Regional power grid coordinated dispatching method considering partition aggregation

Abstract

The invention relates to the technical field of smart power grids, in particular to a regional power grid coordinated dispatching method considering partition aggregation. Comprising the steps of collecting power generation and power consumption area resources; power generation and power consumption area resources are classified; for the concentrated resources, establishing a single resource aggregation model, and aggregating by adopting a k-means method; establishing a distributed resource aggregation model for the distributed resources, and performing partition aggregation on the regional distributed resources; through an optimization algorithm, based on target functions and constraint conditions set for different resource types, coordination between regions is realized. The method has the advantages that resources are aggregated according to resource distribution and operation conditions of a regional power grid, the power grid dispatching difficulty can be effectively reduced, the power grid resource utilization rate is improved, the requirements of a current novel power system are met, and compared with a traditional dispatching method, the method has the advantages of being efficient, flexible, economical and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of smart grids, and particularly to a coordinated dispatching method for regional power grids considering partition aggregation. Background Art

[0002] In modern power systems, distributed power resources (such as distributed generation, energy storage devices, demand response, etc.) have been widely applied. However, how to efficiently integrate these distributed resources and achieve optimal regulation has become the core challenge in the current development of power systems. Existing power grid systems usually rely on centralized dispatching methods, mainly focusing on the centralized optimization of large-scale power generation, and insufficiently considering the diversity and regional characteristics of distributed resources. Specifically, it is reflected in the following aspects:

[0003] Limitations of centralized dispatching strategies: The current dispatching mode centers around large-scale power generation, ignoring the uniqueness of distributed resources in different regions and making it difficult to fully utilize their potential.

[0004] Static demand response model: Most existing demand response strategies are based on fixed rules and fail to adapt to the dynamic changes of user behavior or price fluctuations in the market in real time, reducing the flexibility of the system.

[0005] Single-layer optimization model: Existing dispatching models usually only optimize a certain level of provincial or prefectural power grids and lack the ability of cross-level integrated collaborative optimization, resulting in low efficiency of resource integration and regulation.

[0006] In addition, these existing technologies also have the following specific defects:

[0007] Insufficient utilization of the spatio-temporal characteristics of distributed resources: The characteristics of distributed resources in time (periodicity, volatility) and space (geographical distribution) have not been fully studied and utilized, resulting in lack of pertinence in dispatching strategies and making it difficult to achieve optimal resource allocation.

[0008] Single user-side response mechanism: Existing user response strategies fail to design classification for the different demand characteristics of industrial, commercial, and residential users, and the response mode is too single, resulting in low user participation.

[0009] Insufficient provincial-prefectural collaborative optimization ability: The optimal dispatching between provincial power grids and prefectural power grids is usually carried out independently, lacking a cross-level resource integration mechanism and making it difficult to achieve global optimization.

[0010] Insufficient flexibility of market-based incentives: The current market price signals lack the ability of dynamic adjustment, and the incentive mechanism design is not comprehensive enough, making it difficult to stimulate the potential of users to actively participate in resource regulation.

[0011] Therefore, there is an urgent need for efficient management and optimization of coordinated dispatching of regional power grids with partition aggregation. Summary of the Invention

[0012] In view of the above existing technical problems, the present invention is proposed.

[0013] To solve the above technical problems, the present invention provides the following technical solutions: A coordinated dispatching method for regional power grids considering partition aggregation, which includes collecting power generation and power consumption regional resources;

[0014] Classifying power generation and power consumption regional resources;

[0015] For concentrated resources, establish a single resource aggregation model and use the k-means method for aggregation;

[0016] For distributed resources, establish a distributed resource aggregation model and partition and aggregate the regional distributed resources;

[0017] Through an optimization algorithm, based on the objective functions and constraint conditions set for different resource types, inter-regional coordination is achieved.

[0018] The resource collection includes counting centralized resources and counting distributed resources;

[0019] The concentrated power generation clusters are aggregated as power generation units;

[0020] The concentrated power consumption clusters are aggregated as power consumption units;

[0021] The concentrated power generation / consumption clusters are aggregated as composite aggregates.

