阅读说明：本技术 一种配电网光伏消纳容量提升方法 (Method for improving photovoltaic absorption capacity of power distribution network ) 是由 陈文藻 冯子蛟 杨玉勇 舒孝国 黄金涛 钱晓明 蒋冬强 于 2021-07-22 设计创作，主要内容包括：本发明公开了一种配电网光伏消纳容量提升方法,其特征在于,所述方案包括：通过预先建立的配电网模型对配电网进行全年时序电压仿真,得到配电网各节点全年电压越限情况；对配电网节各点电压进行无功调节,得到初步改善之后的各节点全年电压越限情况,计算节点电压越限度；确定储能定址方案,针对不同方案计算储能容量、年总弃光功率以及各自方案下的节点电压平均波动改善指标以及最大消纳能力指标；根据不同方案下的四个指标,采用层次分析法和灰色关联度分析综合评价每一个方案,得到最终的储能定址定容以及弃光方案。(The invention discloses a method for improving photovoltaic absorption capacity of a power distribution network, which is characterized by comprising the following steps: carrying out annual time sequence voltage simulation on the power distribution network through a pre-established power distribution network model to obtain annual voltage out-of-limit conditions of each node of the power distribution network; performing reactive power regulation on the voltage of each point of the distribution network node to obtain the annual voltage out-of-limit condition of each node after preliminary improvement, and calculating the out-of-limit of the node voltage; determining an energy storage addressing scheme, and calculating energy storage capacity, annual total optical power curtailment, node voltage average fluctuation improvement indexes and maximum absorption capacity indexes under respective schemes aiming at different schemes; and comprehensively evaluating each scheme by adopting an analytic hierarchy process and grey correlation degree analysis according to four indexes under different schemes to obtain a final energy storage addressing constant volume and light abandoning scheme.)

1. A method for improving photovoltaic absorption capacity of a power distribution network is characterized by comprising the following steps:

carrying out annual time sequence voltage simulation on the power distribution network through a pre-established power distribution network model to obtain annual voltage out-of-limit conditions of each node of the power distribution network;

performing reactive power regulation on the voltage of each point of the distribution network node to obtain the annual voltage out-of-limit condition of each node after preliminary improvement, and calculating the out-of-limit of the node voltage;

selecting energy storage installation nodes through a cluster strategy and a node voltage out-of-limit degree to determine to obtain an energy storage addressing scheme, and calculating energy storage capacity, annual total optical power curtailment, node voltage average fluctuation improvement indexes and maximum absorption capacity indexes under respective schemes aiming at different schemes;

and comprehensively evaluating each scheme by adopting an analytic hierarchy process and grey correlation degree analysis according to four indexes under different schemes to obtain a final energy storage addressing constant volume and light abandoning scheme.

2. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 1, wherein the power distribution network model parameters comprise the topological structure, the reference voltage, the reference power, the line impedance of the power distribution network, and the annual load power and photovoltaic power of each node of the power distribution network.

3. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 1, wherein the node voltage out-of-limit is used for measuring the size of the node voltage out-of-limit, and the specific formula is as follows:

in the formula: m is the number of overyear time, U_NmaxIs the upper limit of the safe threshold of the node voltage, U_iIs the voltage at node i.

4. The method for increasing the photovoltaic absorption capacity of the power distribution network according to claim 1, wherein the method for determining the energy storage addressing scheme comprises the following steps:

dividing the power distribution network into a plurality of clusters by taking the electrical distance and the power balance as cluster strategy indexes according to the photovoltaic power and the load power of each node in the power distribution network;

and selecting the first six points with the most serious out-of-limit as energy storage addressing nodes according to the out-of-limit voltage of each node in each cluster, and formulating an energy storage addressing scheme.

5. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 1, wherein the method for calculating the energy storage capacity, the annual total optical power curtailment, the node voltage average fluctuation improvement index under each scheme and the maximum absorption capacity index comprises the following steps:

the method comprises the steps of solving the optimal energy storage capacity configuration and the annual total abandoned optical power under respective schemes by adopting a cooperative game model and taking the sum of the energy storage configuration capacity and the annual abandoned optical power as a characteristic function;

and carrying out annual photovoltaic consumption again under the optimal energy storage configuration capacity and annual total abandoned optical power configuration, and calculating the node voltage average fluctuation improvement index and the maximum consumption capability index after the new optimized consumption of each scheme.

6. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 5, wherein the cooperative game model specifically comprises:

the participants: nc nodes of the energy storage configuration;

the strategy set is as follows: each energy storage configuration node stores electric power, abandoned optical power and reactive power regulation power at the time of the whole year;

characteristic function: the total sum of the total energy storage configuration capacity and the annual total abandoned optical power of the power distribution network is the lowest;

in the formula: nc is the number of energy storage node installations, S_nFor the energy storage installation capacity of node n and the sum of annual total optical power curtailment, Q_tReactive power, P, compensated for node n at time t_ch,tCharging power of the energy storage means at time t, P_dis,tDischarge power of the energy storage device at time t, P_gp,tThe light power is abandoned at the time T, and the time T is the annual time.

7. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 5, wherein the node voltage average fluctuation improvement index is used for measuring the improvement degree of voltage fluctuation after energy storage annual charge and discharge scheduling and light abandoning treatment:

in the formula: nc is the number of nodes, du, configuring the energy storage_iTo the extent that the voltage fluctuation at node i is improved, U_i,t、U_i,t' the voltage value of the node i before and after the improvement at time t, U_i,av、U_i,av' average value before and after improvement of voltage at node i, dU_i、dU_i' the voltage fluctuation before and after the voltage improvement, respectively, and T is the time of the whole year.

8. The method for improving the photovoltaic consumption capacity of the power distribution network according to claim 5, wherein the maximum consumption capacity index is used for measuring the photovoltaic consumption capacity improvement condition of the power distribution network caused by energy storage configuration and light abandonment in a scheme, and the new photovoltaic consumption of the power distribution network is based on the rated capacity, the rated power and the annual total light abandonment power of the energy storage in the scheme.

9. The method for improving the photovoltaic absorption capacity of the power distribution network according to claim 1, wherein the method for evaluating the energy storage addressing capacity fixing and light abandoning scheme comprises the following steps:

determining four index values of different energy storage addressing schemes: the total energy storage configuration capacity, the annual total optical power abandonment, the node voltage average fluctuation improvement and the maximum absorption capacity;

determining the weight of the four indexes by an analytic hierarchy process, determining the gray relevance of different energy storage addressing schemes by the gray relevance, and determining the final energy storage addressing scheme by gray weighting.

Technical Field

The invention relates to a method for improving photovoltaic absorption capacity of a power distribution network, and belongs to the technical field of distributed photovoltaic planning of the power distribution network.

Background

The consumption of traditional energy not only causes the deterioration of the environment, but also continuously exhausts the energy. The new energy has the characteristics of cleanness and reproducibility, and the attention of people is focused on the development of the new energy. In recent years, with the continuous advocated new energy of China and the continuous implementation of various subsidy policies of various new energy, the rapid development of new energy is promoted. Wherein, the energy with the largest installed capacity is photovoltaic energy. According to incomplete statistics, the photovoltaic installed capacity reaches 30.1GW in 2019. However, due to the fluctuation and randomness of the photovoltaic output, the safety and stability of the power distribution network are affected when the power distribution network is accessed in such a large scale. Therefore, it is of great significance to adopt reasonable measures to carry out photovoltaic absorption so as to ensure the safety and stability of the power distribution network.

To date, various countries have conducted extensive and profound research into photovoltaic consumption of power distribution networks. The photovoltaic absorption method is various, such as energy storage equipment, distribution network reconstruction, reactive compensation, power factor adjustment of a photovoltaic inverter, voltage coordination control and the like.

