Distribution network cluster control method, equipment and medium considering photovoltaic heterogeneous characteristics

Abstract

The present invention belongs to the field of distribution network control technology, and specifically relates to a distribution network cluster control method, device and medium that considers the heterogeneous characteristics of photovoltaics. In view of the fact that the cluster division method of the existing distribution network cluster scheduling method is not scientific enough and fails to give full play to the regulation ability of some HDPVs, the present invention adopts the following technical scheme: a distribution network cluster optimization control method that considers the heterogeneous characteristics of distributed photovoltaics, including: constructing and solving a distribution network cluster division model that considers the comprehensive performance indicators of the cluster to obtain the optimal division result of the distribution network cluster; establishing a heterogeneous distributed photovoltaic model under different operating modes; taking the operation cost and voltage deviation as the minimum objective function, establishing a multi-objective optimization model, and setting constraints based on the heterogeneous characteristics of photovoltaics; solving the objective function based on the optimal division result of the distribution network cluster and the constraints. The beneficial effects of the present invention are: dividing the cluster more scientifically and giving full play to the regulation ability of HDPV.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network control, and specifically relates to a distribution network cluster control method, equipment and medium taking into account photovoltaic heterogeneous characteristics. Background Art

[0002] With the increasing penetration of distributed photovoltaic power generation in the distribution network, the operation of the distribution network faces many challenges, among which the problems of reverse power flow and voltage over-limit are particularly significant, which seriously threaten the stability and economy of the distribution network operation. In response to the above problems, if the traditional centralized optimization and control method is adopted, on the one hand, it faces the dual dilemma of difficult information collection and heavy communication burden. On the other hand, due to the differences in capacity, location, and control methods of distributed photovoltaics connected to the distribution network (that is, heterogeneous characteristics), it is difficult to achieve "observable and controllable" operation status of large-scale heterogeneous distributed photovoltaics (HDPV).

[0003] In recent years, the distribution network optimization and control technology based on "intra-group autonomy and inter-group coordination" has become a research hotspot. Distributed energy and loads that are close to each other in geographical location or electrical form or form a complementary relationship are aggregated to form a distributed generation cluster, that is, the distribution network cluster division. Based on the optimal cluster division results, the distribution network cluster is optimized and dispatched, thereby simplifying the complexity of distribution network control.

[0004] However, the existing distribution network cluster control methods still have the following shortcomings: 1. In terms of distribution network cluster division, the existing methods do not take the voltage deviation sensitivity within the cluster into consideration as a comprehensive energy efficiency indicator for distribution network cluster division. After cluster division, the node voltage deviates greatly from the system rated voltage, and the voltage over-limit problem still exists; 2. In terms of distribution network cluster optimization and control, most current studies do not consider the heterogeneous characteristics of distributed photovoltaics. If the traditional, single photovoltaic model is still used to optimize the control of distribution network clusters with high penetration HDPV access, the active and reactive regulation potential of HDPV cannot be fully utilized, and a large amount of energy that cannot be absorbed will be sent back to the large power grid. Network losses will also occur during the return transmission, causing the distribution network voltage to exceed the limit, seriously affecting the economy and stability of the distribution network operation.

[0005] Therefore, improving the distribution network cluster division method and fully considering the heterogeneous characteristics of photovoltaics and the active and reactive coordination capabilities are of great significance to improving the distribution network cluster regulation level. Summary of the invention

[0006] The present invention aims at solving the problem that the cluster division method of the existing distribution network cluster scheduling method is not scientific enough, and does not take into account the heterogeneous characteristics of distributed photovoltaics, and fails to give full play to the shortcomings of the active and reactive regulation capabilities of some HDPVs. The present invention provides a distribution network cluster optimization and control method considering the heterogeneous characteristics of distributed photovoltaics, introduces cluster voltage deviation sensitivity, and improves the distribution network cluster division index modeling method; according to the active and reactive characteristics of HDPV, its operation models under different modes are established; with network loss cost, photovoltaic active reduction cost, reverse transmission penalty cost and voltage deviation as targets, a distribution network cluster optimization and control model considering the heterogeneous characteristics of distributed photovoltaics is established to achieve distributed coordinated optimization among clusters.

[0007] To achieve the above object, the present invention adopts the following technical solution: a distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics, the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics includes:

[0008] Step S1, constructing and solving a distribution network cluster partitioning model considering the comprehensive performance index of the cluster, and obtaining an optimal partitioning result of the distribution network cluster, wherein the comprehensive performance index of the cluster includes the cluster voltage deviation sensitivity;

[0009] Step S2: establishing heterogeneous distributed photovoltaic models under different operation modes according to the communication status of the heterogeneous distributed photovoltaics;

[0010] Step S3, taking network loss costs, photovoltaic active power reduction costs, power reverse transmission penalty costs and voltage deviation as decision variables, and taking minimizing operating costs and voltage deviation as objective functions, a multi-objective optimization model is established;

[0011] Step S4: Based on the optimal division result of the distribution network cluster obtained in step S1 and the heterogeneous distributed photovoltaic model obtained in step S2, determine the constraints of the objective function in the multi-objective optimization model, and solve the objective function to obtain the active power and reactive power output by each heterogeneous distributed photovoltaic system, thereby achieving optimal control of the distribution network cluster.

