15 minutes; and They respectively represent the state change indicator variables of whether the energy storage system connected to node j starts charging and discharging in time period t, with a value of 1 indicating yes and a value of 0 indicating no; and It represents the maximum number of charging and discharging times allowed for the energy storage system connected to node j within the entire scheduling time range.

[0035] Constraints (17) and (18) are power constraints of the energy storage system, constraints (19) to (21) are energy constraints of the energy storage system, and constraints (22) to (27) are charge and discharge state constraints of the energy storage system. Specifically, constraints (17) and (18) represent the maximum charge and discharge power constraints of the energy storage system. Constraint (19) indicates that the remaining power of the energy storage system must be maintained within a certain range, such as 20% to 100% of the capacity. Constraint (20) indicates that the remaining energy of the energy storage system at the end of the last time period and the remaining power at the beginning of the first time period are equal to the initial power of the energy storage system. Constraint (21) represents the dynamic balance equation of the grid-connected point power and the remaining energy of the energy storage system. Constraint (22) indicates that the charging and discharging states of the energy storage system in the same time period are mutually exclusive. Constraints (23) to (25) represent the logical constraints for the charging and discharging states of the energy storage system to start changing; constraints (26) to (27) indicate that the total number of charges and total number of discharges of the energy storage system within the entire scheduling time range does not exceed a certain value, with the aim of limiting the frequent charging and discharging of the energy storage.

[0036] The photovoltaic grid-connected inverter can regulate both active power and reactive power. The model of its operation constraint condition is: (28) (29) (30) in, represents the active power output by the photovoltaic grid-connected inverter connected to node j in time period t; represents the reactive power output by the photovoltaic grid-connected inverter connected to node j in time period t; represents the maximum active power output by the photovoltaic grid-connected inverter connected to node j in time period t; represents the maximum power factor angle of the photovoltaic grid-connected inverter connected at node j, Corresponding minimum power factor , ; represents the apparent power capacity of the PV grid-connected inverter connected to node j;

[0037] The static VAR compensator is connected to the feeder node and regulates the reactive power flow of the feeder by emitting or absorbing reactive power. The model of its operation constraint condition is: (31) in, and They represent the minimum and maximum reactive compensation amounts of the static VAR compensator connected at node j respectively.

[0038] Formula (31) represents the reactive power compensation range of the static VAR compensator.

[0039] S2. Construct the objective function of distribution network optimization control by weighted combination of calculation formulas of multiple optimization objectives.

[0040] The distribution network with a high proportion of distributed photovoltaic can utilize a variety of active and reactive regulation resources on the substation side and the feeder side to achieve multiple operating goals such as reducing distribution network losses, reducing voltage deviations, and improving photovoltaic absorption rates. These goals can be expressed by the calculation formula of minimum network loss, minimum voltage deviation, minimum photovoltaic net power generation cost, minimum energy storage system operation cost, minimum static VAR compensator operation cost, minimum capacitor bank switching cost, and minimum load-regulating transformer gear switching cost, specifically as follows:

[0041] The formula for calculating the minimum network loss is: (A1) in, It represents the total loss of all lines in the distribution network within the entire dispatching time range. Represents the set of all lines.

[0042] The calculation formula for the minimum voltage deviation is: (A2) in, It represents the total voltage deviation of all nodes in the distribution network in each time period; Indicates the total number of time periods within the entire scheduling time range. For example, the day-ahead optimization scheduling range is 24 hours, with 15 minutes per time period. ; BUS represents the set of all nodes in the distribution network; Represents the square of the reference voltage value of node i in time period t, usually 1.

[0043] The formula for calculating the minimum net photovoltaic power generation cost is: (A3) in, represents the net power generation cost of all PV in the entire dispatch time range; Indicates the cost of photovoltaic electricity, which can be 0.25 yuan / kWh; It represents the photovoltaic grid-connected electricity price, which can be 0.391 yuan / kWh; a negative net power generation cost means that photovoltaic grid-connected power generation and electricity sales generate income, and the greater the income, the more photovoltaic power generation.

