ESOP site selection method and device based on Wasserstein distance, terminal equipment and computer readable storage medium

Abstract

The invention discloses an ESOP site selection method and device based on a Wasserstein distance, terminal equipment and a computer readable storage medium, and the method comprises the steps: building a cost optimization model with the minimum comprehensive cost as a target according to the parameters of a power system, building a Wasserstein constraint, an ESOP constraint, a power flow constraint, a system operation constraint and a transformer substation power factor constraint based on the model, and carrying out the optimization of the Wasserstein constraint, the ESOP constraint, the power flow constraint, the system operation constraint and the transformer substation power factor constraint; the model is solved to obtain the current of each line, the voltage of each node, the active power between the nodes and the reactive power between the nodes, the LSI value and the VDI value of each node are calculated according to the solved parameters, and then a target node group is screened out according to the LSI value, the VDI value and a preset threshold value; and connecting each node in the target positive value node group with the corresponding sorted nodes in the target negative value node group to obtain an ESOP site selection combination. By implementing the invention, the efficiency and the energy utilization rate of the power system can be improved.

Description

Technical Field

[0001] The present invention relates to the field of ESOP site selection, and in particular to an ESOP site selection method, device, terminal equipment and computer-readable storage medium based on Wasserstein distance. Background Art

[0002] The widespread access of distributed power sources represented by wind power and photovoltaics to the distribution network can achieve green development and reap considerable environmental benefits. However, the increased penetration and uncertainty of distributed power sources cause changes in the flow direction of distribution network power flows, from the traditional one-way flow to two-way flow. It also causes problems such as voltage over-limit and increased peak-to-valley differences in the system, posing great challenges to the operation and control of the distribution network.

[0003] Based on the above problems, existing research has proposed intelligent soft switches to replace traditional interconnection switches to achieve flexible interconnection between distribution network feeders and make the system operation more reliable. ESOP, as a new type of power electronic device integrating intelligent soft switches and energy storage devices, can not only perform peak shaving and valley filling, but also realize continuous regulation of active / reactive power of distribution networks, and can be flexibly adjusted according to the real-time operating status of distribution networks. However, existing research still has some problems in dealing with variable optimization in ESOP site selection and sizing problems. With the expansion of the scale of distribution networks, the number of decision variables and constraints in the model will also increase rapidly, which is a great challenge to computational efficiency and computational time. Summary of the invention

[0004] The embodiments of the present invention provide an ESOP site selection method, apparatus, terminal equipment and computer-readable storage medium based on Wasserstein distance, which can improve the efficiency and energy utilization of the power system.

[0005] An embodiment of the present invention provides an ESOP site selection method based on Wasserstein distance, comprising:

[0006] Obtain power system parameters; power system parameters include: Wasserstein distance parameter, line resistance parameter, line reactance parameter, coefficient parameter, upper and lower limit parameters, time interval and nominal voltage;

[0007] According to the power system parameters, with the goal of minimizing the comprehensive cost, a cost optimization model is constructed, and Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints, and substation power factor constraints are constructed based on the cost optimization model;

[0008] Under the constraints of Wasserstein constraint, ESOP constraint, power flow constraint, system operation constraint and substation power factor constraint, the current of each line, the voltage of each node, the active power between each node and the reactive power between each node are generated when the comprehensive cost is minimized;

[0009] According to the current of each line, the voltage of each node, the active power between each node and the reactive power between each node, the LSI value and VDI value of each node are calculated;

[0010] According to the LSI value and VDI value of each node and the preset threshold, the target node group is screened out; the target node group includes: a target positive value node group and a target negative value node group;

[0011] Each node in the target positive node group is connected to the correspondingly ordered nodes in the target negative node group to obtain the site selection combination of ESOP.

[0012] Furthermore, the LSI value and VDI value of each node are calculated according to each line current, each node voltage, active power between each node, and reactive power between each node, including:

[0013] The total line loss of each line is calculated based on the line resistance parameters, the voltage of each node, the active power between each node and the reactive power between each node;

[0014] The total route loss of each line is partially derived and summed with respect to the active power between each node and the reactive power between each node, and then normalized to obtain the LSI value of each node;

[0015] Subtract the voltage of each node from the nominal voltage and perform normalization to obtain the VDI value of each node.

[0016] Furthermore, according to the LSI value and VDI value of each node and the preset threshold, the target node group is screened out, including:

[0017] Group all nodes according to the positive and negative LSI values;

[0018] Calculate the absolute weight of the LSI value and VDI value of each node in each group, and sort them from large to small to obtain positive node groups and negative node groups;

[0019] From the negative value node group, select the nodes whose absolute weight is greater than the preset threshold and use them as the target negative value node group;

[0020] From the positive node group, select the nodes with the same order as those in the target negative node group and use them as the target positive node group.