[0012] As an improvement, in step S1, the cluster voltage deviation sensitivity reflects the sensitivity of the node voltage change within the cluster to the reactive power change. The calculation formula of the cluster voltage deviation sensitivity index is:

[0013] (1)

[0014] (2)

[0015] (3)

[0016] Where: f vis the cluster voltage deviation sensitivity, N is the number of nodes in the cluster, N c is the number of clusters, S si For Node i For the node with the highest reactive power sensitivity in the cluster s The voltage sensitivity coefficient, U i For Node i The actual voltage, U N is the improved node rated voltage, S ss The node with the highest reactive power sensitivity in the cluster s For its own voltage sensitivity coefficient, Δ V s With Δ Q s Node s The voltage change and reactive power change of ΔQ i For Node i The reactive power change;

[0017] Improved node voltage rating U N The calculation formula is:

[0018] (4)

[0019] (5)

[0020] (6)

[0021] in, β min and β max are the minimum and maximum load rates of HDPV, or min and or max are the minimum and maximum load rates, a and b They are HDPV load factor β and load factor or The coefficient of P PV,k,t For the k The real-time active power output of heterogeneous distributed photovoltaics. P PV,k,t,max For the k The maximum output of heterogeneous distributed photovoltaics, P L,i,tFor Node i The real-time active load, P L,i,t,max For Node i The maximum active load.

[0022] As an improvement, in step S1, the cluster comprehensive performance index also includes a modularity index based on electrical distance. The larger the modularity index, the higher the degree of internal coupling of the cluster.

[0023] Electrical distance L ij The expression is:

[0024] (7)

[0025] Where: L ij For Node i With Node j The electrical distance between nodes is larger, and the larger the value is, the farther the electrical distance between nodes is. e ij The smaller the size, the smaller the size, and vice versa; d ij For Node j When the active and reactive power changes, the node j Voltage Change and Node i The ratio of voltage change;

[0026] Define node edge weights e ij Distance from electrical L ij The expression of the relationship is:

[0027] (8)

[0028] Modularity function r The expression is:

[0029] (9)

[0030] Where: e ij For Node i and nodes j The weight of the connected edges; nodes i and j When connected e ij= 1. On the contrary e ij =0; m is the sum of the weights of all edges in the entire community network; k i , kj Node i , j The sum of the weights of all connected edges; node i and j In the same cluster d ( i,j )=1, otherwise d ( i,j )=0;

[0031] In step S1, the comprehensive performance index of the cluster also includes the balance degree of active and reactive power within the cluster. The higher the balance degree, the smaller the scale of power transmission between clusters.

[0032] The larger the active and reactive power balance index is, the higher the internal power balance of the cluster is. The formula is:

[0033] (10)

[0034] Where: f s It is the power balance index; f p It is the total active power balance index of the system; f Q It is the total reactive power balance index of the system;

[0035] The expression of the total active power balance index of the system is:

[0036] (11)

[0037] Where: P clu(t),i For cluster k The net power of the scene, T is the duration of the typical time-varying scene, N c is the number of clusters;

[0038] The system reactive power balance index is used to describe the balance relationship between reactive power demand and supply within the cluster, which is expressed as:

[0039] (12)

[0040] (13)

[0041] Where: Q sup,i , Q need,i Respectively i The maximum reactive power output and reactive power demand of each cluster; Q i For cluster iReactive power balance.

[0042] As an improvement, in step S1, the cluster comprehensive performance index f The expression is:

[0043] (14)

[0044] Where: l 1. l 2. l 3 are the weight coefficients of each indicator;

[0045] Different weight coefficients are selected according to different cluster division objectives to obtain the corresponding optimal cluster division results.

