Method for depicting flexible power supply capability of power grid

Abstract

The invention relates to a method for describing the flexible power supply capacity of a power grid, and belongs to the technical field of power grids. The method comprises the following steps: firstly, acquiring daily load data at each feeder outlet from a power utilization information acquisition system of the power distribution network, and preprocessing the acquired data; then respectively establishing a load power evaluation model, an active power distribution network power supply capability probability evaluation model, a power grid maximum power supply capability evaluation model considering section coupling and an evaluation model considering multiple time scales; and then solving the model, and scientifically describing the power supply capability state of the power grid by using the solved total power supply capability and available power supply capability of the power grid. The method effectively reduces the calculation amount of complex power grid power supply capability evaluation, improves the evaluation efficiency, improves the evaluation accuracy, provides an important basis for operation decision making of a power system, and is easy to popularize and apply.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grids, and in particular relates to a method for characterizing the flexible power supply capability of a power grid. Background Art

[0002] The power system is an important infrastructure for the development of modern society. Its main task is to transmit electric energy from the power source to the load through transmission, transformation and distribution, and then the electrical equipment at the load end converts the electric energy into other forms of energy to serve social production and life. In the operation of the power system, in order to ensure the safety and stability of the power grid, it is necessary to accurately evaluate the power supply capacity of the power grid.

[0003] The power supply capacity of the power grid, referred to as power supply capacity, refers to the ability of the power grid to continuously, safely and stably supply power to users in accordance with the prescribed power supply quality within a certain period of time, which is manifested as the reliability of the power grid to supply power to each power consumption point. Due to the large scale and extremely complex power system, and the fact that the operation of the power system is affected by many factors (in addition to its own internal operating mechanism, it is also affected by the external environment), the evaluation of the power supply capacity of the power grid is a very complex task.

[0004] The existing power grid power supply capacity assessment method mainly uses power system simulation software to perform a large number of simulation calculations on the power system to obtain the operating results of the system under different states, thereby evaluating the power supply capacity of the power grid. Although this method can obtain relatively accurate results, it has a huge amount of calculation and requires a large amount of example setting and data processing. The calculation efficiency is low and it is difficult to meet the growing needs of power system analysis and decision-making.

[0005] Therefore, researching and developing a method that can accurately and efficiently evaluate the power supply capacity of the power grid is of great significance for power system operation decision-making, improving the power supply capacity and power supply quality of the power system, and meeting the electricity demand of power users. Summary of the invention

[0006] The purpose of the present invention is to solve the deficiencies of the prior art and to provide a method for characterizing the flexible power supply capability of a power grid.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] A method for characterizing the flexible power supply capability of a power grid comprises the following steps:

[0009] S1. Obtain the daily load data at each feeder outlet, the power generation and purchase declaration information under medium- and long-term transactions, and the operation data and basic data of the distribution network from the distribution network power consumption information collection system, and pre-process the collected daily load data;

[0010] S2. Establish a load power evaluation model, which takes the maximum load carried by the power grid as the objective function and the voltage, thermal limit and stability of the distribution network as the constraint conditions;

[0011] S3. Establish a probability assessment model for the power supply capacity of the active distribution network to assess the probability of the power supply capacity of the active distribution network; the probability assessment model for the power supply capacity of the active distribution network combines the repeated power flow method, the three-point estimation method, and the Cornish-Fisher series to comprehensively assess the power supply capacity of the active distribution network and calculate the power supply capacity distribution function of the active distribution network;

[0012] S4. Establish a maximum power supply capacity assessment model for the power grid considering section coupling. The maximum power supply capacity assessment model for the power grid considering section coupling takes the maximum power supply power increment, the maximum minimum section margin and the most balanced reserve retention as the objective function, and takes the partition power generation equal to the sum of the power generation of all types of power sources in the partition, and the partition power adjustment equal to the sum of the power adjustment of all types of power sources in the partition as the constraint conditions;

[0013] S5. Establish an evaluation model that considers multiple time scales. The evaluation model that considers multiple time scales adopts a two-layer model. The upper layer model takes the maximum social benefit as the objective function, and the obtained transaction volume is passed to the lower layer model; the lower layer model takes the maximum power supply capacity of the distribution network as the objective function;

[0014] S6. Based on the data obtained in step (1), the model constructed in S2 to S5 is solved to obtain the total power supply capacity and available power supply capacity of the power grid.

[0015] Further, preferably, in S1, the preprocessing is normalization processing.

[0016] Furthermore, preferably, the specific method for normalizing the daily load data is to use the maximum value of the daily load curve of the feeder as a normalization factor to convert the original load data into a dimensionless load value.

[0017] Further, preferably, in S2, the load power evaluation model established is as follows:

[0018]

[0019]

[0020] Where: P LSC represents the maximum power supply capacity of the active distribution network under the constraints; P pi represents the initial active load of node i; P ni represents the increase in active load of node i; H represents the load growth multiple; N represents the number of network nodes; V i Represents the voltage of node i; V imax 、Vimin are the maximum and minimum voltage of node i respectively; S ij represents the power passing through the branch between nodes i and j; S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i; P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i, G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; V j represents the voltage at node j.

[0021] Further, preferably, in S3, the method for establishing the probability evaluation model of the power supply capacity of the active distribution network is specifically as follows: if the evaluation object is an active distribution network containing k DGs, each DG needs to estimate 3 sample points for calculation; since the first k-order moments of the discrete distribution of the sample points constructed by all DGs in the distribution network are equal to the first k-order moments of the DG to be calculated, the probability density function is calculated by solving the first k-order moments of the discrete points, and the probability density function is shown in formula (19);

[0022] Construct the sample points corresponding to each DG in the distribution network, and calculate each component as follows:

[0023]

[0024] Where: μ k is the first-order origin moment of the sample point, σ is x k The standard deviation of x k,1 、x k,2 、x k,3 are the estimated 3 sample points of the kth DG, ξ k,1 , k,2 , k,3 are the position parameters corresponding to the three sample points;

[0025] The weight parameters and location parameters corresponding to each sample point satisfy the following equation:

[0026]

[0027] Where: k,l Ml (x k )and The ratio of the power of l (x k ) is the l-th order origin moment of the k-th DG in the distribution network; is the lth power of the standard deviation of the kth DG in the distribution network; ω k,i is the weight parameter of the i-th sample point in the k-th DG; n is the total number of DGs; is the lth power of the location parameter corresponding to the i-th sample point in the k-th DG;

[0028]

[0029] The specific formula of the constraint condition is as follows:

[0030] V imin ≤V i ≤V imax (12)

[0031]

[0032]

[0033] Where V i Represents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; I ij represents the branch current between nodes i and j, I ijmin and I ijmax are the maximum and minimum values ​​of the branch current between nodes i and j respectively; S ij represents the power passing through the branch between nodes i and j, S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i, P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i; G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; S DG(i) represents the power emitted by node i DG, is the maximum output power of DG at node i;

[0034] According to the mathematical properties of the semi-invariant, the first t-order origin moment of the power supply capacity statistic is calculated, that is, the t-order semi-invariant γ of the random variable is expressed by its t-order origin moment μ(γ j ) is obtained, the formula is as follows:

[0035]

[0036] Where: γ t+1 is a t+1-order semi-invariant; γ1 is a 1-order semi-invariant; u(γ) is the t-order origin moment of the random variable element; μ(γ j ) is the calculation result of the j-order origin moment; γ t＝j+1 For the j+1 order semi-invariant;

[0037] The Cornish-Fisher expansion formula is used to obtain the probability density of the random variable y(φ), where φ is the quantile of the power supply y, which is approximately expressed as:

[0038]

[0039] Where: γ3 and γ4 represent the third-order and fourth-order semi-invariants of the random variable y respectively; ξ(φ) represents the standard normal quantile function.

[0040] Further, preferably, in S4, the method for establishing the maximum power supply capacity evaluation model of the power grid considering section coupling is specifically as follows:

[0041] A maximum power supply capacity assessment model for the power grid considering section coupling is established. The model is described in units of zones. The decision variables are determined as the power adjustment amounts of various types of power sources in each zone of the provincial network, and the increase in power generation is stipulated as the positive direction. The value range of the decision variables must meet the following conditions:

[0042]

[0043] Where: Indicates the negative reserve of class S power supply in partition i, which is a positive number; ΔP i.s It represents the power adjustment of the class s power supply in the partition i; Indicates the positive and standby of the s-type power supply in zone i; i g.type represents the set of all power types in partition i; O represents the number of partitions involved in the adjustment;

[0044] At the same time, the following relationships are satisfied: the power generation of a partition is equal to the sum of the power generation of all types of power sources in the partition; the power adjustment of a partition is equal to the sum of the power adjustment of all types of power sources in the partition;

[0045]

[0046] Among them, P i represents the active power of node i; represents the set of s-type power sources belonging to node i; P i.s represents the active power of the s-type power supply at node i; ΔP i Represents the change in active power of node i; ΔP i.s represents the change in active power of the s-type power supply at node i;

[0047] The objective function is as follows:

[0048] (1) Calculate the maximum power supply increment of the large power grid at a certain moment. The formula is as follows:

[0049]

[0050] Among them, ΔP i represents the change in active power of node i; maxf1 represents the maximum power supply power increment; this formula represents the calculation of the maximum power adjustment of the large power grid at a certain moment based on the given large power grid unit startup mode, load and power generation power.

