Active distribution network island division method and system based on information gap decision theory

Abstract

The present invention discloses an active distribution network islanding division method and system based on information gap decision theory, including the following steps: 1. Obtain historical data of distributed power sources and loads at each node in the distribution network, and predict wind power, photovoltaic output and load requirements at each node during the island power supply period; 2. Construct a distribution network islanding division model, and substitute the predicted values ​​of wind power, photovoltaics and loads for solution, and obtain an islanding division scheme and power outage losses based on a deterministic model; 3. Construct an uncertainty model that considers wind, light and load fluctuations based on information gap decision theory; 4. Select corresponding uncertainty models for solution according to different risk management strategies; 5. Obtain the final islanding division scheme. The islanding division scheme obtained by this method can withstand fluctuations in a certain range of distributed power output and load, ensure the stability of each island operation during the island power supply period, and minimize power outage losses.

Description

Technical Field

[0001] The invention belongs to the field of power system dispatching, and in particular relates to an active distribution network islanding division method and system based on information gap decision theory. Background Art

[0002] In recent years, as some extreme weather conditions have occurred frequently, power companies have paid more and more attention to the black start and island division of distribution networks. The widespread application of distributed power sources in distribution networks provides the possibility of island operation of distribution networks. When some lines in the distribution network fail or are under maintenance and the main grid power supply is lost, the distributed power sources in the distribution network are fully utilized to connect the surrounding loads for island power supply to ensure the operation of important loads, which is of great significance to reducing the economic losses caused by power outages and improving power supply reliability. Since the power generation capacity of distributed power sources in the distribution network is limited and does not have the strong voltage and frequency regulation capabilities of the main grid power supply, the power supply balance relationship within the island must be fully considered when dividing the island. The volatility of wind power, photovoltaic power, and load will bring risks to the stable operation of the island power grid.

[0003] There are two types of research on the uncertainty of wind, solar and load: one is based on the scenario method, which requires a large amount of historical data as support to solve the probability distribution of uncertain variables and uses a large number of typical prediction scenarios to represent the possible output of uncertain variables. The solution efficiency is usually low. The other type is to construct an output set of uncertain variables and perform robust optimization on the model. The results obtained by this method are usually too conservative and less economical.

[0004] In addition, in the island partitioning model, since factors such as switch status, load switching, and network flow constraints need to be considered, this problem is a mixed integer nonlinear programming problem. The traditional method of solving it using swarm optimization intelligent algorithms is usually inefficient, takes a long time to solve, and is prone to falling into local optimal solutions. Summary of the invention

[0005] In order to solve the technical problems existing in the background technology, the present invention aims to provide an active distribution network island partitioning method and system based on information gap decision theory, which does not require accurate probability distribution of wind and solar power output and load demand. At the same time, the island partitioning model is modeled as a mixed integer convex optimization model and solved using a commercial solver, with fast solution speed and high accuracy.

[0006] In order to solve the technical problem, the technical solution of the present invention is:

[0007] An active distribution network islanding division method based on information gap decision theory, the method comprising:

[0008] The historical data of distributed power sources and loads at each node in the distribution network are used to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and the demand forecast values ​​of wind power, photovoltaic and load are obtained;

[0009] Taking the minimization of economic losses of power outages during the island power supply period as the objective function, a distribution network island partitioning model is constructed.

[0010] Inputting the demand forecast value into the distribution network island partition model for solving, and obtaining the power outage loss based on the deterministic model;

[0011] Based on the island division method of power outage loss and information gap decision theory of deterministic model, the deviation coefficient of uncertain variables is calculated, and the uncertainty model considering wind, light and load fluctuations is constructed;

[0012] Select decision strategies, set deviation factors, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model;

[0013] Solve the robust model or chance model to determine the final island partitioning solution.

[0014] Furthermore, the final island division plan includes: the division range of each island, the node load recovery status, the branch switch status, and the output status of the distributed power source in each time period.

[0015] Furthermore, the objective function of the distribution network island partitioning model is:

[0016]

[0017] In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t It represents the recovery status of the load at node i during period t. If the load is restored, its value is 1, otherwise it is 0. Δt represents the unit period. The second term represents the loss of the distribution network, P loss Indicates the total network loss of the distribution network in the current period.