[0022] As a preferred solution of the coordinated dispatching method for regional power grids considering partition aggregation of the present invention, wherein: S1. Propose a power generation output aggregation index for a single aggregate and establish a single aggregate aggregation model;

[0023] Among them, the establishment of the single resource aggregation model includes the average output of a single aggregate, the output volatility of a single aggregate, and the output distribution of a single aggregate;

[0024] S2. Normalize the indicators and distinguish positive indicators and negative indicators;

[0025] Among them, the average output of a single aggregate is a positive indicator, and the output volatility of a single aggregate and the skewness of the output distribution of a single aggregate are negative indicators;

[0026]

[0027] In the formula: is the normalized positive indicator, is the normalized negative indicator; X i is the original indicator value; X i,最大 、Xi,最小 They are the maximum and minimum values in the original index values respectively;

[0028] S3. Determine the number of clusters;

[0029] Calculate the index SSE and plot it as a curve. The point where the SSE decline rate in the curve significantly decreases, that is, the point where the curve appears at the elbow, and the corresponding K value is the optimal number of clusters. The SSE calculation formula is as follows

[0030]

[0031] In the formula, SSE is the sum of squared errors of clustering; K is the total number of clustering clusters; C i is the i-th cluster after partitioning; x is the index value after normalization within C i ; μ i is the index value corresponding to the selected clustering center.

[0032] S4. Perform clustering using k-means;

[0033] First, initialize the clustering center and randomly select K data points as the initial clustering centers. These center points represent the center of each cluster;

[0034] Secondly, assign data points to the nearest clustering center. For each data point, calculate its distance from each clustering center and assign it to the cluster where the nearest clustering center is located;

[0035] Furthermore, update the clustering center: for each cluster, calculate the average value of all data points in the cluster and use this average value as the new clustering center;

[0036] Finally, repeat S2 and S3: until the clustering center no longer changes significantly or reaches the predetermined number of iterations.

[0037] As a preferred scheme of the regional power grid coordinated dispatching method considering partition aggregation of the present invention, wherein: the average output of a single aggregator, calculate the average output situation of the single aggregator within a cycle, which is expressed as:

[0038]

[0039] In the formula, P 平均 is the daily average output of a single aggregator; P t is the output of a single aggregator in the t-th time period; T is a cycle, taking 24h;

[0040] The volatility of the output of a single aggregator, use the volatility of the output of a single aggregator to describe the fluctuation level of the output of a single aggregator, which is expressed as:

[0041]

[0042] In the formula, P 波动 is the daily single-aggregator output volatility;

[0043] The single-aggregator output distribution, which can be used to describe the skewness of the single-aggregator output; it is expressed as:

[0044]

[0045] In the formula, SK is the skewness of the single-aggregator output distribution.

[0046] As a preferred embodiment of the regional power grid coordinated dispatching method considering partition aggregation of the present invention, wherein: the regional partitioning method specifically includes:

[0047] S1. Comprehensively considering the energy supply capacity, distribution location, and regional function of the region, initially divide it into K regions, and the kth energy supply region is denoted as M k , which is expressed as:

[0048] M k = [X k , 电源 ; X k , 储能 ; X k , 柔性负荷

[0049] S2. Make a detailed division of the region M k , initialize each region, the number of blocks N i after the detailed division, and determine the objective function and constraints;

[0050] S3. Make an equal-area detailed division of the region M i according to the objective function, and judge whether the constraints are satisfied;

[0051] S4. If satisfied, N i = N i + 1, then return to S2 for further division; if not satisfied, N i = N i - 1, and the corresponding division result is the final detailed division result of the region M i ;

[0052] S5. Repeat the iteration of fitness, non-dominated solution selection, and speed-position update until the maximum number of iterations is reached.

[0053] As a preferred embodiment of the regional power grid coordinated dispatching method considering partition aggregation of the present invention, wherein: the objective function for the regional division of distributed resources includes the daily average cost, daily load characteristics, and regional aggregation degree; ​

[0054] Among them, the daily average cost includes:

[0055] The output of each time period composite aggregate tracks the planned output, while considering the power generation cost and the charging cost.

[0056] The penalty cost is used to represent the degree to which the actual output of the composite aggregate deviates from the planned output.

[0057]

[0058] In the formula, C 电源 is the power generation cost of the power source; C 储能 is the charging cost of the energy storage; C 惩罚 is the penalty cost paid by the composite aggregate to the power grid; is the power generation power of the power source in area i at time period k; X i 电源 indicates whether there is a power source in area i; n i is the daily average charging times of the energy storage; is the energy storage charging power in area i at time period k; indicates whether there is an energy storage device in area i; is the flexible load power in area i at time period k; C S is the unit penalty cost; P k0 is the planned output at time period k;

[0059] The daily load characteristic is represented by the curve W, which is the difference between the actual output and the planned output of the composite aggregate. Then, the quality of the daily load characteristic can be measured by the volatility of the curve L. The smoother the curve W is, the smoother the composite curve is, which is more conducive to power grid dispatching. It is expressed as:

[0060]

[0061] In the formula, f 2 is the volatility of the curve W; σ is the standard deviation; μ is the geometric mean; W k is the difference between the actual output and the planned output of the composite aggregate at time period k; is the arithmetic mean of the output power;

