Energy storage optimization configuration method based on region division

Abstract

The invention discloses an energy storage optimization configuration method based on region division. The method mainly comprises the steps that a node network is partitioned through a spectral clustering algorithm, the spectral clustering algorithm analyzes the electrical connection relation between nodes of a power grid, the power grid is divided into a plurality of areas, the nodes in each area are electrically and closely connected, and the electrical connection between the areas is relatively weak; on the basis of a partition result, a double-layer energy storage configuration method is adopted to optimize energy storage addressing and sizing, and the upper layer determines the required capacity according to the power grid load and renewable energy output condition in each partition; and the lower layer further partitions internal energy storage installation nodes according to the capacity demand of the upper layer. Through the double-layer energy storage configuration method, it is ensured that the energy storage system meets the power demand, meanwhile, the uncertainty of composite fluctuation and new energy output is flexibly dealt with, and improvement of the overall economy of the system and new energy consumption are facilitated.

Description

Technical Field

[0001] The present invention belongs to the field of energy storage system configuration, and in particular relates to an energy storage optimization configuration method based on regional division. Background Art

[0002] With the "energy transition" and the gradual increase in the proportion of renewable energy, the power system faces more and more challenges. Wind power has developed into one of the leading renewable energy sources due to its advantages such as rich resources and environmental protection. However, wind power is affected by a variety of meteorological factors, and is intermittent, volatile and has poor prediction accuracy, which makes it difficult for wind power to participate in the dispatch plan as a stable power source, and other power sources are required to assist in the dispatch plan. In this case, energy storage technology becomes a very valuable resource. It can not only provide backup capacity for the power system, but also absorb excess wind power when wind power cannot be fully absorbed, thereby improving the utilization rate of wind power. Energy storage technology is flexible and can store and release energy according to demand to provide a stable power supply to the power grid.

[0003] The development trend of global energy storage installed capacity is growing. The flexibility of energy storage systems enables them to improve the distribution of power resources on a time scale. However, the configuration of energy storage systems still has problems such as low utilization, high installation and operation costs, a single way of participating in market regulation, and unreasonable configuration. Especially in large-scale power grid systems, how to reasonably plan and configure energy storage to achieve the best economic benefits is an urgent problem to be solved.

[0004] At present, there are some achievements in the optimization configuration of energy storage. According to different optimization goals, typical capacity configuration strategies include the lowest energy storage configuration cost, minimizing transmission line power flow fluctuations, or maximizing energy storage operation benefits. The scale of traditional power grids is complex, but the current methods of site selection and capacity determination of energy storage are usually carried out in a given area, without considering the autonomy of different regions in the system. In order to cope with the large scale and complex operation mode of the power grid, the transmission network is divided into zones, and energy storage is configured according to the characteristics of each region to achieve a reasonable allocation of energy storage resources, making the power dispatch of each region more flexible and efficient, and improving the flexibility and stability of the entire system. With the continuous development of energy storage, energy storage configuration methods based on transmission network partitioning will play an increasingly important role in future power systems. Summary of the invention

[0005] The present invention aims to provide a method for energy storage site selection and capacity determination, which aims to achieve optimal configuration of energy storage resources in the power grid through reasonable energy storage zoning site selection and capacity determination, so as to improve the overall stability and economy. The site selection and capacity determination method is a method for optimizing energy storage configuration based on regional division. The method uses a spectral clustering algorithm to partition the node network. The spectral clustering algorithm divides the power grid into multiple regions by analyzing the electrical connection relationship between the nodes of the power grid. The nodes in each region are tightly electrically connected, while the electrical connection between regions is relatively weak. Then, based on the partition results, a double-layer energy storage configuration method is used to optimize the energy storage site selection and capacity determination. The upper layer determines the required capacity according to the power grid load and renewable energy output in each partition; the lower layer further partitions the internal energy storage installation nodes according to the upper layer capacity requirements. Through the double-layer energy storage configuration method, it is ensured that the energy storage system meets the power demand while flexibly responding to compound fluctuations and the uncertainty of new energy output, which helps to improve the overall economy of the system and the consumption of new energy.

[0006] To solve the above technical problems, the present invention comprises the following steps:

[0007] Step A: Obtain network parameters, load power and wind power in the area to be configured.

[0008] Step B: Divide the area to be configured with energy storage into regions based on the spectral clustering algorithm.

[0009] Step C. Based on the zoning results of step B, a two-layer energy storage configuration model taking into account regional division is constructed to obtain the energy storage site selection and capacity determination results.

[0010] Compared with the prior art, the present invention has the following beneficial effects:

[0011] Most existing methods use a single model to select the site and capacity of energy storage, without considering the regional characteristics of the power grid, network connection structure, etc., and are difficult to adapt to power grids with different structures, different loads and new energy distributions. When faced with complex power grid scenarios, they lack adaptability and specificity.

