A method and system for optimizing electric energy dispatch considering random fluctuations on both sides of source and load

Abstract

The invention discloses an electric energy dispatch optimization method and system considering random fluctuations on both sides of source and load, and specifically relates to the technical field of electric energy dispatch. The technical key points are: performing K-means clustering on a new energy feature set and a load feature set, and optimizing the clustering results, and finally obtaining different clustering scenarios; constructing a source-load double-sided random fluctuation probability model that characterizes the random fluctuations on both sides of source and load under different clustering scenarios; constructing a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations based on historical new energy power generation data and historical load data and combining a multivariate VAR model; constructing an electric energy dispatch model for an electric power system based on pre-constructed constraint conditions and taking the maximum consumption of new energy as an objective function; adjusting the electric energy dispatch model for an electric power system through the source-load double-sided random fluctuation probability model and the dynamic coupling model, and solving the adjusted electric energy dispatch model for an electric power system to obtain an electric energy dispatch plan for the electric power system.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power dispatching, and in particular to an electric energy dispatching optimization method and system taking into account random fluctuations on both sides of a source and a load. Background Art

[0002] Against the backdrop of global energy transformation, the traditional fossil fuel-based power supply system is gradually transforming into a new clean, low-carbon, diversified and complementary power system, and new energy power generation technology has been developed. At present, there are also some uncertainties in the access of new energy to the power system, mainly including the uncertainty of new energy power generation and the uncertainty of user electricity consumption. The uncertainty of new energy power generation is mainly reflected in photovoltaic power generation and wind power generation. Photovoltaic power generation relies on solar panels to convert light energy into electrical energy, which is easily affected by lighting conditions; wind power generation relies on wind power to drive the blades of the wind turbine to rotate and generate electricity, and there are natural instabilities in wind speed and wind direction. The uncertainty of user electricity consumption is mainly reflected in the random changes of residential users and the frequent sudden changes in electricity load; it is also reflected in the limitations of industrial electricity consumption, which is easily restricted by production processes and order conditions.

[0003] In addition, the power supply side and the load side are closely connected through the power grid. The fluctuations on one side will be quickly transmitted to the other side, not only affecting each other at the physical level, but also at the power distribution level. This will lead to unreasonable scheduling, making it difficult to absorb new energy, wasting clean energy, and making the power system unstable.

[0004] Therefore, the present invention aims to provide an electric energy scheduling optimization method and system taking into account random fluctuations on both sides of the source and the load, so as to solve the above-mentioned related problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is that random fluctuations on both sides of the source and the load will affect the dispatch of the power system, causing difficulties in the consumption of new energy, wasting clean energy and making the operation of the power system unstable. The purpose is to provide an electric energy dispatch optimization method and system considering random fluctuations on both sides of the source and the load, by analyzing the random fluctuation characteristics of both sides of the source and the load in different scenarios, to construct a random fluctuation probability model of both sides of the source and the load that characterizes the random fluctuations of both sides of the source and the load in different clustering scenarios, which can better describe the random fluctuations of both sides of the source and the load, and provide a more accurate data basis for the dispatch of the power system; then, based on historical new energy power generation data and historical load data, a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations is constructed in combination with a multivariate VAR model; and through the random fluctuation probability model of both sides of the source and the load and the dynamic coupling model, the electric energy dispatch model of the power system with the maximum consumption of new energy as the goal is adjusted, so as to optimize the influence of the random fluctuations of both sides of the source and the load and the coupling correlation characteristics on the electric energy dispatch of the power system, and improve the comprehensiveness and stability of the dispatch.

[0006] The present invention is achieved through the following technical solutions:

[0007] An energy dispatch optimization method considering random fluctuations on both sides of source and load, the method comprising:

[0008] Obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios;

[0009] Based on the new energy feature set and load feature set and combined with the probability density distribution function, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation in different clustering scenarios.

[0010] Based on historical renewable energy power generation data and historical load data, a dynamic coupling model is constructed to characterize the dynamic coupling relationship between renewable energy fluctuations and load fluctuations in combination with a multivariate VAR model.

[0011] Based on the pre-built constraints, the power system power dispatch model is constructed with the maximum consumption of new energy as the objective function;

[0012] The electric energy dispatching model of the power system is adjusted through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and the adjusted electric energy dispatching model of the power system is solved to obtain the electric energy dispatching plan of the power system.

[0013] Furthermore, historical renewable energy power generation data and historical load data are obtained, wherein the historical renewable energy power generation data include solar power generation data and wind power generation data; the historical load data include historical industrial load data, historical commercial load data and historical residential load data.

[0014] Furthermore, based on the pre-built different scenario types, new energy feature sets and load feature sets are extracted, specifically:

[0015] Acquire meteorological data and special event data to construct different scene types, including sunny scene, cloudy scene, rainy scene, extreme weather scene and special event scene;

[0016] For each scenario type, multiple new energy characteristics are extracted to construct a new energy characteristic set, and multiple load characteristics are extracted to construct a load characteristic set, wherein the multiple new energy characteristics include solar power generation power characteristics and wind power generation power characteristics; the multiple load characteristics include industrial load power characteristics, commercial load power characteristics and residential load power characteristics.