[0062] The regional aggregation degree is considered when selecting the resources of the region, taking into account the spatial location and load density of the region. The spatial distance is expressed as:

[0063] d(i,j) = ω p ‖p i -p j ‖ + ω ρ ‖ρ i -ρ j ‖

[0064] In the formula, ωp and ω ρ are the weights of the spatial position attribute and the load density attribute respectively, and the sum of the two is 1, p i and p j are the position coordinates of regions i and j; ρ i and ρ j are the load densities of regions i and j;

[0065] The degree of regional aggregation is measured by the sum of the maximum fusion spatial distances of m regions, which is expressed as:

[0066]

[0067] In the formula, X i and X h represent whether there are resources in regions i and h participating in the construction of the composite aggregate. If so, it is 1, otherwise it is 0; d(i, h) represents the fusion spatial distance between regions i and h.

[0068] As a preferred solution of the regional power grid coordinated dispatching method considering partition aggregation in the present invention, among them: the constraint conditions for the regional division of distributed resources include composite aggregate scale constraint, spatial distance constraint, composite aggregate output constraint, and resource complementarity constraint;

[0069] Among them, the composite aggregate scale constraint aggregates various resources such as distributed power sources, energy storage devices, and flexible loads into a whole; the composite aggregate scale constraint is expressed as:

[0070]

[0071] In the formula, Mmin is the minimum number of resources, and Qmin is the minimum power consumption in period k;

[0072] The spatial distance constraint is that there is a spatial distance between any two regions, and the spatial distance should be controlled within a certain range to facilitate the dispatching of the composite aggregate, which is expressed as: maxd(i, j) ≤ d 最大

[0073] In the formula, dmax is the maximum spatial distance;

[0074] The resource complementarity constraint considers the complementarity between various resources during the construction of the composite aggregate; the correlation degree between various resources is expressed as:

[0075]

[0076] In the formula, Λ(k) is the weighting function for constructing the active power at different time points; σ xk is the standard deviation of the resource sequence x(k); σ ykis the standard deviation of the resource sequence y(k);

[0077] Then the resource complementary coefficient γ between resources x and y xy is expressed as:

[0078] γ xy = 1 - β xy .

[0079] As a preferred embodiment of the regional power grid coordinated dispatching method considering partition aggregation of the present invention, wherein: the inter-region coordinated optimization method includes establishing a joint dispatching optimization model for a multi-compound aggregation system, and the optimization of multiple aggregations takes the minimum total cost of each aggregation as the objective function;

[0080] Among them, the objective function of the coordinated optimization method is expressed as:

[0081]

[0082] In the formula, C i is the total cost of the aggregation; is the power generation cost; is the power exchange cost; is the default cost; and are the charging and discharging powers of the energy storage h at time t; is the output of the flexible load q at time t; H i and Q i are the numbers of energy storage and flexible loads in the aggregation respectively; ρ d , ρ m , ρ s , ρ q are the unit power generation costs of gas turbines, renewable energy, energy storage and flexible loads respectively; P i,j,t is the interaction power between aggregations i and j at time t; z i,j is the distance between aggregations i and j; ρ 交换 and ρ 违约 are the unit interaction cost and unit default cost respectively;

[0083] Among them, the constraint conditions of the coordinated optimization method include system power constraints, gas turbine constraints and energy storage constraints;

[0084] The system power constraint is expressed as:

[0085]

[0086] The gas turbine constraint is expressed as:

[0087]

[0088] In the formula, and are the minimum and maximum output powers of the gas turbine d, respectively

[0089] The energy storage constraint is expressed as:

[0090]

[0091] In the formula, and are the minimum and maximum charging powers of the energy storage h, respectively; and are the minimum and maximum discharging powers of the energy storage h, respectively;

[0092] The multi-objective particle swarm optimization algorithm is used to solve the joint scheduling optimization model of multiple said composite aggregate systems, including:

[0093] S1. Initialize the population: Randomly generate a group of particles, and each particle has an initial position and velocity;

[0094] S2. Evaluate the fitness of particles: Calculate the fitness of each particle on all objective functions, and update the best position and best fitness of the particle;

[0095] Among them, the velocity of each particle on all objective functions is expressed as:

[0096]

[0097] The position of each particle on all objective functions is expressed as:

[0098] X k+1 = X k + V k+1

[0099] In the formula, ω is the inertia weight; rand 1 and rand 2 are random numbers distributed in the interval [0, 1]; k is the current iteration number; is the position of the individual optimal particle; is the position of the global optimal particle; C 1 and C 2 are constants; V is the particle velocity; X is the particle position;

[0100] S3. Non-dominated solution selection: Use the dominance relationship to judge the superiority and inferiority between particles, and find a group of non-dominated solutions to form the Pareto front;

[0101] S4. Update the velocity and position. According to the best position of the particle and the global best position, update the velocity of the particle and adjust the position of the particle.

