A two-stage partition optimization configuration method and device for voltage regulation resources of a high-density photovoltaic power distribution network, an electronic device, and a storage medium

Abstract

The application discloses a two-stage partition optimization configuration method of voltage regulation resources of a power distribution network containing high-density photovoltaic power, and belongs to the technical field of power coordination configuration planning. The method comprises the following steps: generating high-precision typical photovoltaic scenes and limit photovoltaic scenes based on historical big data; performing two-step site selection of voltage regulation resources by using partition rough selection to configure regions and partition fine selection to configure nodes; and constructing a two-stage power distribution network voltage regulation resource capacity determination model. The configuration method can fully mobilize the reactive power output potential of photovoltaic power, reduce the number of OLTC actions and the configuration capacity, greatly reduce the calculation dimension, improve the algorithm convergence ability, make the configuration and scenes more suitable, and improve the economy, robustness and applicability of the configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of power coordinated configuration planning, and in particular to a two-stage partitioned optimization configuration method, device, electronic equipment and storage medium for voltage regulation resources in a high-density photovoltaic distribution network. Background Art

[0002] With the rapid development of renewable energy, wind, solar, and other new energy sources are being connected to distribution networks at high density and penetration rates. This has led to overcapacity in renewable energy production, with some regions nearing capacity limits. Furthermore, there are issues with distribution network voltage exceeding limits, difficulties in peak regulation, and large voltage fluctuations. Coordinating the allocation of voltage regulation resources is the fundamental solution to these voltage exceeding limits and fluctuations in distribution networks containing high-density photovoltaic power generation.

[0003] Extensive research has been conducted on the coordinated configuration of voltage regulation resources. Chinese patent CN116316666A, "Method and System for Coordinated Optimization of Reactive Compensation Devices and Energy Storage," addresses the coordinated optimization of reactive compensation devices and energy storage; CN118523395A, "Method for Site Selection and Sizing of Energy Storage for High-Proportion Distributed Photovoltaic Grid-Connected Systems," addresses the zoning of site selection and sizing for voltage regulation resources; and CN117374999A, "Method and System for Two-Tier Optimization of Voltage Regulation Resources in Distribution Networks," addresses the coordinated optimization model and algorithm for voltage regulation resource configuration. Most of these studies are limited to single scenarios or typical daily scenarios, with location selection based on experience and configuration using a centralized optimization algorithm. Therefore, a two-stage zoning optimization configuration method for voltage regulation resources in high-density photovoltaic distribution networks is needed to address the low computational speed, poor convergence, poor robustness, and poor applicability of existing coordinated configuration methods for voltage regulation resources. Summary of the Invention

[0004] The purpose of this invention is to propose a two-stage zoning optimization method for voltage regulation resource allocation in a high-density photovoltaic distribution network to address the current technical problems of complex photovoltaic scenario generation methods, empirical or lack of specificity in the selection of regulation resource locations, single capacity configuration scenarios, and limited consideration of extreme scenarios, thereby ensuring stable voltage operation in the distribution network. The method includes the following steps:

[0005] Generate high-precision typical photovoltaic scenarios and extreme photovoltaic scenarios based on historical big data;

[0006] Based on typical and extreme photovoltaic scenarios, voltage regulation resources are sited in two steps using coarse selection of configuration areas within the zones and fine selection of configuration nodes within the zones.

[0007] According to the selected address of the voltage regulation resource, a two-stage distribution network voltage regulation resource sizing model is constructed, the first stage is to optimize the photovoltaic reactive power output of each subarea by time according to the voltage deviation, and the distribution network power flow is updated according to the photovoltaic reactive power output, the second stage is to establish a multi-scene double-layer capacity optimization configuration model considering the capacity / power demand, capacity / power cost, voltage out-of-limit times and degree, and multi-objective of network loss, voltage deviation and fluctuation, wherein the upper model is a configuration model including load regulation transformer, energy storage, SVG and adjustable capacitor, the lower model is a centralized operation optimization model considering multiple scenes, the upper model delivers the voltage regulation resource capacity configuration information to the lower model, and the lower model obtains the optimization result according to the voltage regulation resource capacity configuration information and feeds back to the upper model.

[0008] Further, the typical photovoltaic scene includes four seasons of photovoltaic large and small scenes.

[0009] Further, the constraint expression of photovoltaic reactive power output is as follows:

[0010]

[0011] Wherein, S inv,N represents the rated capacity of the photovoltaic inverter; P PV represents the active power output of the photovoltaic; Q PV represents the reactive power output of the photovoltaic.

[0012] Further, the upper model includes: device capacity / power demand model, capacity / power cost model and voltage out-of-limit times and degree model;

[0013] The calculation formula of the device capacity / power demand model is as follows:

[0014]

[0015] α1+α2+α3+α4=1

[0016] Wherein, N ESS is the number of energy storage installations; N CB is the number of adjustable capacitor CB submodules; N SVG is the number of SVG installations; M device is the capacity / power demand model; is the installed energy storage power of node i; is the installed capacity of energy storage of node i; is the installed capacity of adjustable capacitor of node i; is the installed SVG power of node i; α1, α2, α3, α4 are the installation power of energy storage, the installation capacity of energy storage, the installation capacity of adjustable capacitor and the installation capacity of SVG coefficient;

[0017] The calculation formula of the capacity / power consumption model is as follows:

[0018]

[0019] β1+β2=1

[0020] Wherein, M op is the power consumption model of the distribution network; p j is the proportion of the representative day of each typical scenario in the total number of years; N b , N l are the number of branches and the number of power purchase ports from the upper level respectively; N sen1 and N sen2 are the number of typical scenarios and the number of limit scenarios respectively; and are the power loss of branch λ at time t in scenario j and the input power of external upper grid node i at time t in scenario j respectively; β1 and β2 are the power loss and the power purchase coefficient from the upper level;

[0021] The calculation formula of the voltage out-of-limit times and degree model is as follows:

[0022]

[0023] Wherein, N ove,all represents the total voltage out-of-limit times of n limit scenarios; N ove,n represents the total voltage out-of-limit times of the nth limit scenario; U ove,all represents the voltage out-of-limit degree of n limit scenarios; U ove,n represents the voltage out-of-limit degree of the nth limit scenario; N sen1 and N sen2 are the number of typical scenarios and the number of limit scenarios respectively.