[0021] Furthermore, the cost optimization model is as follows:

[0022] minF=C LOSS +C CUR ;

[0023]

[0024] Among them, F represents the comprehensive cost, C LOSS Represents the system loss cost, C CUR represents the cost of curtailing wind and solar power, represents the line power loss cost, represents the ESOP power loss cost, c l Represents the line power loss cost coefficient, Ω l represents the line set, r ij represents the resistance of line ij, I ij represents the current flowing through line ij, c esop represents the ESOP power loss cost coefficient, Ω ESOP represents the ESOP collection, represents the ESOP power loss, c cur represents the cost coefficient of wind and solar power abandonment, Ω DG Indicates the set of wind and solar power abandonment, Indicates the maximum output power. Represents the penalty cost for curtailing wind and solar power.

[0025] Furthermore, the Wasserstein constraint is as follows:

[0026]

[0027] Among them, d w1 represents the Wasserstein metric under the 1-norm Wasserstein radius, d w∞ represents the Wasserstein metric under the infinite norm Wasserstein radius, represents a data set containing historical data of all distribution networks, ξ represents historical data of a single distribution network in the data set, and Φ represents and ξ, κ1 represents the spherical radius of the Wasserstein distance under the 1-norm, κ ∞ The radius of the sphere representing the infinite norm Wasserstein distance, T d represents the running time set, t represents the running time, P represents the empirical distribution, R represents the real distribution, represents the empirical distribution of historical data of the distribution network, β1 represents the confidence of the 1-norm Wasserstein distance, β ∞represents the confidence of the infinite norm Wasserstein distance, and Ξ represents the uncertainty set.

[0028] Based on the above method embodiment, the present invention provides a corresponding device embodiment, including: a power system parameter acquisition module, a model and constraint condition construction module, a model solving module, a node index value calculation module, a target node screening module, and an ESOP site selection module;

[0029] The power system parameter acquisition module is used to acquire power system parameters; the power system parameters include: Wasserstein distance parameter, line resistance parameter, line reactance parameter, coefficient parameter, upper and lower limit parameters, time interval and nominal voltage;

[0030] Model and constraint condition construction module, which is used to construct a cost optimization model based on power system parameters with the goal of minimizing the comprehensive cost, and to construct Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints, and substation power factor constraints based on the cost optimization model;

[0031] The model solving module is used to generate the current of each line, the voltage of each node, the active power between each node and the reactive power between each node when the comprehensive cost is minimized under the constraints of Wasserstein constraint, ESOP constraint, power flow constraint, system operation constraint and substation power factor constraint;

[0032] The node index value calculation module is used to calculate the LSI value and VDI value of each node according to the current of each line, the voltage of each node, the active power between each node and the reactive power between each node;

[0033] The target node screening module is used to screen out the target node group according to the LSI value and VDI value of each node and the preset threshold; the target node group includes: a target positive value node group and a target negative value node group;

[0034] The ESOP site selection module is used to connect each node in the target positive node group with the correspondingly ordered nodes in the target negative node group to obtain the ESOP site selection combination.

[0035] Further, the node index value calculation module includes: a total route loss calculation unit, a node LSI value calculation unit, and a node VDI value calculation unit;

[0036] A total line loss calculation unit, used to calculate the total line loss of each line according to line resistance parameters, voltages of each node, active power between each node and reactive power between each node;

[0037] A node LSI value calculation unit is used to perform partial derivative summation of the total route loss of each line with respect to the active power between each node and the reactive power between each node, and perform normalization processing to obtain the LSI value of each node;

[0038] The node VDI value calculation unit is used to subtract the voltage of each node from the nominal voltage and perform normalization processing to obtain the VDI value of each node.

[0039] Further, the target node screening module includes: a grouping unit, a sorting unit, a target negative value screening unit, and a target positive value screening unit;

[0040] A grouping unit, used to group all nodes according to the positive and negative values ​​of LSI values;

[0041] A sorting unit is used to calculate the absolute weight of the LSI value and VDI value of each node in each group, and sort them from large to small to obtain positive node groups and negative node groups;

[0042] A target negative value screening unit is used to screen out nodes whose absolute weights are greater than a preset threshold from the negative value node group and use them as the target negative value node group;

[0043] The target positive value screening unit is used to screen out nodes with the same order as those in the target negative value node group from the positive value node group, and use them as the target positive value node group.

[0044] Based on the above-mentioned method embodiment, the present invention provides a corresponding terminal device embodiment, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of the ESOP site selection method based on Wasserstein distance as described in the present invention are implemented.

[0045] Based on the above method embodiment, the present invention provides a corresponding computer-readable storage medium embodiment, including: a stored computer program, which controls the device where the computer-readable storage medium is located to execute the steps of the ESOP site selection method based on Wasserstein distance as described in the present invention when the computer program is running.