[0044] The calculation formula for the minimum operating cost of the energy storage system is: (A4) in, represents the operating cost of all energy storage systems within the entire dispatch time range, It represents the cost per kilowatt-hour of the feeder-side energy storage system, which can be taken as 0.65 yuan / kWh.

[0045] The calculation formula for the minimum operating cost of the static VAR compensator is: (A5) in, It represents the operating cost of all static VAR compensators in the entire dispatch time range; It represents the unit reactive power cost of the static VAR compensator on the feeder side, which can be taken as 0.5 yuan / kVarh.

[0046] The calculation formula for the minimum switching cost of capacitor banks is: (A6) in, It represents the switching cost of all capacitor banks within the entire dispatch time range; Indicates the single switching cost of the capacitor bank.

[0047] The formula for calculating the minimum gear switching cost of the load-regulating transformer is: (A7) in, It represents the gear switching cost of the load-regulating transformer within the entire dispatching time range; It represents the cost of a single increase or decrease of the load-changing transformer.

[0048] Based on the above objective functions, a multi-objective weighted combination distribution network optimization control objective function is proposed as follows: (A8) Where: The objective function of the distribution network optimization control that represents the multi-objective weighted combination, represents the weight of the total network loss, represents the weight of voltage deviation, represents the weight of the net photovoltaic power generation cost, represents the weight of the energy storage system operating cost, represents the weight of the reactive power compensation cost of the static VAR compensator, represents the weight of the capacitor bank switching cost, The weight representing the gear switching cost of the load-regulating transformer; the weights corresponding to each optimization sub-objective function are selected by the distribution network dispatching management personnel based on operating experience.

[0049] S3, constructing an optimization scheduling model that takes all constraint models constructed in step S1 and the objective function constructed in step S2 as constraint conditions to obtain the minimum value of the objective function;

[0050] Combining the constraint model and objective function of the above constraints, a multi-objective active-reactive coordinated optimization dispatch model for distribution networks with a high proportion of distributed photovoltaics is constructed as follows: (M1).

[0051] S4. Convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution to schedule the distribution network.

[0052] The optimization scheduling model M1 is a mixed integer non-convex optimization model, which cannot guarantee the global optimality of the solution algorithm. In order to facilitate the solution, the optimization scheduling model M1 needs to be appropriately simplified. By convexifying or linearizing the non-convex constraints corresponding to formulas (2), (24), (25) and (30) in the optimization scheduling model M1, as well as the calculation formulas (A2) and (A5) containing absolute values, the model can be converted into a mixed integer second-order cone programming model that can solve the optimal solution. The specific processing method is introduced as follows:

[0053] The non-convex constraint condition corresponding to formula (2) is convexified by the second-order cone relaxation approximation. The convexified constraint condition after processing is (2.S) Among them, the operator represents the second-order norm.

[0054] The logical constraints represented by the non-convex constraints corresponding to formulas (24) and (25) contain bilinear terms, which are equivalently processed using linear constraints. The processed linear constraints are: (24.L) (25.L)

[0055] The feasible domain of the non-convex constraint corresponding to formula (30) is a circular area. To facilitate the solution, the intersection of two square areas is used for linear approximation. The linear constraint after processing is: (30.L)

[0056] The absolute value in the calculation formula (A2) is a nonlinear term. Since the calculation formula (A2) is a minimization problem, the auxiliary variable is defined as the absolute value of the difference between the square of the voltage amplitude of node i in time period t and the square of the reference voltage value. to replace , and supplement the constraints, so as to linearize the calculation formula (A2) by removing the absolute value operator. The processed calculation formula is (A2.L)

[0057] The absolute value in the calculation formula (A5) is also a nonlinear term. Since the calculation formula (A5) is a minimization problem, the auxiliary variable defining the absolute value of the reactive compensation amount of the static VAR compensator is to replace , and supplement the constraints, so as to linearize the calculation formula (A5) by removing the absolute value operator. The calculation formula after processing is: (A5.L)

[0058] Through the above simplified processing method, the original optimization model (M1) is transformed into the optimization model (M2), as shown below: (M2)

[0059] The optimization scheduling model M2 is a Mixed Integer Second Order Cone Programming (MISOCP) problem. By calling the commercial solver Cplex / Gurobi / Mosek to solve it, the global optimal solution of the optimization model can be given, thereby obtaining an optimization control scheme for multi-objective active-reactive coordination of a high-proportion distributed photovoltaic distribution network that meets the needs of distribution network dispatching managers.

