Flexibility requirement quantification method, device, equipment and storage medium

Abstract

The present application relates to a flexibility demand quantification method, device, equipment and storage medium. The method includes: determining a set of output curves corresponding to the first period based on the historical power generation data of the first clean energy, the output curve set including multiple maximum output curves of the first period; for each maximum output curve of the first period, determining the corresponding short-term flexibility demand based on the maximum output curve of the first period and a preset flexibility analysis model; predicting the power generation trend of the second clean energy in the second period based on the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand. The use of this method can improve the accuracy of quantifying the long-, medium- and short-term flexibility demands for water, wind, solar and storage in the basin.

Description

Technical Field

[0001] The present application relates to the technical field of clean energy system flexibility quantification, and in particular to a flexibility demand quantification method, device, equipment and storage medium. Background Art

[0002] The volatility and uncertainty of clean energy such as hydropower, wind power and solar energy are reflected on seasonal, monthly, daily and even smaller time scales, and their trends and changes are difficult to predict. In other words, hydropower, wind power and photovoltaic power generation have a high degree of uncertainty and their time distribution is extremely uneven.

[0003] In the existing technology, the flexibility demand determined by expert experience is not accurate enough. Therefore, how to accurately determine the quantitative method of long-, medium- and short-term flexibility demand for water, wind, solar and storage in the river basin is a technical problem that needs to be solved urgently. Summary of the invention

[0004] Based on this, it is necessary to provide a flexibility demand quantification method, device, equipment and storage medium that can accurately quantify long-, medium- and short-term flexibility needs in response to the above technical problems.

[0005] In a first aspect, the present application provides a flexibility demand quantification method, including:

[0006] Determine an output curve set corresponding to a first period according to historical power generation data of the first clean energy source, where the output curve set includes a plurality of maximum output curves of the first period;

[0007] For each maximum output curve of the first period, determine the corresponding short-term flexibility demand according to the maximum output curve of the first period and a preset flexibility analysis model;

[0008] The power generation trend of the second clean energy in the second period is predicted based on the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand.

[0009] In one embodiment, determining the output curve set corresponding to the first period according to the historical power generation data of the first clean energy source includes:

[0010] A power generation matrix is ​​determined according to the historical power generation data of the first clean energy source, wherein the matrix element of the i-th row and j-th column in the power generation matrix is ​​the maximum output value of the j-th day in the i-th month, and the maximum output value represents the maximum power output by the first clean energy source under ideal operating conditions; the power generation matrix is ​​preprocessed, and the preprocessing includes data cleaning and standardization; the processed power generation matrix is ​​clustered according to a preset clustering method, and a set of output curves corresponding to the first period is determined according to the clustering results.

[0011] In one of the embodiments, the clustering method is a hierarchical clustering method, the output curve set corresponding to the first period includes a plurality of typical daily wind and solar maximum output curves within a month, the processed power generation matrix is ​​clustered according to a preset clustering method, and the output curve set corresponding to the first period is determined according to the clustering result, including:

[0012] According to the processed row data of the power generation matrix, multiple output curves are determined; for each output curve, the characteristic vector of the output curve is determined, and the characteristic vector includes the origin moment, the center distance, the skewness and the kurtosis; according to the hierarchical clustering method, each output curve is clustered based on each characteristic vector, and according to the clustering results, multiple typical daily wind and solar maximum output curves within a month are determined. The hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets the preset conditions into one cluster until the target cluster is reached.

[0013] In one of the embodiments, the similarity measurement methods of the hierarchical clustering method include: nearest neighbor clustering, farthest neighbor clustering or average clustering; the nearest neighbor clustering is used in the clustering process, for any two cluster structures, to use the nearest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the farthest neighbor clustering is used in the clustering process, for any two cluster structures, to use the farthest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the average clustering is used in the clustering process, for any two cluster structures, to use the average distance between each sample point in the two cluster structures as the measure of the similarity between the two clusters.