[0012] The innovation of the present invention lies in the reasonable partitioning of the grid node network through the spectral clustering algorithm, the accurate analysis of the electrical connection conditions of each region, the implementation of energy storage site selection and capacity determination based on regional characteristics, the optimization of energy storage resources in the grid, the avoidance of unreasonable layout of energy storage, the better use of energy storage resources, and the actual needs of different regions. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of the flow of the energy storage optimization configuration method based on regional division in the present invention;

[0014] Figure 2 is a flow chart of the spectral clustering algorithm in the present invention;

[0015] Figure 3 It is a schematic diagram of the area division of the power transmission network;

[0016] Figure 4 This is a schematic diagram of equivalent nodes after area division;

[0017] Figure 5 The improved IEEE14-node network area partition results;

[0018] Figure 6 This is a schematic diagram of energy storage capacity planning results;

[0019] Figure 7 Schematic diagram of energy storage site selection results. DETAILED DESCRIPTION

[0020] In order to better illustrate the present invention further below with reference to the accompanying drawings, the present invention comprises the following steps:

[0021] Step A: Obtain the network parameters, load power and wind power of the area to be configured, and calculate the regional division coefficient.

[0022] Step B: Divide the area to be configured with energy storage into regions based on the spectral clustering algorithm.

[0023] Step C. Based on the zoning results of step B, a two-layer energy storage configuration model taking into account regional division is constructed to obtain the energy storage site selection and capacity determination results.

[0024] Step A obtains the network parameters, load power and wind power of the area to be configured, and calculates the regional division coefficient, including:

[0025] Step A1. Use Euclidean distance to measure the similarity of time series and calculate the source-load similarity:

[0026]

[0027] Where P L (t) represents the total load power within a certain area at time t, P W (t) represents the total wind power within a certain area at time t, P Li (t) represents the load power of the ith node at time t, P Wj (t) represents the power of the jth wind farm at time t.

[0028] Step A2. Obtain the line impedance Z between node i and node j ij .

[0029] Step A3. Obtain the line transmission capacity P between node i and node j lij,N .

[0030] Step B: Divide the energy storage area to be configured based on the spectral clustering algorithm. Figure 2 As shown, the specific steps include:

[0031] Step B1: construct the similarity matrix S of the network store relationship based on the comprehensive regional division coefficient. Suppose the topology of the power grid is G = (V, E), V = {v1, v2, ..., v n} represents n nodes, each node vi={P Li ,P Wi} consists of load power and wind power information, E = {e1, e2, ..., e m} represents m power grid lines, each line e i Corresponding node (v i ,v j ).

[0032] Multiply the matrix F of n nodes with load power and wind power n×2 , we can get the matrix B by formula (2)

[0033] B=FF T (2)

[0034] (1) Construct the node connection matrix W

[0035]

[0036] (2) Calculate the similarity matrix

[0037]

[0038] (3) The formula is converted into a Laplace matrix, and the eigenvalues ​​and characteristic matrix are solved to obtain the node

[0039] Feature Space of Point Networks

[0040]

[0041] (4) The feature matrix is ​​divided into k categories by the k-means algorithm to achieve the clustering of the graph. Figure 3 As shown,

[0042] Step C. Based on the partitioning results of step B, the specific steps of constructing a two-layer energy storage configuration model taking into account regional division are:

[0043] Step C1. According to the partitioning result of step B, each partition is regarded as an equivalent node, and a simplified equivalent node network is obtained, such as Figure 4 Shown

[0044] Step C2, determine the required capacity based on the simplified equivalent node network. The objective function of the upper model consists of the operating cost of the thermal power plant, the cost of wind power abandonment, and the construction and operation cost of the energy storage power station.

[0045] F=min(F th +F wind +F storage ) (7)

[0046]

[0047] F storage =F js +F op (10)

[0048]

[0049] In the formula, F represents the total operating cost of the system, F th 、F wind and F storage Respectively represent the operating cost of thermal power plants, the cost of wind power abandonment and the construction and operation cost of energy storage power stations; a i 、b i 、c i They represent the operating cost coefficient of the i-th thermal power plant; T represents the optimal time domain for intraday scheduling; N th Indicates the number of thermal power plants in the region; P th,i,t represents the active power of the i-th wind farm at time t, w j is the penalty coefficient of the jth wind farm, and the size of the wind farm penalty coefficient is determined according to the wind farm grouping result; P windp,j,t and P wind,j,t The predicted power and planned power of the jth wind farm at time t; C fixed is the fixed cost of energy storage, C P represents the unit power cost of energy storage, Indicates the energy storage rated power, C E It represents the unit capacity cost, Indicates the rated capacity of energy storage, T life represents the useful life, r represents the discount rate, k op Represents the energy storage operation and maintenance cost.

[0050] Step C3. The constraints corresponding to the upper capacity configuration model are power balance constraints, thermal power unit output constraints, thermal power unit ramp constraints, wind farm output constraints, line transmission capacity constraints and energy storage power station related constraints.

[0051] (1) Power balance constraints:

[0052]

[0053] Where P load,t Represents the system load power at time t.