[0102] The present invention also provides the following technical solution, including a computer device, comprising a memory and a processor, the memory storing a computer program, wherein when the processor executes the computer program, the steps of a regional power grid coordinated dispatching method considering partition aggregation are implemented.

[0103] The present invention also provides the following technical solution, including a computer-readable storage medium, on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of a regional power grid coordinated dispatching method considering partition aggregation are implemented.

[0104] The beneficial effects of the present invention are as follows: According to the resource distribution and operation status of the regional power grid, aggregating resources can effectively reduce the difficulty of power grid dispatching, improve the utilization rate of power grid resources, meet the requirements of the current new power system, and have advantages such as high efficiency, flexibility, and economy compared with traditional dispatching methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings according to these drawings without creative efforts.

[0106] Figure 1 It is a flowchart of the regional power grid coordinated dispatching method considering partition aggregation in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0107] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification.

[0108] Many specific details are set forth in the following description in order to fully understand the present invention, but the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0109] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0110] Embodiment 1, referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a regional power grid coordinated dispatching method considering partition aggregation, including:

[0111] S1: Collect regional electrical parameters, including the power of power generation equipment and electrical equipment.

[0112] S2: For centralized resources, establish a single resource aggregation model and aggregate using the k-means method.

[0113] S3: For distributed resources, establish a distributed resource aggregation model and partition and aggregate the regional distributed resources.

[0114] S4: Through an optimization algorithm, based on the set objective function and constraint conditions, achieve coordination between regions.

[0115] As a preferred embodiment, the above step S1 is specifically as follows:

[0116] Collect data such as power generation of power generation equipment and power consumption of users, such as data of thermal power plants, centralized / distributed new energy power plants, residential electricity, and industrial electricity. Statistically analyze existing large power generation / consumption units (such as large power plants, photovoltaic / wind power clusters, large factory electricity consumption, etc.), and statistically analyze distributed resources (such as electric vehicles, flexible loads, distributed photovoltaics, etc.).

[0117] Aggregate centralized large power plants and photovoltaic / wind power clusters as power generation units, aggregate centralized factory electricity consumption and residential electricity clusters as electricity consumption units, and aggregate composite power generation / consumption clusters into aggregates such as virtual power plants and microgrids.

[0118] Example 2, refer to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a preferred solution for the coordinated dispatching method of the regional power grid considering partition aggregation.

[0119] Step S2 is specifically as follows:

[0120] For a centralized single-type aggregate (such as a single centralized photovoltaic / wind power plant, etc.), taking photovoltaic as an example, use the k-means method to aggregate a large number of photovoltaic devices.

[0121] Step 1: Propose a photovoltaic power output aggregation index and establish a photovoltaic aggregation model.

[0122] (1) Average photovoltaic power output, calculate the average photovoltaic power output situation within a cycle.

[0123]

[0124] In the formula, P 平均 is the daily average photovoltaic power output; P t is the photovoltaic power output in the t-th time period; T is a cycle, taking 24h.

[0125] (2) Photovoltaic output volatility, which is used to describe the fluctuation level of photovoltaic output.

[0126]

[0127] Wherein, P 波动 is the daily photovoltaic output volatility.

[0128] (3) Photovoltaic output distribution.

[0129] The photovoltaic output distribution can be used to describe the skewness of photovoltaic output.

[0130]

[0131] Wherein, SK is the skewness of photovoltaic output distribution.

[0132] Step 2: Normalize the indicators and distinguish positive indicators and negative indicators (the larger the value of the positive indicator, the better; the smaller the value of the negative indicator, the better). Among them, the average photovoltaic output is a positive indicator, and the photovoltaic output volatility and the skewness of photovoltaic output distribution are negative indicators.

[0133]

[0134] Wherein: are the normalized positive and negative indicators respectively; X i is the original indicator value; X i,最大 , X i,最小 are the maximum and minimum values in the original indicator values respectively.

[0135] Step 3: Determine the number of clusters

[0136] The elbow method is used to determine the number of clusters. Calculate the indicator SSE and draw it into a curve. The point where the SSE decline rate in the curve decreases significantly, that is, the point where the curve appears at the elbow, and the corresponding K value is the optimal number of clusters. The SSE calculation formula is as follows:

[0137]

[0138] Wherein, SSE is the sum of squared errors of clustering; K is the total number of clustering clusters; C i is the i-th cluster after division; x is the normalized indicator value within C i ; μ i is the indicator value corresponding to the selected clustering center.