[0046] As an improvement, in step S2, when the heterogeneous distributed photovoltaic is in offline mode, it does not participate in grid interaction and cannot regulate the distribution network. The heterogeneous distributed photovoltaic offline model is expressed as:

[0047] (15)

[0048] Where: M1 is offline mode, P HDPV and Q HDPV are the real-time output active power and reactive power of heterogeneous distributed photovoltaics, P ( t )and Q ( t ) are the real-time predicted power of HDPV, T For any time, α is the minimum power factor angle of the photovoltaic inverter;

[0049] When the heterogeneous distributed photovoltaic system is in semi-offline mode, only the reactive power regulation function participates in the grid interaction, which can effectively adjust the voltage level of the distribution network, but cannot solve the problem of power reverse transmission. The semi-offline model of heterogeneous distributed photovoltaic system is expressed as:

[0050] (16)

[0051] Where: M2 is the semi-offline mode, Δ Q ( t ) is the real-time regulation amount of heterogeneous distributed photovoltaic reactive power;

[0052] When the heterogeneous distributed photovoltaic system is in online mode, it has active and reactive power regulation functions. Through the active power reduction of HDPV itself and reactive power compensation for the distribution network, the power reverse transmission and voltage over-limit problems of the distribution network can be solved. The online model of heterogeneous distributed photovoltaic system is expressed as:

[0053] (17)

[0054] Where: M3 is online mode, P dec ( t ) is the real-time reduction of active power of heterogeneous distributed photovoltaics.

[0055] As an improvement, in step S3, the objective function of the multi-objective optimization model is expressed as:

[0056] (18)

[0057] (19)

[0058] In the formula, oh 1. oh 2 are the weight coefficients of each indicator; C loss Network loss costs; C pvdec Reduce the cost of active power for heterogeneous distributed photovoltaics; C re Penalty fees for power backhaul; c p is the network loss cost coefficient; I ij,t is the real-time current of the line; r ij is the line impedance; c pv Reduce the cost factor for photovoltaic active power; P PV,i,t,max The maximum predicted photovoltaic output power at each moment; P PV,i,t The real-time photovoltaic output power at each moment; c re is the reverse power penalty coefficient; P re,t is the reverse power; D V is the overall system voltage deviation.

[0059] As an improvement, in step S3, the constraints include heterogeneous distributed photovoltaic operation constraints, voltage constraints, and power flow balance constraints.

[0060] The operation constraints of heterogeneous distributed photovoltaics are expressed as:

[0061] (20)

[0062] (twenty one)

[0063] (twenty two)

[0064] In the formula, M1, M2, and M3 are the three operating modes of heterogeneous distributed photovoltaics. P PV,i,t It is the real-time active power of heterogeneous distributed photovoltaics. P PV,i,t,max is the upper limit of active power output of heterogeneous distributed photovoltaics, Q PV,i,t It is the real-time reactive power of heterogeneous distributed photovoltaics. Q PV,i,t,max The upper limit of reactive power output of heterogeneous distributed photovoltaics;

[0065] The voltage constraint is expressed as:

[0066] (twenty three)

[0067] Where: V min and V max is the lower and upper limits of voltage;

[0068] The power flow balance constraint is expressed as:

[0069] (twenty four)

[0070] Where: P j,PV,t and Q j,PV,t Node j Active and reactive power injected by the connected heterogeneous distributed photovoltaics; P j,L and Q j,L For Node j The connected load, P jk,t and Q jk,t Injection nodes j Active power and reactive power; P ij,t and Q ij,t are the active power and reactive power transmitted on the line respectively; x ij is the line reactance.

[0071] As an improvement, in step S1, a genetic algorithm is used to solve equations (1) to (14) to obtain the optimal division result of the distribution network cluster; in step S4, a synchronous alternating direction multiplier method and a commercial solver are used to solve the objective function.

[0072] A computer device includes a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, the aforementioned distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics is implemented.

[0073] A computer-readable storage medium stores a computer program, which, when executed, implements the aforementioned distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics.

[0074] The beneficial effects of the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics of the present invention are as follows: in terms of distribution network cluster division, the voltage deviation sensitivity is taken into account to improve the distribution network cluster division method in view of the problem that high penetration HDPV access easily causes voltage over-limit in the cluster, and the voltage regulation of each cluster is flexible, the net load distribution is balanced, and the nodes in the cluster are highly aggregated, so as to better adapt to the economic operation and voltage regulation of the distribution network; in terms of distribution network cluster optimization control, the heterogeneous characteristics of distributed photovoltaics are taken into account, and the active and reactive regulation capabilities of HDPV are fully utilized, which alleviates the problems of power flow return and voltage over-limit in the distribution network to the greatest extent, and also improves the reliability and economy of the distribution network cluster operation. Furthermore, when dividing the distribution network cluster, the voltage deviation sensitivity, modularity and active and reactive balance are taken into account. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a distribution network cluster optimization method considering the heterogeneous characteristics of distributed photovoltaics according to an embodiment of the present invention.

[0076] Figure 2 It is an improved IEEE33 node graph.

[0077] Figure 3 It is the distribution network load and HDPV output forecast data.

[0078] Figure 4 This is a diagram of the optimal cluster division result obtained by using the method of the embodiment of the present invention.

[0079] Figure 5 This is the power return situation under different strategies.