The energy storage technology is the most common photovoltaic absorption method, and can absorb an electric load to serve as a load when the voltage is higher; when the voltage is lower, electric power release is carried out to act as a power supply; therefore, the stored energy can improve the voltage waveform and play a role in peak clipping and valley filling.

The power regulation of the photovoltaic inverter is to control the power factor of the photovoltaic grid connection, and the photovoltaic output absorption is carried out in the output peak period so as to avoid the voltage from exceeding the limit and improve the photovoltaic acceptance capacity of the power distribution network. The photovoltaic consumption of the photovoltaic inverter can be divided into three stages, wherein the first stage is reactive power regulation until the maximum power factor or the maximum photovoltaic installation capacity is reached, and the process does not involve photovoltaic active power rejection; the second stage is to keep the reactive power unchanged and discard the active power to reach the maximum power factor; the third phase is the simultaneous dimming and reactive power regulation around the maximum photovoltaic power factor.

The hongmin yao et al make the nodes return to the specified values by means of the photovoltaic inversion processing and the staged control of the inverter, but do not perform further research from the planning perspective; zhao ripples et al carry out site selection and constant volume to the energy storage through L index, because the correlation between the node, it is not the best choice to carry out the energy storage configuration to single index. The Douqiong et al address and fix the volume of the energy storage of the power distribution network with high photovoltaic permeability through a double-layer planning model, but the double-layer planning model is relatively complex, and the energy storage address of the upper layer is random. Dingming et al perform cluster division on the distribution networks according to the correlation among the considered nodes, establish a double-layer coordination site selection constant volume planning model of distributed photovoltaic and energy storage according to the cluster division result, and comprehensively consider the photovoltaic installation and energy storage configuration of the distribution networks.

Disclosure of Invention

The invention aims to provide a method for improving photovoltaic absorption capacity of a power distribution network, which aims to overcome the defects in the prior art.

A method for improving photovoltaic absorption capacity of a power distribution network comprises the following steps:

carrying out annual time sequence voltage simulation on the power distribution network through a pre-established power distribution network model to obtain annual voltage out-of-limit conditions of each node of the power distribution network;

performing reactive power regulation on the voltage of each point of the distribution network node to obtain the annual voltage out-of-limit condition of each node after preliminary improvement, and calculating the out-of-limit of the node voltage;

selecting energy storage installation nodes through a cluster strategy and a node voltage out-of-limit degree to determine to obtain an energy storage addressing scheme, and calculating energy storage capacity, annual total optical power curtailment, node voltage average fluctuation improvement indexes and maximum absorption capacity indexes under respective schemes aiming at different schemes;

and comprehensively evaluating each scheme by adopting an analytic hierarchy process and grey correlation degree analysis according to four indexes under different schemes to obtain a final energy storage addressing constant volume and light abandoning scheme.

Furthermore, the power distribution network model parameters include the topological structure, the reference voltage, the reference power, the line impedance of the power distribution network, and the annual load power and photovoltaic power of each node of the power distribution network.

Further, the node voltage out-of-limit degree is used for measuring the size of the node voltage out-of-limit, and the specific formula is as follows:

in the formula: m is the number of overyear time, U_NmaxIs the upper limit of the safe threshold of the node voltage, U_iIs the voltage at node i.

Further, the method for determining the energy storage addressing scheme comprises the following steps:

dividing the power distribution network into a plurality of clusters by taking the electrical distance and the power balance as cluster strategy indexes according to the photovoltaic power and the load power of each node in the power distribution network;

and selecting the first six points with the most serious out-of-limit as energy storage addressing nodes according to the out-of-limit voltage of each node in each cluster, and formulating an energy storage addressing scheme.