[0017] Furthermore, based on the new energy feature set and the load feature set and combined with the probability density distribution function, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation in different clustering scenarios. Specifically, it is:

[0018] Under different scene types, a normal distribution function, a lognormal distribution function, a Bernoulli distribution function or a truncated normal distribution function is used to describe the probability density distribution of solar power generation characteristics, and multiple solar power generation power fluctuation probability functions are obtained respectively;

[0019] Under different scenario types, the probability density distribution of wind power characteristics is described using one of the Weibull distribution function, the two-parameter gamma distribution function or the generalized Pareto distribution function, and multiple wind power fluctuation probability functions are obtained respectively;

[0020] Under different scene types, a normal distribution function, a mixed Gaussian distribution function or a Poisson distribution function is used to describe the probability density distribution of the residential load power characteristics, and multiple residential load power fluctuation probability functions are obtained respectively;

[0021] Under different scenario types, the normal distribution function is used to describe the probability density distribution of industrial load power characteristics, and multiple industrial load power fluctuation probability functions are obtained respectively;

[0022] Combining multiple solar power generation power fluctuation probability functions, multiple wind power generation power fluctuation probability functions, multiple residential load power fluctuation probability functions and multiple industrial load power fluctuation probability functions, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation under different clustering scenarios.

[0023] Furthermore, based on the historical renewable energy power generation data and historical load data, a dynamic coupling model is constructed in combination with the multivariate VAR model to characterize the dynamic coupling relationship between renewable energy fluctuations and load fluctuations. Specifically, the dynamic coupling model is:

[0024]

[0025] Among them, P solar (t), P solar (t-1) are the solar power generation power at time t or time t-1; P wind (t), P wind (t-1) are the wind power generation power at time t or time t-1 respectively; P res (t), P res (t-1) is the total power consumption of residential users at time t or time t-1 respectively; P ind (t), P ind(t-1) is the total power consumption of industrial users at time t or time t-1; a ij (i=1,2,3,4,j=1,2,3,4) are model coefficients; ∈ 1t ,∈ 2t ,∈ 3t and ∈ 4t They are white noise sequences corresponding to the solar power generation power, wind power generation power, residential power load power and industrial power load power functions respectively.

[0026] Furthermore, the objective function is specifically:

[0027]

[0028] Among them, E NA represents the consumption of new energy; T is the total number of time periods in the dispatch cycle; P N,t is the actual power consumption of renewable energy generation at time t; C penalty is the penalty cost item.

[0029] The present invention also provides an electric energy dispatch optimization system considering random fluctuations on both sides of the source and the load, the system is used in any one of the above-mentioned electric energy dispatch optimization methods considering random fluctuations on both sides of the source and the load, the system comprising:

[0030] The feature clustering module is used to obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios;

[0031] A probability model building module is used to build a source-load double-sided random fluctuation probability model that characterizes the source-load double-sided random fluctuation in different clustering scenarios based on the new energy feature set and the load feature set and combined with the probability density distribution function;

[0032] Dynamic coupling model building module, which is used to build a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations based on historical new energy power generation data and historical load data and combined with a multivariate VAR model;

[0033] The dispatch model building module is used to build a power system power dispatch model based on pre-built constraints and taking the maximum consumption of new energy as the objective function;

[0034] The dispatching scheme generation module is used to adjust the electric energy dispatching model of the power system through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and solve the adjusted electric energy dispatching model of the power system to obtain the electric energy dispatching scheme of the power system.

[0035] The present invention also provides a computer device, comprising a system memory and a processor, wherein the system memory stores a computer program, and the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0036] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods described above are implemented.

[0037] The present invention also provides a computer program product comprising instructions, and when the instructions are executed by a computer device cluster, the computer device cluster executes any of the above methods.

[0038] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0039] In the present invention, by analyzing the random fluctuation characteristics of both sides of the source and load under different scenarios, a random fluctuation probability model of both sides of the source and load that characterizes the random fluctuation of both sides of the source and load under different clustering scenarios is constructed, which can better describe the random fluctuation of both sides of the source and load, and provide a more accurate data basis for power system scheduling; then, based on historical new energy power generation data and historical load data, a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations is constructed in combination with a multivariate VAR model; and through the random fluctuation probability model of both sides of the source and load and the dynamic coupling model, the power system electric energy scheduling model with the maximum new energy consumption as the goal is adjusted, thereby optimizing the influence of the random fluctuations and coupling correlation characteristics of both sides of the source and load on the power system electric energy scheduling, and improving the comprehensiveness and stability of the scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0041] Figure 1 Schematic diagram of a method flow of an electric energy dispatch optimization method considering random fluctuations on both sides of source and load in this embodiment;

[0042] Figure 2 This is a schematic diagram of module connections of an electric energy dispatch optimization system considering random fluctuations on both sides of the source and load in this embodiment;

[0043] Figure 3 It is a structural schematic diagram of a computer device in this embodiment. DETAILED DESCRIPTION

[0044] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0045] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0046] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0047] Example 1

[0048] See also Figure 1 As shown, this embodiment provides an energy scheduling optimization method considering random fluctuations on both sides of the source and the load, and the method includes:

[0049] S1: Obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios;

[0050] It should be noted that the renewable energy power generation data includes various types of power plants, and in this embodiment, it mainly includes solar power generation data and wind power generation data. In other embodiments, it may also include other renewable energy power generation data, which will not be described in detail here; the load data includes the power consumption, electricity consumption, and electricity consumption time of industrial, commercial, and residential users on the load side, and in other embodiments, it may also include other load data, which will not be described in detail here; the data collection frequency can be determined according to actual needs, and data can be collected every 15 minutes, 30 minutes, 60 minutes, or once a day;

[0051] At the same time, it should be noted that after obtaining the historical new energy power generation data and historical load data, the data needs to be preprocessed, including data cleaning and data standardization. This technical means is common knowledge in the field and will not be elaborated here.