[0046] Compared with the prior art, the beneficial effects of the embodiment of this solution are:

[0047] The invention takes the minimum comprehensive cost as the goal, constructs a cost optimization model, and constructs Wasserstein constraints, ESOP constraints, flow constraints, system operation constraints and substation power factor constraints based on the cost optimization model. When solving the minimum comprehensive cost, the current of each line, the voltage of each node, the active power between each node and the reactive power between each node are considered. By introducing the Wasserstein constraint, the cost optimization model can be guided to consider the power flow distance between nodes when selecting the optimal line current distribution, so that the transmission loss and the problem of power load concentration can be effectively reduced, and the efficiency and energy utilization rate of the power system can be improved. Then, according to the solved current of each line, the voltage of each node, the active power between each node and the reactive power between each node, the LSI value and the VDI value of each node are calculated. The LSI value can reflect the sensitivity of the node to the loss. By screening the LSI value and the VDI value, the candidate positions of the node can be reduced, and the calculation efficiency can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of an ESOP site selection method based on Wasserstein distance provided by one embodiment of the present invention;

[0049] Figure 2 It is a structural schematic diagram of a distribution network in a planning area provided by an embodiment of the present invention;

[0050] Figure 3 It is a structural schematic diagram of an ESOP site selection device based on Wasserstein distance provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.

[0053] like Figure 1 As shown, an embodiment of the present invention provides an ESOP site selection method based on Wasserstein distance, the method at least comprising the following steps:

[0054] Step S1: acquiring power system parameters; the power system parameters include: Wasserstein distance parameters, line resistance parameters, line reactance parameters, coefficient parameters, upper and lower limit parameters, time intervals and nominal voltages;

[0055] For step S1, in the present invention, the Wasserstein distance parameters include: the spherical radius κ1 of the Wasserstein distance under the 1-norm, the spherical radius κ of the Wasserstein distance under the infinite norm ∞ , the confidence level of the 1-norm Wasserstein distance β1, and the confidence level of the infinite-norm Wasserstein distance β ∞ ;

[0056] The above line resistance parameters include: the resistance r of line ij ij ;

[0057] The above line reactance parameters include: the reactance X of line ij ij ;

[0058] The above coefficient parameters include: Line power loss cost coefficient c l , ESOP power loss cost coefficient c esop , wind and solar curtailment cost coefficient c cur , loss coefficient ESOP charging efficiency coefficient ρ ch , and ESOP discharge efficiency coefficient ρ dis ;

[0059] The above upper and lower limit parameters include: ESOP active power upper limit ESOP injected reactive power upper limit of node i The lower limit of reactive power injected by ESOP at node i ESOP capacity limit of node i The upper limit of the state of charge ESOP of the e-th ESOP e,max , the lower limit of the charge state of the e-th ESOP ESOP e,min , the upper limit of the charging power of the e-th ESOP The lower limit of the charging power of the e-th ESOP The upper limit of the discharge power of the e-th ESOP The lower limit of the discharge power of the e-th ESOP The upper limit of the voltage amplitude at node i is V i,max , the lower limit of the voltage amplitude at node i is V i,min , the upper limit I of the branch current of branch ij ij,max , the upper limit of active power generated by the substation The lower limit of active power output from the substation Upper limit of reactive power generated by substation The lower limit of reactive power generated by the substation Upper limit of substation power factor angle limit tan And the lower limit of the substation power factor angle limit tan

[0060] The above time interval includes: time interval Δt;

[0061] The above nominal voltages include: Nominal voltage V n .

[0062] The comprehensive cost F and system loss cost C mentioned in the invention LOSS , Cost of curtailing wind and solar power C CUR , Line power loss cost ESOP Power Loss Cost The current I flowing through line ij ij , ESOP power loss Maximum output power The penalty cost of curtailing wind and solar power Wasserstein metric d under 1-norm Wasserstein radius w1 , Wasserstein metric d under infinite norm Wasserstein radius w∞ , a dataset of all distribution network historical data The historical data of a single distribution network in the data set ξ, The joint distribution Φ of and ξ, the empirical distribution P, the true distribution R, and the empirical distribution of historical data of the distribution network Uncertainty set Ξ, discharge power of the e-th ESOP Charging power of the e-th ESOP ESOP injected active power at node i ESOP power loss at node i ESOP injected reactive power at node i The charge state of the e-th ESOP at time t ESOP e,t , the e-th ESOP charging power ESOP Initial Charge State ESOP e,initial , ESOP final charge state ESOP e,final , the active power P of line ij at time t ij,t , reactive power Q of line ij at time t ij,t , the current I flowing through line ij at time t ij,t , the active power P at node i at time t i,t , the voltage U at node i at time t i,t, the voltage amplitude V at node i at time s i,s , the substation generates active power at time s Substation generates reactive power at time s And the substation power factor angle tan These data items are all unknown data items and are calculated in the subsequent process of solving the cost optimization model.

[0063] Step S2: According to the power system parameters, with the goal of minimizing the comprehensive cost, a cost optimization model is constructed, and Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints, and substation power factor constraints are constructed based on the cost optimization model;

[0064] In a preferred embodiment, the cost optimization model is specifically:

[0065] minF=C LOSS +C CUR ;

[0066]

[0067] Among them, F represents the comprehensive cost, C LOSS Represents the system loss cost, C CUR represents the cost of curtailing wind and solar power, represents the line power loss cost, represents the ESOP power loss cost, c l Represents the line power loss cost coefficient, Ω l represents the line set, r ij represents the resistance of line ij, I ij represents the current flowing through line ij, c esop represents the ESOP power loss cost coefficient, Ω ESOP represents the ESOP collection, represents the ESOP power loss, c cur represents the cost coefficient of wind and solar power abandonment, Ω DG Indicates the set of wind and solar power abandonment, Indicates the maximum output power. Represents the penalty cost for curtailing wind and solar power.