[0051] (2) Calculate the limits of each section when the large power grid reaches its maximum power supply capacity. The formula is as follows:

[0052]

[0053] in

[0054]

[0055] Where: maxf2 represents the maximum minimum section margin; M represents the section set; It represents the upper margin after adjustment of section j; It represents the lower margin after adjustment of section j; Indicates the upper limit of section power; P j Indicates the lower limit of section power; P j represents the power of section j after the partition power adjustment; ΔP j represents the power variation of section j; Indicates the power before section adjustment; S j-N S represents the row vector of section sensitivity of all partitions of the large power grid to section j; j-L Represents the row vector of section sensitivity of all the sections to be evaluated in the large power grid to section j; represents the column vector consisting of the ratio of the load of each partition to be evaluated to the total load of all partitions to be evaluated before adjustment; N1 represents the number of partitions to be evaluated; 1 1×Nrepresents a 1×N row vector with all 1s; ΔP N represents the column vector composed of the power adjustment of each partition; C represents the coefficient matrix of the coupling relationship between all the partitions to be evaluated in the large power grid;

[0056] (3) Calculate the ratio of the positive reserve capacity to be retained by each partition after the power adjustment of each type of power supply in each partition to the load of each partition. The ratio is required to be basically the same. The formula is as follows:

[0057]

[0058] Where: minf3 is the most balanced function for backup; α i represents the ratio of partition i’s positive standby to partition load; The adjusted positive and spare parts of partition i; Indicates the active and standby state of partition i before adjustment; is the load before adjustment for partition i; represents the average value of α.

[0059] Further, preferably, in S5, the method for establishing the evaluation model considering multiple time scales is specifically as follows:

[0060] The evaluation model considering multiple time scales includes upper-level model and lower-level model;

[0061] The upper model aims to maximize social benefits, takes the declared prices and declared quantities of power generators and power buyers as decision variables, considers load balance, unit output constraints, node voltage constraints and line flow constraints, and passes the obtained transaction volume to the lower model; the lower model aims to maximize the power supply capacity of the power grid, considers flow constraints, main transformer load transfer constraints, feeder load transfer constraints and main transformer load rate constraints.

[0062] Furthermore, preferably, the objective function of the upper model is:

[0063]

[0064] in,

[0065]

[0066] In the formula, U is the total number of load nodes participating in power trading; T is the user time period; z is the number of units; D u,t B is the estimated demand load of the power sales company at node u at time t; u,t (D u,t ) is the electricity purchase cost of the power sales company at node u at time t; P z,t is the output of unit z at time t; C z,t (P z,t) is the operating cost of unit z at time t; M is the penalty factor for network power flow constraint relaxation; are the forward and reverse power flow relaxation variables of line l respectively; N N The total number of sections quoted by the power sales company; D u,t,n is the estimated demand load of the nth segment at node u of the power sales company at time t; C u,t,n N is the purchase price corresponding to the nth segment number at node u at time t of the power sales company; M P is the total number of sections in the unit quotation; z,t,m is the predicted output of the mth stage of unit z at time t; C z,t,m is the operating price corresponding to the mth section of unit z at time t; N L is the total number of unit lines;

[0067] The constraints are as follows:

[0068] (1) System load balance equation constraint; its formula is:

[0069]

[0070] Where, T j,t is the exchange power between tie line j and the outside of the system at time t; N T is the total number of contact lines; K is the total number of nodes; is the load on node k that does not participate in market transactions at time t;

[0071] (2) Unit output constraint; the formula is:

[0072]

[0073] Where P z,t represents the power output generated by the zth unit at time t;

[0074] represents the minimum power output of the zth unit at time t;

[0075] represents the maximum power output of the zth unit at time t;

[0076] (3) Node voltage constraint; its formula is:

[0077] V kmin ≤V k ≤V kmax (30)

[0078] Where V kmin represents the minimum voltage limit of the kth node; V kmax represents the maximum voltage limit of the kth node; V k is the voltage of the kth node;

[0079] (4) Line flow constraint; its formula is:

[0080]

[0081] Where P l max is the maximum value of power flow transmission of line l; G l-i is the generator output power transfer distribution factor of the node where unit i is located to line l; G l-j is the generator output power transfer distribution factor of the node where the tie line j is located to the line l; G l-k is the generator output power transfer distribution factor of node k to line l; u∈k is the power sales company at node k.

[0082] Furthermore, preferably, the objective function of the lower model is:

[0083]

[0084] in:

[0085] D u,t =D u,t,z +D u,t,r +D u,t,s

[0086]

[0087] In the formula, k' is the load growth multiple; Y k Y is the load growth base at node k; A ASC D is the available power supply capacity of the distribution network; u,t,z , D u,t,r , D u,t,s are the loads of medium- and long-term transactions, day-ahead transactions, and real-time transactions at node u at time t; T TSC Represents the total power supply capacity of the distribution network, that is, the total power supply of all nodes; S k,t represents the power supply of node k at time t; It is the total load of transactions in the medium and long term market; is the total load traded in the day-ahead market; is the total load traded in the live market;

[0088] The constraints are as follows:

[0089] (1) Power flow constraint; its formula is:

[0090] Ai′=I′(35)

[0091] Where A is the node-branch association matrix; i′ is the complex current vector of all branches; I′ is the complex current injection vector of all nodes;

[0092] (2) Main transformer load transfer belt constraint; the formula is:

[0093]

[0094] In the formula, t ij B is the load transferred to main transformer j when main transformer i fails; j is the load on the main transformer j; is the rated capacity of the main transformer j;

[0095] (3) Feeder load transfer constraint; the formula is:

[0096] l mn +F n ≤R Fn (37)

[0097] In the formula, l mn is the load transferred to feeder n when feeder m fails; F is the load on feeder n; R Fn is the rated capacity of feeder n.

[0098] (4) Main transformer load factor constraint; the formula is:

[0099] B imin ≤B i ≤B imax (38)

[0100] In the formula, B i The load factor of the main transformer i, B imin The minimum load factor of the main transformer i, B imax is the maximum load rate of the main transformer i;

[0101] Furthermore, preferably, in order to describe the load balancing, the load balancing degree is defined in combination with the variance for:

[0102]

[0103] In the formula, The average load factor of the main transformer; N B The number of main variables.

[0104] In step (1) of the present invention, in addition to acquiring the daily load data at the outlet of each feeder and the power generation and purchase declaration information under medium- and long-term transactions, the collected data also needs to acquire the following data, including but not limited to: the voltage, resistance, inductance, and capacitance parameters of the distribution network; repeated power flow data, including the power flow on each feeder and the load distribution after power dispatch; three-point estimation data, used to evaluate the power supply capacity of the distribution network, including the reasonable power supply capacity of the main line and branch line, and resource utilization rate;

[0105] Cornish-Fisher series data, including probability indicators and statistical distribution for analyzing power supply capacity; fracture coupling data of distribution network, including voltage and current data and coupling relationship data of each cross-linked section; multi-time scale data, according to the different time frequencies of power supply, collect power supply data of different time scales, such as power consumption data and power supply data of different time scales of year, season, month, day and hour. All these data will be used in the model construction and solution process in the subsequent steps, providing a basis for theoretical research and practical application of power supply capacity of power grid flexibility.

[0106] Formula (18) of the present invention is used to calculate the first t-order origin moments of the power supply capacity statistics, and the origin moments correspond to the previous formulas (9)-(10). Formula (19) uses the Cornish-Fisher expansion formula to solve the probability density function of the random variable (power supply). The purpose of this step is to obtain the power supply capacity distribution function of the active distribution network; y(φ) here corresponds to the first k-order moments of the DG to be calculated in the previous formulas (9)-(10).

[0107] The calculation process of formula (19) belongs to the application of the Cornish-Fisher expansion formula, which is the result obtained by further using the first-order, second-order, third-order and fourth-order semi-invariants of the random variables calculated previously (calculated by formula (18) in the previous step); this formula is used to calculate the quantile y(φ) of the power supply capacity, which is a comprehensive application and output of all the previous calculation steps. Its calculation results can be used to evaluate the power supply capacity of the active distribution network at different probability levels.

[0108] The present invention needs to take into account both system stability and economic benefits, and provides a method for characterizing the flexible power supply capability of a power grid based on multiple time scales, which solves the problems of traditional evaluation methods with large computational complexity, low efficiency, and difficulty in meeting the needs of power system operation decision-making.

[0109] In the present invention, the distribution network power consumption information collection system is an existing system.

[0110] The present invention obtains the total power supply capacity and available power supply capacity of the power grid, and its functions are:

[0111] (1) Optimization of grid operation: Understanding the total power supply capacity and available power supply capacity of the grid provides a scientific basis for grid operation and helps the grid adopt scientific dispatching strategies, such as rationally allocating various power sources, avoiding power shortages, and reducing power outages.

[0112] (2) Providing a basis for grid planning and capacity expansion: By characterizing the total power supply capacity and available power supply capacity of the grid, we can more accurately understand the current operating status and development trend of the grid, provide a basis for grid planning and capacity expansion, and contribute to the healthy and steady development of the power system.

[0113] The present invention can form a method for realizing a multi-dimensional, dynamic and comprehensive characterization of the power supply capacity of the power grid through a load assessment model, a probability assessment of the power supply capacity of the power grid, a maximum power supply capacity assessment model and a multi-time scale assessment model. This method can realize the optimal operation strategy of the distribution network under various actual constraints by controlling various constraints and optimizing at different time scales. The characterization described in the present invention is that through this series of models and algorithms, the power supply capacity of the power grid can be described and analyzed in an all-round and multi-angle manner, and the abstract concept of the power supply capacity of the power grid can be quantified to form specific indicators that can be measured and compared, thereby providing reference information for the planning and design, operation control and optimization and improvement of the power grid.