[0060] The active-reactive coordinated optimization dispatching system of the distribution network containing a high proportion of distributed photovoltaics according to the present invention comprises:

[0061] Constraint model building module: used to build the constraint model of network flow constraints of the distribution network, the operating constraints of reactive voltage regulation equipment on the substation side, and the operating constraints of active and reactive regulation equipment on the feeder side;

[0062] Objective function construction module: used to construct the objective function of distribution network optimization control with multi-objective weighted combination;

[0063] Optimization scheduling model construction module: used to construct an optimization scheduling model that takes all constraint models constructed in step S1 and the objective function constructed in step S2 as constraint conditions to obtain the minimum value of the objective function;

[0064] Solution scheduling module: It is used to convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution results to schedule the distribution network.

[0065] The computer-readable storage medium storing one or more programs described in the present invention includes one or more programs including instructions, and when the instructions are executed by a computing device, the computing device executes any one of the above methods.

[0066] In order to better demonstrate the effect of this method, a simulation example is presented below. Before the example is presented, several indicators of the operating status of the distribution network containing a high proportion of distributed photovoltaics are defined as follows:

[0067] Definition 1: Photovoltaic installed capacity penetration rate

[0068] (I1) Where: represents the photovoltaic installed capacity penetration rate in the distribution network, represents the set of all nodes in the distribution network, represents the photovoltaic active installed capacity at node j, Represents the annual maximum active power of the load at node j.

[0069] Definition 2: Annual average photovoltaic power penetration rate

[0070] (I2) Where: represents the average annual photovoltaic power penetration rate in the distribution network, represents the annual utilization hours of photovoltaic power at node j, Represents the maximum annual utilization hours of the load at node j.

[0071] Definition 3: Amount of photovoltaic power abandoned

[0072] (I3) Where: It indicates the amount of photovoltaic power abandoned by the distribution network during the entire dispatching time range.

[0073] Taking the IEEE 33-node distribution network as a test example, the system topology diagram and the installation locations of the load-controlled voltage-changing transformer OLTC, capacitor bank CB, energy storage system ESS, photovoltaic grid-connected inverter PV, and static VAR compensator SVC are shown in the figure. Figure 2 As shown, Figure 2 The numbers in the figure represent the node numbers. Figure 2At the 8 feeder nodes in the figure, the photovoltaic capacity at each node is 500kVA, and the minimum power factor is 0.98; the total photovoltaic installed capacity is 4WM, the maximum load is 3.715MW, and the photovoltaic installed capacity penetration rate is 107.67%; the annual photovoltaic utilization hours are 1100h, the annual maximum load utilization hours are 3000h, and the annual average photovoltaic power penetration rate is 39.48%; the power prediction curves of photovoltaic and load are as follows Figure 3 As shown; the photovoltaic power cost is 0.25 yuan / kWh, and the electricity price is 0.391 yuan / kWh. The maximum charging and discharging power of the energy storage ESS is 500kW, the charging and discharging efficiency is 0.95, the maximum / minimum / initial SOC is 1 / 0.15 / 0.5 respectively, and the maximum charging and discharging times is one charge and two discharges, that is, a maximum of one charge and two discharges. The capacity of the static VAR compensator SVC is ±500kVar. The single maximum up and down adjustment gear of the on-load tap-changing transformer OLTC, that is, the total number of tap positions Ntaps, is 2, and the maximum up and down adjustment times in the entire dispatching cycle (24 hours) is 5, without considering the up and down adjustment cost of the OLTC tap. The maximum switching times of the capacitor CB in the entire dispatching cycle 24 hours is 3, without considering the CB switching cost. The node voltage deviation allowable range is ±7%, and the maximum line current is 866A. According to the order of magnitude of each objective function and test experience, the weight coefficients in the objective function (A8) are set as follows: the weight coefficient of total network loss is 1, the weight coefficient of total voltage deviation is 100, the weight coefficient of PV net cost is 1, and the weight coefficients of other objective functions are all 0, that is, other objective functions are not considered; Cplex is selected as the solver, MIPGap is set to 1%, and TimeLimit is set to 1 hour.