[0052] Specifically, in this embodiment, historical renewable energy power generation data and historical load data are obtained, wherein the historical renewable energy power generation data includes solar power generation data and wind power generation data; the historical load data includes historical industrial load data, historical commercial load data and historical residential load data;

[0053] Then obtain meteorological data and special event data to construct different scene types, including sunny scene, cloudy scene, rainy scene, extreme weather scene and special event scene;

[0054] It should be noted that, in this embodiment, the meteorological data obtained include weather type, temperature, wind speed, solar radiation intensity, etc., the acquired meteorological data and special event data are cleaned, abnormal data such as error values ​​and missing values ​​in the data are removed, and the data is standardized so that data with different characteristics have the same dimension.

[0055] For each scenario type, multiple new energy characteristics are extracted to construct a new energy characteristic set, and multiple load characteristics are extracted to construct a load characteristic set, wherein the multiple new energy characteristics include solar power generation power characteristics and wind power generation power characteristics; the multiple load characteristics include industrial load power characteristics, commercial load power characteristics and residential load power characteristics; the solar power generation power characteristics include the mean solar power generation power, the standard deviation of solar power generation power, the maximum solar power generation power and the minimum solar power generation power; the wind power generation power characteristics include the mean wind power generation power, the standard deviation of wind power generation power, the maximum wind power generation power and the minimum wind power generation power; the industrial load power characteristics include the mean industrial load power and the standard deviation of industrial load power; the commercial load power characteristics include the mean commercial load power and the standard deviation of commercial load power; the residential load power characteristics include the mean residential load power and the standard deviation of residential load power.

[0056] K-Means clustering is performed on the new energy feature set and the load feature set. Specifically, according to the number of scene types, the value of the cluster number K is preliminarily set to be equal to the number of scene types; in this embodiment, there are five scene types, so the number of clusters is first set to K=5; K data points are randomly selected from the preprocessed data set as the initial cluster centers; these cluster centers are vectors containing both source and load characteristics, and the cluster center vectors are specifically:

[0057] C i =[μ solar-i ,σsolar-i ,μ wind-i ,...,μ load-type-i ,σ load-type-i ], where C i represents the i-th cluster center vector, i=1,2,.,K; μ solar-i represents the mean solar power generation power of the i-th cluster center; σ solar-i represents the standard deviation of solar power generation at the i-th cluster center; μ wind-i represents the mean wind power generation power of the i-th cluster center; μ load-type-i represents a load power mean of the i-th cluster center; σ load-type-i Represents a load power standard deviation of the i-th cluster center.

[0058] Determine the source-load bilateral eigenvectors corresponding to the acquired data. The data point vectors are specifically:

[0059] X=[x solar ,σ solar ,x wind ,...,x solar-type ,σ solar-type ], where X represents the data point vector; x solar Represents the actual power of solar power generation at a specific point in time; σ solar is the standard deviation, which is used to measure the fluctuation characteristics of solar power generation; x wind is the actual power of wind power generation at that time point; x solar-type is the actual power of this type of load at a specific time point, σ solar-type It indicates the degree of fluctuation of the power of this type of load near that time point. It should be noted that the data point vector composed of these variables can fully describe the status of both sides of the source and the load at a specific time point, including the power generation of new energy and the power consumption of various loads, so that the clustering algorithm can divide the data at different time points into appropriate scene type clusters according to the characteristics of these vectors.

[0060] Calculate X to each cluster center C i Here we use the Euclidean distance, and the calculation formula is as follows:

[0061] Where n is the number of features, x j is the jth feature of data point X, c ij is the cluster center C i The jth feature of .

[0062] According to the calculated distance, the data point X is assigned to the cluster to which the cluster center closest to it belongs. That is, if d(X,C m)=min(d(X,C1),d(X,C2),...,d(X,C k )), then assign X to the mth cluster.

[0063] For each cluster, recalculate its cluster center and update the cluster center of each family. The cluster center is the mean vector of all data points in the cluster. For example, for the data point X in the kth cluster, p (p=1,2,...,N k , N k is the number of data points in the kth cluster), the new cluster center Each feature of is calculated as where j is the feature index.

[0064] Finally, the cluster assignment and updating of cluster centers are repeated until the cluster centers no longer change significantly, that is, the preset maximum number of iterations is reached, and the final different clustering scenarios are obtained.

[0065] S2: Based on the new energy feature set and load feature set and combined with the probability density distribution function, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation in different clustering scenarios;

[0066] It should be noted that in this embodiment, the volatility characteristics of both the source and the load are calculated for each cluster (each clustering scenario type); for the new energy side, such as the volatility of solar power generation, the coefficient of variation is used to represent it: This formula represents the relative size of the standard deviation and the mean; for the load side, the coefficient of variation is also used to calculate the volatility of various types of loads: By comparing the volatility indicators of different clusters, the differences in random fluctuations on both sides of new energy and load in different scenarios are analyzed; for example, by comparing the coefficient of variation of solar power generation in sunny scenes and extreme weather scenes, it can be found that the volatility in extreme weather scenes is much greater than that in sunny scenes.