Further, the method for calculating the energy storage capacity, the annual total optical power curtailment, the node voltage average fluctuation improvement index and the maximum absorption capacity index under the respective schemes comprises the following steps:

the method comprises the steps of solving the optimal energy storage capacity configuration and the annual total abandoned optical power under respective schemes by adopting a cooperative game model and taking the sum of the energy storage configuration capacity and the annual abandoned optical power as a characteristic function;

and carrying out annual photovoltaic consumption again under the optimal energy storage configuration capacity and annual total abandoned optical power configuration, and calculating the node voltage average fluctuation improvement index and the maximum consumption capability index after the new optimized consumption of each scheme.

Further, the cooperative game model specifically includes:

the participants: nc nodes of the energy storage configuration;

the strategy set is as follows: each energy storage configuration node stores electric power, abandoned optical power and reactive power regulation power at the time of the whole year;

characteristic function: the total sum of the total energy storage configuration capacity and the annual total abandoned optical power of the power distribution network is the lowest;

in the formula: nc is the number of energy storage node installations, S_nFor the energy storage installation capacity of node n and the sum of annual total optical power curtailment, Q_tReactive power, P, compensated for node n at time t_ch,tCharging power of the energy storage means at time t, P_dis,tDischarge power of the energy storage device at time t, P_gp,tThe light power is abandoned at the time T, and the time T is the annual time.

Further, the node voltage average fluctuation improvement index measures the improvement degree of voltage fluctuation after energy storage annual charge and discharge scheduling and light abandoning treatment:

in the formula: nc is the number of nodes, du, configuring the energy storage_iTo the extent that the voltage fluctuation at node i is improved, U_i,t、U_i,t' the voltage value of the node i before and after the improvement at time t, U_i,av、U_i,av' average value before and after improvement of voltage at node i, dU_i、dU_i' the voltage fluctuation before and after the voltage improvement, respectively, and T is the time of the whole year.

Furthermore, the maximum absorption capacity index is used for measuring the improvement condition of the photovoltaic absorption capacity of the power distribution network caused by energy storage configuration and light abandonment in the scheme, and the new photovoltaic absorption of the power distribution network is established on the basis of the rated capacity, the rated power and the annual total light abandonment power of the energy storage in the scheme.

Further, the evaluation method of the energy storage addressing constant volume and light abandoning scheme comprises the following steps:

determining four index values of different energy storage addressing schemes: the total energy storage configuration capacity, the annual total optical power abandonment, the node voltage average fluctuation improvement and the maximum absorption capacity;

determining the weight of the four indexes by an analytic hierarchy process, determining the gray relevance of different energy storage addressing schemes by the gray relevance, and determining the final energy storage addressing scheme by gray weighting.

Compared with the prior art, the invention has the following beneficial effects: the invention overcomes the complexity of a double-layer algorithm, carries out addressing selection of energy storage in a simpler mode of index classes, considers the correlation characteristics of nodes, cooperatively consumes photovoltaic capacity, carries out energy storage capacity and light abandoning optimization, introduces two new indexes for objectively evaluating the energy storage capacity of different schemes, carries out scheme evaluation by adopting a mode of combining an analytic hierarchy process and a grey correlation degree analysis subjectively and objectively, and can carry out final selection by a final decision maker according to the priority sequence of the four indexes.

Drawings

FIG. 1 is a photovoltaic consumption flow diagram of a photovoltaic inverter and energy storage staging control;

FIG. 2 is a 33 node distribution network diagram of the IEEE standard;

FIG. 3 is a photovoltaic capacity configuration of a node;

FIG. 4 is a year-round voltage value for node 24 before reactive regulation;

FIG. 5 is a year-round voltage value for node 24 after reactive regulation;

FIG. 6 is a graph of voltage violations of corresponding nodes of different clusters.

Detailed Description

In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.

As shown in fig. 1-6, a method for increasing photovoltaic absorption capacity of a distribution network is disclosed, as shown in fig. 1, the method includes:

step 1: establishing a power distribution network model containing a photovoltaic system, and carrying out annual time sequence voltage simulation on the power distribution network to obtain annual voltage out-of-limit conditions of each node of the power distribution network;

the power distribution network model parameters comprise a topological structure of the power distribution network, reference voltage, reference power, line impedance, annual load power of each node of the power distribution network and photovoltaic power.