[0068] In a preferred embodiment, the Wasserstein constraint is specifically:

[0069]

[0070] Among them, d w1 represents the Wasserstein metric under the 1-norm Wasserstein radius, d w∞represents the Wasserstein metric under the infinite norm Wasserstein radius, represents a data set containing historical data of all distribution networks, ξ represents historical data of a single distribution network in the data set, and Φ represents and ξ, k1 represents the spherical radius of the Wasserstein distance under the 1-norm, k ∞ The radius of the sphere representing the infinite norm Wasserstein distance, T d represents the running time set, t represents the running time, P represents the empirical distribution, R represents the real distribution, represents the empirical distribution of historical data of the distribution network, β1 represents the confidence of the 1-norm Wasserstein distance, β ∞ represents the confidence of the infinite norm Wasserstein distance, and Ξ represents the uncertainty set.

[0071] For step S2, the ESOP constraints are:

[0072]

[0073] ESOP e,min ≤ESOP e,t ≤ESOP e,max ;

[0074]

[0075] ESOP e,initial =ESOP e,final ;

[0076] in, represents the discharge power of the e-th ESOP, represents the charging power of the e-th ESOP, represents the ESOP injected active power at node i, represents the ESOP power loss of node i, represents the lower limit of the reactive power injected by ESOP at node i, represents the reactive power injected by ESOP at node i, represents the upper limit of reactive power injected by ESOP at node i, represents the upper limit of ESOP capacity of node i, represents the loss factor, ESOP e,t represents the charge state of the e-th ESOP at time t, ρ ch represents the ESOP charging efficiency coefficient, represents the charging power of the e-th ESOP, ρ dis represents the ESOP discharge efficiency coefficient, Δt represents the time interval, represents the discharge power of the e-th ESOP, ESOP e,min Indicates the lower limit of the charge state of the e-th ESOP, ESOP e,max represents the upper limit of the state of charge of the e-th ESOP, represents the lower limit of the charging power of the e-th ESOP, represents the upper limit of the charging power of the e-th ESOP, represents the lower limit of the discharge power of the e-th ESOP, Indicates the upper limit of the discharge power of the e-th ESOP, ESOP e,initial Indicates the initial charge state of ESOP, ESOP e,final Indicates the final state of charge of the ESOP.

[0077] Power flow constraints, specifically:

[0078]

[0079] Among them, Ω l represents the set of lines, P ij,t represents the active power of line ij at time t, Q ij,t represents the reactive power of line ij at time t, R ij Indicates the resistance value of line ij, I ij,t represents the current flowing through line ij at time t, P i,t represents the active power at node i at time t, U i,t represents the voltage at node i at time t, X ij Represents the reactance value of line ij.

[0080] System constraints, specifically:

[0081] V i,min ≤V i,s ≤V i,max ;

[0082] I ij ≤I ij,max ;

[0083] Among them, V i,min Represents the lower limit of the voltage amplitude at node i, V i,s represents the voltage amplitude at node i at time s, V i,max Represents the upper limit of the voltage amplitude at node i, I ij,max Represents the upper limit of the branch current of branch ij.

[0084] Substation power factor constraints, specifically:

[0085]

[0086] in, Indicates the lower limit of the active power generated by the substation. Indicates the active power generated by the substation at time s, Indicates the upper limit of the active power generated by the substation. Indicates the lower limit of reactive power generated by the substation. Indicates that the substation generates reactive power at time s, Indicates the upper limit of reactive power generated by the substation, tan Indicates the substation power factor angle, tan Indicates the lower limit of the substation power factor angle limit, tan Indicates the upper limit of the substation power factor angle limit.

[0087] According to the power system parameters, a cost optimization model is established with the minimum comprehensive cost as the goal. Then, on the basis of constructing the cost optimization model, Wasserstein constraints, ESOP constraints, flow constraints, system operation constraints and substation power factor constraints are constructed. Through the Wasserstein constraints, while minimizing the cost, the power resources can be reasonably allocated to optimize the supply and demand matching of the power system, thereby avoiding insufficient or excessive power supply; through the ESOP constraints, the system can be guaranteed to operate within an economically reasonable power output range and avoid generator overload or insufficient load; through the flow constraints, the current and voltage of each component in the power system can be ensured to be within a safe range to prevent overload and failure; through the system operation constraints, the generator operating status, transmission line capacity and load balance can be restricted; through the substation power factor constraints, the power factor in the power system can be guaranteed to be within a reasonable range to improve the power quality of the system.