[0014] In one embodiment, the preset flexibility analysis model includes a first sub-model, a second sub-model, and a third sub-model, and the short-term flexibility demand includes an intra-day hourly flexibility demand, a daytime flexibility demand, and a monthly flexibility demand. The corresponding short-term flexibility demand is determined according to the maximum output curve of the first period and the preset flexibility analysis model, including:

[0015] The daily maximum output value is determined according to the maximum output curve of the first cycle; the hourly flexibility demand within the day is determined according to the first sub-model, wherein, in the first sub-model, the hourly flexibility demand within the day is determined according to the daily maximum output value and the preset daily power generation of the first clean energy; the daytime flexibility demand is determined according to the second sub-model, wherein, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy and the preset lower limit of the daily power generation; the monthly flexibility demand is determined according to the third sub-model, wherein, in the third sub-model, the monthly flexibility demand is determined according to the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy, the preset lower limit of the daily power generation and the preset lower limit of the monthly power generation.

[0016] In one embodiment, the statistical model is an autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and the power generation trend of the second clean energy in the second period is predicted based on the historical power generation data of the second clean energy and a preset statistical model, including:

[0017] The historical power generation time series is determined based on the historical power generation data of the second clean energy, and the periodic components are removed from the historical power generation time series to obtain a time series after periodic adjustment. Based on the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, the power generation trend of the second clean energy in the second period is predicted.

[0018] In one embodiment, the flexibility requirement quantification method further includes:

[0019] The short-term flexibility demand and the long-term flexibility demand are input into a preset scheduling model to determine a scheduling strategy according to the output of the scheduling model.

[0020] In a second aspect, the present application also provides a flexibility demand quantification device, including:

[0021] A first determination module is used to determine an output curve set corresponding to a first period according to historical power generation data of a first clean energy source, where the output curve set includes a plurality of maximum output curves of the first period;

[0022] a quantification module, for determining, for each first-period maximum output curve, a corresponding short-term flexibility demand according to the first-period maximum output curve and a preset flexibility analysis model;

[0023] The second determination module is used to predict the power generation trend of the second clean energy in the second period according to the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand.

[0024] In one of the embodiments, the first determination module is specifically used to determine the power generation matrix based on the historical power generation data of the first clean energy, the matrix element in the i-th row and j-th column in the power generation matrix is ​​the maximum output value on the j-th day in the i-th month, and the maximum output value represents the maximum power output of the first clean energy under ideal operating conditions; preprocessing the power generation matrix, the preprocessing includes data cleaning and standardization; clustering the processed power generation matrix according to a preset clustering method, and determining the output curve set corresponding to the first period according to the clustering results.

[0025] In one of the embodiments, the clustering method is a hierarchical clustering method, the set of output curves corresponding to the first period includes multiple typical daily wind and solar maximum output curves within a month, and the first determination module is specifically used to determine the multiple output curves based on the processed row data of the power generation matrix; for each output curve, determine the characteristic vector of the output curve, the characteristic vector includes the origin moment, the center distance, the skewness and the kurtosis; cluster the output curves based on each characteristic vector according to the hierarchical clustering method, and determine multiple typical daily wind and solar maximum output curves within a month according to the clustering results. The hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets the preset conditions into one cluster until the target cluster is reached.

[0026] In one of the embodiments, the similarity measurement methods of the hierarchical clustering method include: nearest neighbor clustering, farthest neighbor clustering or average clustering; the nearest neighbor clustering is used in the clustering process, for any two cluster structures, to use the nearest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the farthest neighbor clustering is used in the clustering process, for any two cluster structures, to use the farthest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the average clustering is used in the clustering process, for any two cluster structures, to use the average distance between each sample point in the two cluster structures as the measure of the similarity between the two clusters.