[0080] Figure 6 This is the voltage distribution diagram of each node of the system when strategy 1 is adopted.

[0081] Figure 7 This is the voltage distribution diagram of each node of the system when strategy 2 is adopted.

[0082] Figure 8 This is the voltage distribution diagram of each node of the system when strategy 3 is adopted. DETAILED DESCRIPTION

[0083] The technical solutions of the embodiments of the present invention are explained and described below, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the protection scope of the present invention.

[0084] See also Figure 1 , the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics in an embodiment of the present invention, the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics includes:

[0085] Step S1, constructing and solving a distribution network cluster partitioning model considering the comprehensive performance index of the cluster, and obtaining an optimal partitioning result of the distribution network cluster, wherein the comprehensive performance index of the cluster includes the cluster voltage deviation sensitivity;

[0086] Step S2: establishing heterogeneous distributed photovoltaic models under different operation modes according to the communication status of the heterogeneous distributed photovoltaics;

[0087] Step S3, taking network loss costs, photovoltaic active power reduction costs, power reverse transmission penalty costs and voltage deviation as decision variables, and taking minimizing operating costs and voltage deviation as objective functions, a multi-objective optimization model is established;

[0088] Step S4: Based on the optimal division result of the distribution network cluster obtained in step S1 and the heterogeneous distributed photovoltaic model obtained in step S2, determine the constraints of the objective function in the multi-objective optimization model, and solve the objective function to obtain the active power and reactive power output by each heterogeneous distributed photovoltaic system, thereby achieving optimal control of the distribution network cluster.

[0089] In this embodiment, in step S1, the cluster voltage deviation sensitivity reflects the sensitivity of the node voltage change within the cluster to the reactive power change. The smaller the cluster voltage deviation sensitivity (when the reactive power change of the node is large and the voltage change of the node is small, the cluster voltage deviation sensitivity is small), the better the voltage quality. The calculation formula of the cluster voltage deviation sensitivity index is:

[0090] (1)

[0091] (2)

[0092] (3)

[0093] Where: f v is the cluster voltage deviation sensitivity, N is the number of nodes in the cluster, N c is the number of clusters,S si For Node i For the node with the highest reactive power sensitivity in the cluster s The voltage sensitivity coefficient, U i For Node i The actual voltage, U N is the improved node rated voltage, S ss The node with the highest reactive power sensitivity in the cluster s For its own voltage sensitivity coefficient, Δ V s With Δ Q s Node s The voltage change and reactive power change of ΔQ i For Node i The reactive power change;

[0094] Among them, the improved node rated voltage U N The calculation formula is:

[0095] (4)

[0096] (5)

[0097] (6)

[0098] in, β min and β max are the minimum and maximum load rates of HDPV, or min and or max are the minimum and maximum load rates, a and b They are HDPV load factor β and load factor or The coefficient of P PV,k,t For the k The real-time active power output of heterogeneous distributed photovoltaics. P PV,k,t,max For the k The maximum output of heterogeneous distributed photovoltaics, P L,i,t For Node i The real-time active load, P L,i,t,max For Node iThe maximum active load.

[0099] In this embodiment, in step S1, the cluster comprehensive performance index also includes a modularity index based on electrical distance. The larger the modularity index is, the higher the degree of internal coupling of the cluster is.

[0100] When calculating the modularity index, the electrical distance is calculated first. L ij The expression is:

[0101] (7)

[0102] Where: L ij For Node i With Node j The electrical distance between nodes is larger, and the larger the value is, the farther the electrical distance between nodes is. The corresponding edge weight in the modularity function is e ij The smaller the better, and vice versa; d ij For Node j When the active and reactive power changes, the node j Voltage Change and Node i The ratio of voltage change;

[0103] Define node edge weights e ij Distance from electrical L ij The expression of the relationship is:

[0104] (8)

[0105] Modularity function r The expression is:

[0106] (9)

[0107] Where: e ij For Node i and nodes j The weight of the connected edges; nodes i and j When connected e ij= 1. On the contrary e ij =0; m is the sum of the weights of all edges in the entire community network; k i , k j Node i ,j The sum of the weights of all connected edges; node i and j In the same cluster d ( i,j )=1, otherwise d ( i,j )=0;

[0108] In step S1, the comprehensive performance index of the cluster also includes the balance degree of active and reactive power within the cluster. The higher the balance degree, the smaller the scale of power transmission between clusters.