[0067] Specifically, in this embodiment, a source-load double-side random fluctuation probability model is constructed to characterize the source-load double-side random fluctuation in different clustering scenarios, specifically:

[0068] Under different scene types, a normal distribution function, a lognormal distribution function, a Bernoulli distribution function or a truncated normal distribution function is used to describe the probability density distribution of solar power generation characteristics, and multiple solar power generation power fluctuation probability functions are obtained respectively;

[0069] Specifically, in this embodiment, in a sunny day scenario, solar power generation is relatively stable, so it is described by a normal distribution, specifically:

[0070] Among them, P solar is the solar power generation power, μ solar is the average value of solar power generation, which is determined by factors such as solar radiation intensity and solar panel conversion efficiency. solar is the standard deviation of solar power generation, reflecting fluctuations caused by small factors such as occasional cloud cover;

[0071] In cloudy scenarios, the solar power generation power drops significantly and fluctuates slightly, and the probability distribution can be approximated by a log-normal distribution, specifically:

[0072] Among them, μ solar and σ solar Determined based on the statistical data of solar power generation on cloudy days, the mean reflects the average level of power generation on cloudy days, and the standard deviation reflects the degree of fluctuation;

[0073] In rainy days, the solar power generation is extremely low, close to zero, and can be described by Bernoulli distribution, specifically:

[0074] Let X represent whether the solar power generation power is zero, P(X=k)=p k (1-p) 1-k , where k = 0, 1, and p is the probability of a small amount of power generation on a rainy day (possibly due to factors such as thin clouds or raindrop reflection), obtained through historical data statistics;

[0075] In extreme weather scenarios, solar power generation is likely to drop sharply or even approach zero under extreme weather conditions such as heavy rain, heavy snow or severe sandstorms. It can be approximated by the truncated normal distribution, specifically:

[0076] Assume that the solar power generation power is normally distributed according to N(μ normal ,σ normal ), in extreme weather conditions, the cutoff point is set to P cut , Among them: When P solar <P cut , P(P solar )=0, when P solar >P cut , where Φ is the cumulative distribution function of the standard normal distribution.

[0077] Under different scenario types, the probability density distribution of wind power characteristics is described using one of the Weibull distribution function, the two-parameter gamma distribution function or the generalized Pareto distribution function, and multiple wind power fluctuation probability functions are obtained respectively;

[0078] Specifically, in this embodiment, in a sunny day scenario, the wind speed is relatively stable but low on a sunny day, and the wind power generation power is described by Weibull distribution, specifically:

[0079] Among them, P wind is the wind power generation, k is the shape parameter, c is the scale parameter, the shape parameter determines the shape of the distribution, and the scale parameter is related to the average level and fluctuation range of the power generation;

[0080] In cloudy scenarios, the wind speed may vary, and the fluctuation of wind power generation is greater than that in sunny days. Therefore, a two-parameter gamma distribution is used to describe it, specifically:

[0081] Among them, α and β are parameters obtained by fitting the wind power generation data on cloudy days, and the gamma distribution can better describe this asymmetric fluctuation.

[0082] In rainy days, the wind speed and direction change in a complex way, and the wind power generation fluctuates violently, which can be described by the two-parameter gamma distribution, specifically:

[0083] Among them, α and β are obtained by fitting the wind power generation data on rainy days, and can adapt to the asymmetric fluctuation of power generation.

[0084] In extreme weather scenarios, wind power varies in complex ways under extreme strong winds such as typhoons and hurricanes. Its probability distribution is described by the generalized Pareto distribution (GPD), specifically:

[0085] Among them, when ξ≠0, When ξ=0, μ is the location parameter, σ is the scale parameter, and ξ is the shape parameter, which is determined by statistical analysis of wind power data under extreme strong wind weather.

[0086] At the same time, it should be noted that in special time scenarios, taking large-scale sports events as an example, the probability distribution of power generation on the renewable energy side is not directly affected by special events and is still described according to the new energy power generation distribution corresponding to the weather at that time. However, if the area where the special event is located has special requirements for power supply, certain scheduling adjustments may be made to renewable energy power generation, and the parameters of the probability distribution may change accordingly.

[0087] Under different scene types, a normal distribution function, a mixed Gaussian distribution function or a Poisson distribution function is used to describe the probability density distribution of the residential load power characteristics, and multiple residential load power fluctuation probability functions are obtained respectively;

[0088] Specifically, in this embodiment, in the sunny day scenario, the resident load has a certain regularity on sunny days. Some residents go out during the day, and the load is relatively low. It starts to increase in the evening. Therefore, a combination of multiple normal distributions is used to describe the load conditions at different time periods, specifically:

[0089] Among them, P load-res is the residential load power, μ res1 and σ res1 is the mean and standard deviation of the residential load during the daytime; μ res2 and σ res2 is the mean and standard deviation of the residential load in the evening and night time, μ res2 >μ res1 , because more electrical equipment is used at night, σ res2 It is also relatively large, reflecting the differences in residents’ electricity consumption behavior at night.

[0090] In cloudy scenes, due to the dim indoor light, residents may use more lighting equipment, and the indoor temperature may be low, the probability of using heating equipment increases, and the load power of residents is described by mixed Gaussian distribution, specifically:

[0091] Among them, ω i is the weight of the i-th Gaussian component, μ i and σ i are the mean and standard deviation of the i-th Gaussian component, respectively. For example, there are two Gaussian components, one corresponding to the normal basic load (such as refrigerators, TVs, etc.), and the other corresponding to the electricity consumption of lighting and heating equipment that is turned on extra due to cloudy days.

[0092] In rainy days, residents' outdoor activities decrease, indoor activities increase, and the usage pattern of electrical equipment changes. The load power of residents may obey a mixed model of Poisson distribution and normal distribution. The Poisson distribution part is used to describe the load increase caused by some discrete events, such as the increase in the probability of using washing machines, dryers and other equipment. The specific probability function is:

[0093] Among them: λ is the average number of events per unit time, which is obtained based on the frequency of residents using these devices on rainy days; the normal distribution part is used to describe the power of basic load and other continuously changing electrical equipment, such as lighting, television, etc.