[0024] Further, the lower layer model comprises: a running power loss model, a running voltage deviation model, a running voltage fluctuation model, and a voltage out-of-limit times and degree model;

[0025] The calculation formula of the running power loss model is as follows:

[0026]

[0027] Wherein, P loss is the daily power loss model; p j is the proportion of the representative day of each typical scenario in the total number of years; is the power loss of branch i at time t in scenario j; N sen1 is the number of typical scenarios; and Nb is the number of branches.

[0028] The calculation formula of the running voltage deviation model is as follows:

[0029]

[0030] wherein, U Δ denotes the daily voltage deviation; N is the number of nodes of the power distribution network; T is the number of hours of the scenario; U i,j,t is the voltage of node i at time t in scenario j; U N is the rated voltage;

[0031] The calculation formula of the operating voltage fluctuation model is as follows:

[0032]

[0033] wherein, U FL denotes the daily voltage fluctuation; is the average voltage value of node i in T hours; U i,j,t is the voltage of node i at time t in scenario j; N is the number of nodes of the power distribution network; T is the number of hours of the scenario;

[0034] The calculation formula of the voltage out-of-limit times and degree model is as follows:

[0035]

[0036] wherein, a i,t denotes the state of voltage out-of-limit; N ove denotes the voltage out-of-limit times of each node in a period; b i,t denotes the voltage out-of-limit amplitude; U ove denotes the voltage out-of-limit degree of each node in a period, U i,t is the voltage of node i at time t; U max is the upper limit of voltage; U min is the lower limit of voltage.

[0037] Further, the constraint condition of the upper layer model is:

[0038]

[0039] wherein, P ESS,min , P ESS,max are the lower limit and upper limit of the rated power of the energy storage installation; E ESS,min , E ESS,max are the lower limit and upper limit of the rated capacity of the energy storage installation; Q SVG,min , Q SVG,max are the lower limit and upper limit of the installation capacity of the SVG; Q CB,min , Q CB,max are the lower limit and upper limit of the installation capacity of the CB; respectively denote the rated power of the energy storage, the rated capacity of the energy storage, the rated capacity of the SVG, and the rated capacity of the CB.

[0040] Further, the constraint conditions of the lower layer model include: power distribution network operation constraints, on-load tap-changing transformer operation constraints, energy storage operation constraints, SVG operation constraints, and adjustable capacitor operation constraints.

[0041] The calculation formula of the power distribution network operation constraints is as follows:

[0042]

[0043] wherein, P i,t and Q i,t are the active power and the reactive power injected by the node i at time t; G ij and B ij are the conductance and the susceptance of the line ij; U i,t is the voltage of the node i at time t; U j,t is the voltage of the node j at time t; θ ij is the phase angle difference between the voltage of the node i and the voltage of the node j; U min and U max are the upper limit and the lower limit of the node voltage; I ij,t is the current of the branch ij at time t; I ij,min and I ij,max are the upper limit and the lower limit of the current of the branch ij;

[0044] The calculation formula of the on-load tap-changing transformer operation constraints is as follows:

[0045]

[0046]

[0047] wherein, r t0 is the transformer ratio at time t0; r t is the transformer ratio at time t; Δr j is the ratio difference between adjacent gear positions of the transformer; and are the up and down gear operation of the transformer, when the up gear is operated, otherwise when the down gear is operated, otherwise is the maximum number of operations allowed by the transformer within T time; T is the regulation time of the transformer;

[0048] The calculation formula of the energy storage operation constraints is as follows:

[0049]

[0050]

[0051]

[0052]

[0053] SOC i,min E i,N ≤E i,t ≤SOC i,max E i,N

[0054] where, and denotes the state of charge of the energy storage at time t at node i; denotes the discharging power of the energy storage at time t; denotes the charging power of the energy storage at time t; and denotes the upper and lower limit constraints of the discharging power of the energy storage; and denotes the upper and lower limit constraints of the charging power of the energy storage;E i,t denotes the capacity of the energy storage at node i at time t;E i,t+1 denotes the capacity of the energy storage at node i at time t+1; and denotes the charging and discharging efficiency of the energy storage;SOC i,min and SOC i,max denotes the minimum and maximum state of charge of the energy storage;E i,N denotes the rated capacity of the energy storage;

[0055] The calculation formula of the SVG operation constraint is as follows:

[0056]

[0057] where, denotes the reactive power output of the SVG at time t at node i; denotes the rated capacity of the SVG at node i;

[0058] The calculation formula of the operation constraint of the adjustable capacitor is as follows:

[0059]

[0060]

[0061]

[0062] where, n i,t is the number of capacitors put into the sub-module at time t at node i; is the maximum number of sub-modules;N op,max is the maximum switching action number of the capacitor bank; denotes the unit capacity of the adjustable capacitor; Ci(t) represents the adjustable capacitor access capacity at node i at time t.

[0063] The application discloses a two-stage partition optimization configuration device of voltage regulation resources in a high-density photovoltaic power distribution network.