[0088] It should be noted that the Wasserstein constraint is based on the Wasserstein distance, where the Wasserstein distance is also called the Earth Mover's Distance (EMD), which is used to indicate the similarity between two distributions. It measures the minimum average distance required to move data from distribution p to distribution q.

[0089] Step S3: Under the constraints of Wasserstein constraint, ESOP constraint, power flow constraint, system operation constraint and substation power factor constraint, generate the current of each line, the voltage of each node, the active power between each node and the reactive power between each node when the comprehensive cost is minimized;

[0090] For step S3, specifically, under the above constraints, with the goal of minimizing the comprehensive cost, the original cost optimization model is transformed into a linear programming model with Wasserstein constraints:

[0091]

[0092] Among them, y represents the decision variable in the first stage, ξ represents the uncertainty factor, and x represents the operation strategy in the second stage;

[0093] st

[0094] y≥0

[0095] y=F3

[0096] G1x ≥ G2-G3y

[0097] J1x=J2-ξ 3t F3-ξ 4t F4

[0098]

[0099] The converted model is a distributed robust model. The CCG algorithm is used for iterative solution to obtain the line current, node voltage, active power between nodes, and reactive power between nodes when the comprehensive cost is minimized.

[0100] Step S4: Calculate the LSI value and VDI value of each node according to each line current, each node voltage, active power between each node, and reactive power between each node;

[0101] In a preferred embodiment, the LSI value and VDI value of each node are calculated according to each line current, each node voltage, active power between each node, and reactive power between each node, including:

[0102] The total line loss of each line is calculated based on the line resistance parameters, the voltage of each node, the active power between each node and the reactive power between each node;

[0103] The total route loss of each line is partially derived and summed with respect to the active power between each node and the reactive power between each node, and then normalized to obtain the LSI value of each node;

[0104] Subtract the voltage of each node from the nominal voltage and perform normalization to obtain the VDI value of each node.

[0105] For step S4, when the comprehensive cost is minimized, the current of each line, the voltage of each node, the active power between each node, and the reactive power between each node are calculated through step S3. After obtaining these parameters, the loss sensitivity index (LSI) is used to determine the candidate nodes that are sensitive to loss. The loss sensitivity index can measure the sensitivity of each node in the power system to the system loss. By calculating the LSI, those nodes that have a greater impact on the system loss can be determined. By determining the candidate nodes that are sensitive to loss, ESOP (Energy Storage and Optimization Placement) can be connected between these sensitive nodes, thereby achieving more efficient energy scheduling and allocation, reducing the energy transmitted over long distances, and reducing transmission losses.

[0106] According to the line resistance parameters, the voltage of each node, the active power between each node and the reactive power between each node, the total line loss of each line is calculated by the following formula:

[0107]

[0108] When the power flow flows from node i to node j, the LSI of node j can be expressed as:

[0109]

[0110] ESOP can not only regulate active power flow, but also provide reactive power for the system. Therefore, the LSI for ESOP site selection should calculate active power and reactive power at the same time. The partial derivatives of active power and reactive power can be summed to obtain:

[0111]

[0112] Among them, P ij,loss represents the total line loss of the line connecting nodes i and j, P ij represents the total active power flowing through line ij, Q ij represents the total reactive power flowing through line ij, r ij Represents line resistance, V j represents the voltage at node j, LSI P,j represents the LSI value of node j relative to active power, LSI q,j Represents the LSI value of node j relative to reactive power, LSI Pq,j represents the LSI value at node j.

[0113] The LSI values ​​of the calculated nodes are: some of them are positive and some are negative. The nodes are divided into positive node group and negative node group according to the positive and negative values ​​of LSI, and normalized by the following formula:

[0114]

[0115] For each node, the difference between the nominal voltage and the actual voltage is calculated and then normalized to obtain the voltage deviation index (VDI) of each node:

[0116] VDI j =V n -V j ;

[0117]

[0118] Among them, LSI Pq,j,norm represents the normalized LSI value at node j, LSI Pq,max Indicates the maximum LSI value of the corresponding group, VDI j,norm represents the normalized VDI value at node j, VDI max Indicates the maximum value of VDIs in all nodes.

[0119] After the VDI value is calculated, it can be used to evaluate the degree to which the voltage of each node deviates from the nominal voltage, thereby judging the voltage stability of each node in the power system. A higher VDI value indicates that the node voltage deviation is large and there may be voltage stability problems.

[0120] Step S5: Filter out a target node group according to the LSI value and VDI value of each node and a preset threshold; the target node group includes: a target positive node group and a target negative node group;

[0121] In a preferred embodiment, the target node group is screened out according to the LSI value and VDI value of each node and a preset threshold, including:

[0122] Group all nodes according to the positive and negative LSI values;

[0123] Calculate the absolute weight of the LSI value and VDI value of each node in each group, and sort them from large to small to obtain positive node groups and negative node groups;

[0124] From the negative value node group, select the nodes whose absolute weight is greater than the preset threshold and use them as the target negative value node group;

[0125] From the positive node group, select the nodes with the same order as those in the target negative node group and use them as the target positive node group.