[0074] Three comparative examples are designed: Example 1 only considers reactive voltage adjustment on the substation side, that is, only OLTC and CB participate in the adjustment; Example 2 only considers active and reactive adjustment on the feeder side, that is, only ESS, SVC, and PV inverter participate in the adjustment; Example 3 considers adjustment on both the substation side and the feeder side, that is, OLTC, CB, ESS, SVC, and PV inverter all participate in the adjustment. The comparison of the example results is as follows: Table 1 Comparison of calculation results

[0075]

[0076] The node voltage and network active power loss in different time periods under different examples are as follows: Figure 4a~Figure 6b shown.

[0077] Through Table 1 and Figure 4a~Figure 6bBy comparing the results of Examples 1 to 3, we can find that: in Examples 1 to 3, the net cost of photovoltaic power generation is negative and equal, and the amount of abandoned light is 0, which means that photovoltaic grid-connected power generation has generated benefits, and the distributed photovoltaic power generation in the distribution network has been fully absorbed by the equipment adjustment on the substation side and / or the feeder side; in Example 1, only the adjustment of the equipment on the substation side can also reduce the voltage deviation, because the load-controlled voltage-changing transformer raises the voltage at the head end of the feeder, making the voltage distribution of the entire feeder closer to the rated voltage as a whole; in Example 2, only the adjustment of the equipment on the feeder side is compared with the adjustment of the equipment on the substation side in Example 1, the voltage deviation is slightly larger, but the network loss is greatly reduced, because the adjustment on the feeder side alone does not raise the voltage at the head end, but through the energy storage system, the photovoltaic power generation equipment is adjusted to meet the requirements of the power grid-connected power generation. The PV grid-connected inverter and static VAR compensator adjust the active and reactive power, thereby changing the line flow and greatly reducing the network loss; at the same time, compared with only adjusting the substation side equipment in Example 1 and only adjusting the feeder side equipment in Example 2, the network loss and voltage deviation are reduced by adjusting the substation side and feeder side equipment in Example 3, indicating that the coordinated regulation of various active and reactive regulation equipment on the substation side and feeder side can not only reduce network losses but also improve voltage distribution, thereby achieving better overall operating performance of the distribution network; the example results verify the effectiveness of the multi-objective active-reactive collaborative optimization scheduling method for distribution networks with a high proportion of distributed photovoltaics proposed in the present invention in assisting distribution network dispatching and control managers in formulating distribution network operation strategies.

Claims

1. A method for coordinated optimization of active and reactive power dispatching in a distribution network containing a high proportion of distributed photovoltaic power generation, characterized in that: The following steps are involved: S1. Constructing a constraint model for the network flow constraints of the distribution network, the operation constraints of the reactive voltage regulation equipment on the substation side, and the operation constraints of the active and reactive regulation equipment on the feeder side; S2. Constructing an objective function of distribution network optimization control by weighted combination of calculation formulas of multiple optimization objectives; S3, constructing an optimization scheduling model, which uses all constraint models constructed in step S1 as constraint conditions to find the minimum value of the objective function constructed in step S2; S4. Convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution to schedule the distribution network.

2. The method for coordinated optimization of active and reactive power in distribution network according to claim 1, characterized in that: In step S1, the network flow includes node voltage, line current and power; the reactive voltage regulation equipment on the substation side includes the load-controlled tap-changing transformer and the capacitor bank; the active and reactive regulation equipment on the feeder side includes the energy storage system, the photovoltaic grid-connected inverter and the static VAR compensator.