[0064] The scene generation module is used for generating high-precision typical photovoltaic scenes and limit photovoltaic scenes based on historical big data.

[0065] The two-step site selection module is used for performing two-step site selection of voltage regulation resources according to the typical photovoltaic scenes and the limit photovoltaic scenes, and performing coarse selection configuration of regions and fine selection configuration of nodes in the regions.

[0066] The double-layer optimization module is used for constructing a two-stage power distribution network voltage regulation resource capacity determination model according to the selected addresses of the voltage regulation resources, performing photovoltaic reactive power output optimization of each partition in the first stage with voltage deviation as a target, and updating power distribution network power flow according to the photovoltaic reactive power output, and establishing a multi-scene double-layer capacity optimization configuration model considering capacity / power demand amount, capacity / power cost amount, voltage out-of-limit times and degree, and network loss, voltage deviation and fluctuation, wherein an upper-layer model is a configuration model of a load voltage regulation transformer, energy storage, SVG and an adjustable capacitor, a lower-layer model is a centralized operation optimization model considering multiple scenes, the upper-layer model feeds capacity configuration information of the voltage regulation resources to the lower-layer model, and the lower-layer model performs optimization calculation to obtain an optimization result and feeds back to the upper-layer model.

[0067] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements each step in the two-stage partition optimization configuration method of voltage regulation resources in a high-density photovoltaic power distribution network when executing the computer program.

[0068] A storage medium has a computer program stored thereon, and each step in the two-stage partition optimization configuration method of voltage regulation resources in a high-density photovoltaic power distribution network is implemented when the computer program is executed by a processor.

[0069] The application has the following beneficial effects:

[0070] 1. The configuration model introduces annual scenes for capacity configuration, so that the configuration is closer to the actual situation, and the model applicability is improved.

[0071] 2. Before the capacity of the special voltage regulation resource is configured, the reactive power output of the photovoltaic inverter is preferentially optimized, so that the photovoltaic resource is fully mobilized to participate in voltage regulation, and the configuration capacity of the special voltage regulation resource is reduced.

[0072] 3. The photovoltaic reactive voltage regulation resource and the special voltage regulation resource are calculated in stages, so that the calculation dimension is greatly reduced, and the convergence ability of the calculation is improved.

[0073] 4. Fully mobilize the reactive output potential of photovoltaic, reduce the number of OLTC actions, and delay equipment wear and tear;

[0074] 5. The lower layer operation model of the double-layer configuration model optimizes typical scenarios and out-of-limit scenarios in parallel, which can make the configuration fully consider various voltage problems of the power distribution network and improve the pertinence and robustness of the model. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 The flowchart of the two-stage partition optimization configuration method of voltage regulation resources of the high-density photovoltaic power distribution network of the present application;

[0076] Figure 2 The flowchart of the cluster structure correction and the cluster center correction;

[0077] Figure 3 The schematic diagram of the upper layer model and the lower layer model. DETAILED DESCRIPTION

[0078] The present application provides a two-stage partition optimization configuration method of voltage regulation resources of a high-density photovoltaic power distribution network, which is further described below in combination with the drawings and specific embodiments.

[0079] Figure 1 The flowchart of the two-stage partition optimization configuration method of voltage regulation resources of the high-density photovoltaic power distribution network of the present application, which comprises the following steps:

[0080] Step S101: Establish high-precision typical photovoltaic scenario and limit photovoltaic scenario models based on historical big data.

[0081] Step S102: According to the typical photovoltaic scenario and the limit photovoltaic scenario, construct a two-step site selection model of voltage regulation resources based on power distribution network configuration partition, the first step is to use partition coarse selection to configure the region, and the second step is to fine select the configuration node in the partition.

[0082] Step S103: According to the selected address of the voltage regulation resource, a two-stage distribution network voltage regulation resource sizing model is constructed, the first stage is to optimize the photovoltaic reactive power output of each subarea by time with voltage deviation as the target, to fully mobilize the potential of photovoltaic spare resource output, reduce subsequent configuration investment, and update the distribution network power flow according to the photovoltaic reactive power output, and the second stage is to propose a multi-scenario double-layer capacity optimization configuration model considering capacity / power demand, capacity / power cost, voltage out-of-limit times and degree, and network loss, voltage deviation and fluctuation and other multi-objectives, the upper layer is the configuration model of on-load voltage regulating transformer (OLTC), energy storage, static var generator (SVG) and adjustable capacitor (CB), and the lower layer is a centralized operation optimization model considering multiple scenarios, the upper layer delivers the voltage regulation resource capacity configuration information to the lower layer model, and the lower layer model obtains the optimization result according to the capacity configuration information and feeds back to the upper layer model.

[0083] The typical scene generation method based on historical data in step S101 includes a cluster structure correction part and a cluster center correction part. The typical scene includes photovoltaic large and small generation scenes in four seasons.

[0084] The cluster structure correction part step includes:

[0085] First, collect the required historical data of solar radiation in the year and aggregate it by season; second, use the fuzzy C-means clustering algorithm to cluster each season, with 2 clusters for each season representing photovoltaic large and small generation; after clustering, calculate the silhouette coefficient (SC) of each particle and set a margin ε, in the particles with a silhouette coefficient less than the margin ε, select the particle with the smallest silhouette coefficient to belong to the adjacent class, and then calculate the silhouette coefficient in a loop until the silhouette coefficient is greater than the margin ε. The process of using cluster structure correction is shown in S201 of Figure 2 .