[0126] For step S5, all nodes are grouped according to the positive and negative values ​​of the LSI value, and then for each node, the absolute value of the LSI value and the VDI value is calculated and used as the weight. Then, the nodes in each group are sorted from large to small according to the weight, and the positive node group is obtained according to the sorted results. and negative node groups From the negative node group, select the nodes whose absolute weight is greater than the preset threshold and use these nodes as the target negative node group. From the positive node group, filter out nodes with the same order as those in the target negative node group, and use these nodes as the target positive node group By screening out the target positive node group and the target negative node group through the absolute weights of the LSI value and the VDI value, the stability problems existing in the power system can be identified, processed and adjusted in a targeted manner.

[0127] Step S6: Connect each node in the target positive node group with the correspondingly ordered nodes in the target negative node group to obtain the site selection combination of ESOP.

[0128] For step S6, each node in the target positive node group is connected to the correspondingly ordered node in the target negative node group to obtain potential paired nodes in the system. The nodes in the sequence are node 11 and node 14, and the target is the node group The nodes in the sequence are node 16 and node 32, then the site selection combinations of ESOP are: node 11 connected to node 16, node 14 connected to node 32.

[0129] The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and typical cases:

[0130] This embodiment is described by taking a typical IEEE 33-node system as the planning object. Figure 2 The figure shows the structure diagram of the distribution network in the planning area. Two groups of photovoltaic power generation are selected as examples of access to the distribution network, and the access nodes are nodes 21 and 28. Four groups of wind power generation are selected as examples of access to the distribution network, and the access nodes are nodes 11, 14, 16 and 32. The following six schemes are compared and analyzed:

[0131] Solution 1: No ESOP or distributed generation connected to the distribution network;

[0132] Solution 2: Connecting to distributed power sources without considering the power factor constraints of substations;

[0133] Solution 3: Consider the power factor constraint of the substation and connect to the distributed power source;

[0134] Solution 4: ESOP is connected to the tie switch, and the capacity of each port is not considered separately;

[0135] Solution 5: ESOP is connected to the tie switch, and the capacity of each port is considered separately;

[0136] Solution 6: The ESOP site selection and capacity optimization method based on Wasserstein proposed in the present invention.

[0137] The system loss costs, wind and solar power abandonment costs, and substation power factors of the six schemes are calculated respectively (Table 1), and the site selection and sizing results of ESOP are compared (Table 2).

[0138] Table 1. Cost of the solution

[0139] plan <![CDATA[C LOSS (10,000 yuan / year)]]> <![CDATA[C CUR (10,000 yuan / year)]]> Substation power factor <![CDATA[C total (10,000 yuan / year)]]> 1 364.63 0 0.861 364.63 2 274.87 38.62 0.546 313.49 3 267.34 53.89 0.844 321.23 4 253.92 16.03 0.946 269.95 5 248.04 13.28 0.949 261.32 6 220.84 7.29 0.951 228.13

[0140] Table 2 E-SOP planning scheme

[0141] plan ESOP Location ESOP capacity (kVA) Energy storage capacity (kVA) 1 - - - 2 - - - 3 - - - 4 25-29,18-33,9-15,21-8,12-22 1230 450 5 25-29,18-33,9-15,21-8,12-22 1215 470 6 30-16,25-14 945 400

[0142] When the power factor constraint of the substation is applied in Scheme 3, the cost of wind and solar power abandonment is 538,900 yuan, while the cost of wind and solar power abandonment is 386,200 yuan in Scheme 2. Since this power factor constraint forces the mathematical model to maintain a certain active and reactive exchange ratio of the substation, the reactive power compensation of the system is insufficient and the distributed generation cannot be dispatched at full load.

[0143] However, in Schemes 4 to 6, since ESOP is installed in the system and sufficient reactive power compensation is provided, the cost of wind and solar power abandonment is greatly reduced, and the power factor of the substation is also significantly improved. In Scheme 6, the method proposed by the present invention achieves the lowest comprehensive cost.

[0144] In terms of reducing power loss, after adopting the method proposed in the present invention, the system loss cost considering the line and ESOP power loss is reduced from 2.6784 million yuan in Scheme 3 to 2.2084 million yuan in Scheme 6, and the reduction of system loss is more significant. At the same time, by using the method proposed in Scheme 6, the optimal substation power factor (between 0.93 and 1) can be obtained. Although the power factor results of Schemes 5 and 6 are similar, Scheme 6 requires less ESOP capacity than Scheme 5, as shown in Table 2.