[0018] Furthermore, the constraints of the distribution network island partitioning model include: network flow constraints, wind and solar constraints, energy storage constraints, network topology constraints and operation constraints;

[0019] The network flow constraints are as follows:

[0020]

[0021]

[0022]

[0023] In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij in time period t; M is a sufficiently large positive number. Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation;

[0024] The scenery constraints are as follows:

[0025] 0≤P pw,t ≤P pw,max,t (5)

[0026] -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ (6)

[0027] Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t represents the upper limit of wind and solar power output during period t, and γ represents the maximum adjustable power factor angle of wind and solar power;

[0028] The energy storage constraints are as follows:

[0029]

[0030] SOC min ≤SOC t ≤SOC max (8)

[0031] SOC t+1 =SOC t +ηP ch(t)Δt-P dis (t)Δt / η (9)

[0032] Where P ch (t) and P dis (t) are the energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage, respectively, and η is the energy storage charging and discharging efficiency;

[0033] The network topology constraints are as follows:

[0034]

[0035]

[0036]

[0037]

[0038] In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island partition variable, 1 means branch ij belongs to island s, α ij Indicates the open and closed state of branch ij, 1 indicates that the branch is closed, 0 indicates that the branch is disconnected, and |S| indicates the number of distribution network islands. Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirement; There are bilinear terms in formula (11), and its linear transformation is as shown in formula (14):

[0039]

[0040] The operating constraints are as follows:

[0041] U min ≤U i,t ≤U max (15)

[0042] I min ≤I ij,t ≤I max (16)

[0043] Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

[0044] Further, the calculation of the uncertainty variable to obtain the deviation coefficient specifically includes:

[0045] The net load P is determined as the difference between the user's electricity demand and the wind and solar power output. The fluctuation range of the net load P is as follows:

[0046]

[0047] In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1;

[0048] Determine the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying:

[0049]

[0050] τ1+τ2+τ3=1 (19)

[0051] By using the maximum standard deviation of historical data to measure the fluctuation range of each uncertain variable, the deviation coefficient corresponding to each uncertain variable can be determined;

[0052]

[0053] In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ max It means taking the standard deviation of each hourly data of the variable for 30 days, and then taking the maximum value of the standard deviation, and taking the δ of each uncertain variable max The value is used as the basis for determining the coefficient of deviation;

[0054] If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then:

[0055] τ1:τ2:τ3=δ1:δ2:δ3 (21)

[0056] Combining equations (18) and (19), we can obtain the deviation coefficient ξ corresponding to wind power, photovoltaic power and load: wind , pv , load .

[0057] Furthermore, the uncertainty model considering wind, light and load fluctuations is constructed based on the information gap decision theory, specifically:

[0058] Robust Model:

[0059]

[0060] In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0. wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 load is the predicted value. d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage when the uncertain variables of wind, solar and load take fixed values;

[0061] Opportunity Model:

[0062]

[0063] In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating f r The degree of deviation from f0.

[0064] An active distribution network islanding division system based on information gap decision theory, the system comprising:

[0065] The prediction module is used to use the historical data of distributed power sources and loads at each node in the distribution network to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and obtain the demand forecast values ​​of wind power, photovoltaic and load;

[0066] A construction module is used to construct a distribution network island partitioning model with the objective function of minimizing the economic loss of power outage in the distribution network during the island power supply period;

[0067] A first solution module is used to input the demand forecast value into the distribution network island partition model for solution to obtain the power outage loss based on the deterministic model;

[0068] The calculation construction module is used to calculate the island division method of power outage loss and information gap decision theory based on the deterministic model, calculate the deviation coefficient of uncertain variables, and build an uncertainty model considering wind, light, and load fluctuations;

[0069] The determination module is used to select the decision strategy, set the deviation factor, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model;

[0070] The second solution module is used to solve the robust model or the chance model to determine the final island partitioning solution.

[0071] Further, the construction module includes: a construction subunit;

[0072] The construction subunit is used to construct a distribution network island partition model with network power flow, wind and solar, energy storage, network topology and operation as constraints and the minimum economic loss of power outage in the distribution network during the island power supply period as the objective function;

[0073] The objective function of the distribution network island partitioning model is:

[0074]

[0075] In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t It represents the recovery status of the load at node i during period t. If the load is restored, its value is 1, otherwise it is 0. Δt represents the unit period. The second term represents the loss of the distribution network, P loss Indicates the total network loss of the distribution network in the current period;

[0076] The network flow constraints are as follows:

[0077]

[0078]

[0079]

[0080] In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij in time period t; M is a sufficiently large positive number. Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation;

[0081] The scenery constraints are as follows:

[0082] 0≤P pw,t ≤P pw,max,t (5)

[0083] -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ (6)

[0084] Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t represents the upper limit of wind and solar power output during period t, and γ represents the maximum adjustable power factor angle of wind and solar power;

[0085] The energy storage constraints are as follows:

[0086]

[0087] SOC min ≤SOC t ≤SOC max (8)

[0088] SOC t+1 =SOCt +ηP ch (t)Δt-P dis (t)Δt / η (9)

[0089] Where P ch (t) and P dis (t) are the energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage, respectively, and η is the energy storage charging and discharging efficiency;

[0090] The network topology constraints are as follows:

[0091]

[0092]

[0093]

[0094]

