A data-driven method and system for quantifying grid flexibility margins

Abstract

The application discloses a data-driven power grid flexibility margin quantification method and system, and relates to the technical field of power grid operation flexibility. The method comprises the following steps: based on the Copula theory, constructing a wind-solar output space-time correlation set according to historical wind power generation data sets and historical photovoltaic power generation data sets; using a scenario reduction method based on a probability distance to reduce the wind-solar output space-time correlation set to obtain a net load typical scenario, and then establishing a power grid flexibility demand model, a power supply side flexibility supply model and a demand side flexibility supply model; based on a flexibility supply-demand balance constraint, taking the sum of a power supply side consumption cost of the power grid, a demand side discharge loss compensation cost of the power grid and a penalty cost corresponding to a power grid flexibility shortage as the minimum as the target, establishing a data-driven optimization scheduling model, and solving to obtain optimal scheduling results of the power grid for the power supply side and the demand side, and then calculating the flexibility margin of the power grid. The application solves the uncertainty problem of power grid flexibility balance.

Description

Technical Field

[0001] This invention relates to the field of power grid operation flexibility technology, and in particular to a data-driven method and system for quantifying power grid flexibility margin. Background Technology

[0002] The rapid development of renewable energy power generation equipment and grid connection technologies has led to the gradual formation of a new type of power grid dominated by new energy sources. The random volatility of a high proportion of new energy sources significantly increases the difficulty of grid operation and dispatch. Traditional flexibility resources, such as conventional generating units, are no longer sufficient to effectively meet the rapidly increasing flexibility demands of the system. The grid suffers from severe flexibility deficiencies in certain time periods, making demand-side flexibility a core research focus in grid dispatch and operation. Quantifying the demand for flexibility resources has become a key research issue. The International Energy Agency (IEA) and the North American Electric Reliability Association define flexibility as: the ability of a power grid's flexibility resources to meet flexibility demands over a given time scale. Quantifying flexibility demand and supply is crucial for analyzing grid flexibility margins. In the field of grid flexibility margin quantification, the offsetting and superposition effects of load and new energy sources are typically considered, with the system's flexibility demand represented by the first-order difference of the net load. Wind and solar loads are uncertain, and existing technologies use joint probability distribution functions to describe the randomness of the net load. These probability-based techniques cannot accurately provide the grid's flexibility demand and regulation capacity for different time periods and are limited by the probability distribution of uncertainties. In addition, existing techniques for quantifying flexibility requirements rarely consider the spatiotemporal correlation of wind and solar power output, resulting in inaccurate descriptions of power output sets. Summary of the Invention

[0003] The purpose of this invention is to provide a data-driven method and system for quantifying power grid flexibility margin, which optimizes and quantifies power grid flexibility margin, and solves the uncertainty problem of power grid flexibility balance.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A data-driven method for quantifying grid flexibility margin includes:

[0006] Obtain historical wind power generation datasets and historical photovoltaic power generation datasets;

[0007] Based on Copula theory, a spatiotemporal correlation set of wind and solar power output is constructed according to the historical wind power generation dataset and the historical photovoltaic power generation dataset.

[0008] The scene reduction method based on probabilistic distance is used to reduce the spatiotemporal correlation set of wind and solar power output to obtain typical net load scenarios.

[0009] Based on the aforementioned typical net load scenario, a power grid flexibility demand model is established;

[0010] Establish a power supply flexibility model for the power grid and a demand-side flexibility supply model for the power grid;

[0011] Based on the power grid flexibility demand model, the power source flexibility supply model, and the demand-side flexibility supply model, the power grid flexibility deficit is calculated; the power grid flexibility deficit represents the flexibility supply and demand balance constraint.

[0012] Based on the aforementioned flexibility supply and demand balance constraint, a data-driven optimization scheduling model is established with the objective of minimizing the sum of the power grid's power source consumption cost, the power grid's demand-side discharge loss compensation cost, and the penalty cost corresponding to the power grid's flexibility deficit.

[0013] The data-driven optimization scheduling model is solved to obtain the optimal scheduling results of the power grid for the power supply side and the demand side.

[0014] The power grid flexibility margin is calculated based on the optimal scheduling results.

[0015] Optionally, based on Copula theory, a spatiotemporal correlation set of wind and solar power output is constructed according to the historical wind power generation dataset and the historical photovoltaic power generation dataset, specifically including:

[0016] Based on the kernel density estimation method, the probability density of wind power output in each time period is calculated according to the historical wind power generation dataset, and the probability density of photovoltaic power output in each time period is calculated according to the historical photovoltaic power generation dataset.