[0094] In extreme weather scenarios, residents' load will change significantly in extreme weather such as severe cold, extreme heat, or rainstorms and floods. Taking severe cold weather as an example, residents use a lot of heating equipment, and the load power may follow a normal distribution N(μ) that is shifted to the right. resident-extreme ,σ resident-extreme ), μ resident-extreme >μresident-normal , μ resident-normal is the normal weather mean, σ resident-extreme For extreme weather such as rainstorms and floods that may affect residents' lives, the load power of residents may show a multi-peak distribution, which can be described by a mixed Gaussian distribution, with multiple Gaussian components corresponding to different power consumption modes.

[0095] Under different scenario types, the normal distribution function is used to describe the probability density distribution of industrial load power characteristics, and multiple industrial load power fluctuation probability functions are obtained respectively;

[0096] In the sunny day scenario, the industrial load on sunny days mainly depends on the production plan. Assuming that the industrial production process is stable, the industrial load power follows a normal distribution, specifically:

[0097] Among them, P load-ind is the industrial load power, μ ind is the mean value of industrial load power, which is determined by the operating power and quantity of production equipment, σ ind is the standard deviation, which reflects the fluctuations caused by small factors such as equipment failure maintenance and product batch changes;

[0098] In cloudy scenarios, the industrial production process is not directly affected by cloudy weather. Its load power distribution may be similar to that on sunny days, but there are still slight changes and it follows a normal distribution, specifically:

[0099] Among them, μ ind-cloudy Slightly higher than the average value on sunny days, σ ind-cloudy There may also be a slight increase, as obtained through statistical analysis of industrial load data on cloudy days;

[0100] In the rainy day scenario, if the enterprise does not stop or reduce production due to the rain, the industrial load power distribution is still mainly normal distribution. However, since rainy days may affect the transportation of raw materials, warehouse management and other links, the operating time and power of some equipment will change. Specifically:

[0101] Among them, μ ind-roin and σ ind-rain Adjust and determine based on the actual production situation and equipment operation of the enterprise on rainy days;

[0102] In extreme weather conditions, for some weather-sensitive industries, extreme weather may cause production suspension or production reduction. When production is suspended, the industrial load power approaches a lower fixed value, represented by the Dirac function δ; when production is reduced but not suspended, the industrial load power obeys a new normal distribution N(μ ind-extreme ,σ ind-extreme), whose mean and standard deviation are determined according to the degree of production reduction and equipment operation. For industrial sectors that are not directly affected by extreme weather, their load power distribution may not change much, but considering that emergency measures will lead to small probability fluctuations, a small probability normal distribution component is added to describe it;

[0103] In special time scenarios, such as during large-scale sports events, the load on the venue and surrounding areas will increase dramatically. Here, a mixed distribution is used to describe that in the event venue area, the load power follows a normal distribution with a very high peak value, specifically:

[0104] Among them, μ event-venue is the average value of the venue load power during the event, which is determined by the event scale, venue facilities, etc., σ event-venue It reflects the fluctuations in the use of equipment in the venue. In the surrounding areas, the load power may obey a mixed distribution of normal distribution and Poisson distribution. Poisson distribution is used to describe the increase in some discrete commercial activities caused by the gathering of spectators, and normal distribution is used to describe the load changes caused by the indirect impact of the event on surrounding residents and commercial facilities.

[0105] Combining multiple solar power generation power fluctuation probability functions, multiple wind power generation power fluctuation probability functions, multiple residential load power fluctuation probability functions and multiple industrial load power fluctuation probability functions, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation under different clustering scenarios.

[0106] S3: Based on the historical renewable energy power generation data and historical load data, a dynamic coupling model is constructed to characterize the dynamic coupling relationship between renewable energy fluctuations and load fluctuations in combination with the multivariate VAR model. Specifically, the dynamic coupling model is as follows:

[0107]

[0108] Among them, P solar (t), P solar (t-1) are the solar power generation power at time t or time t-1; P wind (t), P wind (t-1) are the wind power generation power at time t or time t-1 respectively; P res (t), P res (t-1) is the total power consumption of residential users at time t or time t-1 respectively; P ind (t), P ind (t-1) is the total power consumption of industrial users at time t or time t-1; a ij (i=1,2,3,4,j=1,2,3,4) are model coefficients, for example, a 12Represents the wind power generation sequence; ∈ it (i=1,2,3,4) is a white noise sequence variable, ∈ 1t ,∈ 2t ,∈ 3t and ∈ 4t They are white noise sequences corresponding to the solar power generation power, wind power generation power, residential power load power and industrial power load power functions respectively.

[0109] S4: Based on the pre-built constraints, the power system power dispatch model is constructed with the maximum consumption of new energy as the objective function;

[0110] Specifically, in this embodiment, the objective function is:

[0111] Among them, E NA represents the consumption of new energy; T is the total number of time periods in the dispatch cycle; P N,t is the actual power consumption of renewable energy generation at time t; C penalty is the penalty cost item.

[0112] It should be noted that, in this embodiment, the penalty cost item is used to balance the potential negative effects of the system caused by excessive pursuit of new energy consumption, covering power generation costs, reliability risk costs, etc., and is specifically composed as follows:

[0113] Where: C G (P G,t ) is the cost of traditional thermal power, hydropower, etc., C R (P R,t ) is the backup cost. Given the uncertainty of both the source and the load, it is necessary to reserve backup capacity. Its cost is linearly related to the backup power and is set as C R (P R,t )=a R P R,t Among them: a R is the standby cost coefficient, P R,t is the standby power generation at time t; C WA (P WA,t ) Due to the volatility of renewable energy generation, there is a tendency for excess power generation to occur but the power cannot be connected to the grid. WA (P WA,t )=a WA P WA,t , where a WA P is the penalty coefficient for wind and solar power abandonment, WA,t is the power of wind and solar power abandoned at time t; C LS (P LS,t ) is the loss cost generated when power supply cannot meet load demand, C LS (PLS,t )=a LS P LS,t where a LS is the load loss cost coefficient, P LS,t The unmet load power at time t.