[0109] The larger the active and reactive power balance index is, the higher the internal power balance of the cluster is. The formula is:

[0110] (10)

[0111] Where: f s It is the power balance index; f p It is the total active power balance index of the system; f Q It is the total reactive power balance index of the system;

[0112] The expression of the total active power balance index of the system is:

[0113] (11)

[0114] Where: P clu(t),i For cluster k The net power is the active power output of photovoltaic power in the cluster minus the active load demand. T is the duration of the typical time-varying scenario. N c is the number of clusters;

[0115] The system reactive power balance index is used to describe the balance relationship between reactive power demand and supply within the cluster, which is expressed as:

[0116] (12)

[0117] (13)

[0118] Where: Q sup,i , Q need,i Respectively i The maximum reactive power output and reactive power demand of each cluster; Q i For cluster i Reactive power balance.

[0119] In this embodiment, in step S1, the cluster comprehensive performance index f The expression is:

[0120] (14)

[0121] Where: l 1. l 2. l 3 are the weight coefficients of each indicator;

[0122] Different weight coefficients are selected according to different cluster division objectives to obtain the corresponding optimal cluster division results.

[0123] In this embodiment, in step S1, a genetic algorithm is used to solve equations (1) to (14) to obtain the optimal division result of the distribution network cluster.

[0124] In this embodiment, in step S2, when the heterogeneous distributed photovoltaic is in offline mode, it does not participate in grid interaction and cannot regulate the distribution network. The heterogeneous distributed photovoltaic offline model is expressed as:

[0125] (15)

[0126] Where: M1 is offline mode, P HDPV and Q HDPV are the real-time output active power and reactive power of heterogeneous distributed photovoltaics, P ( t )and Q ( t ) are the real-time predicted power of HDPV, T For any time, α is the minimum power factor angle of the photovoltaic inverter;

[0127] When the heterogeneous distributed photovoltaic system is in semi-offline mode, only the reactive power regulation function participates in the grid interaction, which can effectively adjust the voltage level of the distribution network, but cannot solve the problem of power reverse transmission. The semi-offline model of heterogeneous distributed photovoltaic system is expressed as:

[0128] (16)

[0129] Where: M2 is the semi-offline mode, Δ Q ( t ) is the real-time regulation amount of heterogeneous distributed photovoltaic reactive power;

[0130] When the heterogeneous distributed photovoltaic system is in online mode, it has active and reactive power regulation functions. Through the active power reduction of HDPV itself and reactive power compensation for the distribution network, the power reverse transmission and voltage over-limit problems of the distribution network can be solved. The online model of heterogeneous distributed photovoltaic system is expressed as:

[0131] (17)

[0132] Where: M3 is online mode, P dec ( t ) is the real-time reduction of active power of heterogeneous distributed photovoltaics.

[0133] In this embodiment, in step S3, the objective function of the multi-objective optimization model is expressed as:

[0134] (18)

[0135] (19)

[0136] In the formula, oh 1. oh 2 are the weight coefficients of each indicator; C loss Network loss costs; C pvdec Reduce the cost of active power for heterogeneous distributed photovoltaics; C re Penalty fees for power backhaul; c p is the network loss cost coefficient; I ij,t is the real-time current of the line; r ij is the line impedance; c pv Reduce the cost factor for photovoltaic active power; P PV,i,t,max The maximum predicted photovoltaic output power at each moment; P PV,i,t The real-time photovoltaic output power at each moment; c re is the reverse power penalty coefficient; P re,t is the reverse power; D V is the overall system voltage deviation.

[0137] In this embodiment, in step S3, the constraints include heterogeneous distributed photovoltaic operation constraints, voltage constraints, and power flow balance constraints.

[0138] The operation constraints of heterogeneous distributed photovoltaics are expressed as:

[0139] (20)

[0140] (twenty one)

[0141] (twenty two)

[0142] In the formula, M1, M2, and M3 are the three operating modes of heterogeneous distributed photovoltaics. P PV,i,t It is the real-time active power of heterogeneous distributed photovoltaics. P PV,i,t,max is the upper limit of active power output of heterogeneous distributed photovoltaics, Q PV,i,t It is the real-time reactive power of heterogeneous distributed photovoltaics. Q PV,i,t,max The upper limit of reactive power output of heterogeneous distributed photovoltaics;

[0143] The voltage constraint is expressed as:

[0144] (twenty three)

[0145] Where: V min and V max is the lower and upper limits of voltage;

[0146] The power flow balance constraint is expressed as:

[0147] (twenty four)

[0148] Where: P j,PV,t and Q j,PV,t Node j Active and reactive power injected by the connected heterogeneous distributed photovoltaics; P j,L and Q j,L For Node j The connected load, P jk,t and Q jk,t Injection nodes j Active power and reactive power; P ij,t and Q ij,t are the active power and reactive power transmitted on the line respectively; x ij is the line reactance.

[0149] In this embodiment, in step S4, the objective function is solved using a synchronous alternating direction multiplier method and a commercial solver.

[0150] The distribution network cluster optimization control method according to an embodiment of the present invention is described in more detail below.