[0095] In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island partition variable, 1 means branch ij belongs to island s, α ij Indicates the open and closed state of branch ij, 1 indicates that the branch is closed, 0 indicates that the branch is disconnected, and |S| indicates the number of distribution network islands. Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirement; There are bilinear terms in formula (11), and its linear transformation is as shown in formula (14):

[0096]

[0097] The operating constraints are as follows:

[0098] U min ≤U i,t ≤Umax (15)

[0099] I min ≤I ij,t ≤I max (16)

[0100] Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

[0101] Furthermore, the calculation building module includes: a calculation subunit, which is used to calculate the deviation coefficient of the uncertain variables of wind, light and load, specifically including:

[0102] The net load P is determined as the difference between the user's electricity demand and the wind and solar power output. The fluctuation range of the net load P is as follows:

[0103]

[0104] In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1;

[0105] Determine the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying:

[0106]

[0107] τ1+τ2+τ3=1 (19)

[0108] By using the maximum standard deviation of historical data to measure the fluctuation range of each uncertain variable, the deviation coefficient corresponding to each uncertain variable can be determined;

[0109]

[0110] In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ max It means taking the standard deviation of each hourly data of the variable for 30 days, and then taking the maximum value of the standard deviation, and taking the δ of each uncertain variable max The value is used as the basis for determining the coefficient of deviation;

[0111] If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then:

[0112] τ1:τ2:τ3=δ1:δ2:δ3 (21)

[0113] Combining equations (18) and (19), we can obtain the deviation coefficient ξ corresponding to wind power, photovoltaic power and load: wind , pv , load .

[0114] Furthermore, the calculation building module further includes: a first building unit and a second building unit, which are used to build an uncertainty model considering wind, light and load fluctuations, specifically:

[0115] The first construction unit is used to construct a robust model:

[0116]

[0117] In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0. wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 load is the predicted value. d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage when the uncertain variables of wind, solar and load take fixed values;

[0118] The second building block is used for the chance model:

[0119]

[0120] In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating fr The degree of deviation from f0.

[0121] A computer storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are used to implement any of the above methods when executed by a processor.

[0122] Compared with the prior art, the advantages of the present invention are:

[0123] 1. The traditional island partitioning scheme does not consider the uncertainty of distributed power sources and loads in the distribution network. All parameters of the model are solved under certain conditions. The resulting scheme may not be implemented when facing large fluctuations in distributed power sources and loads, resulting in additional economic losses.

[0124] 2. Compared with the scenario method, the method proposed in the present invention does not need to obtain the probability distribution of uncertain parameters and does not require a large amount of historical data as data support.

[0125] 3. Compared with the traditional robust optimization scheme, the active distribution network islanding method based on the information gap decision theory proposed in the present invention can pre-set the target value, so that the final decision scheme is no worse than the preset minimum acceptable result within the fluctuation range of the uncertain variables. It avoids the overly conservative decision-making of the traditional robust optimization scheme, and can also cope with a certain degree of fluctuation of the uncertain variables. BRIEF DESCRIPTION OF THE DRAWINGS

[0126] Figure 1 It is a flowchart of the active distribution network islanding division method based on information gap decision theory;

[0127] Figure 2 It is the topological diagram of the test distribution network structure;

[0128] Figure 3 This is the result diagram of distribution network island division. DETAILED DESCRIPTION

[0129] The specific implementation mode of the present invention is described below in conjunction with embodiments:

[0130] It should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0131] At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" cited in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments to their relative relationships should be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0132] Embodiment 1:

[0133] The present invention provides an active distribution network islanding division method based on information gap decision theory, and the specific steps are as follows:

[0134] Step 1: Taking the minimization of the economic loss of power outage in the distribution network during the island power supply period as the objective function, and taking the network flow constraint, wind and solar constraint, energy storage constraint, network topology constraint, operation constraint and other constraints as constraints, a distribution network island partition model is constructed. The model is a mixed integer second-order cone optimization model. Wind power, photovoltaic, load forecast values ​​and distribution network structure data are brought into the model for solution to obtain the economic loss f0 of power outage in the distribution network based on wind and solar and load forecast values.

[0135] Step 2: Based on the information gap decision theory, an uncertainty model is constructed that takes into account the fluctuations of wind, light, and load. According to different risk management strategies, it can be divided into a robust model and an opportunity model. Among them, the robust model corresponds to the risk avoidance strategy, and the opportunity model corresponds to the risk preference strategy.

[0136] Step 3: The decision maker selects a decision strategy and sets a deviation factor to determine the maximum economic loss that can be tolerated due to power outage (robust model) or the minimum economic loss that can be pursued due to power outage (opportunity model).

[0137] Step 4: Solve the robust / opportunistic model and determine the island division plan, including the division range of each island, node load recovery status, branch switch status, and distributed power generation output in each period.