[0114] At the same time, the pre-built constraints are as follows:

[0115] 1. Power balance constraints:

[0116] P G,t +P N,t +P R,t =P L,t +P LS,t +P loss,t

[0117] Where: P G,t is the power generation of the traditional power generation equipment at time t; P L,t is the power load power predicted at time t; P loss,t is the power loss of the power grid at time t, which can be estimated based on the power grid topology and line parameters, or obtained through historical data statistics.

[0118] 2. Power generation equipment constraints:

[0119] Power upper and lower limits: Thermal power: P T,min ≤P T,t ≤P T,max ; Water and electricity: P H,min ≤P H,t ≤P H,max ; New energy power generation: P N,min ≤P N,t ≤P N,max , each parameter limits the minimum and maximum power generation of the corresponding power generation equipment at time t.

[0120] Ramp rate constraint: Thermal power: -ΔP T,down ≤P T,t -P T,t-1 ≤ΔP T,up ; Hydropower: -ΔP H,down ≤P H,t -P H,t-1 ≤ΔP H,up ; New energy power generation: -ΔP N,down ≤P N,t -P N,t-1 ≤ΔP N,up , where ΔP i,down and ΔP i,up They are respectively the downward and upward climbing rate limit values ​​of device i.

[0121] 3. Constraints on renewable energy power generation: renewable energy power generation is constrained by resource conditions. Solar power generation needs to consider light intensity and weather changes, while wind power generation is constrained by wind speed and direction. Set a lower limit on renewable energy power generation to ensure a certain level of power generation reliability. N,t >P N,min,t , where P N,min,t Determined based on resource characteristics and power generation equipment performance.

[0122] Power consumption constraints: Among them, P N,avail,t is the renewable energy power generation at time t, and γ is the consumption ratio.

[0123] 4. Load demand constraint: It is necessary to meet the basic demand of power load and take into account the uncertainty and volatility of load. Suppose the confidence interval of the predicted load is: [(1-α)P L,t ,(1+β)P L,t ], the actual power supply needs to satisfy: (1-α)P L,t ≤P L,t +P LS,t ≤(1+β)P L,t , where α and β are the error coefficients of load forecasting.

[0124] 5. Dynamic coupling constraints: Based on the dynamic coupling model of the source and load sides, new constraints are added to reflect the mutual influence between the two sides:

[0125] in, is the original load plan without power generation fluctuation interference, λ t and μ t The coefficients determined by the dynamic coupling model are dynamically adjusted according to different scenarios, promoting real-time adaptation of load and power generation and improving the consumption rate of new energy.

[0126] 6. Grid operation constraints: Voltage constraints: V min ≤V t ≤V max , ensure that the voltage of each node meets the equipment operation requirements; Line flow constraint: I L,min ≤I L,t ≤I L,max , to prevent line overload from causing tripping, burning and other accidents, V t is the grid voltage at time t, I L,t is the power flow circuit current at time t, and the upper and lower limit parameters are set according to the rated value and safety standards of the power grid equipment.

[0127] S5: The electric energy dispatching model of the power system is adjusted through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and the adjusted electric energy dispatching model of the power system is solved to obtain the electric energy dispatching plan of the power system.

[0128] Specifically, in this embodiment, the objective function is first adjusted, and the expectation of the power generation of renewable energy in each scenario is calculated according to the probability model of the power generation of renewable energy in different scenarios. For example, solar power generation in sunny scenarios obeys normal distribution, while in cloudy scenarios it may obey log-normal distribution, etc. Suppose there are S scenarios, is the power generated by renewable energy at time t under scenario s, with probability p s , then the expected power generation of new energy is:

[0129] According to the above formula, P in the objective function N,t Replace with E[P N,t ], so that the objective function can more accurately consider the long-term average consumption of renewable energy power generation.

[0130] At the same time, considering the risk cost brought by the fluctuation of renewable energy power generation, the probability model is used to calculate the penalty cost caused by insufficient power generation. For example, in scenario s, when That is, when the minimum renewable energy power meets the load demand in the scenario, an additional penalty cost will be incurred. This penalty cost can be calculated by integrating the probability distribution, such as:

[0131] in, is the probability density function of the renewable energy power generation at time t under scenario s, c r is the penalty cost coefficient corresponding to unit power shortage, and C risk Add to the penalty cost C penalty In the objective function, the risk of fluctuations in renewable energy generation can be reflected.

[0132] For the load side, the load power expectation is also calculated based on the probability model of load power under different scenarios, specifically:

[0133] The P in the constraint condition L,t and P in the objective function LS,t Adjustments are made based on this expectation. For example, in calculating the load loss cost C LS (P LS,t ), more accurately consider the expected load fluctuations and avoid over- or under-estimating the load loss costs. Consider the uncertainty costs caused by load fluctuations, such as when the actual load exceeds expectations and additional generation resources are needed to meet demand. This uncertainty cost is calculated through the load probability model and added to C penalty In this paper, the economic impact of fluctuations on both the source and the load sides is more comprehensively considered together with the risk cost of fluctuations in renewable energy power generation.