[0114] Compared with the prior art, the present invention has the following beneficial effects:

[0115] 1. Improved evaluation efficiency: Since the present invention adopts a multi-time scale evaluation model and carries out detailed design and optimization for the upper and lower models respectively, it can effectively reduce the computational complexity of complex power grid power supply capacity evaluation and improve evaluation efficiency.

[0116] 2. Improved evaluation accuracy: The present invention takes into account multiple practical factors such as flow constraints, voltage constraints, unit output, etc. in the characterization of the power supply capacity of the power grid, which can more accurately reflect the actual operating status of the power grid, thereby improving the evaluation accuracy.

[0117] 3. Provides decision-making basis: The power grid power supply capacity evaluation model of the present invention can provide an important basis for power system operation decision-making, which is conducive to improving the power supply capacity and power supply quality of the power system and meeting the power demand of power users.

[0118] 4. Dealing with the complexity of modern power systems: In a large power grid environment, due to the large scale and extreme complexity of the system, there are many factors that affect the power supply capacity. The multi-time scale evaluation model of the present invention can better deal with this problem, comprehensively consider various factors, and obtain more comprehensive and accurate results. BRIEF DESCRIPTION OF THE DRAWINGS

[0119] Figure 1A flow chart of a method for characterizing the flexible power supply capability of a power grid according to the present invention;

[0120] Figure 2 A flow chart for calculating the power supply capacity of the power distribution network of the present invention;

[0121] Figure 3 This is a maximum power distribution calculation flow chart of the present invention;

[0122] Figure 4 The present invention adopts the primal dual interior point method and the improved repeated power flow method to solve the algorithm flow chart of the model. DETAILED DESCRIPTION

[0123] The present invention is further described in detail below in conjunction with embodiments.

[0124] Those skilled in the art will appreciate that the following examples are only used to illustrate the present invention and should not be considered to limit the scope of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or the product specifications are used. If the manufacturer of the materials or equipment used is not specified, they are all conventional products that can be purchased.

[0125] Example 1

[0126] A method for characterizing the flexible power supply capability of a power grid, characterized by comprising the following steps:

[0127] S1. Obtain the daily load data at each feeder outlet, the power generation and purchase declaration information under medium- and long-term transactions, and the operation data and basic data of the distribution network from the distribution network power consumption information collection system, and pre-process the collected daily load data;

[0128] S2. Establish a load power evaluation model, which takes the maximum load carried by the power grid as the objective function and the voltage, thermal limit and stability of the distribution network as the constraint conditions;

[0129] S3. Establish a probability assessment model for the power supply capacity of the active distribution network to assess the probability of the power supply capacity of the active distribution network; the probability assessment model for the power supply capacity of the active distribution network combines the repeated power flow method, the three-point estimation method, and the Cornish-Fisher series to comprehensively assess the power supply capacity of the active distribution network and calculate the power supply capacity distribution function of the active distribution network;

[0130] S4. Establish a maximum power supply capacity assessment model for the power grid considering section coupling. The maximum power supply capacity assessment model for the power grid considering section coupling takes the maximum power supply power increment, the maximum minimum section margin and the most balanced reserve retention as the objective function, and takes the partition power generation equal to the sum of the power generation of all types of power sources in the partition, and the partition power adjustment equal to the sum of the power adjustment of all types of power sources in the partition as the constraint conditions;

[0131] S5. Establish an evaluation model that considers multiple time scales. The evaluation model that considers multiple time scales adopts a two-layer model. The upper layer model takes the maximum social benefit as the objective function, and the obtained transaction volume is passed to the lower layer model; the lower layer model takes the maximum power supply capacity of the distribution network as the objective function;

[0132] S6. Based on the data obtained in step (1), the model constructed in S2 to S5 is solved to obtain the total power supply capacity and available power supply capacity of the power grid.

[0133] Example 2

[0134] A method for characterizing the flexible power supply capability of a power grid comprises the following steps:

[0135] S1. Obtain the daily load data at each feeder outlet, the power generation and purchase declaration information under medium- and long-term transactions, and the operation data and basic data of the distribution network from the distribution network power consumption information collection system, and pre-process the collected daily load data;

[0136] S2. Establish a load power evaluation model, which takes the maximum load carried by the power grid as the objective function and the voltage, thermal limit and stability of the distribution network as the constraint conditions;

[0137] S3. Establish a probability assessment model for the power supply capacity of the active distribution network to assess the probability of the power supply capacity of the active distribution network; the probability assessment model for the power supply capacity of the active distribution network combines the repeated power flow method, the three-point estimation method, and the Cornish-Fisher series to comprehensively assess the power supply capacity of the active distribution network and calculate the power supply capacity distribution function of the active distribution network;

[0138] S4. Establish a maximum power supply capacity assessment model for the power grid considering section coupling. The maximum power supply capacity assessment model for the power grid considering section coupling takes the maximum power supply power increment, the maximum minimum section margin and the most balanced reserve retention as the objective function, and takes the partition power generation equal to the sum of the power generation of all types of power sources in the partition, and the partition power adjustment equal to the sum of the power adjustment of all types of power sources in the partition as the constraint conditions;

[0139] S5. Establish an evaluation model that considers multiple time scales. The evaluation model that considers multiple time scales adopts a two-layer model. The upper layer model takes the maximum social benefit as the objective function, and the obtained transaction volume is passed to the lower layer model; the lower layer model takes the maximum power supply capacity of the distribution network as the objective function;

[0140] S6. Based on the data obtained in step (1), the model constructed in S2 to S5 is solved to obtain the total power supply capacity and available power supply capacity of the power grid.

[0141] In S1, the preprocessing is normalization processing.

[0142] The specific method for normalizing daily load data is to use the maximum value of the feeder daily load curve as the normalization factor to convert the original load data into a dimensionless load value.

[0143] In S2, the load power evaluation model established is as follows:

[0144]

[0145]

[0146] Where: P LSC represents the maximum power supply capacity of the active distribution network under the constraints; P pi represents the initial active load of node i; P ni represents the increase in active load of node i; H represents the load growth multiple; N represents the number of network nodes; V i Represents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; S ij represents the power passing through the branch between nodes i and j; S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i; P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i, G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; V j represents the voltage at node j.

[0147] In S3, the establishment method of the probability evaluation model of the power supply capacity of the active distribution network is as follows: if the evaluation object is an active distribution network containing k DGs, each DG needs to estimate 3 sample points for calculation; since the first k-order moments of the discrete distribution of the sample points constructed by all DGs in the distribution network are equal to the first k-order moments of the DG to be calculated, the probability density function is calculated by solving the first k-order moments of the discrete points. The probability density function is shown in formula (19);

[0148] Construct the sample points corresponding to each DG in the distribution network, and calculate each component as follows:

[0149]

[0150] Where: μ k is the first-order origin moment of the sample point, σ is x k The standard deviation of x k,1 、x k,2 、x k,3 are the estimated 3 sample points of the kth DG, ξ k,1 , k,2 , k,3 are the position parameters corresponding to the three sample points;

[0151] The weight parameters and location parameters corresponding to each sample point satisfy the following equation:

[0152]

[0153] Where: k,l M l (x k )and The ratio of the power of l (x k ) is the l-th order origin moment of the k-th DG in the distribution network; is the lth power of the standard deviation of the kth DG in the distribution network; ω k,i is the weight parameter of the i-th sample point in the k-th DG; n is the total number of DGs; is the lth power of the location parameter corresponding to the i-th sample point in the k-th DG;

[0154]

[0155] The specific formula of the constraint condition is as follows:

[0156] V imin ≤V i ≤V imax (12)

[0157]

[0158] Where V i Represents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; I ij represents the branch current between nodes i and j, and are the maximum and minimum values ​​of the branch current between nodes i and j respectively; S ij represents the power passing through the branch between nodes i and j, S ijmaxis the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i, P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i; G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; S DG(i) represents the power emitted by node i DG, is the maximum output power of DG at node i;

[0159] According to the mathematical properties of the semi-invariant, the first t-order origin moment of the power supply capacity statistic is calculated, that is, the t-order semi-invariant γ of the random variable is expressed by its t-order origin moment μ(γ j ) is obtained, the formula is as follows:

[0160]

[0161] Where: γ t+1 is a t+1-order semi-invariant; γ1 is a 1-order semi-invariant; u(γ) is the t-order origin moment of the random variable element; μ(γ j ) is the calculation result of the j-order origin moment; γ t＝j+1 For the j+1 order semi-invariant;

[0162] The Cornish-Fisher expansion formula is used to obtain the probability density of the random variable y(φ), where φ is the quantile of the power supply y, which is approximately expressed as:

[0163]

[0164] Where: γ3 and γ4 represent the third-order and fourth-order semi-invariants of the random variable y respectively; ξ(φ) represents the standard normal quantile function.