[0086] The cluster center correction step includes:

[0087] Using the cluster structure correction result, the improved particle swarm algorithm is used to revise the cluster center of each cluster with the minimum intra-cluster tightness as the fitness, so that the cluster center better represents the solar radiation level of the class. The process of using cluster center correction method is shown in S202 of Figure 2 . The expression of intra-cluster tightness is as follows:

[0088]

[0089] Wherein, CP is the intra-cluster tightness; n is the number of particles in the cluster; c is the cluster center; c i is the particle value.

[0090] The extreme photovoltaic scenario in step S101 is generated from historical data to obtain a summary curve of light radiation in each season, and then the upper and lower envelopes of the summary curve of light radiation in each season are obtained respectively. The ideal expression of converting light radiation into photovoltaic output power is as follows:

[0091]

[0092] wherein, P PV,N is the rated power of photovoltaic; S is the actual light intensity; S N is the rated light intensity; P PV is the actual photovoltaic output power.

[0093] The two-step locating model based on voltage regulation resources in the power distribution network in step S102 includes: using a partition coarse selection configuration region model and a partition-based fine selection configuration node model.

[0094] The partition in the partition coarse selection configuration region model is divided into a reactive power dynamic partition model based on typical and extreme scenarios and a comprehensive dynamic partition model. The reactive power dynamic partition model only considers the reactive power influence and is used for configuring SVG. The SVG is preferentially selected to configure the reactive power partition with poor balance degree. The comprehensive dynamic partition model considers both the reactive power influence and the active power influence and is used for configuring energy storage. Due to the high-density distribution characteristics of photovoltaic, at least one energy storage is configured in each comprehensive partition.

[0095] The dynamic partition model includes: comprehensive index construction considering structural index and functional index and annual power distribution network partition model based on seasonal partition.

[0096] The comprehensive index construction step considering the structural index and the functional index in seasons includes: constructing the structural index of the partition by constructing the module degree function with the voltage sensitivity index value, and constructing the functional index of the partition by considering the active power and reactive power balance degree in the region.

[0097] The structural index and the functional index are respectively given weights by using the analytic hierarchy process to generate the weighted comprehensive index. Since each season contains two typical scenarios, and each typical scenario contains 24h photovoltaic and load output, the time-varying nature of the voltage sensitivity and the regional balance degree and the different probabilities of each scenario are considered. The seasonal comprehensive index should be given in the form of expectation, and the genetic algorithm is used to solve the partition results of each season.

[0098] Specifically, the structural index, the functional index and the comprehensive index described above can be determined by formulas (3)-(14):

[0099]

[0100]

[0101]

[0102] where S Pδ , S Qδ , S PV , S QV represents power-phase angle sensitivity and power-voltage sensitivity; is the active-voltage sensitivity of node i to node j in a quarter; T is a period, usually T = 24; n s is the number of scenarios; p s is the probability of scenario s; is the active-voltage sensitivity of node i to node j at time t; is the sensitivity distance of node i to node j; is the electrical distance between i and j; γ P is the active-voltage sensitivity index, the reactive-voltage sensitivity index is the same as the active-voltage sensitivity index; m is the sum of edge weights of the network; k i represents the sum of edge weights of all nodes.

[0103]

[0104]

[0105] where c is the partition category; n c is the number of nodes in c area; N c is the number of partitions; s is the typical scenario category; n s is the number of typical scenarios in each quarter; is the active balance degree of c area under s scenario; T is the period of typical scenario, usually T = 24; is the active power of distributed photovoltaic of node i in c area at time t under s scenario; is the active power of load of node i in c area at time t under s scenario; p s is the probability of the s-th scenario in a quarter.

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] where t1 is the most serious scenario of voltage over-limit in the limit scenario; t2 is the most serious scenario of voltage under-limit in the limit scenario, and its model is the same as that of t1 scenario; i is the most serious node of voltage over-limit in c area; is the node i out-of-limit value at time t1; is the voltage margin of all distributed photovoltaic reactive idle capacity management nodes i in region c at time t1; is the number of distributed photovoltaics in region c; S ij is the voltage-reactive sensitivity of node i to node j; Q j is the maximum reactive idle capacity of node j; is the reactive margin index of region c at time t1; is the reactive margin index of the distribution network at time t1; N c is the number of partitions; p Q is the reactive margin index of the distribution network.

[0112]

[0113] α1+α2+α3+α4=1 (14)

[0114] wherein, δ ESS and δ SVG are the partition comprehensive indexes of energy storage and SVG respectively; α1, α2, α3, α4 are index coefficients determined by the analytic hierarchy process.

[0115] The annual distribution network partition model of seasonal partition in step S102 refers to that for the boundary node whose partition attribution changes with seasons, the probability of belonging to each partition is calculated according to the number of days belonging to different partitions, and finally the region with large attribution probability is attributed to complete the distribution network partition in units of years.

[0116] The fine selection configuration node model based on partition in step S102 includes: the specific installation node of SVG is determined by constructing three indexes of tightness and voltage deviation of the degree and voltage sensitivity in graph theory and using the TOPSIS method based on game theory weighting evaluation; the specific installation node of energy storage device is determined by constructing four indexes of comprehensive tightness of degree and comprehensive sensitivity, whether it is a photovoltaic node, voltage deviation and using the TOPSIS method based on game theory weighting evaluation.

[0117] Specifically, the above-mentioned SVG positioning and ESS positioning can be determined by formulas (15)-(18):

[0118]

[0119]

[0120] wherein, is the degree of the node, when the node i is connected with the node j, a ij =1, otherwise, a ij =0; N Bis the number of nodes in the region; is the node density; is the average value of the spatial electrical distance of node i and node j in a season; is the voltage deviation of the node; U i is the node voltage; U N is the rated voltage of the node; POS SVG is the basis for site selection of a specific node, and the larger the POS value of the node, the more suitable the point is for installing SVG; α1, α2, and α3 are weights obtained according to the analytic hierarchy process.