[0139] Step 4: The k-means clustering steps are as follows:

[0140] Step 4.1: Initialize the clustering center, and randomly select K data points as the initial clustering centers. These center points represent the centers of each cluster.

[0141] Step 4.2: Assign data points to the nearest cluster center. For each data point, calculate its distance from each cluster center and assign it to the cluster where the nearest cluster center is located.

[0142] Step 4.3: Update the cluster centers: For each cluster, calculate the average value of all data points in the cluster and use this average value as the new cluster center.

[0143] Step 4.4: Repeat Steps 2 and 3: Repeat this process until the cluster centers no longer change significantly or a predetermined number of iterations is reached.

[0144] As a preferred embodiment, Step S3 above is specifically as follows:

[0145] Step 1: Considering the energy supply capacity, distribution location, and regional function of the region comprehensively, initially divide it into K regions, and the k-th region is denoted as M k

[0146] M k = [X k，电源 ; X k，储能 ; X k，柔性负荷

[0147] Step 2: Conduct a detailed division of region M k Initialize the number of blocks N for the detailed division of each region i , and determine the objective function and constraint conditions;

[0148] Step 3: Conduct an equal-area detailed division of region M i according to the objective function, and determine whether the constraint conditions are satisfied;

[0149] Step 4:, if satisfied, N i = N i + 1, and re-enter Step 2 for further division; if not satisfied, N i = N i - 1, and the corresponding division result is the final detailed division result of region M i .

[0150] The objective function of the region division is as follows

[0151] (1) Average daily cost

[0152] Track the planned output of the virtual power plant at each time period, and at the same time consider the power generation cost and charging cost. The penalty cost is used to represent the degree of deviation of the actual output of the virtual power plant from the planned output.

[0153]

[0154] In the formula, C 电源The power generation cost of the power source; C 储能 The charging cost of the energy storage; C 惩罚 The penalty cost paid by the virtual power plant to the power grid; The power generation power of the power source in area i at time period k; X i 电源 Whether there is a power source in area i; n i The average daily charging times of the energy storage; The charging power of the energy storage in area i at time period k; Whether there is an energy storage device in area i; The flexible load power in area i at time period k; C S The unit penalty cost; P k0 The scheduled output at time period k.

[0155] (2) Daily load characteristics

[0156] Use the curve W to represent the difference between the actual output and the scheduled output of the virtual power plant. Then, the quality of the daily load characteristics can be measured by the volatility of the curve L. The smoother the curve W, the smoother the composite curve, which is more conducive to power grid dispatching.

[0157]

[0158] In the formula, f 2 Is the volatility of the curve W; σ is the standard deviation; μ is the geometric mean; W k Is the difference between the actual output and the scheduled output of the virtual power plant at time period k; Is the arithmetic mean of the output power.

[0159] (3) Degree of regional aggregation

[0160] When selecting the resources of the region, it is necessary to consider the spatial location and load density of the region. The spatial distance is

[0161] d(i,j) = ω p ‖p i -p j ‖ + ω ρ ‖ρ i -ρ j ‖

[0162] In the formula, ω p And ω ρ Are the weights of the spatial location attribute and the load density attribute respectively, and the sum of the two is 1. p i And p j Are the position coordinates of area i and area j; ρ i And ρ j Are the load densities of area i and area j.

[0163] Use the sum of the maximum fusion space distances of m regions to measure the degree of regional aggregation

[0164]

[0165] Wherein, X i , X h represent whether there are resources in regions i and h participating in the construction of the virtual power plant. If so, it is 1; otherwise, it is 0; d(i, h) represents the fusion space distance between regions i and h.

[0166] Constraint conditions

[0167] (1) Virtual power plant scale constraint

[0168] The virtual power plant can aggregate various resources such as distributed power sources, energy storage devices, and flexible loads into a whole, enabling it to participate in the operation of the electricity market and the ancillary service market. The number of resources and the electricity consumption scale used to construct the virtual power plant should not be too small.

[0169]

[0170] Wherein, Mmin is the minimum number of resources, and Qmin is the minimum electricity consumption in period k

[0171] (2) Spatial distance constraint.

[0172] There is a spatial distance between any two regions, and the spatial distance should be controlled within a certain range to facilitate the dispatching of the virtual power plant.

[0173] max d(i, j) ≤ d 最大

[0174] Wherein, d 最大 is the maximum spatial distance

[0175] (3) Virtual power plant output constraint.

[0176]

[0177] Wherein, P k最小 and P k最大 are respectively the minimum and maximum values of the output of the virtual power plant in period k.