[0165] In S4, the method for establishing the maximum power supply capacity evaluation model of the power grid considering section coupling is as follows:

[0166] A maximum power supply capacity assessment model for the power grid considering section coupling is established. The model is described in units of zones. The decision variables are determined as the power adjustment amounts of various types of power sources in each zone of the provincial network, and the increase in power generation is stipulated as the positive direction. The value range of the decision variables must meet the following conditions:

[0167]

[0168] Where: Indicates the negative reserve of class S power supply in partition i, which is a positive number; ΔP i.s It represents the power adjustment of the class s power supply in the partition i; Indicates the positive and standby of the s-type power supply in zone i; i g.type represents the set of all power types in partition i; O represents the number of partitions involved in the adjustment;

[0169] At the same time, the following relationships are satisfied: the power generation of a partition is equal to the sum of the power generation of all types of power sources in the partition; the power adjustment of a partition is equal to the sum of the power adjustment of all types of power sources in the partition;

[0170]

[0171] Among them, P i represents the active power of node i; represents the set of s-type power sources belonging to node i; P i.s represents the active power of the s-type power supply at node i; ΔP i Represents the change in active power of node i; ΔP i.s represents the change in active power of the s-type power supply at node i;

[0172] The objective function is as follows:

[0173] (1) Calculate the maximum power supply increment of the large power grid at a certain moment. The formula is as follows:

[0174]

[0175] Among them, ΔP i represents the change in active power of node i; maxf1 represents the maximum power supply power increment; this formula represents the calculation of the maximum power adjustment of the large power grid at a certain moment based on the given large power grid unit startup mode, load and power generation power.

[0176] (2) Calculate the limits of each section when the large power grid reaches its maximum power supply capacity. The formula is as follows:

[0177]

[0178] in

[0179]

[0180] Where: maxf2 represents the maximum minimum section margin; M represents the section set; It represents the upper margin after adjustment of section j; It represents the lower margin after adjustment of section j; Indicates the upper limit of section power; P j Indicates the lower limit of section power; P j represents the power of section j after the partition power adjustment; ΔP j represents the power variation of section j; Indicates the power before section adjustment; S j-N S represents the row vector of section sensitivity of all partitions of the large power grid to section j; j-L Represents the row vector of section sensitivity of all the sections to be evaluated in the large power grid to section j; represents the column vector consisting of the ratio of the load of each partition to be evaluated to the total load of all partitions to be evaluated before adjustment; N1 represents the number of partitions to be evaluated; 1 1×N represents a 1×N row vector with all 1s; ΔP N represents the column vector composed of the power adjustment of each partition; C represents the coefficient matrix of the coupling relationship between all the partitions to be evaluated in the large power grid;

[0181] (3) Calculate the ratio of the positive reserve capacity to be retained by each partition after the power adjustment of each type of power supply in each partition to the load of each partition. The ratio is required to be basically the same. The formula is as follows:

[0182]

[0183] Where: minf3 is the most balanced function for backup; α i represents the ratio of partition i’s positive standby to partition load; The adjusted positive and spare parts of partition i; Indicates the active and standby state of partition i before adjustment; is the load before adjustment for partition i; represents the average value of α.

[0184] In S5, the method for establishing the evaluation model considering multiple time scales is as follows:

[0185] The evaluation model considering multiple time scales includes upper-level model and lower-level model;

[0186] The upper model aims to maximize social benefits, takes the declared prices and declared quantities of power generators and power buyers as decision variables, considers load balance, unit output constraints, node voltage constraints and line flow constraints, and passes the obtained transaction volume to the lower model; the lower model aims to maximize the power supply capacity of the power grid, considers flow constraints, main transformer load transfer constraints, feeder load transfer constraints and main transformer load rate constraints.

[0187] The objective function of the upper model is:

[0188]

[0189] in,

[0190]

[0191] In the formula, U is the total number of load nodes participating in power trading; T is the user time period; z is the number of units; D u,t B is the estimated demand load of the power sales company at node u at time t; u,t (D u,t ) is the electricity purchase cost of the power sales company at node u at time t; P z,t is the output of unit z at time t; C z,t (P z,t ) is the operating cost of unit z at time t; M is the penalty factor for network power flow constraint relaxation; are the forward and reverse power flow relaxation variables of line l respectively; N N The total number of sections quoted by the power sales company; D u,t,n is the estimated demand load of the nth segment at node u of the power sales company at time t; C u,t,n N is the purchase price corresponding to the nth segment number at node u at time t of the power sales company; M P is the total number of sections in the unit quotation; z,t,m is the predicted output of the mth stage of unit z at time t; C z,t,m is the operating price corresponding to the mth section of unit z at time t; N L is the total number of unit lines;

[0192] The constraints are as follows:

[0193] (1) System load balance equation constraint; its formula is:

[0194]

[0195] Where, T j,t is the exchange power between tie line j and the outside of the system at time t; N T is the total number of contact lines; K is the total number of nodes; is the load on node k that does not participate in market transactions at time t;

[0196] (2) Unit output constraint; the formula is:

[0197]

[0198] Where P z,t represents the power output generated by the zth unit at time t;

[0199] represents the minimum power output of the zth unit at time t;

[0200] represents the maximum power output of the zth unit at time t;

[0201] (3) Node voltage constraint; its formula is:

[0202] V kmin ≤V k ≤V kmax (30)

[0203] Where V kmin represents the minimum voltage limit of the kth node; V kmax represents the maximum voltage limit of the kth node; V k is the voltage of the kth node;

[0204] (4) Line flow constraint; its formula is:

[0205]

[0206] Where P l max is the maximum value of power flow transmission of line l; G l-i is the generator output power transfer distribution factor of the node where unit i is located to line l; G l-j is the generator output power transfer distribution factor of the node where the tie line j is located to the line l; G l-k is the generator output power transfer distribution factor of node k to line l; u∈k is the power sales company at node k.

[0207] The objective function of the lower model is:

[0208]

[0209] in:

[0210] D u,t =D u,t,z +D u,t,r +D u,t,s

[0211]

[0212] In the formula, k' is the load growth multiple; Y k Y is the load growth base at node k; A ASC D is the available power supply capacity of the distribution network; u,t,z , D u,t,r , D u,t,s are the loads of medium- and long-term transactions, day-ahead transactions, and real-time transactions at node u at time t; T TSC Represents the total power supply capacity of the distribution network, that is, the total power supply of all nodes; S k,t represents the power supply of node k at time t; It is the total load of transactions in the medium and long term market; is the total load traded in the day-ahead market; is the total load traded in the live market;

[0213] The constraints are as follows:

[0214] (1) Power flow constraint; its formula is:

[0215] Ai′=I′ (35)

[0216] Where A is the node-branch association matrix; i′ is the complex current vector of all branches; I′ is the complex current injection vector of all nodes;

[0217] (2) Main transformer load transfer belt constraint; the formula is:

[0218]

[0219] In the formula, t ij B is the load transferred to main transformer j when main transformer i fails; j is the load on the main transformer j; is the rated capacity of the main transformer j;

[0220] (3) Feeder load transfer constraint; the formula is:

[0221] l mn +F n ≤R Fn (37)

[0222] In the formula, l mn is the load transferred to feeder n when feeder m fails; F is the load on feeder n; R Fn is the rated capacity of feeder n.

[0223] (4) Main transformer load factor constraint; the formula is:

[0224] B imin ≤B i ≤B imax (38)

[0225] In the formula, B i The load factor of the main transformer i, B imin The minimum load factor of the main transformer i, B imax is the maximum load rate of the main transformer i;

[0226] To describe the load balancing, the load balancing degree is defined in combination with the variance for:

[0227]

[0228] In the formula, The average load factor of the main transformer; N B The number of main variables.

[0229] Example 3

[0230] A method for characterizing the flexible power supply capability of a power grid comprises the following steps:

[0231] S1. Obtain the daily load data at each feeder outlet from the distribution network power consumption information collection system and collect the power generation and purchase declaration information under medium- and long-term transactions, and pre-process the collected data. The pre-processing uses normalization to convert the original load data into dimensionless values;

[0232] S2. Model establishment:

[0233] S201. Establishment of load power evaluation model:

[0234] The load power assessment (LSC) model is established to assess the maximum load that the distribution network can carry under a series of constraints. The model takes into account multiple factors such as the grid structure, current load, and load growth pattern. After the grid structure is determined, the maximum power delivered by the network (i.e. the maximum power supply capacity of the active distribution network) is calculated according to the size and growth pattern of the load at each node by satisfying the constraints such as the voltage, thermal limit and stability of the distribution network. The model mainly considers the increase and decrease of the load on the node, and the calculation result is the sum of the active power of all nodes in the network. The typical mathematical model expression is as follows:

[0235]

[0236]

[0237] Where: P LSC represents the maximum power supply capacity of the active distribution network under the constraints; P pi represents the initial active load of node i; P ni represents the increase in active load of node i; H represents the load growth multiple; N represents the number of network nodes; V iRepresents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; S ij represents the power passing through the branch between nodes i and j; S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i; P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i, G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; V j represents the voltage at node j.

[0238] The concept of power supply capacity is derived from transmission capacity. In the transmission network, voltage constraints, thermal stability and stability limits are the main factors affecting the transmission capacity. In the distribution network, we do not need to consider the stability limit, but only need to pay attention to the thermal limit of the branch, which is used as the upper limit of the branch capacity. At the same time, the rated power of each node transformer determines the upper limit of the node load capacity. We assume that the upper-level power grid can provide sufficient power. Based on these conditions, the maximum power that the distribution network can transmit while meeting the constraints can be expressed by the objective function of formula (7).