[0017] For any given time period, the wind power distribution function is determined based on the wind power output probability density, and the photovoltaic distribution function is determined based on the photovoltaic output probability density.

[0018] Based on the Frank-Copula function, a joint wind-solar probability distribution function is constructed according to the wind power distribution function and the photovoltaic distribution function;

[0019] A random numerical set of marginal distribution functions for wind and solar variables is generated using the aforementioned wind-solar joint probability distribution function;

[0020] The random numerical set of the marginal distribution function of the wind and solar variables is inverted using the inverse function to obtain the spatiotemporal correlation set of wind and solar power output.

[0021] To achieve the above objectives, the present invention also provides the following technical solutions:

[0022] A data-driven power grid flexibility margin quantification system, comprising:

[0023] The dataset acquisition module is used to acquire historical wind power generation datasets and historical photovoltaic power generation datasets;

[0024] The wind and solar power output spatiotemporal correlation set construction module is used to construct the wind and solar power output spatiotemporal correlation set based on the Copula theory and the historical wind power generation dataset and the historical photovoltaic power generation dataset.

[0025] The typical net load scenario construction module is used to reduce the spatiotemporal correlation set of wind and solar power output using a scenario reduction method based on probabilistic distance to obtain typical net load scenarios.

[0026] The power grid demand model construction module is used to establish a power grid flexibility demand model based on the typical net load scenario.

[0027] The supply model construction module is used to establish the power supply flexibility supply model and the demand-side flexibility supply model of the power grid.

[0028] The grid flexibility deficit calculation module is used to calculate the grid flexibility deficit based on the grid flexibility demand model, the power source flexibility supply model, and the demand side flexibility supply model; the grid flexibility deficit represents the flexibility supply and demand balance constraint.

[0029] The data-driven optimization scheduling model construction module is used to establish a data-driven optimization scheduling model based on the flexibility supply and demand balance constraints, with the goal of minimizing the sum of the power supply side consumption cost of the power grid, the demand side discharge loss compensation cost of the power grid, and the penalty cost corresponding to the flexibility deficit of the power grid.

[0030] The optimal scheduling determination module is used to solve the data-driven optimal scheduling model to obtain the optimal scheduling results of the power grid for the power supply side and the demand side.

[0031] The power grid flexibility margin calculation module is used to calculate the power grid flexibility margin based on the optimal scheduling result.

[0032] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0033] This invention discloses a data-driven method and system for quantifying grid flexibility margin. Based on historical wind power and photovoltaic power generation datasets, a spatiotemporal correlation set of wind and solar power output is constructed, fully considering the uncertainty and correlation of wind and solar load fluctuations. Compared with existing flexibility margin quantification technologies, it achieves a compromise in robustness for grid flexibility margin quantification during the day-ahead dispatch phase. Considering the spatiotemporal correlation of wind and solar power output, a scenario reduction method based on probabilistic distance is used to reduce the spatiotemporal correlation set of wind and solar power output to obtain typical net load scenarios, thereby quantifying flexibility demand and fully considering net load fluctuations under each scenario. Grid flexibility demand is quantified from typical scenarios, and a flexible adjustment characteristic model of the power supply side and demand side is established to quantify the flexibility supply capacity of each resource. Based on the flexibility supply and demand balance constraint, a data-driven optimal dispatch model is established with the objective of minimizing the sum of the power supply side consumption cost of the grid, the discharge loss compensation cost of the power supply side of the grid, and the penalty cost corresponding to the grid flexibility deficit. The model is solved to obtain the optimal dispatch result of the grid for the power supply side and demand side. Finally, the grid flexibility margin can be calculated based on the optimal dispatch result. Based on this invention, dispatchers can switch and adjust flexibility resources according to the flexibility margin of each time period, and respond more flexibly to power fluctuations that occur within a day or day. It has important application and reference value for scientific research institutions and power grid dispatching agencies in dealing with the supply and demand balance of power grid flexibility. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart illustrating the data-driven power grid flexibility margin quantification method of the present invention.

[0036] Figure 2 This is a schematic diagram illustrating the flexibility requirements of the present invention.

[0037] Figure 3 This is a schematic diagram of the data-driven power grid flexibility margin quantification system of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] The purpose of this invention is to provide a data-driven method and system for quantifying grid flexibility margin, which has a certain degree of universality for new power systems dominated by new energy sources and has practical significance for quantifying grid flexibility margin at different stages of development.