3. The method for coordinated optimization of active and reactive power dispatching in distribution network according to claim 2, characterized in that: The model of the network flow constraint of the distribution network is: （1） （2） （3） （4） （5） （6） in, and They represent the square of the voltage amplitude at the head node i and the terminal node j of line ij in time period t, and Respectively represent the resistance and reactance values ​​of line ij, and They represent the active power and reactive power of line ij in time period t, respectively. represents the square of the current value of line ij in time period t, and They represent the active power and reactive power of line jk in time period t respectively, line jk represents the line with the first end node j and the last end node k, and set J represents the set of all lines with the first end node j; and They represent the net active load and net reactive load of node j in time period t, respectively. and They represent the actual active load and reactive load of node j in time period t, and They represent the active power and reactive power generated by the photovoltaic grid-connected inverter connected to node j in time period t, and They represent the charging and discharging active powers of the energy storage grid-connected point connected to node j in time period t, represents the reactive compensation power of the static VAR compensator connected to node j in time period t, represents the reactive compensation amount of the capacitor bank connected at node j in time period t, and They represent the minimum and maximum values ​​of the square of the node voltage, and They represent the minimum and maximum values ​​of the square of the line current respectively.

4. The method for coordinated optimization of active and reactive power in distribution network according to claim 3, characterized in that: The model of the operating constraints of the load-controlled tap-changing transformer is: （7） （8） （9） （10） （11） （12） Where Ntaps represents the total number of tap positions of the load-controlled tap-changing transformer. The state indicator variable indicating whether the tap position of the load-controlled tap-changing transformer in time period t is at the i-th tap, It represents the square of the node voltage value of the substation bus node sub in time period t, Indicates the transformer ratio corresponding to the tap position of the load-controlled tap-changing transformer at the ith tap. and The state indicator variables respectively indicate whether the tap position of the load-regulating transformer is adjusted up or down in time period t compared with the previous time period, It represents the maximum gear interval of the tap position of the load-controlled voltage-changing transformer for a single increase or decrease, T represents the set of all time periods within the entire scheduling time range, Indicates the total number of maximum up- and down-regulations of the load-controlled tap-changing transformer within the entire dispatching time range; The model of the operating constraints of the capacitor bank is: （13） （14） （15） （16） in, It represents the switch-on / off status indicator variable of the capacitor bank connected at node j in time period t, represents the reactive capacity of the capacitor bank connected at node j, The state indicator variable indicating whether the capacitor bank connected to node j starts to be put into or out of service in time period t, It represents the maximum number of times the capacitor bank connected to node j is allowed to be switched on and off within the entire scheduling time range.

5. The method for coordinated optimization of active and reactive power dispatching in distribution network according to claim 4, characterized in that: The model of the operating constraints of the energy storage system is: （17） （18） （19） （20） （21） （22） （23） （24） （25） （26） （27） in, and They represent the charging and discharging power of the energy storage system connected to node j in time period t, and They respectively represent the state indicator variables of whether the energy storage system connected to node j is charging or discharging in time period t, represents the maximum charging and discharging power of the energy storage system connected to node j, represents the capacity of the energy storage system connected to node j, and They represent the minimum and maximum ratios of the actual power and capacity of the energy storage system connected to node j, represents the remaining power of the ESS connected to node j after the end of time period t, Represents the last time period of the energy storage system connected to node j in the entire scheduling time range The remaining power after the end, represents the initial remaining power of the ESS connected to node j at the beginning of the first time period, represents the ratio of the initial actual power and capacity of the energy storage system connected to node j, and They represent the charging efficiency and discharging efficiency of the energy storage system connected at node j, Indicates the length of each time period, and They respectively represent the state change indicator variables of whether the energy storage system connected to node j starts charging and discharging in time period t, and represents the maximum number of charging and discharging times allowed for the energy storage system connected to node j within the entire scheduling time range; The model of the operating constraints of the photovoltaic grid-connected inverter is: （28） （29） （30） in, represents the active power output by the photovoltaic grid-connected inverter connected to node j in time period t; represents the reactive power output by the photovoltaic grid-connected inverter connected to node j in time period t; represents the maximum active power output by the photovoltaic grid-connected inverter connected to node j in time period t, represents the maximum power factor angle of the photovoltaic grid-connected inverter connected at node j, represents the apparent power capacity of the PV grid-connected inverter connected to node j; The model of the operating constraints of the static VAR compensator is: （31） in, and They represent the minimum and maximum reactive compensation powers of the static VAR compensator connected at node j respectively; represents the reactive compensation power of the static VAR compensator connected to node j in time period t.