[0121]

[0122]

[0123] wherein, a ij = 1, otherwise, ij = 0; N B is the number of nodes in the region; is the node density; and is the average value of the active and reactive spatial electrical distance of node i and node j in a season; U i is the node voltage; U N is the rated voltage of the node; indicates the voltage deviation of the node; is the indication of whether photovoltaic installation is present; POS ESS is the basis for site selection of a specific node, and the larger the POS value of the node, the more suitable the point is for installing energy storage; α1, α2, α3, and α4 are weights obtained according to the analytic hierarchy process.

[0124] The specific content of the two-stage distribution network voltage regulation resource sizing method in step S103 is as follows: in the first stage, the transformer gear position is kept in the initial state, the voltage deviation is taken as the target to optimize the photovoltaic reactive power output of each subarea in each hour, the potential of the spare resource output of the photovoltaic is fully mobilized, the subsequent configuration investment is reduced, and the distribution network power flow is updated according to the photovoltaic reactive power output; in the second stage, the constructed multi-scenario double-layer capacity optimization configuration model is used to centrally optimize the OLTC, the adjustable capacitor CB, the SVG, and the active and reactive power output of the energy storage to solve the problems of network loss, voltage deviation and fluctuation, and voltage out-of-limit. Each subarea in the first stage refers to a subarea in each hour considering the structural index and the reactive functional index in each hour under each typical scenario. The expression of the photovoltaic reactive power output constraint is as follows:

[0125]

[0126] wherein, S inv,NP represents the rated capacity of the photovoltaic inverter; P PV Q represents the active power output of the photovoltaic; Q PV Q represents the reactive power output of the photovoltaic.

[0127] The multi-scenario double-layer capacity optimization configuration model in step S103 comprises an upper-layer target optimization model and a lower-layer target optimization model, and the detailed model is as shown in Figure 3 S401 and S402. The upper-layer model adopts a particle swarm algorithm, and the lower-layer model adopts a genetic algorithm for solving.

[0128] The target of the upper-layer optimization model comprises a device capacity / power demand model, a capacity / power cost model, and a voltage out-of-limit times and degree model. The smaller the device capacity / power demand, the smaller the capacity / power cost, and the smaller the voltage out-of-limit times and degree, the better the configuration effect and economy.

[0129] Specifically, the capacity / power demand model, the capacity / power cost model, and the voltage out-of-limit times and degree model can be determined by formulas (20)-(24):

[0130]

[0131] α1+α2+α3+α4=1 (21)

[0132] Wherein, N ESS is the number of energy storage installations; N CB is the number of adjustable capacitor CB sub-modules; N SVG is the number of SVG installations; M device is the capacity / power demand model; is the energy storage power installed at node i; is the energy storage installation capacity at node i; is the adjustable capacitor installation capacity at node i; is the SVG power installed at node i; and α1, α2, α3, and α4 are the energy storage installation power, the energy storage installation capacity, the adjustable capacitor installation capacity, and the SVG installation capacity coefficients, which are determined by the cost of each installation device, and the larger the unit capacity cost, the larger the corresponding coefficient.

[0133]

[0134] β1+β2=1 (23)

[0135] Wherein, M op is the distribution network capacity power cost model; p j is the proportion of each typical scene representative day in the total number of years; N b , and N l are the number of distribution network branches and the number of power purchase ports from the upper level, respectively; and Nsen1 and N sen2 are the number of typical scenarios and extreme scenarios, respectively; and are the power loss of branch λ at time t under scenario j and the input power of external upper grid node i at time t under scenario j, respectively; β1, β2 are the power loss and power purchase from the upper grid coefficient, which is determined by the size of the unit power cost, the larger the unit capacity cost, the larger the corresponding coefficient.

[0136]

[0137] wherein, N ove,all represents the total voltage out-of-limit times of n extreme scenarios; N ove,n represents the total voltage out-of-limit times of the nth extreme scenario; U ove,all represents the voltage out-of-limit degree of n extreme scenarios; U ove,n represents the voltage out-of-limit degree of the nth extreme scenario; N sen1 and N sen2 are the number of typical scenarios and extreme scenarios, respectively.

[0138] The objective of the lower optimization model includes the operation loss model, the operation voltage deviation model, the operation voltage fluctuation model, and the voltage out-of-limit times and degree model.

[0139] The lower optimization refers to parallel optimization of typical scenarios and extreme scenarios. The typical scenarios are optimized and calculated by taking the operation loss model, the operation voltage deviation model, and the operation voltage fluctuation model as the objective, and the extreme scenarios are optimized and calculated by taking the voltage out-of-limit times and degree model as the objective.

[0140] Specifically, the operation loss model, the operation voltage deviation model, the operation voltage fluctuation model, and the voltage out-of-limit times and degree model in the lower optimization model can be determined by (25)-(28):

[0141]

[0142] wherein, P loss is the daily loss model; p j is the proportion of each typical scenario representative day in the total number of years; is the power loss of branch i at time t under scenario j; N sen1 is the number of typical scenarios; Nb is the number of branches.

[0143]

[0144] wherein, U Δ represents the daily voltage deviation; N is the number of distribution network nodes; T is the number of hours of the scenario; U i,j,t is the voltage of node i at time t under scenario j.N is the rated voltage.

[0145]

[0146] wherein U FL denotes the daily voltage fluctuation; is the average voltage value of node i in T hours; U i,j,t is the voltage of node i at time t in scenario j; N is the number of nodes of the power distribution network; and T is the number of hours of the scenario.