[0178] (4) Resource complementarity constraint.

[0179] When constructing the virtual power plant, the complementarity between various resources should be fully considered. The correlation degree between various resources is:

[0180]

[0181] Wherein, Λ(k) is the weighting function for constructing the active power at different time points; σxk is the standard deviation of the resource sequence x(k); σ yk is the standard deviation of the resource sequence y(k).

[0182] Then the resource complementarity coefficient between resources x and y is:

[0183] γ xy = 1 - β xy

[0184] Step 5: Repeat the iteration of fitness, non-dominated solution selection, and speed and position update until the maximum number of iterations is reached.

[0185] Example 3, referring to Figure 1 , is the first embodiment of the present invention, and this embodiment provides a preferred solution for a coordinated dispatching method of a regional power grid considering partition aggregation.

[0186] The above step S4 is specifically as follows:

[0187] Establish a joint dispatching optimization model for a multi-virtual power plant system, and the optimization of multiple aggregations takes the total cost of each aggregation as the objective function;

[0188]

[0189]

[0190] In the formula, C i is the total cost of the aggregation; is the power generation cost; is the power exchange cost; is the default cost; and are the charging and discharging powers of the energy storage h at time t; is the output of the flexible load q at time t; H i and Q i are the numbers of energy storage and flexible loads in the aggregation respectively; ρ d , ρ m , ρ s , ρ q are the unit power generation costs of gas turbines, renewable energy, energy storage, and flexible loads respectively; P i,j,t is the interaction power between aggregations i and j at time t; z i,j is the distance between aggregations i and j; ρ 交换 and ρ 违约 are the unit interaction cost and unit default cost respectively.

[0191] Constraint conditions

[0192] (1) System power constraint.

[0193]

[0194] (2) Gas turbine constraint.

[0195]

[0196] In the formula, and are the minimum and maximum power outputs of gas turbine d respectively.

[0197] (3) Energy storage constraint.

[0198]

[0199]

[0200] In the formula, and are the minimum and maximum charging powers of energy storage h respectively; and are the minimum and maximum discharging powers of energy storage h respectively.

[0201] The multi-objective particle swarm optimization algorithm is used to solve the above model, and the steps are as follows:

[0202] Step 1: Initialize the population: Randomly generate a group of particles, and each particle has an initial position and velocity.

[0203] Step 2: Evaluate the fitness of particles: Calculate the fitness of each particle on all objective functions, and update the best position and best fitness of the particle.

[0204]

[0205] X k+1 = X k + V k+1

[0206] In the formula, ω is the inertia weight; rand 1 and rand 2 are random numbers distributed in the interval [0, 1]; k is the current iteration number; is the position of the individual optimal particle; is the position of the global optimal particle; C 1 and C 2 are constants; V is the particle velocity; X is the particle position.

[0207] Step 3: Non-dominated solution selection: Use the domination relationship to judge the superiority and inferiority between particles, and find a set of non-dominated solutions to form the Pareto front.

[0208] Step 4: Update the velocity and position. Update the velocity of the particle and adjust the position of the particle according to the particle's best position and the global best position.

[0209] It is important to note that the construction and arrangement of the present application shown in multiple different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those skilled in the art who refer to this disclosure should readily understand that many modifications are possible without materially departing from the novel teachings and advantages of the subject matter described in this application (e.g., changes in the dimensions, scales, structures, shapes and proportions of various elements, as well as parameter values (e.g., temperature, pressure, etc.), installation arrangements, use of materials, colors, orientations, etc.). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of an element may be inverted or otherwise changed, and the nature or number or position of discrete elements may be altered or changed. Accordingly, all such modifications are intended to be included within the scope of the present invention. The order or sequence of any process or method steps may be altered or reordered according to alternative embodiments. In the claims, any clause of "means-plus-function" is intended to cover the structures that perform the recited function herein, and not only structural equivalents but also equivalent structures. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to a particular embodiment, but extends to various modifications that still fall within the scope of the appended claims.

[0210] In addition, in order to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the currently contemplated best mode of carrying out the present invention or those features that are not relevant to implementing the present invention).

[0211] It should be understood that in the development of any actual implementation, as in any engineering or design project, numerous specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without undue experimentation, such development efforts will be a routine task of design, manufacture and production.

[0212] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. A regional power grid coordinated dispatching method considering partition aggregation, characterized by: include, Collect resources from power generation and power consumption areas; Classify resources in power generation and power consumption areas; For concentrated resources, a single resource aggregation model is established, and the k-means method is used for aggregation; For distributed resources, a distributed resource aggregation model is established to aggregate regional distributed resources by region; Through optimization algorithms, coordination among regions is achieved based on objective functions and constraints set for different resource types.