The annual time sequence voltage simulation is to perform load flow calculation according to the load power of the power distribution network nodes and the photovoltaic power at each moment of the year to obtain the voltage value of each node at each moment of the year.

Step 2, utilizing an inverter of the photovoltaic access node to perform reactive power regulation on the voltage of the node to obtain the annual voltage out-of-limit condition of each node after preliminary improvement, and calculating the out-of-limit of the voltage of the node;

the reactive power regulation of the inverter of the photovoltaic access node is to compensate the reactive power of the node by using the inverter until the maximum power factor or the maximum inversion capacity is reached, and the photovoltaic active power is not abandoned at this stage.

The node voltage out-of-limit degree is used for measuring the size of the node voltage out-of-limit:

in the formula: m is the number of overyear time, U_NmaxIs the upper limit of the safe threshold of the node voltage, U_iIs the voltage at node i.

Selecting energy storage installation nodes by using a cluster strategy and the node voltage out-of-limit degree to obtain a plurality of energy storage addressing schemes; the method comprises the following specific steps:

(31) dividing the power distribution network into a plurality of clusters by taking the electrical distance and the power balance as cluster strategy indexes according to the photovoltaic power and the load power of each node in the power distribution network;

(32) selecting the first points with the most serious out-of-limit as energy storage addressing nodes according to the out-of-limit voltage of each node in each cluster, and formulating an energy storage addressing scheme;

the cluster strategy is a strategy for regionally dividing the power distribution network by considering strong and weak relevance among nodes. The relevance between clusters is weaker, and the relevance in the clusters is stronger; the specific clustering strategy is to divide the power distribution network into a plurality of areas, and one area is a cluster. When a certain node in the area changes, other nodes in the area change more greatly, and nodes in other areas change less greatly, so that the node relevance in the cluster is stronger, and the node relevance outside the cluster is weaker.

And 4, solving the optimal energy storage configuration capacity and the annual total optical power under the condition of meeting the constraint condition of the multiple energy storage addressing schemes by adopting a cooperative game model, and calculating the node voltage average fluctuation improvement index and the maximum absorption capacity index under the respective schemes. Comprises the following steps:

(41) the method comprises the steps of solving optimal energy storage capacity configuration and annual total abandoned optical power under respective schemes by adopting a cooperative game model and taking the sum of energy storage capacity and annual abandoned optical power as a characteristic function;

(42) carrying out annual photovoltaic consumption again under the optimal energy storage configuration capacity and annual total abandoned optical power configuration, and calculating a node voltage average fluctuation improvement index and a maximum consumption capability index after the new round of optimized consumption of each scheme;

the cooperative game model comprises:

(1) the participants: nc nodes of the energy storage configuration.

(2) The strategy set is as follows: the stored electric power and the abandoned optical power of each energy storage configuration node at the time of the year

And adjusting the power in a reactive mode.

(3) Characteristic function: the total energy storage configuration and the annual optical power abandonment sum of the power distribution network are the lowest.

The constraint conditions of the cooperation model are as follows:

(1) energy storage charge and discharge power constraint

At the same time, the energy storage charging and discharging power can not be carried out simultaneously:

P_ch,t·P_dis,t＝0

in the formula: p_ch,tThe charging power of the energy storage device at the moment t; p_dis,tIs the discharge power of the energy storage device at time t.

(2) Energy storage state of charge confinement

In order to prolong the service life of the electric storage equipment, the state of charge of the electric storage equipment needs to be ensured within a certain range, and the state of charge of the electric storage device at the moment t is related to the state of charge of the electric storage device at the moment t-1 and the charging and discharging power at the moment t:

S_emin≤S_e,t≤S_emax

in the formula: s_e,tState of charge at time t, S_e,t-1Is the state of charge at time t-1, η_echFor the charging efficiency of the electricity storage device, η_edisFor the discharge efficiency of the electricity storage device, S_NeFor the rated capacity of the electricity storage apparatus, S_emin、S_emaxThe minimum value and the maximum value of the energy storage charge state are shown, and delta t is unit scheduling time.