[0027] In one embodiment, the preset flexibility analysis model includes a first sub-model, a second sub-model and a third sub-model, the short-term flexibility demand includes intraday hourly flexibility demand, daytime flexibility demand and monthly flexibility demand, and the quantification module is specifically used to determine the daily maximum output value according to the maximum output curve of the first cycle; determine the intraday hourly flexibility demand according to the first sub-model, wherein, in the first sub-model, the intraday hourly flexibility demand is determined according to the daily maximum output value and the preset daily power generation of the first clean energy; determine the daytime flexibility demand according to the second sub-model, wherein, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the intraday hourly flexibility demand, the preset daily power generation of the first clean energy and the preset lower limit of the daily power generation; determine the monthly flexibility demand according to the third sub-model, wherein, in the third sub-model, the monthly flexibility demand is determined according to the daily maximum output value, the intraday hourly flexibility demand, the preset daily power generation of the first clean energy, the preset lower limit of the daily power generation and the preset lower limit of the monthly power generation.

[0028] In one of the embodiments, the statistical model is an autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and the second determination module is specifically used to determine the historical power generation time series based on the historical power generation data of the second clean energy, remove the periodic components from the historical power generation time series, and obtain a time series after periodic adjustment; based on the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, predict the power generation trend of the second clean energy in the second period.

[0029] In one of the embodiments, the flexibility demand quantification device also includes a scheduling module, which is used to input short-term flexibility demand and long-term flexibility demand into a preset scheduling model and determine a scheduling strategy according to the output of the scheduling model.

[0030] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any of the methods described in the first aspect is implemented.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any of the methods described in the first aspect above.

[0032] In a fifth aspect, the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements any of the methods described in the first aspect above.

[0033] The above flexibility demand quantification method, device, equipment and storage medium determine the output curve set corresponding to the first period according to the historical power generation data of the first clean energy, and the output curve set includes multiple maximum output curves of the first period; for each maximum output curve of the first period, the corresponding short-term flexibility demand is determined according to the maximum output curve of the first period and the preset flexibility analysis model; the power generation trend of the second clean energy in the second period is predicted according to the historical power generation data of the second clean energy and the preset statistical model to determine the long-term flexibility demand. By determining the maximum output curves corresponding to multiple first periods from the historical power generation data, and then determining the short-term flexibility demand based on the maximum output curve, and predicting the trend of hydropower, wind and solar power generation according to the historical power generation sequence, the long-term flexibility demand is obtained. In this way, based on the flexibility of the first clean energy and the second clean energy at different time scales, the flexibility demand can be accurately quantified, and the operation of the power system can be better evaluated and planned, which is helpful to formulate effective scheduling strategies and emergency plans, improve the adaptability and response speed of the system, and thus ensure continuous and stable power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A schematic diagram of a flow chart of a flexibility requirement quantification method in one embodiment;

[0036] Figure 2 A schematic diagram of a flow chart of a step of determining a set of output curves corresponding to a first cycle in an embodiment;

[0037] Figure 3 It is a flowchart of the step of determining a set of output curves corresponding to a first cycle in another embodiment;

[0038] Figure 4 A schematic diagram of a hierarchical clustering process in one embodiment;

[0039] Figure 5 Schematic diagram of a set of output curves corresponding to the first cycle in an embodiment;

[0040] Figure 6 A statistical box plot of the maximum output type in one embodiment;

[0041] Figure 7 A flowchart of a step of determining a corresponding short-term flexibility requirement in one embodiment;

[0042] Figure 8 A schematic diagram of intra-day hourly flexibility demand in a first sub-model in one embodiment;

[0043] Fig. 9 A schematic diagram of daytime flexibility demand in the second sub-model in one embodiment;

[0044] Fig.10 A schematic diagram of monthly flexibility demand in the third sub-model in one embodiment;

[0045] Fig.11 A schematic diagram of a flow chart of a step of predicting a power generation trend of a second clean energy source in a second cycle in an embodiment;

[0046] Fig.12 A schematic flow chart of a method for quantifying flexibility requirements in another embodiment;

[0047] Fig.13 A structural block diagram of a flexibility requirement quantification device in one embodiment;