[0040] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] Example 1

[0042] like Figure 1 As shown, this embodiment provides a data-driven method for quantifying grid flexibility margin, including:

[0043] Step 100: Obtain historical wind power generation dataset and historical photovoltaic power generation dataset.

[0044] Step 200: Based on Copula theory, construct a spatiotemporal correlation set of wind and solar power output according to the historical wind power generation dataset and the historical photovoltaic power generation dataset.

[0045] Specifically, Copula theory can describe the complex, nonlinear correlations between random variables of wind and light. If the random variables of wind and light (X,Y) follow marginal distributions F(x) and G(y), then there exists a unique Copula function C(·) such that H(x,y)=C(F(x),G(y)).

[0046] Here, H(·) is the joint bivariate distribution function of the wind and solar random variables (X,Y) with marginal distributions F(x) and G(y). Taking the partial derivative of H(·) yields the joint probability density function of the wind and solar random variables (X,Y); X represents the wind power random variable, and Y represents the photovoltaic random variable. Random variables refer to all values ​​taken within the sampling period.

[0047] h(x,y)=c(F(x),G(y))(f(x),g(y)).

[0048] Where c(·) is the Copula probability density function, and f(x) and g(y) are the probability density functions of the random variable (X,Y).

[0049] The selection of the Copula function is related to the accuracy of the wind and solar power output set. Common Copula functions used to describe the correlation between wind and solar power include the normal Cupula function, the Frank-Copula function, and the t-Copula function. The Euclidean distance method is used to determine their goodness of fit, thereby selecting the optimal Copula function.

[0050] Let (x) i ,yi Let F(i) = 1, 2, ..., n be a sample of a two-dimensional random variable. n (x i ) and G n (y i Let and represent the empirical distribution functions of the two random variables, respectively. The empirical Copula function of the sample can be expressed as:

[0051]

[0052] Where, u=F(x) i ), v = G(y i ); u = F(x) i ), v = G(y i ), representing the cumulative probability distribution of the sample (x,y); I [·] For a function, when F n (x i When )≤u, there exists otherwise Similarly.

[0053] The optimal Copula function is determined by the squared Euclidean distance, which is defined as:

[0054]

[0055] Among them, u i =F n (x i ), v i =G n (y i );C e (·) represents the empirical Copula function.

[0056] The magnitude of the selected squared Euclidean distance reflects the closeness of various Copula function models to the empirical Copula function; the smaller the value, the better the fit of the selected Copula function. The selection of the Copula function is related to the accuracy of the wind and solar power output set. Considering the complementary characteristics between wind and solar power outputs, the Frank-Copula function is chosen for description.

[0057] Based on the above, step 200 specifically includes:

[0058] 1) Based on the kernel density estimation method, the probability density of wind power output in each time period is calculated using the historical wind power generation dataset, and the probability density of photovoltaic power output in each time period is calculated using the historical photovoltaic power generation dataset. Specifically, the following kernel density estimation function is used to calculate the probability density f of wind power output in each time period. W,t (x W,tPhotovoltaic power output probability density f PV,t (x PV,t ):

[0059]

[0060]

[0061] 2) For any given time period, determine the wind power distribution function based on the wind power output probability density, and determine the photovoltaic distribution function based on the photovoltaic output probability density; specifically, calculate the wind power distribution function F using the following formula. W,t (x W,t ), photovoltaic distribution function F PV,t (x PV,t ):

[0062]

[0063]

[0064] Where n represents the sample size; h represents the smoothing coefficient; and K(·) represents the kernel function, which is generally a symmetric unimodal probability density function.

[0065] 3) Based on the Frank-Copula function, construct the wind-solar joint probability distribution function H(x) according to the wind power distribution function and the photovoltaic distribution function. PV ,x W ).

[0066] 4) Generate a set of random values ​​for the marginal distribution function of wind and solar variables using the aforementioned wind-solar joint probability distribution function; specifically, generate a set of random values ​​{u} for the marginal distribution function of wind and solar variables with spatiotemporal correlation using the following formula. PV,t ,u W,t}:

[0067]

[0068] Among them, a n A random number between 0 and 1.

[0069] 5) For the random numerical set of the marginal distribution function of the wind and solar variables, the inverse function is used to obtain the spatiotemporal correlation set of wind and solar power output, as shown in the following formula:

[0070]

[0071] Step 300: The spatiotemporal correlation set of wind and solar power output is reduced using a scenario reduction method based on probabilistic distance to obtain typical net load scenarios. This involves quantifying grid flexibility requirements using both interval and scenario methods. Considering the time-scale characteristics, directionality, and state dependence of flexibility, grid flexibility requirements are quantified from typical scenarios. The spatiotemporal correlation set of wind and solar power output includes multiple scenarios, each containing both wind power output data and solar power output data.