(3) Waste light confinement

The curtailment power needs to be less than the photovoltaic power generated at a certain moment:

P_gp,t≤P_pv,t

in the formula: p_gp,tIs the optical power, P, of the waste light at time t_pv,tIs the generated photovoltaic power at time t.

(4) Node voltage constraint

U_min≤U_i≤U_max

In the formula: u shape_minIs the lower limit value of the safety range of the node voltage; u shape_maxIs the upper limit value of the safe range of the node voltage; u shape_iIs the voltage at node i.

The node voltage average fluctuation improvement index is used for measuring the improvement degree of voltage fluctuation after energy storage annual charging and discharging scheduling and light abandoning treatment:

in the formula: nc is the number of nodes, du, configuring the energy storage_iTo the extent that the voltage fluctuation at node i is improved, U_i,t、U_i,t' the voltage value of the node i before and after the improvement at time t, U_i,av、U_i,av' average value before and after improvement of voltage at node i, dU_i、dU_i' the voltage fluctuation before and after the voltage improvement, respectively, and T is the time of the whole year.

The maximum absorption capacity index is used for measuring the improvement condition of the photovoltaic absorption capacity of the power distribution network caused by energy storage configuration and light abandonment in the scheme, and the new photovoltaic absorption of the power distribution network is established on the basis of the rated capacity, the rated power and the annual total light abandonment power of the energy storage in the scheme. The maximum absorption capacity is solved by adopting a PSO optimization algorithm, and the adaptive value of the PSO particles is solved by using an adaptive value solving model.

The specific implementation steps are as follows:

(1) configuring relevant parameters of PSO particles, wherein the relevant parameters are the population number and the iteration number of the particles

Number, acceleration coefficient, weight coefficient;

(2) randomly initializing the position and speed of PSO particles with different PV-amps

Filling capacity;

(3) substituting the position of each particle of the PSO into an adaptive value solving model;

(4) inputting relevant parameters of the solving model, wherein the relevant parameters are reference voltage and reference work of the power distribution network

The method comprises the following steps of (1) calculating the power, the topological structure of a power distribution network, the impedance of a power transmission circuit, the load power of each node at each moment all the year around, and calculating the photovoltaic output power of each node at each moment all the year around at the position (photovoltaic capacity);

(5) carrying out trend solution on the (4), and finding out-of-limit nodes and out-of-limit time;

(6) under the condition of meeting the constraint condition, the energy storage and light abandon regulation is carried out when the out-of-limit node is out-of-limit,

the node voltage at the moment is ensured to be recovered to be within a normal range, and if the node voltage cannot be recovered to a normal voltage value after energy storage and light rejection regulation, the adaptive value is returned to be 0; otherwise, returning the adaptive value as the photovoltaic installation capacity.

(7) Updating position and speed of PSO with maximum photovoltaic installation capacity as objective function until full

And (4) the condition of termination is satisfied, and the maximum photovoltaic installation capacity is output.

And (4) solving the annual photovoltaic power output at the moment:

P_pv,i＝CO_pv,tC_pv,i

in the formula: CO 2_pv,tThe photovoltaic output coefficient is obtained at the moment t; c_pv,iIs the photovoltaic installation capacity of node i.