[0134] Then the constraints are adjusted, specifically the probabilistic adjustment of power balance constraints and the adaptive adjustment of power generation equipment constraints and scenarios;

[0135] The traditional power balance constraint P G,t +P N,t +P R,t =P L,t +P LS,t +P loss,t Probabilistic adjustments can be made based on the probability models of both the source and the load in different scenarios. For example, considering the probability distribution of power generation and load power in each scenario, the power balance constraint is set to be valid at a certain confidence level. Taking the normal distribution as an example, suppose the power generation of new energy is Normal distribution Load power Normal distribution At a 95% confidence level, the power balance constraint can be expressed as:

[0136]

[0137] In terms of power generation equipment constraints, the characteristics of renewable energy generation in different scenarios are considered to adjust the upper and lower limits of power and the ramp rate constraints. For example, in a sunny day scenario, the upper limit of solar power generation power P N,max It may be higher in cloudy scenes, but lower in cloudy scenes. According to the probability model of renewable energy power generation in different scenes, the appropriate upper and lower limits of power generation equipment and ramp rate limit in each scene are determined. For example, in scene s, the upper and lower limits of renewable energy power generation power become The ramp rate constraint becomes in and It is a parameter determined based on the probability model of renewable energy power generation under the scenario and the actual operating characteristics.

[0138] Finally, linear programming algorithm, nonlinear programming algorithm and intelligent optimization algorithm are selected to solve the adjusted power system electric energy dispatching model and obtain the power system electric energy dispatching plan.

[0139] It should be noted that, in this embodiment, linear programming algorithms, nonlinear programming algorithms and intelligent optimization algorithms are used to solve the scheduling model, which are conventional technical means in the art and will not be described in detail here.

[0140] Specifically, in this embodiment, by analyzing the random fluctuation characteristics of both sides of the source and load under different scenarios, a random fluctuation probability model of both sides of the source and load that characterizes the random fluctuations of both sides of the source and load under different clustering scenarios is constructed, which can better describe the random fluctuations of both sides of the source and load, and provide a more accurate data basis for power system scheduling; then, based on historical new energy power generation data and historical load data, a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations is constructed in combination with a multivariate VAR model; and through the random fluctuation probability model of both sides of the source and load and the dynamic coupling model, the power system electric energy scheduling model with the maximum new energy consumption as the goal is adjusted, thereby optimizing the influence of the random fluctuations and coupling correlation characteristics of both sides of the source and load on the power system electric energy scheduling, and improving the comprehensiveness and stability of the scheduling.

[0141] Example 2

[0142] See also Figure 2 As shown, the present invention also provides an electric energy dispatch optimization system considering random fluctuations on both sides of the source and the load, the system is used in any one of the above-mentioned electric energy dispatch optimization methods considering random fluctuations on both sides of the source and the load, the system comprising:

[0143] The feature clustering module 100 is used to obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios;

[0144] A probability model building module 200 is used to build a source-load double-sided random fluctuation probability model that characterizes the source-load double-sided random fluctuation under different clustering scenarios based on a new energy feature set and a load feature set and in combination with a probability density distribution function;

[0145] A dynamic coupling model building module 300 is used to build a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations based on historical new energy power generation data and historical load data in combination with a multivariate VAR model;

[0146] The dispatch model building module 400 is used to build a power system power dispatch model based on pre-built constraints and taking the maximum consumption of new energy as the objective function;

[0147] The dispatching scheme generating module 500 is used to adjust the electric energy dispatching model of the power system through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and solve the adjusted electric energy dispatching model of the power system to obtain the electric energy dispatching scheme of the power system.

[0148] It should be noted that the modules in the system of Example 2 correspond to the steps in the method of Example 1. The steps in the method of Example 1 have been described in detail in Example 1. In this Example 2, the contents of the modules in the system will not be described in detail.

[0149] Example 3

[0150] See also Figure 3 As shown, this embodiment further provides a computer device, including a system memory 1005 and a processor 1001, wherein the system memory 1005 stores a computer program, and the processor 1001 implements the steps of any of the above methods when executing the computer program.

[0151] It should be noted that the processor 1001 is used to execute the steps in the above method embodiments according to the instructions in the program code. Alternatively, the processor 1001 implements the functions of each module / unit in the above system / device embodiments when executing the computer program.

[0152] Specifically, in this embodiment, the computer program may be divided into one or more modules / units, one or more modules / units are stored in the system memory 1005, and are executed by the processor 1001 to complete the present application. One or more modules / units may be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0153] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor 1001 and a system memory 1005. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include an input / output device 1003, a network access device 1002, a bus 1006, etc.

[0154] The processor 1001 may be a central processing unit (CPU), or other general-purpose processors 1001, digital signal processors 1001 (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor 1001 may be a microprocessor or any conventional processor, etc.

[0155] The system memory 1005 may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The system memory 1005 may also be a storage device 1004 of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the terminal device. Further, the system memory 1005 may also include both the internal storage unit of the terminal device and the storage device 1004. The system memory 1005 is used to store computer programs and other programs and data required by the terminal device. The system memory 1005 may also be used to temporarily store data that has been output or is to be output.

[0156] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0157] Example 4

[0158] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0159] Among them, the computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, system or device, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk. Random Access Memory (RAM), Read-Only Memory (ROM), Erasable Programmable Read Only Memory (EPROM), registers, hard disks, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above, or any other form of computer-readable storage medium known in the art.

[0160] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In an embodiment of the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system, or device.

[0161] Example 5

[0162] This embodiment also provides a computer program product including instructions. When the instructions are executed by a computer device cluster, the computer device cluster executes the method described in Embodiment 1.