[0147]

[0148] wherein a i,t denotes the state of voltage out-of-limit; N ove denotes the number of times of voltage out-of-limit of each node in a period; b i,t denotes the amplitude of voltage out-of-limit; U ove denotes the degree of voltage out-of-limit of each node in a period.

[0149] The constraint conditions of the upper model include: energy storage rated power constraint, energy storage rated capacity constraint, SVG rated capacity constraint, and adjustable capacitor rated capacity constraint. Specifically, the upper constraint model can be determined by (29):

[0150]

[0151] wherein P ESS,min , P ESS,max are the lower limit and upper limit of the rated power of energy storage installation; E ESS,min , E ESS,max are the lower limit and upper limit of the rated capacity of energy storage installation; Q SVG,min , Q SVG,max are the lower limit and upper limit of the installed capacity of SVG; Q CB,min , Q CB,max are the lower limit and upper limit of the installed capacity of CB; respectively denote the rated power of energy storage, the rated capacity of energy storage, the rated capacity of SVG, and the rated capacity of CB.

[0152] The constraint conditions of the lower model include: power distribution network operation equality and inequality constraints, OLTC operation constraints, energy storage operation constraints, SVG operation constraints, and adjustable capacitor CB operation constraints.

[0153] Specifically, the power distribution network operation equality and inequality constraints, the OLTC operation constraints, the energy storage operation constraints, the SVG operation constraints, and the adjustable capacitor CB operation constraints model can be determined by (30)-(34):

[0154]

[0155] where P i,t and Q i,t are the active and reactive power injected at node i at time t; G ij and B ij are the conductance and susceptance of line ij; U i,t is the voltage at node i at time t; U j,t is the voltage at node j at time t; θ ij is the phase angle difference between the voltage at node i and node j; U min and U max are the upper and lower limits of the node voltage; I ij,t is the current of branch ij at time t; I ij,min and I ij,max are the upper and lower limits of the current of branch ij.

[0156]

[0157] where r t0 is the transformer ratio at time t0; r t is the transformer ratio at time t; Δr j is the ratio difference between adjacent steps of the transformer; and are the up and down step operation of the transformer, when up, otherwise down, otherwise is the maximum number of operations allowed for the transformer in T time; T is the regulation time of the transformer.

[0158]

[0159] where, and represent the charge and discharge state of the energy storage at node i at time t; represents the discharging power of the energy storage at time t; represents the charging power of the energy storage at time t; and represent the upper and lower limit constraints of the discharging power of the energy storage; and represent the upper and lower limit constraints of the charging power of the energy storage; E i,t represents the capacity of the energy storage at node i at time t; E i,t+1 represents the capacity of the energy storage at node i at time t+1; and represent the charging and discharging efficiency of the energy storage; SOC i,min and SOC i,max represent the minimum and maximum state of charge of the energy storage; E i,Nrepresents the rated capacity of energy storage.

[0160]

[0161] wherein, represents the reactive power output of the SVG at node i at time t; represents the rated capacity of the SVG at node i.

[0162]

[0163] wherein, n i,t is the number of capacitors put into the sub-module by node i at time t; is the maximum number of sub-modules; N op,max is the maximum switching action number of the capacitor bank; represents the unit capacity of the adjustable capacitor; represents the CB access capacity at node i at time t.

[0164] The embodiment also includes a two-stage partition optimization configuration device for voltage regulation resources in a high-density photovoltaic power distribution network, comprising:

[0165] a scene generation module for generating high-precision typical photovoltaic scenes and limit photovoltaic scenes based on historical big data;

[0166] a two-step site selection module for two-step site selection of voltage regulation resources by using coarse selection of configuration regions and fine selection of nodes within the regions according to the typical photovoltaic scenes and the limit photovoltaic scenes;

[0167] a double-layer optimization module for constructing a two-stage power distribution network voltage regulation resource capacity model according to the selected address of the voltage regulation resource, the first stage being to optimize the photovoltaic reactive power output of each partition by time with voltage deviation as the target, and the second stage being to establish a multi-scenario double-layer capacity optimization configuration model considering capacity / power demand amount, capacity / power cost amount, voltage out-of-limit times and degree, and multi-objective of network loss, voltage deviation and fluctuation, wherein the upper model is a configuration model including load regulation transformer, energy storage, SVG and adjustable capacitor, the lower model is a centralized operation optimization model considering multiple scenarios, the upper model delivers voltage regulation resource capacity configuration information to the lower model, and the lower model obtains optimization results by optimization calculation according to the voltage regulation resource capacity configuration information and feeds back to the upper model.

[0168] This embodiment also includes an electronic device and a storage medium. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable by the processor. When the processor executes the computer program, it implements each step of the two-stage partitioning optimization method for configuring voltage regulation resources in a high-density photovoltaic distribution network. The storage medium stores the computer program. When the processor executes the computer program, it implements each step of the two-stage partitioning optimization method for configuring voltage regulation resources in a high-density photovoltaic distribution network.

[0169] In summary, the method for coordinated optimization configuration of voltage regulation resources in a high-density photovoltaic distribution network provided by the present invention can not only fully mobilize the reactive power output potential of photovoltaics, reduce the number of OLTC operations, and reduce the configuration capacity to improve economy, but also greatly reduce the computational dimension, improve convergence capability, and make the configuration more appropriate for the scenario.

[0170] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal translation scripting language JavaScript, etc.

[0171] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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 steps in the process. 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.

[0172] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work 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 1the function specified in the one or more blocks.

[0173] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide processes for implementing the flows Figure 1 the flow or flows and / or blocks Figure 1 the function specified in the one or more blocks.