[0072] Step 300 specifically includes:

[0073] 1) Calculate the Euclidean distance d(s) between any two scenes in the spatiotemporal correlation set of the solar and wind power output. (n) ,s (m) ).

[0074] 2) For each scenario, based on the probability p of the scenario's occurrence. (n) The probability distance corresponding to the scene is determined by the Euclidean distance d between the scene and any other scene in the spatiotemporal correlation set of the solar power output.

[0075] 3) Calculate the sum of probability distances corresponding to the scenario based on the multiple probability distances corresponding to the scenario.

[0076] 4) Select and remove the scenario with the smallest probability distance sum from the multiple scenarios. Specifically, the scenario with the smallest probability distance sum can be selected according to the following formula:

[0077]

[0078] Among them, s( k′ ) represents the scenario with the smallest probability distance, and K represents the number of scenarios.

[0079] 5) Change the occurrence probability of the marked scene; the marked scene is the scene with the smallest probability distance to the scene being removed.

[0080] 6) Determine whether the number of scenarios after removing the scenario with the smallest probability distance meets the preset value requirement; if the preset value requirement is not met, return to step 1) above; continuously iterate and remove scenarios by calculating the "minimum probability distance", and the remaining scenarios are the typical scenarios. That is, if the preset value requirement is met, then all scenarios after removing the scenario with the smallest probability distance are determined as typical scenarios of net load.

[0081] Step 400: Based on the typical net load scenario, establish a power grid flexibility demand model.

[0082] For the k-th typical net load scenario, the load model considering the maximum fluctuation error can be expressed as:

[0083]

[0084] in, This represents the predicted load at time t under the typical scenario of the k-th net load; and Let ε represent the upper and lower limits of the fluctuation at time t under the typical scenario of the k-th net load, respectively; L This represents the maximum prediction error coefficient for the load.

[0085] like Figure 2 As shown, under a given time scale τ, the flexibility requirements for a typical scenario of the k-th net load can be divided into three cases:

[0086] 1) When At that time, there is only a need for upward flexibility.

[0087] 2) When At that time, there is only a need for downward flexibility.

[0088] 3) When At that time, there is a need for two-way flexibility, that is, there is a need for both upward and downward flexibility.

[0089] Considering the time coupling relationship, load fluctuations can be represented by the first-order difference between adjacent time periods, as follows:

[0090]

[0091] The quantification process for the flexibility demand generated by wind and solar power is similar to that for load, but it is important to note the difference between the signs of the upstream and downstream flexibility demands and the load. Therefore, the grid flexibility demand model is as follows:

[0092]

[0093] in, and Let them represent the upward and downward flexibility requirements generated by the net load at time t under the typical scenario of the k-th net load; and These represent the upward and downward flexibility requirements generated by the load at time t under the typical scenario of the k-th net load; and These represent the upward and downward flexibility demands of photovoltaic power output at time t under the typical scenario of the k-th net load; and These represent the upward and downward flexibility requirements of wind power output at time t, respectively, under the typical scenario of the k-th net load.

[0094] Step 500: Establish a power supply-side flexibility supply model and a demand-side flexibility supply model for the power grid to quantify the flexibility supply capacity of each resource. When the power grid generates flexibility demand due to random fluctuations in net load, it is necessary to mobilize the adjustment capabilities of various flexibility resources to provide flexibility in order to meet the peak shaving and ramp-up requirements of net load.

[0095] Thermal power units will serve as the power source flexibility supply for the power grid, while electric vehicles will serve as the demand-side flexibility supply for the power grid.

[0096] The power supply flexibility model is as follows:

[0097]

[0098]

[0099] in, and These represent the upward and downward flexibility supply of thermal power units at any given time. and These represent the upward and downward ramp rates of the thermal power unit, respectively. and P represents the maximum and minimum technical output of the thermal power unit, respectively. G,t τ represents the technical output of the thermal power unit at a given moment; τ is the time scale.

[0100] The demand-side flexibility supply model is as follows:

[0101]

[0102]

[0103] in, and P represents the upward and downward flexibility supply of the cluster of electric vehicles at time t, respectively; EV,t E represents the charging and discharging power of the cluster of electric vehicles at time t. EV,t This represents the battery level of the cluster of electric vehicles at time t; and Let represent the maximum and minimum charging / discharging power of the cluster of electric vehicles at time t, respectively. and Let represent the maximum and minimum battery levels of the cluster of electric vehicles at time t, respectively.