[0151] The improved IEEE 33 node distribution network is used as an example. The specific node diagram of the improved IEEE 33 node distribution network can be found in Figure 2 The distribution network related parameters are shown in Table 1. The distribution network load and the output forecast data of each PV (photovoltaic) of HDPV can be found in Figure 3 .

[0152] Table 1 Related parameters

[0153]

[0154] 1. Distribution network cluster division results

[0155] In order to verify the effectiveness of the proposed clustering method, three different weights are set. Different weight combinations will have different clustering results. The indicator weights and clustering results of each scheme are shown in Table 2.

[0156] Table 2 The impact of different weights on cluster division

[0157]

[0158] From Table 2, we can see that: ① In terms of modularity, category 3 is 6.1% lower than category 1 and 1.8% higher than category 2; when all indicators are considered comprehensively, category 3 is 36% higher than category 4 and 52.66% higher than category 5; ② In terms of active and reactive balance, the calculation results of the designed number of groups are not much different; ③ In terms of cluster voltage deviation, category 3 is 15.22% higher than category 4 and 23.33% lower than category 5; ④ In terms of comprehensive cluster division indicators, the comprehensive cluster division index values ​​of category 1 and category 2 are lower than those of the other four groups. At the same time, although the comprehensive cluster division index values ​​of categories 3 to 5 are not much different, isolated nodes appear in categories 4 and 5.

[0159] In the process of weight adjustment, the strengthening of functional indicators (power balance) makes the cluster division more detailed. Although it significantly improves the functional performance, it sacrifices the cluster structure performance. In extreme cases, isolated node clusters may be generated. When the importance of structural indicators (modularity) is increased, the cluster division pays more attention to the degree of internal coupling, which effectively improves the structural performance. At the same time, the functional indicators and stability indicators (cluster voltage deviation sensitivity) remain relatively stable and are only slightly disturbed. In addition, in real scenarios, if a control terminal is installed separately at an isolated node, the construction cost will increase significantly, which does not meet the economic requirements of the distribution network transformation. Therefore, category 3 is selected, and the weight is l 1=0.4, l 2=0.3, l The cluster division result of 3=0.3 is taken as the optimal cluster. The optimal cluster division result is as follows Figure 4 shown.

[0160] 2. Analysis of distribution network optimization and control results

[0161] In order to compare the impact of cluster regulation and its heterogeneous distributed photovoltaic operation characteristics on the operation of the distribution network, the following three strategies are adopted for simulation.

[0162] Strategy 1: No distribution network cluster division is performed, and the active and reactive regulation characteristics of heterogeneous distributed photovoltaics are not considered, serving as a blank control group.

[0163] Strategy 2: Based on the traditional cluster division results (i.e., category 2 in Table 2), the active and reactive regulation characteristics of heterogeneous distributed photovoltaics are considered as the traditional cluster control strategy. The existing method does not consider the active and reactive regulation characteristics of heterogeneous distributed photovoltaics. Strategy 2 is not an existing method and is only used to compare and illustrate the effect of the method of this embodiment.

[0164] Strategy 3: Based on the optimal cluster division result, the active and reactive regulation characteristics of heterogeneous distributed photovoltaics are considered, that is, the regulation strategy of the present invention.

[0165] 1) Network loss and economic analysis

[0166] The costs under the three strategies are shown in Table 3.

[0167] Table 3 Cost calculation results under three strategies

[0168]

[0169] From the data in Table 3, we can see that: ① In terms of network loss optimization, the network loss value under the strategy of the present invention is the smallest, which is reduced by 73.17% and 14.10% compared with strategy 1 and strategy 2 respectively; ② In terms of active power reduction and power reverse transmission, since the strategy of the present invention takes into account the operating characteristics of heterogeneous distributed photovoltaics, in order to ensure the safety of the power grid, the active power output of heterogeneous distributed photovoltaics is reduced, so that the power grid return situation is improved; ③ In terms of comprehensive cost, such as Figure 5 As shown in the figure, the strategy of the present invention reduces the active power of heterogeneous distributed photovoltaics, which increases the active power reduction cost compared with strategy 1 and strategy 2, but the reverse transmission penalty cost is relatively reduced; ④ In terms of comprehensive cost, the strategy of the present invention reduces 12.39% and 4.90% respectively compared with strategy 1 and strategy 2. This is because the strategy of the present invention takes into account functional indicators of different degrees on the basis of structural indicators, which makes the comprehensive cost of the system show a downward trend.

[0170] Therefore, compared with the existing strategies, the strategy proposed in the present invention has obvious loss reduction and power reverse optimization effects, can achieve the effectiveness of regulation, and ensure the safe and economical operation of the distribution network.