2. The method for coordinated dispatching of regional power grids considering partition aggregation as claimed in claim 1, characterized in that: The collection resources include statistical centralized resources and statistical distributed resources; Centralized power generation clusters are aggregated as power generation units; Centralized electricity consumption clusters are aggregated as electricity consumption units; Centralized generation / consumption clusters are aggregated as composite aggregates.

3. The method for coordinated dispatching of regional power grids considering partition aggregation as claimed in claim 1, characterized in that: The single aggregate is aggregated using the k-means method, including: S1. Propose the aggregation index of single aggregate power generation output and establish a single aggregate aggregation model; The establishment of the single resource aggregation model includes the average single aggregate output, the single aggregate output fluctuation rate and the single aggregate output distribution; S2. Normalize the indicators and distinguish between positive and negative indicators; Among them, the average single aggregate output is a positive indicator, and the single aggregate output volatility and the single aggregate output distribution skewness are negative indicators; Where: is the normalized positive indicator, is the normalized negative indicator; X i is the original index value; X i,最大 , X i,最小 They are the maximum and minimum values ​​of the original indicator values ​​respectively; S3, determine the number of clusters; The indicator SSE (sum of the squared errors) is calculated and plotted into a curve. The rate of decrease of SSE in the curve is significantly reduced, that is, the point where the curve appears at the elbow, and the corresponding K value is the optimal number of clusters; the SSE calculation formula is as follows: In the formula, SSE is the sum of square errors of clustering; K is the sum of the number of clusters; C i is the i-th cluster after partition; x is C i The index value after internal normalization; μ i is the index value corresponding to the selected cluster center; S4, clustering using k-means; It includes first initializing the cluster center and randomly selecting K data points as the initial cluster center; these center points represent the center of each cluster; Secondly, assign data points to the nearest cluster center. For each data point, calculate its distance to each cluster center and assign it to the cluster with the nearest cluster center. Then update the cluster center: for each cluster, calculate the average value of all data points in the cluster and use this average value as the new cluster center; Finally, repeat S2 and S3 until the cluster center no longer changes significantly or the predetermined number of iterations is reached.

4. The method for coordinated dispatching of regional power grids considering partition aggregation as claimed in claim 3, characterized in that: The average single aggregate output is calculated within a cycle and is expressed as: Where P 平均 is the daily average output of a single aggregate; P t is the output of a single polymer in the tth time period; T is a period, which is 24 hours; The single aggregate output fluctuation rate is used to describe the fluctuation level of the single aggregate output, which is expressed as: Where P 波动 It is the daily output volatility of a single aggregate; A single aggregate output distribution, wherein the single aggregate output distribution can be used to describe the degree of skewness of the single aggregate output; It is expressed as: Where SK is the output distribution skewness of a single aggregate.

5. The method for coordinated dispatching of regional power grids considering zone aggregation as claimed in claim 4, characterized in that: The regional zoning method specifically includes: S1. Considering the energy supply capacity, distribution location and regional function of the region, it is initially divided into K blocks, and the kth energy supply area is recorded as M k , which is expressed as: M k =[X k , 电源 ;X k , 储能 ;X k , flexible load] S2, for area M k Perform detailed division, initialize each area, and divide the number of blocks N into i , determine the objective function and constraints; S3, according to the objective function of the region M i Perform detailed division of equal areas to determine whether the constraints are met; S4. If satisfied, N i =N i +1, then return to S2 for further division; if not satisfied, N i =N i -1, the corresponding division result is area M i The final detailed division result; S5. Repeat iterative fitness, non-dominated solution selection and velocity position update until the maximum number of iterations is reached.