[0104] Step 600: Based on the grid flexibility demand model, the power source flexibility supply model, and the demand-side flexibility supply model, calculate the grid flexibility deficit; the grid flexibility deficit represents the flexibility supply and demand balance constraint. The grid flexibility deficit includes the grid upward flexibility deficit and the grid downward flexibility deficit.

[0105] Step 600 specifically includes:

[0106] 1) Calculate the grid upward flexibility supply at time t based on the upward flexibility supply of the thermal power unit at time t and the upward flexibility supply of the cluster electric vehicle at time t.

[0107] 2) Calculate the grid down-flexibility supply at time t based on the down-flexibility supply of thermal power units at time t and the down-flexibility supply of clustered electric vehicles at time t.

[0108] 3) Based on the power grid flexibility demand model, determine the upward and downward flexibility demands generated by the net load at time t.

[0109] 4) Determine the grid upward flexibility deficit based on the grid upward flexibility supply and the upward flexibility demand generated by the net load.

[0110] 5) Determine the grid down-flexibility deficit based on the grid down-flexibility supply and the down-flexibility demand generated by the net load.

[0111] Specifically, firstly, by integrating the flexibility supply capabilities of thermal power units and electric vehicles, the total flexibility supply capability of the power grid can be obtained. The calculation process is as follows:

[0112]

[0113]

[0114] in, and These represent the upward and downward flexibility supply of the power grid at time t, respectively.

[0115] Secondly, the grid flexibility margin can be expressed as the difference between flexibility demand and flexibility supply. When demand exceeds supply, a flexibility deficit will occur, as shown in the following formula:

[0116]

[0117]

[0118]

[0119] in, and These represent the upward and downward flexibility margins of the power grid, respectively. and These represent the upward and downward flexibility deficits of the power grid at time t, respectively.

[0120] Under the constraint of supply and demand balance for flexibility, the participation of load-side resources such as electric vehicles in demand response provides a certain ramp-up capacity, effectively reducing net load fluctuations and improving the grid's capacity to accommodate renewable energy and its ability to respond to emergencies. Considering the flexibility of demand-side resources, the overall flexibility of the grid is improved, and its flexibility margin is significantly enhanced.

[0121] Step 700: Based on the aforementioned flexibility supply and demand balance constraint, a data-driven optimization scheduling model is established with the objective of minimizing the sum of the power grid's power source consumption cost, the power grid's demand-side discharge loss compensation cost, and the penalty cost corresponding to the power grid's flexibility deficit.

[0122] The data-driven optimization scheduling model is specifically as follows:

[0123]

[0124]

[0125] Among them, C G,t.k C EV,t.k C represents the coal consumption cost of thermal power units and the discharge loss compensation cost of clustered electric vehicles at time t under the typical scenario of the k-th net load of the power grid; defc,t,k P represents the penalty cost corresponding to the grid flexibility deficit at time t under the typical scenario of the k-th net load; a, b, and c are the cost coefficients of thermal power units; G,t,k K represents the technical output of a thermal power unit at time t under the typical net load scenario of the kth time. EV To compensate for the unit cost of discharge of electric vehicles in the cluster; K represents the discharge power of the cluster of electric vehicles at time t under a typical scenario with the kth net load; defc For flexibility deficit penalty factor; π k This represents the probability of the k-th typical net load scenario occurring. and These represent the upward and downward flexibility deficits of the power grid at time t, respectively.

[0126] Step 800: Solve the data-driven optimization scheduling model to obtain the optimal scheduling results of the power grid for the power supply side and the demand side. The optimal scheduling results of the power grid for the power supply side and the demand side include the technical output allocated by the power grid to the power supply side and the charging and discharging power allocated by the power grid to the demand side.

[0127] Step 900: Calculate the grid flexibility margin based on the optimal scheduling result. Specifically, calculate the total grid flexibility supply using the flexibility supply formulas for thermal power units and electric vehicles given above, and then compare it with the grid flexibility demand to obtain the grid flexibility margin.

[0128] Preferably, the method for quantifying grid flexibility margin further includes introducing a flexibility adjustment factor to evaluate the contribution of different flexibility resources to ensuring grid flexibility balance. Specifically:

[0129] 1) Based on the power supply flexibility supply model, calculate the upward flexibility supply and downward flexibility supply of the power supply side according to the technical output allocated by the power grid to the power supply side.

[0130] 2) Based on the demand-side flexibility supply model, and based on the charging and discharging power allocated by the power grid to the demand side, calculate the upward flexibility supply and downward flexibility supply on the demand side.