6. The method for coordinated optimization of active and reactive power dispatching in distribution network according to claim 5, characterized in that: In step S2, the objective function is composed of weighted calculation formulas for minimizing network loss, minimizing voltage deviation, minimizing photovoltaic net power generation cost, minimizing energy storage system operating cost, minimizing static VAR compensator operating cost, minimizing capacitor bank switching cost, and minimizing load-controlled voltage-changing transformer gear switching cost.

7. The method for coordinated optimization of active and reactive power dispatching in distribution network according to claim 6, characterized in that: The calculation formula for the minimum network loss is: （A1） in, It represents the total loss of all lines in the distribution network within the entire dispatching time range. Represents the set of all lines; The calculation formula for the minimum voltage deviation is: （A2） in, It represents the total voltage deviation of all nodes in the distribution network in each time period. It represents the total number of time periods in the entire dispatching time range, BUS represents the set of all nodes in the distribution network, represents the square of the reference voltage value of node i in time period t; The formula for calculating the minimum net photovoltaic power generation cost is: （A3） in, represents the net power generation cost of all photovoltaics in the entire scheduling time range, represents the photovoltaic electricity cost, Indicates the photovoltaic grid-connected electricity price; The calculation formula for the minimum operating cost of the energy storage system is: （A4） in, represents the operating cost of all energy storage systems within the entire dispatch time range, Indicates the electricity cost of the feeder-side energy storage system; The calculation formula for the minimum operating cost of the static VAR compensator is: （A5） in, represents the operating cost of all static VAR compensators in the entire dispatch time range, It represents the unit reactive power cost of the static VAR compensator on the feeder side; The calculation formula for the minimum switching cost of capacitor banks is: （A6） in, It represents the switching cost of all capacitor banks within the entire dispatch time range; Indicates the single switching cost of the capacitor bank; The formula for calculating the minimum switching cost of the load-regulating transformer is: （A7） in, It represents the gear switching cost of the load-regulating transformer within the entire dispatching time range; It represents the cost of a single increase or decrease of the load-changing transformer.

8. The method for coordinated optimization of active and reactive power dispatching in distribution network according to claim 7, characterized in that: Step S4 also includes: convexifying the non-convex constraint condition corresponding to formula (2) by second-order cone relaxation approximation. The convex constraint condition after processing is: （2.S） The non-convex constraints corresponding to formulas (24) and (25) are equivalently processed using linear constraints. The processed linear constraints are: （24.L） （25.L） The circular feasible region corresponding to the non-convex constraint condition corresponding to formula (30) is linearized and approximated by the intersection of two square regions. The linear constraint condition after processing is: （30.L） Using auxiliary variables to replace , and add constraints to linearize the calculation formula (A2). The calculated formula after processing is: <h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr"> (A2.L) in, is the absolute value of the difference between the square of the voltage amplitude of node i in time period t and the square of the reference voltage value; Using auxiliary variables replace , and add constraints to linearize the calculation formula (A5). The calculated formula after processing is: <h2 style=";text-align:left;direction:ltr"> <h2 style=";text-align:left;direction:ltr"> (A5.L) in, is the absolute value of the reactive compensation amount of the static VAR compensator.

9. A distribution network active-reactive coordinated optimization dispatching system containing a high proportion of distributed photovoltaics, characterized in that: The system comprises: Constraint model building module: a constraint model used to build the network flow constraints of the distribution network, the operating constraints of the reactive voltage regulation equipment on the substation side, and the operating constraints of the active and reactive regulation equipment on the feeder side; Objective function construction module: used to construct the objective function of distribution network optimization control with multi-objective weighted combination; Optimization scheduling model construction module: used to construct an optimization scheduling model, which uses all constraint models constructed by the constraint model construction module as constraint conditions to find the minimum value of the objective function constructed by the objective function construction module; Solution scheduling module: It is used to convexify and linearize the non-convex constraints and objective functions containing absolute values ​​in the optimization scheduling model, transform the optimization scheduling model into a mixed integer second-order cone programming model and solve it, and use the solution results to schedule the distribution network.

10. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions which, when executed by a computing device, cause the computing device to perform any one of the methods according to claims 1 to 8.

Images