[0145] like Figure 3 As shown, based on the above method embodiment, a corresponding device embodiment is provided;

[0146] An embodiment of the present invention provides an ESOP site selection device based on Wasserstein distance, comprising: a power system parameter acquisition module, a model and constraint condition construction module, a model solving module, a node index value calculation module, a target node screening module, and an ESOP site selection module;

[0147] The power system parameter acquisition module is used to acquire power system parameters; the power system parameters include: Wasserstein distance parameter, line resistance parameter, line reactance parameter, coefficient parameter, upper and lower limit parameters, time interval and nominal voltage;

[0148] Model and constraint condition construction module, which is used to construct a cost optimization model based on power system parameters with the goal of minimizing the comprehensive cost, and to construct Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints, and substation power factor constraints based on the cost optimization model;

[0149] The model solving module is used to generate the current of each line, the voltage of each node, the active power between each node and the reactive power between each node when the comprehensive cost is minimized under the constraints of Wasserstein constraint, ESOP constraint, power flow constraint, system operation constraint and substation power factor constraint;

[0150] The node index value calculation module is used to calculate the LSI value and VDI value of each node according to the current of each line, the voltage of each node, the active power between each node and the reactive power between each node;

[0151] The target node screening module is used to screen out the target node group according to the LSI value and VDI value of each node and the preset threshold; the target node group includes: a target positive value node group and a target negative value node group;

[0152] The ESOP site selection module is used to connect each node in the target positive node group with the correspondingly ordered nodes in the target negative node group to obtain the ESOP site selection combination.

[0153] In a preferred embodiment, the node index value calculation module includes: a total route loss calculation unit, a node LSI value calculation unit, and a node VDI value calculation unit;

[0154] A total line loss calculation unit, used to calculate the total line loss of each line according to line resistance parameters, voltages of each node, active power between each node and reactive power between each node;

[0155] A node LSI value calculation unit is used to perform partial derivative summation of the total route loss of each line with respect to the active power between each node and the reactive power between each node, and perform normalization processing to obtain the LSI value of each node;

[0156] The node VDI value calculation unit is used to subtract the voltage of each node from the nominal voltage and perform normalization processing to obtain the VDI value of each node.

[0157] In a preferred embodiment, the target node screening module includes: a grouping unit, a sorting unit, a target negative value screening unit, and a target positive value screening unit;

[0158] A grouping unit, used to group all nodes according to the positive and negative values ​​of LSI values;

[0159] A sorting unit is used to calculate the absolute weight of the LSI value and VDI value of each node in each group, and sort them from large to small to obtain positive node groups and negative node groups;

[0160] A target negative value screening unit is used to screen out nodes whose absolute weights are greater than a preset threshold from the negative value node group and use them as the target negative value node group;

[0161] The target positive value screening unit is used to screen out nodes with the same order as those in the target negative value node group from the positive value node group, and use them as the target positive value node group.

[0162] It can be understood that the above-mentioned device item embodiments correspond to the method item embodiments of the present invention, and can implement the ESOP site selection method based on Wasserstein distance provided by any of the above-mentioned method item embodiments of the present invention.

[0163] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art may understand and implement the present invention without creative work.

[0164] Based on the above-mentioned embodiment of the ESOP site selection method based on Wasserstein distance, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the ESOP site selection method based on Wasserstein distance of any embodiment of the present invention is implemented.

[0165] Exemplarily, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more module elements may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0166] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0167] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0168] Based on the above method embodiment, another embodiment is provided: another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the ESOP site selection method based on Wasserstein distance described in any one of the above method embodiments of the present invention.

[0169] Wherein, the module / unit integrated with the ESOP site selection device / terminal equipment based on Wasserstein distance, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0170] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An ESOP site selection method based on Wasserstein distance, characterized in that: include: Obtain power system parameters; The power system parameters include: Wasserstein distance parameter, line resistance parameter, line reactance parameter, coefficient parameter, upper and lower limit parameters, time interval and nominal voltage; According to the power system parameters, a cost optimization model is constructed with the goal of minimizing the comprehensive cost, and Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints and substation power factor constraints are constructed based on the cost optimization model; Under the constraints of the Wasserstein constraint, ESOP constraint, power flow constraint, system operation constraint and substation power factor constraint, generate the current of each line, the voltage of each node, the active power between each node and the reactive power between each node when the comprehensive cost is minimized; Calculate the LSI value and VDI value of each node according to the line current, the node voltage, the active power between each node and the reactive power between each node; According to the LSI value and VDI value of each node and the preset threshold, a target node group is screened out; the target node group includes: a target positive node group and a target negative node group; Each node in the target positive node group is connected to the correspondingly ordered nodes in the target negative node group to obtain a site selection combination of ESOP.

2. The ESOP site selection method based on Wasserstein distance according to claim 1, characterized in that: According to the line currents, node voltages, active power between nodes, and reactive power between nodes, the LSI value and VDI value of each node are calculated, including: The total line loss of each line is calculated based on the line resistance parameters, the voltage of each node, the active power between each node and the reactive power between each node; The total route loss of each line is partially derived and summed with respect to the active power between each node and the reactive power between each node, and normalized to obtain the LSI value of each node; Subtract the voltage of each node from the nominal voltage and perform normalization to obtain the VDI value of each node.