[0138] The island partition model described in step 1 takes the minimization of economic losses during power outages during the island power supply period as the objective function, as shown below:

[0139]

[0140] In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t It represents the recovery status of the load at node i during period t. If the load is restored, its value is 1, otherwise it is 0. Δt represents the unit period. The second term represents the loss of the distribution network, P loss Indicates the total network loss of the distribution network in the current period.

[0141] The constraints of the island partition model in step 1 include network flow constraints, wind and solar constraints, energy storage constraints, network topology constraints, and operation constraints as follows:

[0142] Network flow constraints:

[0143]

[0144]

[0145]

[0146] In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij during period t; M is a sufficiently large positive number. Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation.

[0147] Scenery constraints:

[0148] 0≤P pw,t ≤P pw,max,t (5)

[0149] -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ (6)

[0150] Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t It represents the upper limit of wind and solar power output in period t, and γ represents the maximum adjustable power factor angle of wind and solar power.

[0151] Energy storage constraints:

[0152]

[0153] SOC min ≤SOC t ≤SOC max (8)

[0154] SOC t+1 =SOC t +ηP ch (t)Δt-P dis (t)Δt / η (9)

[0155] Where P ch (t) and P dis (t) are the energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage respectively, and η is the energy storage charging and discharging efficiency.

[0156] Network topology constraints:

[0157]

[0158]

[0159]

[0160]

[0161] In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island partition variable, 1 means branch ij belongs to island s, α ijIndicates the open and closed state of branch ij, 1 indicates that the branch is closed, 0 indicates that the branch is disconnected, and |S| indicates the number of distribution network islands. Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirements. There are bilinear terms in formula (11), and their linear transformation is shown in formula (14):

[0162]

[0163] Operation constraints:

[0164] U min ≤U i,t ≤U max (15)

[0165] I min ≤I ij,t ≤I max (16)

[0166] Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

[0167] Step 2 intends to build an island partition model based on information gap decision theory. The uncertain variables in the model include wind power, photovoltaic output and load. Net load P is defined as the difference between user power demand and wind and photovoltaic output. The fluctuation range of net load P is as follows:

[0168]

[0169] In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1.

[0170] The present invention defines the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying:

[0171]

[0172] τ1+τ2+τ3=1 (19)

[0173] The present invention uses historical data and the maximum standard deviation of each data to measure the fluctuation range of each uncertain variable, thereby determining the deviation coefficient corresponding to each uncertain variable.

[0174]

[0175] In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ max It means taking the standard deviation of each hourly data of the variable for 30 days and then taking the maximum standard deviation. max The value is used as the basis for determining the coefficient of deviation.

[0176] If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then:

[0177] τ1:τ2:τ3=δ1:δ2:δ3 (21)

[0178] Combining equations (18) and (19), the deviation coefficient ξ corresponding to wind power, photovoltaic power, and load can be obtained: wind , pv , load .

[0179] The uncertainty model considering wind, light and load fluctuations based on the information gap decision theory described in step 2 is as follows:

[0180] Robust Model:

[0181]

[0182] In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0. wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 loadis the predicted value. d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage for island division when uncertain variables such as wind, solar, and load take fixed values.

[0183] Opportunity Model:

[0184]

[0185] In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating f r The degree of deviation from f0.

[0186] Embodiment 2:

[0187] This embodiment uses the 69-node system of the PG&E company's distribution network in the United States as a test example for simulation testing. The system structure is as follows: Figure 2 There are distributed power sources at nodes 5, 19, 36, and 62. The power sources at nodes 5 and 36 are wind-storage combined systems, and the power sources at nodes 19 and 62 are photovoltaic-storage combined systems. Each power source has a certain voltage and frequency regulation capability and can serve as the main power source of the island. The parameters of distributed power sources and energy storage in the distribution network are shown in Table 1.

[0188] Table 1 Wind and solar power generation parameters

[0189] Power Type Maximum active power / kW Maximum reactive power / kvar Minimum power factor Wind power 500 / 0.9 Photovoltaic 400 / 0.9

[0190] Table 2 Energy storage system parameters

[0191] Maximum charge / discharge power / kW Energy storage capacity / kWh Min. / max. SOC / pu Charge and discharge efficiency 500 1000 0.05 / 0.95 0.95

[0192] Scene setup:

[0193] When the distribution network is disconnected from the main grid and loses the power supply of the main grid, the distributed power sources in the distribution network are used for short-term island power supply, and the island power supply duration is set to 6 hours. Considering the coordination and cooperation of different distributed power control strategies, this example sets the number of pre-divided islands to 4, and each island is powered by 4 distributed power sources as the main power supply, and the island fusion scenario is not considered.