[0239] The power supply capacity of a distribution network with a simple network structure can be directly calculated using the repeated power flow method. The calculation process is to make the current active load P pi As the benchmark load, set the active load growth benchmark value P of each node ni, the initial load is continuously increased according to the set step length k, and the flow calculation is performed on the state after the load is changed to verify whether the constraints are met until the limit is exceeded. In the algorithm of repeated flow, k is usually self-adjusted by means of variable step length, and the iteration ends when k is reduced to below the required accuracy. The limit load value calculated at the end of the iteration is the maximum power supply capacity of the current distribution network. Regarding the selection of step length k, if the step length is too large, the result error will be large. If the step length is too small, there will be a lot of invalid calculations, which makes the convergence speed very slow. Therefore, the variable step length is generally used for calculation. Usually, k is searched forward step by step using the binary search method. If the constraint condition does not exceed the limit after increasing the load by multiples of k, continue to use the original step length k as the step length search. If the limit is exceeded after increasing the load, the power is reduced to the previous value before the limit is exceeded, and the step length is reduced to k / 2. Repeat the iteration until the step length is less than the set convergence accuracy. The detailed steps are as follows, such as Figure 2 shown.

[0240] (1) Input the starting step size k and the convergence criterion C;

[0241] (2) Input active load growth reference value P ni , original active load P pi ;

[0242] (3) If k is greater than e, proceed to the next step of calculation; if k is less than e, the calculation ends;

[0243] (4)P L =P pi +kp ni ;

[0244] (5) Use P L As a new network active load, it is brought into the network for power flow calculation to determine whether the constraints are met;

[0245] (6) If the constraints are met, let P L =P pi , go to step (4), otherwise proceed to the next step;

[0246] (7) Reduce the step length by half, that is, k = k / 2, and go to step (3).

[0247] S202. Probabilistic assessment of power supply capacity of active distribution network

[0248] This step will combine the repeated power flow method, three-point estimation method, and Cornish-Fisher series to comprehensively evaluate the power supply capacity of the active distribution network and calculate the power supply capacity distribution function of the active distribution network. In view of the uncertainty of the research object, the distribution function of the random variable is obtained by calculating the deterministic power flow of the estimation point;

[0249] If the evaluation object is an active distribution network (distributed generation) containing k DGs, each DG needs to estimate three sample points for calculation (sample point 1: DG expected minimum output; sample point 2: DG expected maximum output; sample point 3: DG expected expected output). According to the knowledge of probability statistics, the first k-order moments of the discrete distribution of the sample points constructed by all DGs in the distribution network are equal to the first k-order moments of the DG to be calculated. By solving the first k-order moments of the discrete points, the probability density function is calculated.

[0250] Construct the sample points corresponding to each DG in the distribution network, and calculate each component as follows:

[0251]

[0252] Where: μ k is the first-order origin moment of the sample point, σ is x k The standard deviation of x k,1 、x k,2 、x k,3 are the estimated 3 sample points of the kth DG, ξ k,1 , k,2 , k,3 are the position parameters corresponding to the three sample points;

[0253] The weight parameters and location parameters corresponding to each sample point satisfy the following equation:

[0254]

[0255] Where: k,l M l (x k )and The ratio of the power of l (x k ) is the l-th order origin moment of the k-th DG in the distribution network; is the lth power of the standard deviation of the kth DG in the distribution network; ω k,i is the weight parameter of the i-th sample point in the k-th DG; n is the total number of DGs; is the lth power of the location parameter corresponding to the i-th sample point in the k-th DG;

[0256]

[0257] In the formula, λ 1,l =0,λ 2,l =1,λ 3,l =2 is x k The skewness of the distribution function represents the deviation from the normal distribution; 4,l For x k The kurtosis represents the slope of the distribution function around the mean.

[0258] The constraints mainly consider voltage and power flow constraints. In addition, the uncertainty of DG output, line loss and load fluctuation need to be considered. The specific formula is as follows:

[0259] V imin ≤V i ≤V imax (12)

[0260]

[0261] Where V i Represents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; I ij represents the branch current between nodes i and j, and are the maximum and minimum values ​​of the branch current between nodes i and j respectively; S ij represents the power passing through the branch between nodes i and j, S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i, P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i; G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; S DG(i) represents the power emitted by node i DG, is the maximum output power of DG at node i.

[0262] like Figure 3 As shown, the specific calculation process for the maximum power distribution is as follows:

[0263] 1. Input the original parameters of the power grid and give the initial active load P of each node pi , if there are n DGs, use the three-point estimation method to take out 3n samples respectively to determine the DG power;

[0264] Three-point estimation is a statistical method used to evaluate unknown quantities. It is one of the simplest ways to accurately estimate a value, using the most optimistic, most pessimistic, and most likely estimates to predict the outcome.

[0265] For a given grid original parameters (such as line impedance, node voltage, etc.), here we assume that there are n distributed generation systems (DGs), and we need to determine the power of these DGs.

[0266] First, for each DG, we assign it an initial active load as a baseline. This will serve as our "most likely" estimate.

[0267] Next, we can make the most optimistic and pessimistic estimates for each DG. The most optimistic estimate may assume that all conditions are ideal, such as low load demand and good weather conditions, so that the DG can produce the maximum output power. On the contrary, the most pessimistic estimate may assume that all conditions are bad, such as high load demand and bad weather conditions, so that the DG can only produce the minimum output.

[0268] Now, we have 3 estimates (most optimistic, most likely, most pessimistic) for each DG, with a total of 3n samples. These samples can be used for subsequent statistical analysis, such as calculating the average power, power distribution range, confidence interval, etc. At the same time, this also provides input data for determining the optimal DG power output configuration.

[0269] 2. Set the growth active load P of each node ni and the load growth step length k, while giving the calculation accuracy e;

[0270] 3. Calculate the new load value of each node, bring in the network parameters and perform power flow calculation to determine whether the calculation results meet the constraints. If so, repeat step 3; if not, proceed to step 4.

[0271] 4. Use P L =P pi +kp ni Actual load of each node as the next power flow calculation;

[0272] 5. Determine whether k is smaller than the calculation precision. If k is less than e, implement step 7; if k is greater than e, implement step 6;

[0273] 6. Shorten the step length k, so that the load growth step length changes from k to k = k / 2. Substitute into step 4 and update the loads of all nodes in the network to perform power flow calculation again to determine whether the calculation results meet the constraints. If the constraints are met, return to step 5; if not, first set P L =P pi+kp ni , you need to return to step 5 to continue iterating;

[0274] 7. After the calculation is completed, the power supply power of the three sample points corresponding to each DG is obtained, and the statistical characteristic value of the power supply power is calculated.

[0275] According to the mathematical properties of the semi-invariant, the first t-order origin moment of the power supply capability statistic can be calculated, that is, the t-order semi-invariant γ of the random variable can be expressed by its t-order origin moment μ(γ j ) is obtained, the formula is as follows:

[0276]

[0277] Where: γ t+1 represents the t+1 order semi-invariant, t=1,2,…; the Cornish-Fisher expansion formula is used to obtain the probability density of the random variable y(φ), φ is the quantile of the power supply y, which is approximately expressed as:

[0278]

[0279] Where: ξ(φ)=Φ -l (φ), Φ represents the standard normal distribution N(0,1); using y(φ) = F -l The power supply capacity reflects the load carrying capacity of the distribution network and is an indicator of the safety of the distribution network. The power supply capacity is evaluated using the probabilistic power flow model.

[0280] S203, establishing a maximum power supply capacity assessment model for a power grid considering section coupling, wherein the maximum power supply capacity assessment model for a power grid considering section coupling takes the maximum power supply power increment, the maximum minimum section margin, and the most balanced reserve retention as objective functions, and takes the partition power generation equal to the sum of the power generation of all types of power sources in the partition, and the partition power adjustment equal to the sum of the power adjustment of all types of power sources in the partition as constraint conditions;

[0281] Specifically:

[0282] A maximum power supply capacity assessment model for the power grid considering section coupling is established, and the model is described in units of partitions.

[0283] The decision variables are determined as the power adjustment of each type of power supply in each sub-area of ​​the provincial network, and the positive direction is defined as the increase in power generation. The value range of the decision variables must meet the following conditions:

[0284]

[0285] Where: Indicates the negative reserve of class S power supply in partition i, which is a positive number; ΔPi.s It represents the power adjustment of the class s power supply in the partition i; Indicates the positive and standby of the s-type power supply in zone i; i g.type represents the set of all power types in partition i; O represents the number of partitions involved in the adjustment.

[0286] At the same time, the following relationships are satisfied: the partition power generation is equal to the sum of the power generation of each type of power source in the partition; the partition power adjustment is equal to the sum of the power adjustment of each type of power source in the partition.

[0287]

[0288] Among them, P i represents the active power of node i; represents the set of s-type power sources belonging to node i; P i.s represents the active power of the s-type power supply at node i; ΔP i Represents the change in active power of node i; ΔP i.s represents the change in active power of the s-type power supply at node i;

[0289] The objective function is as follows:

[0290] (1) Based on the current startup mode, load and power generation of the large power grid units, the power supply power increment is guaranteed to be the largest. By adjusting the output of various types of power sources, the maximum power adjustment amount of the large power grid at a certain moment is calculated.

[0291]

[0292] Among them, ΔP i represents the change in active power of node i; maxf1 represents the maximum power supply power increment; this formula represents the calculation of the maximum power adjustment of the large power grid at a certain moment based on the given large power grid unit startup mode, load and power generation power.

[0293] (2) Minimum section margin is maximum. In actual operation, it is necessary to ensure that each section has a certain stability margin to ensure the safe and stable operation of the power grid. When the large power grid reaches its maximum power supply capacity, each section still has a certain limit.