[0174] Although preferred embodiments of the application have been described, those skilled in the art will be able to make additional modifications and variations to the described embodiments without departing from the spirit and scope of the application. Accordingly, the appended claims are intended to encompass all such modifications and variations as falling within the scope of the application.

[0175] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A two-stage partition optimization configuration method for voltage regulation resources in a high-density photovoltaic power distribution network, characterized in that, The method comprises the following steps: generate high-precision typical photovoltaic scenes and limit photovoltaic scenes based on historical big data; according to the typical photovoltaic scenes and the limit photovoltaic scenes, use the two-step site selection of the partition rough selection configuration region and the partition fine selection configuration node to select the address of the voltage regulation resource; according to the selected address of the voltage regulation resource, construct a two-stage distribution network voltage regulation resource capacity configuration model, the first stage is to optimize the photovoltaic reactive power output of each partition by time according to the voltage deviation, and update the distribution network power flow according to the photovoltaic reactive power output, the second stage is to establish a multi-scene double-layer capacity optimization configuration model considering the capacity / power demand, capacity / power cost, voltage out-of-limit times and degree, network loss, voltage deviation and fluctuation multi-objective, the multi-scene double-layer capacity optimization configuration model comprises an upper target optimization model and a lower target optimization model; the upper target optimization model comprises a device capacity / power demand model, a capacity / power cost model and a voltage out-of-limit times and degree model; the smaller the device capacity / power demand, the smaller the capacity / power cost and the smaller the voltage out-of-limit times and degree, the better the configuration effect and economy; the lower target optimization model comprises an operating network loss model, an operating voltage deviation model, an operating voltage fluctuation model and a voltage out-of-limit times and degree model; the lower target optimization refers to the parallel optimization of the typical scene and the limit scene, the typical scene is optimized and calculated according to the operating network loss model, the operating voltage deviation model and the operating voltage fluctuation model, and the limit scene is optimized and calculated according to the voltage out-of-limit times and degree model; the upper target optimization model feeds the voltage regulation resource capacity configuration information to the lower target optimization model, and the lower target optimization model performs optimization calculation according to the voltage regulation resource capacity configuration information to obtain an optimization result and feed back to the upper target optimization model.

2. The two-stage partition optimization configuration method of voltage regulation resources in a high-density photovoltaic power distribution network according to claim 1, characterized in that, The typical photovoltaic scene comprises photovoltaic large and small scenes in four seasons.

3. The two-stage partition optimization configuration method of voltage regulation resources in a high-density photovoltaic power distribution network according to claim 1 or 2, characterized in that, The constraint expression of the photovoltaic reactive power output is as follows: where S inv,N represents the rated capacity of the photovoltaic inverter; P PV represents the active power of the photovoltaic; Q PV represents the reactive power of the photovoltaic.

4. The two-stage partitioning optimization configuration method of voltage regulation resources for high-density photovoltaic power distribution networks according to claim 3, characterized in that, The calculation formula of the device capacity / power demand model is as follows: α1+α2+α3+α4＝1 where N ESS is the number of energy storage installations; N CB is the number of adjustable capacitor CB sub-modules; N SVG is the number of SVG installations; M device is the capacity / power demand model; is the energy storage power installed at node i; is the energy storage installed capacity at node i; is the adjustable capacitor installed capacity at node i; is the SVG power installed at node i; a1, a2, a3, a4 are the energy storage installed power, energy storage installed capacity, adjustable capacitor installed capacity and SVG installed capacity coefficients; The calculation formula of the capacity / power cost model is as follows: β1+β2＝1 where M op is the power consumption model of distribution network capacity; p j is the proportion of each typical scenario representing the total number of days; N b , N l are the number of branches and the number of power purchase ports from the upper level, respectively; N sen1 and N sen2 are the number of typical scenarios and limit scenarios, respectively; and are the network loss power of branch λ at time t in scenario j and the input electric power of external upper grid node i at time t in scenario j, respectively; β1, β2 are the network loss power and power purchase power coefficients from the upper level. The calculation formula of the voltage out-of-limit times and degree model is as follows: wherein N ove,all represents the total number of voltage out-of-limit times of n limit scenarios; N ove,n represents the total number of voltage out-of-limit times of the nth limit scenario; U ove,all represents the voltage out-of-limit degree of n limit scenarios; U ove,n represents the voltage out-of-limit degree of the nth limit scenario; N sen1 and N sen2 are the number of typical scenarios and limit scenarios, respectively.

5. The two-stage partitioning optimization configuration method of voltage regulation resources for high-density photovoltaic power distribution networks according to claim 4, characterized in that, The calculation formula of the operating network loss model is as follows: where P loss is the daily network loss model; p j is the proportion of the representative days of each typical scenario in the total annual days; is the network loss power of branch i at time t in scenario j; N sen1 is the number of typical scenarios; Nb is the number of branches; The calculation formula of the operating voltage deviation model is as follows: where U Δ represents the daily voltage deviation; N is the number of nodes of the power distribution network; T is the number of hours of the scenario; U i,j,t is the voltage of node i at time t in scenario j; U N is the rated voltage; The calculation formula of the operating voltage fluctuation model is as follows: where U FL indicates the daily voltage fluctuation; is the average voltage value of node i in T hours; U i,j,t is the voltage of node i at time t in scenario j; N is the number of nodes in the distribution network; and T is the number of hours of the scenario. The calculation formula of the voltage out-of-limit times and degree model is as follows: wherein, a i,t represents the state of voltage out-of-limit; N ove represents the number of times of voltage out-of-limit of each node in a period; b i,t represents the amplitude of voltage out-of-limit; U ove represents the degree of voltage out-of-limit of each node in a period; U i,t is the voltage of node i at time t; U max is the upper limit of voltage; U min is the lower limit of voltage.