3. The ESOP site selection method based on Wasserstein distance according to claim 2, characterized in that: According to the LSI value and VDI value of each node and the preset threshold, the target node group is screened out, including: Group all nodes according to the positive and negative LSI values; Calculate the absolute weight of the LSI value and VDI value of each node in each group, and sort them from large to small to obtain positive node groups and negative node groups; From the negative value node group, select the nodes whose absolute weight is greater than the preset threshold and use them as the target negative value node group; Nodes with the same order as those in the target negative node group are screened out from the positive node group and used as the target positive node group.

4. The ESOP site selection method based on Wasserstein distance according to claim 3, characterized in that: The cost optimization model is specifically: minF=C LOSS +C CUR ; Among them, F represents comprehensive cost, C LOSS Represents the system loss cost, C CUR represents the cost of curtailing wind and solar power, represents the line power loss cost, represents the ESOP power loss cost, c l represents the line power loss cost coefficient, Ω l represents the line set, r ij represents the resistance of line ij, I ij represents the current flowing through line ij, c esop represents the ESOP power loss cost coefficient, Ω ESOP represents the ESOP collection, represents the ESOP power loss, c cur represents the cost coefficient of wind and solar power abandonment, Ω DG Indicates the set of wind and solar power abandonment, Indicates the maximum output power. Represents the penalty cost for curtailing wind and solar power.

5. The ESOP site selection method based on Wasserstein distance according to claim 4, characterized in that: The Wasserstein constraint is specifically: Among them, d w1 represents the Wasserstein metric under the 1-norm Wasserstein radius, d w∞ represents the Wasserstein metric under the infinite norm Wasserstein radius, represents a data set containing all distribution network historical data, ξ represents the historical data of a single distribution network in the data set, and Φ represents and ξ, κ1 represents the spherical radius of the Wasserstein distance under the 1-norm, κ ∞ The radius of the sphere representing the infinite norm Wasserstein distance, T d represents the running time set, t represents the running time, P represents the empirical distribution, R represents the real distribution, represents the empirical distribution of historical data of the distribution network, β1 represents the confidence of the 1-norm Wasserstein distance, β ∞ represents the confidence of the infinite norm Wasserstein distance, and Ξ represents the uncertainty set.

6. An ESOP site selection device based on Wasserstein distance, characterized in that: include: Power system parameter acquisition module, model and constraint condition construction module, model solving module, node index value calculation module, target node screening module, and ESOP site selection module; The power system parameter acquisition module is used to acquire power system parameters; The power system parameters include: Wasserstein distance parameter, line resistance parameter, line reactance parameter, coefficient parameter, upper and lower limit parameters, time interval and nominal voltage; The model and constraint condition construction module is used to construct a cost optimization model based on power system parameters with the goal of minimizing the comprehensive cost, and to construct Wasserstein constraints, ESOP constraints, power flow constraints, system operation constraints, and substation power factor constraints based on the cost optimization model; The model solving module is used to generate the current of each line, the voltage of each node, the active power between each node and the reactive power between each node when the comprehensive cost is minimized under the constraints of the Wasserstein constraint, the ESOP constraint, the power flow constraint, the system operation constraint and the substation power factor constraint; The node index value calculation module is used to calculate the LSI value and VDI value of each node according to the current of each line, the voltage of each node, the active power between each node and the reactive power between each node; The target node screening module is used to screen out a target node group according to the LSI value and VDI value of each node and a preset threshold; the target node group includes: a target positive value node group and a target negative value node group; The ESOP site selection module is used to connect each node in the target positive node group with the correspondingly ordered nodes in the target negative node group to obtain an ESOP site selection combination.

7. The ESOP site selection device based on Wasserstein distance according to claim 6, characterized in that: The node index value calculation module includes: a total route loss calculation unit, a node LSI value calculation unit, and a node VDI value calculation unit; The total line loss calculation unit is used to calculate the total line loss of each line according to the line resistance parameter, the voltage of each node, the active power between each node and the reactive power between each node; The node LSI value calculation unit is used to perform partial derivative summation of the total route loss of each line with respect to the active power between each node and the reactive power between each node, and perform normalization processing to obtain the LSI value of each node; The node VDI value calculation unit is used to subtract the voltage of each node from the nominal voltage and perform normalization processing to obtain the VDI value of each node.

8. The ESOP site selection device based on Wasserstein distance according to claim 7, characterized in that: The target node screening module includes: a grouping unit, a sorting unit, a target negative value screening unit, and a target positive value screening unit; The grouping unit is used to group all nodes according to positive and negative LSI values; The sorting unit is used to calculate the absolute weight of the LSI value and the VDI value of each node in each group, and sort them from large to small to obtain a positive node group and a negative node group; The target negative value screening unit is used to screen out nodes whose absolute weights are greater than a preset threshold from the negative value node group, and use them as the target negative value node group; The target positive value screening unit is used to screen out nodes with the same order as those in the target negative value node group from the positive value node group, and use them as the target positive value node group.

9. A terminal device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the ESOP site selection method based on Wasserstein distance as described in any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium, characterized in that: include: A stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the ESOP site selection method based on Wasserstein distance as described in any one of claims 1 to 5.

Images