[0294]

[0295] in

[0296]

[0297] Where: maxf2 represents the maximum minimum section margin; M represents the section set; It represents the upper margin after adjustment of section j; It represents the lower margin after adjustment of section j; Indicates the upper limit of section power; P j Indicates the lower limit of section power; P j represents the power of section j after the partition power adjustment; ΔP j represents the power variation of section j; Indicates the power before section adjustment; S j-N S represents the row vector of section sensitivity of all partitions of the large power grid to section j; j-L Represents the row vector of section sensitivity of all the sections to be evaluated in the large power grid to section j; represents the column vector consisting of the ratio of the load of each partition to be evaluated to the total load of all partitions to be evaluated before adjustment; N1 represents the number of partitions to be evaluated; 1 1×N represents a 1×N row vector with all 1s; ΔP N represents the column vector composed of the power adjustment of each partition; C represents the coefficient matrix of the coupling relationship between all the partitions to be evaluated in the large power grid;

[0298] (3) The most balanced reserve

[0299] After the power of each type of power source in each partition is adjusted, it is hoped that the ratio of the positive reserve capacity retained by each partition to the load of each partition is basically the same. The setting of this goal can make the partitions with more positive reserve and lower load in the large power grid increase more power, and the partitions with less positive reserve and more load increase less power, thereby ensuring a more reasonable and balanced distribution of positive reserve in the power grid.

[0300]

[0301] Where: minf3 is the most balanced function for backup; α i represents the ratio of partition i’s positive standby to partition load; The adjusted positive and spare parts of partition i; Indicates the active and standby state of partition i before adjustment; is the load before adjustment for partition i; represents the average value of α.

[0302] (1) Processing of the objective function

[0303] For f2, it can be transformed into a linear objective function by introducing auxiliary variables. The specific form is as follows

[0304]

[0305] in, represents the power adjustment of partition j;

[0306] j represents each partition, belonging to the partition set M;

[0307] * indicates the adjustment type, which can be "up" or "down", i.e., adjusting the power generation up or down.

[0308] M represents the set of all partitions

[0309] (2) Processing of Constraints

[0310] The first type of cross-section coupling constraints are linearized. In summary, the constructed maximum power supply capacity evaluation model is transformed into a multi-objective mixed integer programming model.

[0311] Linearize the objective function: In practice, the nonlinear objective function is converted into a linear objective function by introducing auxiliary variables and adding additional constraints, or at least a linear approximation can be obtained.

[0312] Linearize the constraints: For those constraints that are linear, use them directly. For those that are nonlinear, use the Taylor series method to linearize them.

[0313] With a linear objective function and linear constraints, we can regard the problem as a standard linear programming problem or a mixed integer linear programming problem.

[0314] S204. Evaluation model considering multiple time scales:

[0315] Includes upper model and lower model;

[0316] The upper model aims to maximize social benefits, takes the declared prices and declared quantities of power generators and power buyers as decision variables, considers load balance and unit output constraints, and passes the obtained transaction volume to the lower power supply capacity model. The objective function is:

[0317]

[0318] in,

[0319]

[0320] Where U is the load node participating in power trading; T is the user period. For medium- and long-term trading, T is set to 1; for day-ahead trading, T is set to 24; for real-time trading, T is set to 96; z is the number of units; D u,t B is the estimated demand load of the power sales company at node u at time t; u,t (D u,t ) is the electricity purchase cost of the power sales company at node u at time t; P i,t is the output of unit i at time t; C i,t (P i,t) is the operating cost of unit i at time t; M is the penalty factor for network power flow constraint relaxation (determined based on expert experience and the specific situation of the problem. An initial value of 10 is adopted); are the forward and reverse power flow relaxation variables of line l respectively; N N The total number of sections quoted by the power sales company; D u,t,n is the estimated demand load of the nth segment at node u of the power sales company at time t; C u,t,n N is the purchase price corresponding to the nth segment number at node u at time t of the power sales company; M P is the total number of sections in the unit quotation; i,t,m is the predicted output of the mth stage of unit i at time t; C i,t,m is the operating price corresponding to the mth section of unit i at time t; N L is the total number of unit lines;

[0321] The constraints are as follows:

[0322] (1) System load balance equation constraint; its formula is:

[0323]

[0324] Where, T j,t is the exchange power between tie line j and the outside of the system at time t; N T is the total number of contact lines; K is the total number of nodes; is the load on node k that does not participate in market transactions at time t;

[0325] (2) Unit output constraint; the formula is:

[0326]

[0327] Where P z,t represents the power output generated by the zth unit at time t;

[0328] represents the minimum power output of the zth unit at time t;

[0329] represents the maximum power output of the zth unit at time t;

[0330] (3) Node voltage constraint; its formula is:

[0331] V kmin ≤V k ≤V kmax (30)

[0332] Where V kmin represents the minimum voltage limit of the kth node; V kmaxrepresents the maximum voltage limit of the kth node; V k is the voltage of the kth node;

[0333] (4) Line flow constraint. The formula is:

[0334]

[0335] Where P l max is the maximum value of power flow transmission of line l; G l-i is the generator output power transfer distribution factor of the node where unit i is located to line l; G l-j is the generator output power transfer distribution factor of the node where the tie line j is located to the line l; G l-k is the power transfer distribution factor of the generator output from node k to line l; u∈k is the power sales company at node k. The specific values ​​of these distribution factors depend on the specific layout and parameters of the power grid, as well as the operating status of the grid. For example, if the grid is highly coupled, a power change at one node will cause a large change in the power of other nodes with the same transfer factor. On the other hand, if the grid is decentralized, a power change at one node will only cause a small range of power changes at nearby nodes. These transfer distribution factors will take values ​​between 0 and 1. A transfer factor of 1 indicates a strong coupling relationship between nodes or lines, while a transfer factor close to 0 indicates a weak coupling relationship.

[0336] The lower model is an optimization process for the maximum power supply capacity of the distribution network based on the load of each node participating in the power market transaction obtained by the upper model. The lower model takes the maximum power supply capacity of the power grid as the goal, and considers the power flow constraints, main transformer load transfer constraints, feeder load transfer constraints and main transformer load rate constraints.

[0337] The objective function is:

[0338]

[0339] in:

[0340] D u,t =D u,t,z +D u,t,r +D u,t,s

[0341]

[0342] In the formula, k' is the load growth multiple; Y k Y is the load growth base at node k; A ASC D is the available power supply capacity of the distribution network; u,t,z , D u,t,r , D u,t,sare the loads of medium- and long-term transactions, day-ahead transactions, and real-time transactions at node u at time t; T TSC Represents the total power supply capacity of the distribution network, that is, the total power supply of all nodes; S k,t represents the power supply of node k at time t; It is the total load of transactions in the medium and long term market; is the total load traded in the day-ahead market; is the total load traded in the live market;

[0343] The constraints are as follows:

[0344] (1) Power flow constraint. The formula is:

[0345] Ai′=I′ (35)

[0346] Where A is the node-branch association matrix; i′ is the complex current vector of all branches; and I′ is the complex current injection vector of all nodes.

[0347] The node / branch association matrix (also called the node incidence matrix) is a mathematical tool used to describe the topology of a power system or the grid structure (which nodes are connected through which branches).

[0348] In the node / branch association matrix, rows usually represent nodes and columns usually represent branches. If there is a specific starting point of a branch at a certain node, the corresponding matrix element is -1; if there is an end point of the branch at a certain node, the corresponding matrix element is 1; if the branch does not exist at the node, the corresponding matrix element is 0. In other words, the association matrix describes which incoming and outgoing branches exist at each node. For a simple example, suppose we have 3 nodes (node ​​1, node 2, node 3) and two branches (branch 1 from node 1 to node 2, branch 2 from node 2 to node 3). Then the corresponding node / branch association matrix is

[0349]

[0350] (2) Main transformer load transfer belt constraint. The formula is:

[0351]

[0352] In the formula, t ij B is the load transferred to main transformer j when main transformer i fails; j is the load on the main transformer j; is the rated capacity of the main transformer j.

[0353] (3) Feeder load transfer constraint. The formula is:

[0354] lmn +F n ≤R Fn (37)

[0355] In the formula, l mn is the load transferred to feeder n when feeder m fails; F is the load on feeder n; R Fn is the rated capacity of feeder n.

[0356] (4) Main transformer load factor constraint. The formula is:

[0357] B imin ≤B i ≤B imax (38)

[0358] In the formula, B i B is the load factor of the main transformer i. imin The minimum load factor of the main transformer i, B imax The maximum value of the load factor of the main transformer i.

[0359] The main transformer load rate constraint is used to ensure that the operating load of the main transformer is within a safe and effective range. When the load is too high, the main transformer will overheat or be damaged, which will affect the stability of the power grid and the reliability of power supply. On the contrary, if the load is too low, it may lead to energy waste and reduce the efficiency of equipment use.

[0360] Therefore, the main purpose of this constraint is to maintain the stability and reliability of the power grid, while also ensuring the effective use of energy to a certain extent.

[0361] To describe the load balancing, the load balancing degree is defined in combination with the variance for:

[0362]

[0363] In the formula, The average load factor of the main transformer; N B The number of main variables.

[0364] Load balancing refers to the load distribution of each unit or main equipment in the power system. In the actual operation of the power system, in order to improve the economy and stability of the system, it is necessary to distribute the load of each unit as evenly as possible.

[0365] Variance is an important concept in statistics, which is used to describe the degree of dispersion of a set of data, that is, the fluctuation of the data. Here, the load balance degree is defined in combination with the variance, which is based on the load rate of each main transformer, and its average value and variance are calculated to depict the load balance state of the entire system.

[0366] Therefore, this load balance is actually an important indicator for measuring the operation of the power system. An excellent operating state requires not only that the total load of the power system be met as much as possible, but also that the load of individual units be balanced as much as possible to avoid the situation where some units are overloaded while other units are underloaded. This is very important for the stable operation of the power system and the life of the equipment. The load balance here can better reflect this point.