6. The two-stage partitioning optimization configuration method of voltage regulating resources for high-density photovoltaic power distribution networks according to claim 4, characterized in that, The constraint condition of the upper target optimization model is as follows: wherein P ESS,min , P ESS,max are the lower and upper limits of the energy storage installation rated power; E ESS,min , E ESS,max are the lower and upper limits of the energy storage installation rated capacity; Q SVG,min , Q SVG,max are the lower and upper limits of the SVG installation capacity; Q CB,min , Q CB,max are the lower and upper limits of the CB installation capacity; respectively refer to the energy storage rated power, the energy storage rated capacity, the SVG rated capacity and the CB rated capacity.

7. The two-stage partitioning optimization configuration method of voltage regulating resources in a high-density photovoltaic power distribution network according to claim 5, characterized in that, The constraint condition of the lower target optimization model comprises: distribution network operation constraint, on-load voltage regulating transformer operation constraint, energy storage operation constraint, SVG operation constraint and adjustable capacitor operation constraint; The calculation formula of the distribution network operation constraint is as follows: where P i,t and Q i,t are the active and reactive power injected at node i at time t; G ij and B ij are the conductance and susceptance of line ij; U i,t is the voltage at node i at time t; U j,t is the voltage at node j at time t; θ ij is the phase angle difference between the voltages at node i and node j; U min and U max are the upper and lower limits of the node voltage; I ij,t is the current in branch ij at time t; I ij,min and I ij,max are the upper and lower limits of the current in branch ij. The calculation formula of the on-load voltage regulating transformer operation constraint is as follows: wherein r t0 is the transformer ratio at time t0; r t is the transformer ratio at time t; Δr j is the ratio difference between adjacent transformer steps; and are the up and down step operations of the transformer, respectively, when up-stepping, otherwise when down-stepping, otherwise is the maximum number of operations allowed by the transformer in time T; T is the regulation time of the transformer. The calculation formula of the energy storage operation constraint is as follows: SOC i,min E i,N ≤E i,t ≤SOC i,max E i,N wherein, and denotes the state of charge of the energy storage at time t for node i; denotes the discharging power of the energy storage at time t; denotes the charging power of the energy storage at time t; and denotes the upper and lower bound constraints for the discharging power of the energy storage; and denotes the upper and lower bound constraints for the charging power of the energy storage;E i,t denotes the energy storage capacity at time t for node i;E i,t+1 denotes the energy storage capacity at time t+1 for node i; and denotes the charging and discharging efficiency of the energy storage; SOC i,min and SOC i,max denotes the minimum and maximum state of charge of the energy storage;E i,N denotes the rated capacity of the energy storage; The calculation formula of the SVG operation constraint is as follows: wherein, QSVG,i(t) represents the reactive power output of the SVG at node i at time t; QSVG,i(t) represents the reactive power output of the SVG at node i at time t; The calculation formula of the adjustable capacitor operation constraint is as follows: wherein n i,t is the number of capacitors of the sub-module put into by node i at time t; is the maximum number of sub-modules; N op,max is the maximum switching action number of the capacitor bank; represents the unit capacity of the adjustable capacitor; represents the adjustable capacitor access capacity at time t at node i.

8. A two-stage partition optimization configuration device for voltage regulation resources in a high-density photovoltaic power distribution network, characterized in that, The method comprises the following steps: a scene generation module is arranged to generate high-precision typical photovoltaic scenes and limit photovoltaic scenes based on historical big data; a two-step site selection module is arranged to use the two-step site selection of the partition rough selection configuration region and the partition fine selection configuration node to select the address of the voltage regulation resource according to the typical photovoltaic scenes and the limit photovoltaic scenes; The double-layer optimization module is configured to construct a two-stage distribution network voltage regulation resource sizing model according to the selected address of the voltage regulation resource, the first stage is to optimize the photovoltaic reactive power output of each subarea in time with voltage deviation as the target, and the distribution network power flow is updated according to the photovoltaic reactive power output, and the second stage is to establish a multi-scenario double-layer capacity optimization configuration model considering capacity / power demand, capacity / power cost, voltage out-of-limit times and degree, and multi-objective of network loss, voltage deviation and fluctuation, the multi-scenario double-layer capacity optimization configuration model includes an upper-layer target optimization model and a lower-layer target optimization model; the upper-layer target optimization model includes a device capacity / power demand model, a capacity / power cost model and a voltage out-of-limit times and degree model; the smaller the device capacity / power demand, the smaller the capacity / power cost and the smaller the voltage out-of-limit times and degree represent the better configuration effect and economy; The lower-layer target optimization model includes an operating network loss model, an operating voltage deviation model, an operating voltage fluctuation model and a voltage out-of-limit times and degree model; The lower-layer target optimization refers to parallel optimization of a typical scenario and a limit scenario, the typical scenario is optimized and calculated with the operating network loss model, the operating voltage deviation model and the operating voltage fluctuation model as the targets, and the limit scenario is optimized and calculated with the voltage out-of-limit times and degree model as the target; The upper-layer target optimization model delivers voltage regulation resource capacity configuration information to the lower-layer target optimization model, the lower-layer target optimization model performs optimization calculation according to the voltage regulation resource capacity configuration information to obtain an optimization result and feeds back to the upper-layer target optimization model.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize each step in the two-stage subarea optimization configuration method of the voltage regulation resource of the high-density photovoltaic distribution network as claimed in any one of claims 1-7.

10. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to realize each step in the two-stage subarea optimization configuration method of the voltage regulation resource of the high-density photovoltaic distribution network as claimed in any one of claims 1-7.

Images