[0163] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing power dispatch considering random fluctuations on both sides of source and load, characterized in that: include: Obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios; Based on the new energy feature set and load feature set and combined with the probability density distribution function, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation in different clustering scenarios. Based on historical renewable energy power generation data and historical load data, a dynamic coupling model is constructed to characterize the dynamic coupling relationship between renewable energy fluctuations and load fluctuations in combination with a multivariate VAR model. Based on the pre-built constraints, the power system power dispatch model is constructed with the maximum consumption of new energy as the objective function; The electric energy dispatching model of the power system is adjusted through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and the adjusted electric energy dispatching model of the power system is solved to obtain the electric energy dispatching plan of the power system.

2. The method for optimizing electric energy dispatch considering random fluctuations on both sides of source and load according to claim 1 is characterized in that: Obtain historical renewable energy power generation data and historical load data, wherein the historical renewable energy power generation data includes solar power generation data and wind power generation data; the historical load data includes historical industrial load data, historical commercial load data and historical residential load data.

3. The method for optimizing electric energy dispatch considering random fluctuations on both sides of source and load according to claim 1 is characterized in that: Extract new energy feature sets and load feature sets based on different pre-built scenario types, specifically: Acquire meteorological data and special event data to construct different scene types, including sunny scene, cloudy scene, rainy scene, extreme weather scene and special event scene; For each scenario type, multiple new energy characteristics are extracted to construct a new energy characteristic set, and multiple load characteristics are extracted to construct a load characteristic set, wherein the multiple new energy characteristics include solar power generation power characteristics and wind power generation power characteristics; the multiple load characteristics include industrial load power characteristics, commercial load power characteristics and residential load power characteristics.

4. The method for optimizing electric energy dispatch considering random fluctuations on both sides of source and load according to claim 3 is characterized in that: Based on the new energy feature set and load feature set and combined with the probability density distribution function, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation under different clustering scenarios. Specifically, it is: Under different scene types, a normal distribution function, a lognormal distribution function, a Bernoulli distribution function or a truncated normal distribution function is used to describe the probability density distribution of solar power generation characteristics, and multiple solar power generation power fluctuation probability functions are obtained respectively; Under different scenario types, the probability density distribution of wind power characteristics is described using one of the Weibull distribution function, the two-parameter gamma distribution function or the generalized Pareto distribution function, and multiple wind power fluctuation probability functions are obtained respectively; Under different scene types, a normal distribution function, a mixed Gaussian distribution function or a Poisson distribution function is used to describe the probability density distribution of the residential load power characteristics, and multiple residential load power fluctuation probability functions are obtained respectively; Under different scenario types, the normal distribution function is used to describe the probability density distribution of industrial load power characteristics, and multiple industrial load power fluctuation probability functions are obtained respectively; Combining multiple solar power generation power fluctuation probability functions, multiple wind power generation power fluctuation probability functions, multiple residential load power fluctuation probability functions and multiple industrial load power fluctuation probability functions, a source-load double-sided random fluctuation probability model is constructed to characterize the source-load double-sided random fluctuation under different clustering scenarios.

5. The method for optimizing electric energy dispatch considering random fluctuations on both sides of source and load according to claim 1 is characterized in that: Based on the historical renewable energy power generation data and historical load data, a dynamic coupling model is constructed to characterize the dynamic coupling relationship between renewable energy fluctuations and load fluctuations in combination with the multivariate VAR model. The dynamic coupling model is specifically as follows: Among them, P solar (t), P solar (t-1) are the solar power generation power at time t or time t-1; P wind (t), P wind (t-1) are the wind power generation power at time t or time t-1 respectively; P res (t), P res (t-1) is the total power consumption of residential users at time t or time t-1 respectively; P ind (t), P ind (t-1) is the total power consumption of industrial users at time t or time t-1; a ij (i=1,2,3,4,j=1,2,3,4) are model coefficients; ∈ 1t ,∈ 2t ,∈ 3t and ∈ 4t They are white noise sequences corresponding to the solar power generation power, wind power generation power, residential power load power and industrial power load power functions respectively.

6. The method for optimizing electric energy dispatch considering random fluctuations on both sides of source and load according to claim 1, characterized in that: The specific objective function is: Among them, E NA represents the consumption of new energy; T is the total number of time periods in the dispatch cycle; P N,t is the actual power consumption of renewable energy generation at time t; C penalty is the penalty cost item.

7. An energy dispatch optimization system considering random fluctuations on both sides of source and load, characterized in that: The system is used in an electric energy dispatch optimization method considering random fluctuations on both sides of source and load as described in any one of claims 1 to 6, and the system comprises: The feature clustering module is used to obtain historical renewable energy power generation data and historical load data, extract renewable energy feature sets and load feature sets based on pre-built different scenario types, perform K-Means clustering on the renewable energy feature sets and load feature sets, and optimize the clustering results to finally obtain different clustering scenarios; A probability model building module is used to build a source-load double-sided random fluctuation probability model that characterizes the source-load double-sided random fluctuation in different clustering scenarios based on the new energy feature set and the load feature set and combined with the probability density distribution function; Dynamic coupling model building module, which is used to build a dynamic coupling model that characterizes the dynamic coupling relationship between new energy fluctuations and load fluctuations based on historical new energy power generation data and historical load data and combined with a multivariate VAR model; The dispatch model building module is used to build a power system power dispatch model based on pre-built constraints and taking the maximum consumption of new energy as the objective function; The dispatching scheme generation module is used to adjust the electric energy dispatching model of the power system through the random fluctuation probability model of the source and load sides and the dynamic coupling model, and solve the adjusted electric energy dispatching model of the power system to obtain the electric energy dispatching scheme of the power system.

8. A computer device comprising a system memory and a processor, wherein the system memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising instructions, characterized in that When the instructions are executed by a computer device cluster, the computer device cluster executes the method according to any one of claims 1 to 6.

Images