The constraint condition of (6):

firstly, energy storage charging and discharging conditions:

the charging and discharging power for regulating the out-of-limit voltage must not exceed the rated power:

in the formula: p_NRated power for energy storage

Charging constraint conditions of energy storage:

S_emin≤S_e,t≤S_emax

abandon light constraint condition:

in the formula: p_gp,t,iIs the optical power of the node i at time t, P_sum,iFor node i total optical power discard (i.e. total optical power discard in the scheme)

And 5, comprehensively evaluating each scheme by adopting an analytic hierarchy process and grey correlation degree analysis by combining four indexes under different schemes to obtain the optimal energy storage addressing constant volume and light abandoning scheme. Comprises the following steps:

(51) determining four index values of different energy storage addressing schemes: the total energy storage installation capacity, the annual total optical power abandonment, the node voltage average fluctuation improvement and the maximum absorption capacity;

(52) determining the weight of the four indexes by an analytic hierarchy process, determining the gray relevance of different energy storage addressing schemes by the gray relevance, and determining the final energy storage addressing scheme by gray weighting.

The comprehensive evaluation steps are as follows:

(1) normalizing sample sequences and finding optimal reference sequences

If a total of m schemes, n indexes, are set, the sample matrix X composed of sample sequences is (X)_ij)_mnNormalization is performed and a reference sample is found.

d_j＝opt(c_ij)

In the formula: x is the number of_ijRepresents the ith scheme, and the corresponding value i is 1, 2.. multidot.m under the jth index; j is 1,2, n, d_jAnd (4) selecting the optimal sample in the j index (selecting the maximum value or the minimum value under the j index according to the actual situation).

(2) Calculating grey correlation degree

Definition of Δ_ijFor the normalized ith solution, the deviation of the value on the j index from the optimal value on the j index is:

Δ_ij＝|c_ij-d_j|

after normalization, defining the gray correlation epsilon between the value of the ith scheme on the j indexes and the j best index_ijIs composed of

In the formula: ρ is a resolution coefficient, and is generally selected to be 0.5.

(3) Determining weights by analytic hierarchy process

Constructing a judgment matrix between the considered indexes, performing consistency check on the considered matrix, and passing

The consistency test is carried out to obtain the weight omega ═ omega₁,ω₂,…,ω_j)。

(4) Calculating a weighted relevance

(5) And sorting according to the magnitude of the association degree, and selecting an optimal scheme.

γ_iThe larger the value of (b) indicates that the ith scheme is associated with the optimal sample to a greater extent, and therefore, the optimal scheme selection can be performed according to the associated extent.

Example 2

For an annual time sequence voltage analysis of a power distribution network installed with 6MW of photovoltaic capacity, a topological structure diagram of the power distribution network is shown in fig. 2, a specific node photovoltaic capacity configuration is shown in fig. 3, and since the photovoltaic capacity installed on the node 24 is the largest, the voltage out-of-limit degree is also the highest, as shown in fig. 4.

Therefore, the inverter and energy storage combination strategy is needed to be utilized to enable the node to recover to the specified value, so that the completeness and the stability of the power distribution network are ensured.

1. Preliminary regulation of an inverter

After the initial regulation of the inverter reactive power, the degree of out-of-limit at node 24 is significantly reduced, as shown in fig. 5.

2. Energy storage addressing scheme

And (4) according to the voltage condition of each node simulated all the year round, selecting an energy storage addressing scheme by combining the node voltage out-of-limit index and the cluster division strategy. The cluster division uses the electrical distance as a structural index to measure the correlation degree of the change of the node voltage amplitude; and taking the power balance degree as a power index to measure the influence of active power change and reactive power change on the node.

According to the standard of cluster division and considering the photovoltaic installation capacity and the position, the standard power distribution network is divided into four clusters, and the node voltage out-of-limit degree in each cluster is shown in fig. 6

Four clusters are divided in total, and the nodes of the cluster 1 with out-of-limit voltage are 23, 24 and 25; there is no voltage off-limit node in cluster 2; nodes in cluster 3 at which the voltage is out of limit are 5, 6, 7, 8, 9, 26, 27, 28, 29, 30, 33; the voltage violation nodes in the cluster 4 are 10, 11, 12.