[0131] 3) Based on the power grid flexibility demand model, determine the upward and downward flexibility demands generated by the net load.

[0132] 4) Calculate the upward flexibility adjustment factor on the power supply side based on the upward flexibility supply on the power supply side and the upward flexibility demand generated by the net load; calculate the downward flexibility adjustment factor on the power supply side based on the downward flexibility supply on the power supply side and the downward flexibility demand generated by the net load.

[0133] 5) Calculate the demand-side upward flexibility adjustment factor based on the upward flexibility supply on the demand side and the upward flexibility demand generated by the net load; calculate the demand-side downward flexibility adjustment factor based on the downward flexibility supply on the demand side and the downward flexibility demand generated by the net load.

[0134] Furthermore, the flexible adjustment factor can be calculated using the following formula:

[0135]

[0136] in, and These are the upward and downward flexibility adjustment factors for various flexibility resources, representing the proportion of the grid flexibility demand that a flexibility resource undertakes at any given time, reflecting the degree of contribution of different flexibility resources to ensuring the grid flexibility balance. For example, the upward flexibility adjustment factor on the power source side characterizes the degree of contribution of upward flexibility resources on the power source side to the grid flexibility balance.

[0137] Example 2

[0138] like Figure 3As shown, in order to implement the technical solution in Embodiment 1, this embodiment provides a data-driven power grid flexibility margin quantification system, characterized in that the system includes:

[0139] The dataset acquisition module 101 is used to acquire historical wind power generation datasets and historical photovoltaic power generation datasets.

[0140] The wind and solar power output spatiotemporal correlation set construction module 201 is used to construct a wind and solar power output spatiotemporal correlation set based on Copula theory and the historical wind power generation dataset and the historical photovoltaic power generation dataset.

[0141] The typical net load scenario construction module 301 is used to reduce the spatiotemporal correlation set of wind and solar power output using a scenario reduction method based on probabilistic distance to obtain typical net load scenarios.

[0142] The power grid demand model construction module 401 is used to establish a power grid flexibility demand model based on the typical net load scenario.

[0143] The supply model construction module 501 is used to establish the power supply flexibility supply model and the demand-side flexibility supply model of the power grid.

[0144] The grid flexibility deficit calculation module 601 is used to calculate the grid flexibility deficit based on the grid flexibility demand model, the power source flexibility supply model, and the demand side flexibility supply model; the grid flexibility deficit represents the flexibility supply and demand balance constraint.

[0145] The data-driven optimization scheduling model construction module 701 is used to establish a data-driven optimization scheduling model based on the flexibility supply and demand balance constraints, with the goal of minimizing the sum of the power supply side consumption cost of the power grid, the demand side discharge loss compensation cost of the power grid, and the penalty cost corresponding to the flexibility deficit of the power grid.

[0146] The optimal scheduling determination module 801 is used to solve the data-driven optimal scheduling model to obtain the optimal scheduling results of the power grid for the power supply side and the demand side.

[0147] The power grid flexibility margin calculation module 901 is used to calculate the power grid flexibility margin based on the optimal scheduling result.

[0148] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0149] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A data-driven method for quantifying power grid flexibility margin, characterized in that, The methods include: Obtain historical wind power generation datasets and historical photovoltaic power generation datasets; Based on Copula theory, a spatiotemporal correlation set of wind and solar power output is constructed according to the historical wind power generation dataset and the historical photovoltaic power generation dataset. The scene reduction method based on probabilistic distance is used to reduce the spatiotemporal correlation set of wind and solar power output to obtain typical net load scenarios. Based on the aforementioned typical net load scenario, a power grid flexibility demand model is established; The power grid flexibility requirement model is as follows: in, and They represent the first In a typical net load scenario The upward and downward flexibility requirements generated by the net load at any given moment; and They represent the first In a typical net load scenario The upward and downward flexibility requirements generated by the load at any given moment; and They represent the first In a typical net load scenario The upward and downward flexibility demands generated by photovoltaic power output at all times; and They represent the first In a typical net load scenario The upward and downward flexibility requirements generated by wind power output at all times; and The calculation formula is as follows: in, Indicates the first In a typical net load scenario The predicted load at a given time; and They represent the first In a typical net load scenario The upper limit and lower limit of the fluctuation at any given time; The maximum prediction error coefficient for the load; Establish a power supply flexibility model for the power grid and a demand-side flexibility supply model for the power grid; Thermal power units will be used as the power source flexibility supply for the power grid, while electric vehicles will be used as the demand side flexibility supply for the power grid. The power supply flexibility model is as follows: in, and These respectively represent the thermal power units at The supply of upward and downward flexibility at any given moment; and These represent the upward and downward ramp rates of the thermal power unit, respectively. and These represent the maximum and minimum technical output of the thermal power unit, respectively. Indicates that thermal power units are in Technological contributions at all times; Time scale; The demand-side flexibility supply model is as follows: in, and These represent the cluster of electric vehicles in The supply of upward and downward flexibility at any given moment; Indicates that cluster electric vehicles are in The charging and discharging power at any given time Indicates that cluster electric vehicles are in Battery level at any given moment; and These represent the cluster of electric vehicles in The maximum and minimum charge / discharge power at any given time; and These represent the cluster of electric vehicles in Maximum and minimum battery level at any given time; Based on the power grid flexibility demand model, the power source flexibility supply model, and the demand-side flexibility supply model, the power grid flexibility deficit is calculated; the power grid flexibility deficit represents the flexibility supply and demand balance constraint. Based on the aforementioned flexibility supply and demand balance constraint, a data-driven optimization scheduling model is established with the objective of minimizing the sum of the power grid's power source consumption cost, the power grid's demand-side discharge loss compensation cost, and the penalty cost corresponding to the power grid's flexibility deficit. The data-driven optimization scheduling model is solved to obtain the optimal scheduling results of the power grid for the power supply side and the demand side. The power grid flexibility margin is calculated based on the optimal scheduling results.