[0194] The wind and solar power output and load forecast values ​​are brought into the island model constructed in step 1 to perform island division under deterministic conditions. The division results are as follows: Figure 3 As shown in Figure 2, under deterministic conditions, the economic loss caused by power outages due to island power supply is 13879

[0195] Island division method based on information gap decision theory:

[0196] First, the maximum standard deviation of wind power, photovoltaic power, and load is calculated based on historical data to determine the corresponding deviation coefficient ξ wind , pv , load In this example, we calculate ξ wind =0.48ξ、ξ pv =0.42ξ、ξ load =0.1ξ.

[0197] Robust Model: Setting the Robust Bias Factor is 0.2, then the maximum acceptable economic loss from power outage is 13879×(1+0.2)=16654.8, which is brought into the robust model for solution, and the solution is ξ=0.3226, that is, the maximum fluctuation range of net load that the distribution network can withstand is 32.26%.

[0198] Chance Model: Setting Chance Bias Factor If ξ is 0.2, the minimum economic loss from power outage is 13879×(1-0.2)=11103.2. When it is brought into the opportunity model for solution, we get ξ=0.433, that is, when the net load of the distribution network changes in a favorable direction, the minimum fluctuation range is 43.3%.

[0199] In order to reflect the superiority of the active distribution network islanding method based on information gap decision theory proposed in the present invention, the traditional robust method is used to perform islanding of the distribution network for comparison. According to the requirements of my country's new energy prediction error standard, it is assumed that the actual output fluctuation deviation of wind power and photovoltaic power is 15% of the predicted value, and the load fluctuation range is 5%. The islanding division scheme is solved in the "worst scenario", and the economic loss results of power outages are shown in Table 3.

[0200] Table 3

[0201]

[0202] It can be seen from Table 3 that the active distribution network islanding method based on information gap decision theory proposed in the present invention can set the robust / opportunity deviation factor, select the economic loss of power outage within the tolerable range, obtain the response interval of uncertainty variables such as wind, solar, and load, and more accurately characterize the fluctuation range of uncertainty variables, thereby avoiding the disadvantage of overly conservative traditional robust optimization schemes and providing a decision-making basis for dispatchers.

[0203] Embodiment 3:

[0204] In order to better implement the above method, this embodiment provides an active distribution network islanding division system based on information gap decision theory;

[0205] For example, an active distribution network islanding division system based on information gap decision theory includes:

[0206] The prediction module is used to use the historical data of distributed power sources and loads at each node in the distribution network to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and obtain the demand forecast values ​​of wind power, photovoltaic and load;

[0207] A construction module is used to construct a distribution network island partitioning model with the objective function of minimizing the economic loss of power outage in the distribution network during the island power supply period;

[0208] A first solution module is used to input the demand forecast value into the distribution network island partition model for solution to obtain the power outage loss based on the deterministic model;

[0209] The calculation construction module is used to calculate the island division method of power outage loss and information gap decision theory based on the deterministic model, calculate the deviation coefficient of uncertain variables, and build an uncertainty model considering wind, light, and load fluctuations;

[0210] The determination module is used to select the decision strategy, set the deviation factor, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model;

[0211] The second solution module is used to solve the robust model or the chance model to determine the final island partitioning solution.

[0212] Further, the construction module includes: a construction subunit;

[0213] The construction subunit is used to construct a distribution network island partition model with network power flow, wind and solar, energy storage, network topology and operation as constraints and the minimum economic loss of power outage in the distribution network during the island power supply period as the objective function;

[0214] The objective function of the distribution network island partitioning model is:

[0215]

[0216] In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t It represents the recovery status of the load at node i during period t. If the load is restored, its value is 1, otherwise it is 0. Δt represents the unit period. The second term represents the loss of the distribution network, P loss Indicates the total network loss of the distribution network in the current period;

[0217] The network flow constraints are as follows:

[0218]

[0219]

[0220]

[0221] In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij in time period t; M is a sufficiently large positive number. Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation;

[0222] The scenery constraints are as follows:

[0223] 0≤P pw,t ≤P pw,max,t (5)

[0224] -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ (6)

[0225] Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t represents the upper limit of wind and solar power output during period t, and γ represents the maximum adjustable power factor angle of wind and solar power;

[0226] The energy storage constraints are as follows:

[0227]

[0228] SOC min ≤SOC t ≤SOC max(8)

[0229] SOC t+1 =SOC t +ηP ch (t)Δt-P dis (t)Δt / η (9)

[0230] Where P ch (t) and P dis (t) are the energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage, respectively, and η is the energy storage charging and discharging efficiency;

[0231] The network topology constraints are as follows:

[0232]

[0233]

[0234]

[0235]

[0236] In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island partition variable, 1 means branch ij belongs to island s, α ij Indicates the open and closed state of branch ij, 1 indicates that the branch is closed, 0 indicates that the branch is disconnected, and |S| indicates the number of distribution network islands. Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirement; There are bilinear terms in formula (11), and its linear transformation is as shown in formula (14):