[0171] 2) Voltage distribution and reactive power optimization analysis

[0172] Figure 6 to Figure 8 The voltage distribution under the three strategies is shown in Table 4. The vertical coordinate V (pu) is the voltage, and pu represents the per-unit value. Table 4 shows the highest voltage, lowest voltage and overall voltage deviation rate under the three strategies. Figure 6 to Figure 8 It can be seen that in Strategy 1 and Strategy 2, the maximum and minimum values ​​of the distribution network voltage are 1.097 and 0.901 respectively, and the voltage deviation rate exceeds the ±7% range allowed by the national standard. In Strategy 2 based on the results of traditional cluster division, although the voltage deviation rate is reduced by 12.97% compared with Strategy 1 by adjusting the reactive power of heterogeneous distributed photovoltaics, there is still a voltage over-limit situation. In the strategy of the present invention, after the active and reactive power of heterogeneous distributed photovoltaics are adjusted, the maximum value of the distribution network voltage is reduced to 1.07, and the minimum value is increased to 1, completely eliminating the distribution network voltage over-limit situation. At the same time, in the strategy of the present invention, the voltage deviation rate of the distribution network is 2.88%, which is 74.58% and 70.79% lower than that of Strategy 1 and Strategy 2, respectively, and meets the standard range.

[0173] An embodiment of the present invention also provides a computer device, including a processor and a storage medium, wherein a computer program is stored in the storage medium, and when the computer program is executed by the processor, the aforementioned distribution network cluster optimization and control method is implemented.

[0174] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed, the aforementioned distribution network cluster optimization and control method is implemented.

[0175] Table 4 Voltage distribution under three strategies

[0176]

[0177] The above is only a specific implementation of the invention, but the protection scope of the invention is not limited thereto. Those skilled in the art should understand that the invention includes but is not limited to the contents described in the above specific implementation. Any modification that does not deviate from the functional and structural principles of the invention will be included in the scope of the claims.

Claims

1. A distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics, characterized by: The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics includes: Step S1, constructing and solving a distribution network cluster partitioning model considering the comprehensive performance index of the cluster, and obtaining an optimal partitioning result of the distribution network cluster, wherein the comprehensive performance index of the cluster includes the cluster voltage deviation sensitivity; Step S2: establishing heterogeneous distributed photovoltaic models under different operation modes according to the communication status of the heterogeneous distributed photovoltaics; Step S3, taking network loss costs, photovoltaic active power reduction costs, power reverse transmission penalty costs and voltage deviation as decision variables, and taking minimizing operating costs and voltage deviation as objective functions, a multi-objective optimization model is established; Step S4: Based on the optimal division result of the distribution network cluster obtained in step S1 and the heterogeneous distributed photovoltaic model obtained in step S2, determine the constraint conditions of the objective function in the multi-objective optimization model, and solve the objective function to obtain the active power and reactive power output by each heterogeneous distributed photovoltaic, thereby realizing the optimization and regulation of the distribution network cluster; In step S1, the cluster voltage deviation sensitivity reflects the sensitivity of the node voltage change within the cluster to the reactive power change. The calculation formula of the cluster voltage deviation sensitivity index is: (1) (2) (3) Where: φ v is the cluster voltage deviation sensitivity, N is the number of nodes in the cluster, N c is the number of clusters, S si For Node i For the node with the highest reactive power sensitivity in the cluster s The voltage sensitivity coefficient, U i For Node i The actual voltage, U N is the improved node rated voltage, S ss The node with the highest reactive power sensitivity in the cluster s For its own voltage sensitivity coefficient, Δ V s With Δ Q s Node s The voltage change and reactive power change, Δ Q i For Node i The reactive power change; Improved node voltage rating U N The calculation formula is: (4) (5) (6) in, β min and β max are the minimum and maximum load rates of HDPV, η min and η max are the minimum and maximum load rates, a and b They are HDPV load factor β and load factor η The coefficient of P PV,k,t For the k The real-time active power output of heterogeneous distributed photovoltaics. P PV,k,t,max For the k The maximum output of heterogeneous distributed photovoltaics, P L,i,t For Node i The real-time active load, P L,i,t,max For Node i The maximum active load.

2. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 1 is characterized by: In step S1, the cluster comprehensive performance index also includes a modularity index based on electrical distance. The larger the modularity index, the higher the degree of internal coupling of the cluster. Electrical distance L ij The expression is: (7) Where: L ij For Node i With Node j The electrical distance between nodes is larger, and the larger the value is, the farther the electrical distance between nodes is. e ij The smaller the size, the smaller the size, and vice versa; d ij For Node j When the active and reactive power changes, the node j Voltage Change and Node i The ratio of voltage change; Define node edge weights e ij Distance from electrical L ij The expression of the relationship is: (8) Modularity function ρ The expression is: (9) Where: e ij For Node i and nodes j The weights of the connected edges; m is the sum of the weights of all edges in the entire community network; k i , k j Node i , j The sum of the weights of all connected edges; node i and j In the same cluster δ ( i,j )=1, otherwise δ ( i,j )=0; In step S1, the cluster comprehensive performance index also includes the balance index of active and reactive power within the cluster. The larger the power balance index is, the higher the degree of power balance within the cluster is. The formula is: (10) Where: φ s It is the power balance index; φ p It is the total active power balance index of the system; φ Q It is the total reactive power balance index of the system; The expression of the total active power balance index of the system is: (11) Where: For cluster k The net power characteristics of the time-varying scenario are shown in Figure 2. N c is the number of clusters; The system reactive power balance index is used to describe the balance relationship between reactive power demand and supply within the cluster, which is expressed as: (12) (13) Where: , Respectively k The maximum reactive power output and reactive power demand of each cluster; For cluster k Reactive power balance.

3. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 2 is characterized by: In step S1, the cluster comprehensive performance index φ The expression is: (14) Where: λ 1. λ 2. λ 3 are the weight coefficients of each indicator; Different weight coefficients are selected according to different cluster division objectives to obtain the corresponding optimal cluster division results.

4. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 1 is characterized by: In step S2, when the heterogeneous distributed photovoltaic is in offline mode, it does not participate in grid interaction and cannot regulate the distribution network. The heterogeneous distributed photovoltaic offline model is expressed as: (15) Where: M1 is offline mode, P HDPV and Q HDPV are the real-time output active power and reactive power of heterogeneous distributed photovoltaics, P ( t )and Q ( t ) are the real-time predicted active power and real-time predicted reactive power of HDPV respectively, T For any time, α is the minimum power factor angle of the photovoltaic inverter; When the heterogeneous distributed photovoltaic is in semi-offline mode, only the reactive power regulation function participates in the grid interaction. The semi-offline model of heterogeneous distributed photovoltaic is expressed as: (16) Where: M2 is the semi-offline mode, Δ Q ( t ) is the real-time regulation amount of heterogeneous distributed photovoltaic reactive power; When heterogeneous distributed photovoltaics are in online mode, they have active and reactive power regulation functions. The online model of heterogeneous distributed photovoltaics is expressed as: (17) Where: M3 is online mode, P dec ( t ) is the real-time reduction of active power of heterogeneous distributed photovoltaics.

5. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 4 is characterized by: In step S3, the objective function of the multi-objective optimization model is expressed as: (18) (19) In the formula, ω 1. ω 2 are the weight coefficients of each indicator; C loss Network loss costs; C pvdec Reduce the cost of active power for heterogeneous distributed photovoltaics; C re Penalty fees for power backhaul; c p is the network loss cost coefficient; I ij,t is the real-time current of the line; r ij is the line impedance; c pv Reduce the cost factor for photovoltaic active power; P PV,i,t,max The upper limit of active power output of heterogeneous distributed photovoltaics; P PV,i,t The real-time photovoltaic output power at each moment; c re is the reverse power penalty coefficient; P re,t is the reverse power; D V is the overall system voltage deviation.

6. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 5 is characterized by: In step S4, the constraints include heterogeneous distributed photovoltaic operation constraints, voltage constraints, and power flow balance constraints. The operation constraints of heterogeneous distributed photovoltaics are expressed as: (20) (21) (22) In the formula, M1, M2, and M3 are the three operating modes of heterogeneous distributed photovoltaics. P PV,i,t It is the real-time active power of heterogeneous distributed photovoltaics. P PV,i,t,max is the upper limit of active power output of heterogeneous distributed photovoltaics, Q PV,i,t It is the real-time reactive power of heterogeneous distributed photovoltaics. Q PV,i,t,max The upper limit of reactive power output of heterogeneous distributed photovoltaics; The voltage constraint is expressed as: (23) Where: V min and V max is the lower and upper limits of voltage; The power flow balance constraint is expressed as: (24) Where: P j,PV,t and Q j,PV,t Node j Active and reactive power injected by the connected heterogeneous distributed photovoltaics; P j,d,t and Q j,d,t For Node j The connected load, P jk,t and Q jk,t Injection nodes j Active power and reactive power; P ij,t and Q ij,t are the active power and reactive power transmitted on the line respectively; x ij is the line reactance.

7. The distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics according to claim 1 is characterized by: In step S1, a genetic algorithm is used to solve and obtain the optimal division result of the distribution network cluster; in step S4, a synchronous alternating direction multiplier method and a commercial solver are used to solve the objective function.

8. A computer device comprising a processor and a storage medium, wherein the storage medium stores a computer program, wherein: When the computer program is executed by a processor, the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics as described in any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the distribution network cluster optimization control method considering the heterogeneous characteristics of distributed photovoltaics as described in any one of claims 1 to 7 is implemented.

Images