[0054] (2) Output constraints of thermal power units:

[0055]

[0056] In the formula, and Indicates the minimum and maximum output of a thermal power plant.

[0057] (3) Output ramp constraints of thermal power units:

[0058]

[0059] In the formula, and Represents the upper and lower climbing thresholds of the i-th thermal power plant.

[0060] (4) Wind farm output constraints:

[0061] 0≤P wind,j,t ≤P windp,j,t (16)

[0062] (5) Line transmission capacity constraints

[0063] P l,min ≤P l,t ≤P l,max (17)

[0064] (6) Constraints related to energy storage power stations:

[0065] When the energy storage power station is charging, it meets the following requirements:

[0066]

[0067] When the energy storage power station discharges, it meets the following conditions:

[0068]

[0069] In the formula, SOC t Represents the state of charge of the energy storage power station at time t; η ch and η dis Respectively represent the charging efficiency and discharging efficiency of the energy storage power station; p ch,t and p dis,t Indicates the charging and discharging power of the energy storage power station; E n Represents the rated capacity of the energy storage power station; η ch and η dis Indicates the charging and discharging status of the energy storage power station, which can be 0 or 1.

[0070] SOC min≤SOC(t)≤SOC max (20)

[0071] P dis,zmin ≤P dis,z,t ≤P dis,zmax (twenty one)

[0072] P ch,zmin ≤P ch,z,t ≤P ch,zmax (twenty two)

[0073] η dis *η ch =0 (23)

[0074] Step C4. Based on the above model, the energy storage capacity required for each partition after regional division is solved to determine the configuration capacity of the energy storage system.

[0075] Step C5. Based on the solved energy storage capacity, the lower-level energy storage configuration model aims to determine the energy storage installation nodes within each partition. The present invention uses the power transmission distribution factor and the Theil entropy of power fluctuation as the objective function to determine the optimal installation location of the energy storage.

[0076] The power transmission distribution factor indicates the impact of branch power changes on other road power changes. The higher the coefficient value, the higher the energy storage capacity required in the area to smooth out the fluctuations of new energy. It can be expressed by formula (24):

[0077] express:

[0078]

[0079] Theil entropy of power fluctuation indicates the impact of node power fluctuation on system power flow. The larger the coefficient is, the higher the possibility that the power flow distribution is uneven and heavy when the power fluctuation occurs at the node, and the node needs to be configured with energy storage.

[0080]

[0081] Where Δf i m represents the active power change of line i, It indicates that the active power of line i changes when node i changes.

[0082] The objective function of energy storage site selection is expressed as

[0083] F down =max(ηF i +μT br,j ) (26)

[0084] Step C6. The constraints corresponding to the lower-level capacity configuration model are power balance constraints within the region, thermal power unit output constraints, thermal power unit ramp constraints, wind farm output constraints, line transmission capacity constraints, and energy storage power station-related constraints. The constraint expressions are the same as those in formulas (14)-(24).

[0085] Example

[0086] This embodiment takes the IEEE 14-node network as an example to implement the energy storage optimization configuration method based on regional division. The IEEE 14-node network is improved, and wind turbines are connected at nodes 3 and 8. A similarity matrix is ​​constructed based on the correlation degree of source-load similarity, line impedance, and line transmission capacity, and spectral clustering partitioning is performed. The partitioning results are shown in Table 1. Figure 5 .

[0087]

[0088] based on Figure 5 The regional division results of the energy storage capacity optimization configuration results proposed in this patent are as follows Figure 6 As shown, the site selection results Figure 7 shown.

Claims

1. A method for optimizing energy storage configuration based on regional division, characterized in that: The method comprises the following steps: Step 1. Obtain the network parameters, load power and wind power of the area to be configured; Step 2: Divide the energy storage area to be configured based on the spectral clustering algorithm; Step 3. Based on the zoning results of step 2, a two-layer energy storage configuration model taking into account regional division is constructed to obtain the energy storage site selection and capacity determination results.

2. The method according to claim 1, characterized in that: The spectral clustering algorithm is used to divide the grid nodes into regions, and based on the regional division results, the energy storage installation capacity and location are optimized through the energy storage double-layer configuration method.

3. The method according to claim 1, characterized in that The spectral clustering algorithm includes constructing a similarity matrix, obtaining the feature space of the nodes through eigenvalue decomposition, and then partitioning the power grid nodes using the k-means algorithm.

4. The method according to claim 1, characterized in that: The required energy storage capacity is determined by simplifying the equivalent node network. The objective function includes the operating cost of thermal power units, the cost of wind curtailment, and the construction and operation cost of the energy storage power station. The constraints include power balance constraints, thermal power unit output constraints, wind farm output constraints, line transmission capacity constraints, and energy storage power station related constraints.

5. The method according to claim 1, characterized in that The lower model determines the optimal installation location of energy storage through the power transmission distribution factor and the Theil entropy of power fluctuation.

6. The method according to claim 1, characterized in that By solving the upper and lower models, the energy storage capacity and optimal installation location of each partition are determined, thereby achieving optimal configuration of energy storage.

Images