[0237]

[0238] The operating constraints are as follows:

[0239] U min ≤U i,t ≤U max (15)

[0240] I min ≤I ij,t ≤I max (16)

[0241] Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

[0242] Furthermore, the calculation building module includes: a calculation subunit, which is used to calculate the deviation coefficient of the uncertain variables of wind, light and load, specifically including:

[0243] The net load P is determined as the difference between the user's electricity demand and the wind and solar power output. The fluctuation range of the net load P is as follows:

[0244]

[0245] In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1;

[0246] Determine the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying:

[0247]

[0248] τ1+τ2+τ3=1 (19)

[0249] By using the maximum standard deviation of historical data to measure the fluctuation range of each uncertain variable, the deviation coefficient corresponding to each uncertain variable can be determined;

[0250]

[0251] In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ maxIt means taking the standard deviation of each hourly data of the variable for 30 days, and then taking the maximum value of the standard deviation, and taking the δ of each uncertain variable max The value is used as the basis for determining the coefficient of deviation;

[0252] If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then:

[0253] τ1:τ2:τ3=δ1:δ2:δ3 (21)

[0254] Combining equations (18) and (19), we can obtain the deviation coefficient ξ corresponding to wind power, photovoltaic power and load: wind , pv , load .

[0255] Furthermore, the calculation building module further includes: a first building unit and a second building unit, which are used to build an uncertainty model considering wind, light and load fluctuations, specifically:

[0256] The first construction unit is used to construct a robust model:

[0257]

[0258] In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0. wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 load is the predicted value. d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage when the uncertain variables of wind, solar and load take fixed values;

[0259] The second building block is used for the chance model:

[0260]

[0261] In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating f r The degree of deviation from f0.

[0262] Embodiment 4:

[0263] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0264] To this end, an embodiment of the present invention provides a storage medium storing a plurality of instructions, which can be loaded by a processor to execute the steps in any active distribution network islanding method based on information gap decision theory provided by an embodiment of the present invention.

[0265] For example, the instruction can execute the following steps:

[0266] An active distribution network islanding division method based on information gap decision theory, the method comprising:

[0267] The historical data of distributed power sources and loads at each node in the distribution network are used to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and the demand forecast values ​​of wind power, photovoltaic and load are obtained;

[0268] Taking the minimization of economic losses of power outages during the island power supply period as the objective function, a distribution network island partitioning model is constructed.

[0269] Inputting the demand forecast value into the distribution network island partition model for solving, and obtaining the power outage loss based on the deterministic model;

[0270] Based on the island division method of power outage loss and information gap decision theory of deterministic model, the deviation coefficient of uncertain variables is calculated, and the uncertainty model considering wind, light and load fluctuations is constructed;

[0271] Select decision strategies, set deviation factors, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model;

[0272] Solve the robust model or chance model to determine the final island partitioning solution.

[0273] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0274] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0275] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0276] The preferred embodiments of the present invention are described in detail above, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

[0277] Many other changes and modifications may be made without departing from the concept and scope of the present invention.It should be understood that the present invention is not limited to the specific embodiments, and the scope of the present invention is defined by the appended claims.

Claims

1. An active distribution network islanding method based on information gap decision theory is characterized by: The method comprises: The historical data of distributed power sources and loads at each node in the distribution network are used to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and the demand forecast values ​​of wind power, photovoltaic and load are obtained; Taking the minimization of economic losses of power outages during the island power supply period as the objective function, a distribution network island partitioning model is constructed. Inputting the demand forecast value into the distribution network island partition model for solving, and obtaining the power outage loss based on the deterministic model; Based on the island division method of power outage loss and information gap decision theory of deterministic model, the deviation coefficient of uncertain variables is calculated, and the uncertainty model considering wind, light and load fluctuations is constructed; Select decision strategies, set deviation factors, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model; Solve the robust model or chance model to determine the final island partitioning solution; The final island division plan includes: the division range of each island, the node load recovery status, the branch switch status, and the output of the distributed power source in each period; The calculation of the deviation coefficient of the uncertain variable specifically includes: The net load P is determined as the difference between the user's electricity demand and the wind and solar power output. The fluctuation range of the net load P is as follows: In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1; Determine the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying: τ1+τ2+τ3=1(19) By using the maximum standard deviation of historical data to measure the fluctuation range of each uncertain variable, the deviation coefficient corresponding to each uncertain variable can be determined; In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ max It means taking the standard deviation of each hourly data of the variable for 30 days, and then taking the maximum value of the standard deviation, and taking the δ of each uncertain variable max The value is used as the basis for determining the coefficient of deviation; If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then: τ1:τ2:τ3=δ1:δ2:δ3(21) Combining equations (18) and (19), we can obtain the deviation coefficient ξ corresponding to wind power, photovoltaic power and load: wind , pv , ξl oa d.