[0367] In this model, while comprehensively optimizing social benefits and power supply capacity of the power grid, adding load balance as an indicator can make the entire optimization more comprehensive. It not only takes into account the general direction of social benefits and power supply capacity, but also pays attention to the detailed operating status, making the model more feasible in real applications.

[0368] S3. Model solution

[0369] like Figure 4 As shown in the figure, the original dual interior point method and the improved repeated power flow method are combined to solve the two algorithms in a mixed way, which can improve both the calculation speed and the accuracy of the calculation. Figure 4 .

[0370] (1) Collect information on power generation and purchase declarations under medium- and long-term transactions;

[0371] (2) Using PDIPM to solve the medium- and long-term transaction electricity;

[0372] (3) Decomposition of medium- and long-term transaction volume to obtain D u,t,z ;

[0373] (4) Let k′=1, h=h0, Y k =S k,t +D u,t,z Among them, D u,t,z represents the amount of electricity purchased by user u at time t; h0 represents the initial value, Y k represents the electric energy demand at time k; S k,t represents the power supply of node k at time t;

[0374] (5)T TSC =S s,t +D u,t +k′Y k ; Among them, S s,t represents the reported power generation of power source s at time t; D u,t represents the amount of electricity purchased by user u at time t; k′Y k} represents the amount of electric energy that the power supply needs to provide at time k; k′ is a coefficient set according to the actual operation of the power grid and the power supply capacity of the power supply;

[0375] (6) Whether the flow exceeds the limit. If so, h = h / 2, and the calculation returns to h < ε. h < ε is the judgment condition, which means that when the change of the variable h is less than a sufficiently small positive number ε, the algorithm is considered to have converged and the iterative calculation can be stopped. ε is 0.001.

[0376] (7) If h < ε, then output the medium- and long-term transaction T TSC and A ASC , combined with A ASC Update the power generation and purchase declaration information under day-ahead transactions.

[0377] The same logic is used to calculate day-ahead transactions and real-time transactions.

[0378] The solution of the present invention is not limited to the above method, as long as the method can achieve the solution.

[0379] Application Examples

[0380] Specific embodiment: Assume a small or medium-sized power distribution network, which has a 1MW generator, 10 resistive loads, a maximum current transmission of 1000A, and voltage upper and lower limits of 10kV and 3kV. We use the characterization method of the power grid flexibility power supply capability provided by the present invention to evaluate it.

[0381] In the process of building the model, we obtained the daily load data at each feeder outlet from the distribution network power consumption information collection system. The data is as follows (to simplify the calculation, it is assumed that the load is in a 24-hour stable state):

[0382] Load 1: 80kW

[0383] Load 2: 100kW

[0384] Load 3: 70kW

[0385] Load 4: 90kW

[0386] Load 5: 110kW

[0387] Load 6: 85kW

[0388] Load 7: 120kW

[0389] Load 8: 75kW

[0390] Load 9: 95kW

[0391] Load 10: 80kW

[0392] Based on the above data, we established a load power assessment model, taking into account the grid structure, current load and load growth pattern. The model assessment results show that under the constraints, the maximum load that the distribution network can carry is 930kW, which is much smaller than the maximum output of the generator of 1MW.

[0393] Then, we evaluate the probability of the power supply capacity of the active distribution network by using the model and calculate the power supply capacity distribution function of the active distribution network. The results show that with a probability of 90%, the power supply capacity of the distribution network is 925kW.

[0394] A hybrid solution method combining the original dual interior point method and the improved repeated power flow method is used to maximize the power supply capacity of the distribution network and meet various constraints, including voltage, power flow constraints, line losses and load fluctuations.

[0395] Next, a maximum power supply capacity evaluation model for the power grid considering section coupling was established. Through this model, the maximum power supply capacity and adjustment amount of each node in the power grid were determined under the premise of meeting all constraints. At the same time, the current limit of each section when the power grid reaches the maximum power supply capacity, as well as the ratio of the positive reserve capacity and load retained by each partition in this case, were calculated. The conclusions are as follows:

[0396] 1. The maximum weighted increase in power supply capacity is 920kW, which is the current maximum power output of the network.

[0397] 2. The minimum margin in the power grid is at section 8, which is 70A. At the same time, the minimum margin of the entire power grid is also 70A.

[0398] 3. At maximum power supply capacity, the ratio of reserve power to load retained in each partition is close to uniform.

[0399] Next, based on multiple time scales such as medium- and long-term, day-ahead and real-time, and taking into account the load balance of the power grid, unit output constraints, node voltage constraints and line flow constraints, an upper-level model was established and solved, and the results were passed to the lower-level model for further solution.

[0400] The conclusion is that under the premise of maximum social benefit, the output of the generator meets all the loads of the distribution network, and at the same time, the voltage of each node of the power grid meets the constraints. On this basis, the power supply capacity of the power grid is comprehensively evaluated using the lower-level model, and the conclusion is that under the premise of meeting all constraints, the output of the generator can meet the total load of the distribution network.

[0401] In general, whether it is evaluated at a single time scale or a comprehensive evaluation at multiple time scales, the power supply capacity of the distribution network can meet the actual electricity demand and show high stability and reliability.

[0402] The power supply capacity assessment of existing technologies is generally determined only based on the capacity of the generator and the load distribution, ignoring the impact of various uncertain factors, and often only focusing on the overall power supply capacity of the power grid, ignoring the power supply capacity of the local distribution network, resulting in overestimation or underestimation of the power supply capacity.

[0403] Compared with the prior art, the method of the present invention more accurately evaluates the power supply capacity of the distribution network and takes into account more uncertain factors. In addition, the method of the present invention can also provide more effective support for power grid optimization and operation decision-making. For example, in actual operation, the power supply capacity distribution function and probability provided by the method of the present invention can be used to reasonably allocate and dispatch the output of the generator to achieve supply and demand balance and improve the operation efficiency of the power grid.

[0404] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for characterizing the flexible power supply capability of a power grid, characterized in that: The steps include: S1. Obtain the daily load data at each feeder outlet, the power generation and purchase declaration information under medium- and long-term transactions, and the operation data and basic data of the distribution network from the distribution network power consumption information collection system, and pre-process the collected daily load data; S2. Establish a load power evaluation model, which takes the maximum load carried by the power grid as the objective function and the voltage, thermal limit and stability of the distribution network as the constraint conditions; S3. Establish a probability assessment model for the power supply capacity of the active distribution network to assess the probability of the power supply capacity of the active distribution network; the probability assessment model for the power supply capacity of the active distribution network combines the repeated power flow method, the three-point estimation method, and the Cornish-Fisher series to comprehensively assess the power supply capacity of the active distribution network and calculate the power supply capacity distribution function of the active distribution network; S4. Establish a maximum power supply capacity assessment model for the power grid considering section coupling. The maximum power supply capacity assessment model for the power grid considering section coupling takes the maximum power supply power increment, the maximum minimum section margin and the most balanced reserve retention as the objective function, and takes the partition power generation equal to the sum of the power generation of all types of power sources in the partition, and the partition power adjustment equal to the sum of the power adjustment of all types of power sources in the partition as the constraint conditions; S5. Establish an evaluation model that considers multiple time scales. The evaluation model that considers multiple time scales adopts a two-layer model. The upper layer model takes the maximum social benefit as the objective function, and the obtained transaction volume is passed to the lower layer model; the lower layer model takes the maximum power supply capacity of the distribution network as the objective function; S6. Based on the data obtained in step (1), the model constructed in S2 to S5 is solved to obtain the total power supply capacity and available power supply capacity of the power grid.

2. The method for characterizing the flexible power supply capability of a power grid according to claim 1, characterized in that: In S1, the preprocessing is normalization processing.

3. The method for characterizing the flexible power supply capability of a power grid according to claim 2, characterized in that: The specific method for normalizing daily load data is to use the maximum value of the feeder daily load curve as the normalization factor to convert the original load data into a dimensionless load value.

4. The method for characterizing the flexible power supply capability of a power grid according to claim 1, characterized in that: In S2, the load power evaluation model established is as follows: Where: P LSC represents the maximum power supply capacity of the active distribution network under the constraints; P pi represents the initial active load of node i; P ni represents the increase in active load of node i; H represents the load growth multiple; N represents the number of network nodes; V i Represents the voltage of node i; V imax 、V imin are the maximum and minimum voltage of node i respectively; S ij represents the power passing through the branch between nodes i and j; S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i; P imax is the maximum capacity of the active load of the transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i, G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; V j represents the voltage at node j.