6. The method for coordinated dispatching of regional power grids considering zone aggregation as claimed in claim 5, characterized in that: The objective function for the regional division of distributed resources includes daily average cost, daily load characteristics and regional aggregation degree; The average daily cost includes: The output of the composite aggregate in each period tracks the planned output, taking into account the cost of power generation and charging costs; the penalty cost is used to indicate the degree to which the actual output of the composite aggregate deviates from the planned output; In the formula, C 电源 is the power generation cost of the power source; C 储能 is the charging cost of energy storage; C 惩罚 penalty costs paid to the grid for composite aggregates; is the power generation power of region i in period k; Is there a power source in area i? i The average daily charging times for energy storage; is the energy storage charging power in area i during period k; is whether there is an energy storage device in area i; is the flexible load power in area i during period k; C S is the unit penalty cost; P k0 Contribute to the plan for time period k; The daily load characteristic is represented by curve W, which is the difference between the actual output and the planned output of the composite polymer. The quality of the daily load characteristic can be measured by the volatility of curve L. The more stable the curve W is, the more stable the composite curve is, which is more conducive to grid dispatching. It is expressed as: Where f2 is the volatility of curve W; σ is the standard deviation; μ is the geometric mean; W k is the difference between the actual output and the planned output of the composite aggregate in period k; is the arithmetic mean of the output power; The regional aggregation degree is to consider the spatial location and load density of the region when selecting the resources of the region. The spatial distance is expressed as: d(i,j)=ω p ‖p i -p j ‖+ω ρ ‖r i -r j ‖ In the formula, ω p and ω ρ are the weights of spatial location attribute and load density attribute, the sum of which is 1, p i and p j are the position coordinates of region i and region j; ρ i and ρ j is the load density of area i and area j; The sum of the maximum fusion space distances of m regions is used to measure the degree of regional aggregation, which is expressed as: Where, X i , X h It represents whether there are resources in regions i and h participating in the construction of the composite aggregate, which is 1 if yes and 0 otherwise; d(i,h) represents the fusion space distance between regions i and h.

7. The method for coordinated dispatching of regional power grids considering zone aggregation as claimed in claim 6, characterized in that: The constraints on the regional division of distributed resources include composite aggregate scale constraints, spatial distance constraints, composite aggregate output constraints, and resource complementarity constraints; The composite aggregate scale constraint aggregates multiple resources such as distributed power sources, energy storage devices and flexible load aggregation into a whole; the composite aggregate scale constraint is expressed as: In the formula, Mmin is the minimum number of resources, Qmin is the minimum power consumption in time period k; The spatial distance constraint means that there is a spatial distance between any two regions, and the spatial distance should be controlled within a certain range to facilitate the scheduling of the composite aggregate, which is expressed as: maxd(i,j)≤d 最大 Where, d 最大 is the maximum spatial distance; The resource complementarity constraint considers the complementarity between multiple resources when constructing a composite aggregate; the correlation between various resources is expressed as: Where Λ(k) is the weighted function for constructing active power at different time points; σ xk is the standard deviation of the resource sequence x(k); yk is the standard deviation of the resource sequence y(k); Then the resource complementarity coefficient γ between resources x and y is xy It is expressed as: c xy =1-β xy 。 8. The method for coordinated dispatching of regional power grids considering partition aggregation as claimed in claim 7, characterized in that: The inter-regional coordination optimization method includes establishing a multi-composite polymer system joint scheduling optimization model, and the optimization of the multi-polymer takes the optimal total cost of each polymer as the objective function; The objective function of the coordinated optimization method is expressed as: In the formula, C i is the total cost of the aggregate; For the cost of electricity generation; is the power exchange cost; is the cost of breach of contract; and is the charging and discharging power of energy storage h in period t; is the output of the flexible load q in period t; H i and Q i are the amount of energy storage and flexible load in the polymer, respectively; ρ d , m , s , q are the unit power generation costs of gas turbines, renewable energy, energy storage and flexible loads respectively; P i,j,t is the interaction power between aggregates i and j at time t; z i,j is the distance between aggregates i and j; ρ 交换 and ρ 违约 are the unit interaction cost and the unit breach cost respectively; The constraints of the coordinated optimization method include system power constraints, gas turbine constraints and energy storage constraints; The system power constraint is expressed as: The gas turbine constraints are expressed as: In the formula, and are the minimum and maximum output of gas turbine d, respectively; The energy storage constraint is expressed as: In the formula, and are the minimum and maximum charging powers of energy storage h, respectively; and are the minimum and maximum discharge powers of energy storage h, respectively; The multi-objective particle swarm algorithm is used to solve the joint scheduling optimization model of multiple composite polymer systems, including: S1. Initialize the population: randomly generate a group of particles, each particle has an initial position and velocity; S2. Evaluate particle fitness: calculate the fitness of each particle on all objective functions, and update the optimal position and optimal fitness of the particle; Among them, the speed of each particle on all objective functions is expressed as: The position of each particle on all objective functions is expressed as: X k+1 =X k +V k+1 Where ω is the inertia weight; rand1 and rand2 are random numbers distributed in the interval [0,1]; k is the current number of iterations; is the individual optimal particle position; is the global optimal particle position; C1 and C2 are constants; V is the particle velocity; X is the particle position; S3, non-dominated solution selection: Use the dominance relationship to judge the pros and cons of particles and find a set of non-dominated solutions to form the Pareto frontier; S4, update speed and position: according to the particle's optimal position and the global optimal position, update the particle's speed and adjust the particle's position.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a method for collaborative optimization of distributed power resources as described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for collaborative optimization of distributed power resources as described in any one of claims 1 to 9 are implemented.

Images