2. The active distribution network islanding division method based on information gap decision theory according to claim 1 is characterized in that: The objective function of the distribution network island partitioning model is: In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t Indicates the recovery status of the load at node i during period t. If the load is restored to power, its value is 1, otherwise it is 0. Δt represents the unit period; The second term represents the distribution network loss, P loss Indicates the total network loss of the distribution network in the current period.

3. The active distribution network islanding division method based on information gap decision theory according to claim 1 is characterized in that: The constraints of the distribution network island partitioning model include: network flow constraints, wind and solar constraints, energy storage constraints, network topology constraints and operation constraints; The network flow constraints are as follows: In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij in period t; M is a sufficiently large positive number; Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation; The scenery constraints are as follows: 0≤P pw,t ≤P pw,max,t (5) -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ(6) Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t represents the upper limit of wind and solar power output during period t, and γ represents the maximum adjustable power factor angle of wind and solar power; The energy storage constraints are as follows: SOC min ≤SOC t ≤SOC max (8) SOCIETY t+1 =SOC t +ηP ch (t)Δt-P dis (t)Δt / η(9) Where P ch (t) and P dis (t) are energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage, respectively, and η is the energy storage charging and discharging efficiency; The network topology constraints are as follows: In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island partition variable, 1 means branch ij belongs to island s, α ij represents the open and closed state of branch ij, 1 represents branch closed, 0 represents branch disconnected, |S| represents the number of distribution network islands; Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirement; There is a bilinear term in Formula (11), and its linear transformation is shown in Formula (14): The operating constraints are as follows: IN min ≤U i,t ≤U max (15) I min ≤I ij,t ≤I max (16) Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

4. The active distribution network islanding division method based on information gap decision theory according to claim 1 is characterized in that: The uncertainty model considering wind, light and load fluctuations is constructed based on the information gap decision theory, specifically: Robust Model: In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0; ξ wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 load is the predicted value; d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage when the uncertain variables of wind, solar and load take fixed values; Opportunity Model: In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating f r The degree of deviation from f0.

5. An active distribution network island partitioning system based on information gap decision theory is characterized by: The system comprises: The prediction module is used to use the historical data of distributed power sources and loads at each node in the distribution network to predict the wind power, photovoltaic output and load demand at each node during the island power supply period, and obtain the demand forecast values ​​of wind power, photovoltaic and load; A construction module is used to construct a distribution network island partitioning model with the objective function of minimizing the economic loss of power outage in the distribution network during the island power supply period; A first solution module is used to input the demand forecast value into the distribution network island partition model for solution to obtain the power outage loss based on the deterministic model; The calculation construction module is used to calculate the island division method of power outage loss and information gap decision theory based on the deterministic model, calculate the deviation coefficient of uncertain variables, and build an uncertainty model considering wind, light, and load fluctuations; The determination module is used to select the decision strategy, set the deviation factor, and determine whether the uncertainty model considering wind, light, and load fluctuations is a robust model or a chance model; The second solution module is used to solve the robust model or the chance model to determine the final island partitioning solution.