2. The data-driven method for quantifying grid flexibility margin according to claim 1, characterized in that, Based on Copula theory, a spatiotemporal correlation set of wind and solar power output is constructed according to the historical wind power generation dataset and the historical photovoltaic power generation dataset, specifically including: Based on the kernel density estimation method, the probability density of wind power output in each time period is calculated according to the historical wind power generation dataset, and the probability density of photovoltaic power output in each time period is calculated according to the historical photovoltaic power generation dataset. For any given time period, the wind power distribution function is determined based on the wind power output probability density, and the photovoltaic distribution function is determined based on the photovoltaic output probability density. Based on the Frank-Copula function, a joint wind-solar probability distribution function is constructed according to the wind power distribution function and the photovoltaic distribution function; A random numerical set of marginal distribution functions for wind and solar variables is generated using the aforementioned wind-solar joint probability distribution function; The random numerical set of the marginal distribution function of the wind and solar variables is inverted using the inverse function to obtain the spatiotemporal correlation set of wind and solar power output.

3. The data-driven method for quantifying grid flexibility margin according to claim 1, characterized in that, The spatiotemporal correlation of wind and solar power output includes multiple scenarios, each of which includes wind power output data and solar power output data; The method of scene reduction based on probabilistic distance is used to reduce the spatiotemporal correlation set of wind and solar power output to obtain typical net load scenarios, specifically including: Calculate the Euclidean distance between any two scenes in the spatiotemporal correlation set of the solar and wind power output; For each scenario, the probability distance corresponding to the scenario is determined based on the occurrence probability of the scenario and the Euclidean distance between the scenario and any other scenario in the spatiotemporal correlation set of the wind and solar power output. Calculate the sum of probability distances corresponding to the scenario based on the multiple probability distances corresponding to the scenario; The scenario with the smallest probability distance sum is selected from multiple scenarios and then eliminated. Change the occurrence probability of the marked scene; the marked scene is the scene with the smallest probability distance to the scene being removed; Determine whether the number of scenes after removing the scene with the smallest probability distance meets the preset value requirement; If the preset numerical requirements are not met, return to the step of calculating the Euclidean distance between any two scenes in the spatiotemporal correlation set of the wind and solar power output; If the preset numerical requirements are met, all scenarios after removing the scenario with the smallest probability distance will be determined as typical net load scenarios.

4. The data-driven method for quantifying grid flexibility margin according to claim 1, characterized in that, The grid flexibility deficit includes the grid upward flexibility deficit and the grid downward flexibility deficit; Based on the aforementioned grid flexibility demand model, the power source flexibility supply model, and the demand-side flexibility supply model, the grid flexibility deficit is calculated, specifically including: According to the aforementioned thermal power unit The upward flexibility supply at any time and the cluster of electric vehicles in The upward flexibility of the supply of time, calculation The power grid's upward flexibility supply at all times; According to thermal power units Downward flexibility supply and cluster electric vehicles in time Downward flexibility supply in real time, calculation The power grid's flexible supply at any time; Based on the aforementioned power grid flexibility demand model, it is determined that in The upward and downward flexibility requirements generated by the net load at any given moment; The grid upward flexibility deficit is determined based on the grid upward flexibility supply and the upward flexibility demand generated by net load. The grid down-flexibility deficit is determined based on the grid down-flexibility supply and the down-flexibility demand generated by the net load.