Considering the characteristics of the clusters, firstly, the initial selection is carried out according to the voltage out-of-limit condition in the clusters, and then the comprehensive judgment is carried out by combining the relationship among the clusters. If the voltage out-of-limit degree difference between adjacent nodes in the cluster is large, the node with serious voltage out-of-limit drives the node with small voltage out-of-limit degree. Therefore, as long as the energy storage configuration is carried out at the node with serious out-of-limit voltage, the voltage is reduced, and the voltage of the node with small out-of-limit degree of the node is naturally reduced. After the nodes in the clusters are screened, the relevance among the clusters is considered, if the maximum voltage out-of-limit degree in the clusters is far smaller than the minimum out-of-limit degree in the rest clusters, the cluster voltage out-of-limit is considered to be driven by the rest clusters, the rest cluster voltages are mainly controlled, and the voltage of the clusters returns to a normal value.

The voltage out-of-limit degree of the node 23 and the nodes 24 and 25 in the cluster 1 is relatively large, the voltage out-of-limit of the nodes 24 and 25 is serious, and the voltage out-of-limit of the node 23 can be considered to be driven by the nodes 24 and 25, so that energy storage configuration is performed on the nodes 24 and 25. Cluster 2 has no voltage violations and therefore is not considered. The voltages of the nodes 6, 26, 27 and 28 in the cluster 3 are different from those of the rest nodes, so that an energy storage device needs to be installed. The maximum voltage out-of-limit of the nodes in the cluster 4 is lower, and therefore, it can be considered that the node voltage out-of-limit of the cluster 4 is driven by the rest of the clusters, and therefore, the cluster 4 does not perform energy storage configuration.

According to the node energy storage configuration, the nodes 24, 25, 6, 27, 26 and 28 are proposed as the alternative nodes of the energy storage configuration. And based on the degree of out-of-limit, the following energy storage configuration scheme is drawn up, as shown in table 3.

TABLE 1 schema location and number determination

Scheme(s)	Configuring a node
		A	24
II	24、25
		III	24、25、27
Fourthly	24、25、27、26
		Five of them	24、25、27、26、6
Six ingredients	24、25、27、26、6、28

3. Energy storage constant volume scheme

And (4) carrying out cooperative game according to the energy storage scheme to obtain the optimal energy storage configuration and light abandoning capacity. Results as shown in table 2, the rejection power was satisfactory as the photovoltaic power generated all year round was 8800 MW.

TABLE 2 energy storage configuration and optical rejection power

As can be seen from table 2, in the first scheme, a single node performs energy storage configuration, the energy storage configuration capacity is relatively small, but the annual total abandoned optical power is the highest, and since the single node cannot be cooperatively absorbed by other nodes, the absorption capacity under the energy storage and abandoned optical power is also the lowest; in the sixth scheme, energy storage configuration is carried out on six nodes, the total energy storage configuration capacity is the highest, the annual total abandoned light power is reduced, and the photovoltaic absorption level is improved to the maximum extent due to the mutual cooperation of energy storage and abandoned light among the nodes. In different schemes, because the node energy storage configurations are different, but the light abandoning and the total capacity of the energy storage configurations are different, the more the number of the energy storage configuration nodes is, the larger the total energy storage configuration capacity is, but the new round of absorption capacity is improved; the average node improvement level has no direct relation with the number of nodes, but the node voltage fluctuation is obviously improved due to the regulation effect of energy storage and light abandoning.

5. Multi-index evaluation

Comprehensively evaluating the four indexes according to the different schemes to select the most suitable scheme, taking the correlation between different indexes and the importance degree between the indexes into consideration, and comprehensively evaluating the indexes by adopting an analytic hierarchy process and the grey correlation degree indexes, wherein the results are shown in Table 3

TABLE 3 comprehensive results of multiple index evaluation

The scheme 6 has the highest comprehensive score, namely the best scheme, and the consumption capacity of the power distribution network can be improved due to the fact that the nodes in the scheme 6 have more energy storage configurations and are coordinated to work. When the required photovoltaic capacity of the power distribution network needs to be further expanded, 1.72MW photovoltaic capacity additional expansion can be carried out.

The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.