5. The method for characterizing the flexible power supply capability of a power grid according to claim 1, characterized in that: In S3, the establishment method of the probability evaluation model of the power supply capacity of the active distribution network is as follows: if the evaluation object is an active distribution network containing k DGs, each DG needs to estimate 3 sample points for calculation; since the first k-order moments of the discrete distribution of the sample points constructed by all DGs in the distribution network are equal to the first k-order moments of the DG to be calculated, the probability density function is calculated by solving the first k-order moments of the discrete points. The probability density function is shown in formula (19); Construct the sample points corresponding to each DG in the distribution network, and calculate each component as follows: Where: μ k is the first-order origin moment of the sample point, σ is x k The standard deviation of x k,1 、x k,2 、x k,3 are the estimated 3 sample points of the kth DG, ξ k,1 , k,2 , k,3 are the position parameters corresponding to the three sample points; The weight parameters and location parameters corresponding to each sample point satisfy the following equation: Where: k,l M l (x k )and The ratio of the lth power; M l (x k ) is the l-th order origin moment of the k-th DG in the distribution network; is the lth power of the standard deviation of the kth DG in the distribution network; ω k,i is the weight parameter of the i-th sample point in the k-th DG; n is the total number of DGs; is the lth power of the location parameter corresponding to the i-th sample point in the k-th DG; The specific formula of the constraint condition is as follows: In imin ≤V i ≤V imax (12) I ijmin ≤I ij ≤I ijmax (13) Where V i Represents the voltage of node i; V imax 、V imin They are the maximum voltage and minimum voltage of node i respectively; I ij represents the branch current between nodes i and j, and are the maximum and minimum values ​​of the branch current between nodes i and j respectively; S ij represents the power passing through the branch between nodes i and j, S ijmax is the maximum power passing through the branch between nodes i and j; P i represents the active load of the transformer at node i, P imax is the maximum capacity of active load of transformer at node i; Q i represents the reactive load of the transformer at node i; Q imax is the maximum reactive load capacity of the transformer at node i; P G(i) represents the active power emitted by node i; P L(i) represents the active load of node i; Q G(i) represents the reactive power emitted by node i; Q L(i) represents the reactive load of node i; G ij , B ij ,θ ij represents the conductance, susceptance, and phase angle between i and j; S DG(i) represents the power emitted by node i DG, is the maximum output power of DG at node i; According to the mathematical properties of the semi-invariant, the first t-order origin moment of the power supply capacity statistic is calculated, that is, the t-order semi-invariant γ of the random variable is expressed by its t-order origin moment μ(γ j ) is obtained, the formula is as follows: Where: γ t+1 is a t+1-order semi-invariant; γ1 is a 1-order semi-invariant; u(γ) is the t-order origin moment of the random variable element; μ(γ j ) is the calculation result of the j-order origin moment; γ t＝j+1 For the j+1 order semi-invariant; The Cornish-Fisher expansion formula is used to obtain the probability density of the random variable y(φ), where φ is the quantile of the power supply y, which is approximately expressed as: Where: γ3 and γ4 represent the third-order and fourth-order semi-invariants of the random variable y respectively; ξ(φ) represents the standard normal quantile function.

6. The method for characterizing the flexible power supply capability of a power grid according to claim 1, characterized in that: In S4, the method for establishing the maximum power supply capacity evaluation model of the power grid considering section coupling is as follows: A maximum power supply capacity assessment model considering section coupling is established. The model is described in units of zones. The decision variables are determined as the power adjustment of each type of power source in each zone of the provincial network, and the increase in power generation is stipulated as the positive direction. The value range of the decision variables must meet the following conditions: Where: Indicates the negative reserve of class S power supply in partition i, which is a positive number; ΔP i.s It represents the power adjustment of the class s power supply in the partition i; Indicates the positive and standby of the s-type power supply in zone i; i g.type represents the set of all power types in partition i; O represents the number of partitions involved in the adjustment; At the same time, the following relationships are satisfied: the power generation of a partition is equal to the sum of the power generation of all types of power sources in the partition; the power adjustment of a partition is equal to the sum of the power adjustment of all types of power sources in the partition; Among them, P i represents the active power of node i; represents the set of s-type power sources belonging to node i; P i.s represents the active power of the s-type power supply at node i; ΔP i Represents the change in active power of node i; ΔP i.s represents the change in active power of the s-type power supply at node i; The objective function is as follows: (1) Calculate the maximum power supply increment of the large power grid at a certain moment. The formula is as follows: Where ΔP i represents the change in active power of node i; maxf1 represents the maximum power supply power increment; this formula represents the calculation of the maximum power adjustment of the large power grid at a certain moment based on the given large power grid unit startup mode, load and power generation power. (2) Calculate the limits of each section when the large power grid reaches its maximum power supply capacity. The formula is as follows: in Where: maxf2 represents the maximum minimum section margin; M represents the section set; It represents the upper margin after adjustment of section j; It represents the lower margin after adjustment of section j; Indicates the upper limit of section power; P j Indicates the lower limit of section power; P j represents the power of section j after the partition power adjustment; ΔP j represents the power variation of section j; Indicates the power before section adjustment; S j-N S represents the row vector of section sensitivity of all partitions of the large power grid to section j; j-L Represents the row vector of section sensitivity of all the sections to be evaluated in the large power grid to section j; represents the column vector consisting of the ratio of the load of each partition to be evaluated to the total load of all partitions to be evaluated before adjustment; N1 represents the number of partitions to be evaluated; 1 1×N represents a 1×N row vector with all 1s; ΔP N represents the column vector composed of the power adjustment of each partition; C represents the coefficient matrix of the coupling relationship between all the partitions to be evaluated in the large power grid; (3) Calculate the ratio of the positive reserve capacity to be retained by each partition after the power adjustment of each type of power supply in each partition to the load of each partition. The ratio is required to be basically the same. The formula is as follows: Where: minf3 is the most balanced function for backup; α i represents the ratio of partition i’s positive standby to partition load; The adjusted positive and spare parts of partition i; Indicates the active and standby state of partition i before adjustment; is the load before adjustment for partition i; represents the average value of α.

7. The method for characterizing the flexible power supply capability of a power grid according to claim 1, characterized in that: In S5, the method for establishing the evaluation model considering multiple time scales is as follows: The evaluation model considering multiple time scales includes upper-level model and lower-level model; The upper model aims to maximize social benefits, takes the declared prices and declared quantities of power generators and power buyers as decision variables, considers load balance, unit output constraints, node voltage constraints and line flow constraints, and passes the obtained transaction volume to the lower model; the lower model aims to maximize the power supply capacity of the power grid, considers flow constraints, main transformer load transfer constraints, feeder load transfer constraints and main transformer load rate constraints.

8. The method for characterizing the flexible power supply capability of a power grid according to claim 7, characterized in that: The objective function of the upper model is: in, In the formula, U is the total number of load nodes participating in power trading; T is the user time period; z is the number of units; D u,t B is the estimated demand load of the power sales company at node u at time t; u,t (D u,t ) is the electricity purchase cost of the power sales company at node u at time t; P z,t is the output of unit z at time t; C z,t (P z,t ) is the operating cost of unit z at time t; M is the penalty factor for network power flow constraint relaxation; are the forward and reverse power flow relaxation variables of line l respectively; N N The total number of sections quoted by the power sales company; D u,t,n is the estimated demand load of the nth segment at node u of the power sales company at time t; C u,t,n N is the purchase price corresponding to the nth segment number at node u at time t of the power sales company; M P is the total number of sections in the unit quotation; z,t,m is the predicted output of the mth stage of unit z at time t; C z,t,m is the operating price corresponding to the mth section of unit z at time t; N L is the total number of unit lines; The constraints are as follows: (1) System load balance equation constraint; its formula is: Where, T j,t is the exchange power between tie line j and the outside of the system at time t; N T is the total number of contact lines; K is the total number of nodes; is the load on node k that does not participate in market transactions at time t; (2) Unit output constraint; the formula is: Where P z,t represents the power output generated by the zth unit at time t; represents the minimum power output of the zth unit at time t; represents the maximum power output of the zth unit at time t; (3) Node voltage constraint; its formula is: In kmin ≤V k ≤V kmax (30) Where V kmin represents the minimum voltage limit of the kth node; V kmax represents the maximum voltage limit of the kth node; V k is the voltage of the kth node; (4) Line flow constraint; its formula is: Where P l max is the maximum value of power flow transmission of line l; G l-i is the generator output power transfer distribution factor of the node where unit i is located to line l; G l-j is the generator output power transfer distribution factor of the node where the tie line j is located to the line l; G l-k is the generator output power transfer distribution factor of node k to line l; u∈k is the power sales company at node k.

9. The method for characterizing the flexible power supply capability of a power grid according to claim 7, characterized in that: The objective function of the lower model is: in: D u,t =D u,t,z +D u,t,r +D u,t,s In the formula, k' is the load growth multiple; Y k Y is the load growth base at node k; A ASC D is the available power supply capacity of the distribution network; u,t,z , D u,t,r , D u,t,s are the loads of medium- and long-term transactions, day-ahead transactions, and real-time transactions at node u at time t; T TSC Represents the total power supply capacity of the distribution network, that is, the total power supply of all nodes; S k,t represents the power supply of node k at time t; It is the total load of transactions in the medium and long term market; is the total load traded in the day-ahead market; is the total load traded in the live market; The constraints are as follows: (1) Power flow constraint; its formula is: Ai′=I′ (35) Where A is the node-branch association matrix; i′ is the complex current vector of all branches; I′ is the complex current injection vector of all nodes; (2) Main transformer load transfer belt constraint; the formula is: Where, t ij B is the load transferred to main transformer j when main transformer i fails; j is the load on the main transformer j; is the rated capacity of the main transformer j; (3) Feeder load transfer constraint; the formula is: the mn +F n ≤R Fn (37) In the formula, l mn is the load transferred to feeder n when feeder m fails; F is the load on feeder n; R Fn is the rated capacity of feeder n. (4) Main transformer load factor constraint; the formula is: B imin ≤B i ≤B imax (38) In the formula, B i The load factor of the main transformer i, B imin The minimum load factor of the main transformer i, B imax is the maximum load rate of the main transformer i;.

10. The method for characterizing the flexible power supply capability of a power grid according to claim 9, characterized in that: To describe the load balancing, the load balancing degree is defined in combination with the variance for: In the formula, The average load factor of the main transformer; N B The number of main variables.

Images