6. The active distribution network islanding division system based on information gap decision theory according to claim 5 is characterized in that: The building module comprises: a building subunit; The construction subunit is used to construct a distribution network island partition model with network power flow, wind and solar, energy storage, network topology and operation as constraints and the minimum economic loss of power outage in the distribution network during the island power supply period as the objective function; The objective function of the distribution network island partitioning model is: In the formula, the first term represents the economic loss of load removal, T is the total power supply period of the island, N is the set of load nodes, ω i P represents the economic loss of power outage per unit load at node i, load,i is the active load demand of node i, Y i,t It represents the recovery status of the load at node i during period t. If the load is restored, its value is 1, otherwise it is 0. Δt represents the unit period. The second term represents the loss of the distribution network, P loss Indicates the total network loss of the distribution network in the current period; The network flow constraints are as follows: In the formula, u (j) is the set of all nodes upstream of node j; v (j) is the set of all nodes downstream of node j; P ij,t and Q ij,t are the active power and reactive power of branch ij in period t respectively; P DER,j,t and Q DER,j,t are the total active power and total reactive power output by the distributed generation at node j during period t; P D,j,t and Q D,j,t is the actual active and reactive power of the load at node j during period t; R ij and X ij are the resistance and reactance of branch ij respectively; α ij Indicates the open / closed state of branch ij, 1 indicates that the branch is closed, and 0 indicates that the branch is open; U sqr,i,t is the square of the voltage at node i during period t; I sqr,ij,t is the square of the current of branch ij in period t; M is a sufficiently large positive number; Formula (2) represents the node power balance equation, Formula (3) represents the relationship between adjacent node voltages, and Formula (4) represents the branch power definition after second-order cone relaxation; The scenery constraints are as follows: 0≤P pw,t ≤P pw,max,t (5) -P pw,max,t tanγ≤Q pw,t ≤P pw,max,t tanγ(6) Where P pw,t and Q pw,t P represents the actual active and reactive power of wind and solar power in period t, pw,max,t represents the upper limit of wind and solar power output during period t, and γ represents the maximum adjustable power factor angle of wind and solar power; The energy storage constraints are as follows: SOC min ≤SOC t ≤SOC max (8) SOCIETY t+1 =SOC t +ηP ch (t)Δt-P dis (t)Δt / η(9) Where P ch (t) and P dis (t) are energy storage charging / discharging power, P ch,max and P dis,max are the maximum charging / discharging power of energy storage, s ch,t and dis,t They are energy storage charge / discharge flags, 1 means the energy storage is in charge / discharge state, SOC t , SOC min , SOC max are the current, minimum and maximum charges of energy storage, respectively, and η is the energy storage charging and discharging efficiency; The network topology constraints are as follows: In the formula, S represents the set of distribution network islands, E represents the set of distribution network branches, and N represents the total number of distribution network nodes; c is represents the node island partition variable, 1 means node i belongs to island s, l s ij Represents the branch island division variable, 1 means branch ij belongs to island s, α ij represents the open and closed state of branch ij, 1 represents branch closed, 0 represents branch disconnected, |S| represents the number of distribution network islands; Formula (10) indicates that each node of the distribution network belongs to only one island; Formula (11) indicates that when a line ij belongs to a certain island, then the nodes i and j at both ends of the line must also belong to this island at the same time; Formula (12) indicates that when line ij does not belong to any island, line ij is in a disconnected state; Formula (13) indicates that the relationship between the number of distribution network nodes and the number of branches meets the radial requirement; There is a bilinear term in Formula (11), and its linear transformation is shown in Formula (14): The operating constraints are as follows: IN min ≤U i,t ≤U max (15) I min ≤I ij,t ≤I max (16) Where U i,t represents the voltage of node i during period t, U max , U min Indicates the upper and lower limits of node voltage, I ij,t represents the branch ij current during period t, I max ,I min Indicates the upper and lower limits of branch current.

7. The active distribution network islanding division system based on information gap decision theory according to claim 5 is characterized in that: The calculation building module includes: a calculation subunit, which is used to calculate the deviation coefficient of the uncertain variables of wind, light and load, specifically including: The net load P is determined as the difference between the user's electricity demand and the wind and solar power output. The fluctuation range of the net load P is as follows: In the formula, represents the predicted value of net load, ξ is the deviation coefficient of net load, 0<ξ<1; Determine the deviation coefficient corresponding to wind power, photovoltaic power and load as ξ wind , pv , load , the corresponding deviation coefficient weights are τ1, τ2, τ3, satisfying: τ1+τ2+τ3=1(19) By using the maximum standard deviation of historical data to measure the fluctuation range of each uncertain variable, the deviation coefficient corresponding to each uncertain variable can be determined; In the formula, x i,T represents the historical value of the variable in the Tth period on the i-th day, It represents the average value of the historical value of the variable in the Tth period in 30 days, δ max It means taking the standard deviation of each hourly data of the variable for 30 days, and then taking the maximum value of the standard deviation, and taking the δ of each uncertain variable max The value is used as the basis for determining the coefficient of deviation; If we obtain the δ of wind power, photovoltaic power and load in the distribution network max The values ​​are δ1, δ2, and δ3 respectively, then: τ1:τ2:τ3=δ1:δ2:δ3(21) Combining equations (18) and (19), we can obtain the deviation coefficient ξ corresponding to wind power, photovoltaic power and load: wind , pv , load .

8. The active distribution network islanding division system based on information gap decision theory according to claim 7 is characterized in that: The calculation building module further includes: a first building unit and a second building unit, which are used to build an uncertainty model considering wind, light and load fluctuations, specifically: The first construction unit is used to construct a robust model: In the formula, ξ represents the fluctuation range of the uncertain parameters, f0 represents the economic loss of power outage caused by island division based on wind and solar power output and load forecast values, and f c represents the maximum economic loss of power outage that decision makers can accept, is the deviation factor, indicating f c The degree of deviation from f0; ξ wind , pv , load are the deviation coefficients of wind power, photovoltaic power and load respectively, P wind , P pv , P load Represent the actual values ​​of wind power, photovoltaic power and load respectively, P 0 wind , P 0 pv , P 0 load is the predicted value; d represents the decision variable in the model, f(P wind , P pv , P load , d) function represents the objective function of economic loss of power outage when the uncertain variables of wind, solar and load take fixed values; The second building block is used for the chance model: In the formula, f r represents the minimum economic loss of power outage sought by decision makers, is the deviation factor, indicating f r The degree of deviation from f0.

Images