5. The data-driven method for quantifying grid flexibility margin according to claim 1, characterized in that, Thermal power units are used as the power source side for the grid's flexibility supply, and the power source side consumption cost is the coal consumption cost of the thermal power units; electric vehicles are used as the demand side for the grid's flexibility supply, and the demand side discharge loss compensation cost is the discharge loss compensation cost of the electric vehicle cluster; the grid flexibility deficit includes the grid's upward flexibility deficit and the grid's downward flexibility deficit. The data-driven optimization scheduling model is specifically as follows: in, , These represent the power grid at the [number]th [year]. A typical net load scenario The coal consumption cost of thermal power units at any given time, and the compensation cost for discharge loss of cluster electric vehicles; For the power grid in the first A typical net load scenario The penalty cost corresponding to the grid flexibility deficit at any given moment; , , This represents the cost coefficient for thermal power units. Indicates the thermal power unit in the first A typical net load scenario Technology output at all times To compensate for the unit cost of discharge of electric vehicles in the cluster; For cluster electric vehicles in the first In a typical net load scenario Discharge power at any given moment; A penalty factor for flexibility deficit; For the first The probability of a typical net load scenario occurring; and They represent in The grid's upward and downward flexibility deficits at any given time.

6. The data-driven method for quantifying grid flexibility margin according to claim 1, characterized in that, The optimal scheduling results of the power grid for the power supply side and the demand side include the technical output allocated by the power grid to the power supply side and the charging and discharging power allocated by the power grid to the demand side. The method for quantifying grid flexibility margin also includes: Based on the power supply flexibility supply model, the upward flexibility supply and downward flexibility supply of the power supply side are calculated according to the technical output allocated by the power grid to the power supply side. Based on the demand-side flexibility supply model, and based on the charging and discharging power allocated by the power grid to the demand side, the upward flexibility supply and downward flexibility supply on the demand side are calculated. Based on the power grid flexibility demand model, determine the upward and downward flexibility demands generated by the net load; Calculate the upward flexibility adjustment factor on the power supply side based on the upward flexibility supply on the power supply side and the upward flexibility demand generated by the net load; calculate the downward flexibility adjustment factor on the power supply side based on the downward flexibility supply on the power supply side and the downward flexibility demand generated by the net load. Calculate the upward flexibility adjustment factor on the demand side based on the upward flexibility supply on the demand side and the upward flexibility demand generated by the net load; calculate the downward flexibility adjustment factor on the demand side based on the downward flexibility supply on the demand side and the downward flexibility demand generated by the net load. The power supply-side flexible adjustment factor characterizes the degree of contribution of power supply-side flexible resources to the grid flexibility balance.

7. A data-driven grid flexibility margin quantification system, employing the data-driven grid flexibility margin quantification method according to any one of claims 1-6, characterized in that, The system includes: The dataset acquisition module is used to acquire historical wind power generation datasets and historical photovoltaic power generation datasets; The wind and solar power output spatiotemporal correlation set construction module is used to construct the wind and solar power output spatiotemporal correlation set based on the Copula theory and the historical wind power generation dataset and the historical photovoltaic power generation dataset. The typical net load scenario construction module is used to reduce the spatiotemporal correlation set of wind and solar power output using a scenario reduction method based on probabilistic distance to obtain typical net load scenarios. The power grid demand model construction module is used to establish a power grid flexibility demand model based on the typical net load scenario. The supply model construction module is used to establish the power supply flexibility supply model and the demand-side flexibility supply model of the power grid. The grid flexibility deficit calculation module is used to calculate the grid flexibility deficit based on the grid flexibility demand model, the power source flexibility supply model, and the demand side flexibility supply model; the grid flexibility deficit represents the flexibility supply and demand balance constraint. The data-driven optimization scheduling model construction module is used to establish a data-driven optimization scheduling model based on the flexibility supply and demand balance constraints, with the goal of minimizing the sum of the power supply side consumption cost of the power grid, the demand side discharge loss compensation cost of the power grid, and the penalty cost corresponding to the flexibility deficit of the power grid. The optimal scheduling determination module is used to solve the data-driven optimal scheduling model to obtain the optimal scheduling results of the power grid for the power supply side and the demand side. The power grid flexibility margin calculation module is used to calculate the power grid flexibility margin based on the optimal scheduling result.

Images