[0048] Fig.14 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0050] In order to cope with global warming and ensure energy security, it has become a global consensus to optimize the energy structure and develop and utilize renewable energy. The International Energy Agency pointed out that under the concept of "achieving net zero by 2050", nearly 90% of the world's electricity comes from renewable energy, of which solar and wind power generation account for nearly 70%. my country has clearly pointed out that it is necessary to accelerate the development of "water, wind, light and storage" clean energy bases and build a new power system with a high proportion of renewable energy. Weather and climate-driven wind power generation and photovoltaic power generation have significant spatial and temporal variability. In the new power system with new energy as the main body, a large amount of energy storage resources are needed to cope with extreme weather events such as sudden drop in wind speed, continuous rainy weather, high temperature, drought, and adjust the seasonal and diurnal fluctuations of wind and solar power generation. At present, my country is in a critical period of green and low-carbon energy transformation and development. With the rapid increase in wind and solar penetration, the construction of energy storage that matches the scale of new energy requires huge cost investment. There are hundreds of hydropower stations built in the Jinsha River, Yalong River, Dadu River, Wujiang River, Lancang River, Yarlung Zangbo River and other river basins in southwest my country. Hydropower units have the ability to respond quickly in the short term and can switch from shutdown to full-load power generation within a few minutes to cope with the uncertainty of wind and solar output and changes in load demand. Hydropower station reservoirs can also be transformed into pumped-storage power station reservoirs to provide more flexibility. Therefore, relying on the construction of pumped storage in the river basin, optimizing the allocation of wind and solar new energy on a large scale, and rationally integrating the multi-time scale flexibility adjustment capabilities of conventional hydropower and pumped-storage power stations, through integrated and large-scale development to achieve complementary advantages, improve the absorption and storage capacity of renewable energy, and provide 100% clean electricity for the power grid.

[0051] Hydropower, wind power and solar power are largely dependent on meteorological conditions, such as rainfall, evaporation, wind speed, sunshine intensity and ambient temperature. The volatility and uncertainty of these energy sources are reflected in the seasonal, monthly, daily and even smaller time scales, and their trends and changes are difficult to predict. Therefore, hydropower, wind power and photovoltaic power generation are highly uncertain, and their time distribution is extremely uneven, which is inconsistent with the stability of electricity demand. The grid connection of large-scale power stations will bring significant fluctuations and uncertainties to the power supply side, increasing the difficulty of the power system to maintain stable power supply at seasonal, monthly, daily, hourly and higher time resolutions.

[0052] In the existing technology, the flexibility demand determined by expert experience is not accurate enough. Therefore, how to accurately determine the quantitative method of long-, medium- and short-term flexibility demand for water, wind, solar and storage in the river basin is a technical problem that needs to be solved urgently.

[0053] In view of this, the present application provides a method for accurately determining the long-, medium- and short-term flexibility demand quantification for water, wind, solar and storage in a river basin. The flexibility demand quantification method provided in the embodiment of the present application may be executed by a flexibility demand quantification device, which may be implemented by software, hardware, or a combination of software and hardware. It may be embedded in or independent of a processor in a computer device in the form of hardware, or it may be stored in a memory in a computer device in the form of software. In the following method embodiments, the execution subject is a computer device as an example for explanation, wherein the computer device may be a server or a desktop computer. The embodiment of the present application does not limit the specific type of the computer device.

[0054] In an exemplary embodiment, Figure 1 As shown, a flexibility demand quantification method is provided, including the following steps 101 to 103. Among them:

[0055] Step 101: determine a set of output curves corresponding to a first period according to historical power generation data of a first clean energy source.

[0056] The output curve set includes a plurality of maximum output curves of the first period.

[0057] Optionally, the first clean energy may be wind energy, solar energy, etc., which is not limited in the embodiment of the present application. The historical power generation data of the first clean energy may be historical wind power generation data, historical photovoltaic power generation data, or the sum of historical wind and photovoltaic power generation.

[0058] Optionally, the first period may be a preset period of time, for example, it may be a year, a month or a day; the maximum output curve corresponding to the first period may be a maximum value curve of the first clean energy output per day within a month, a maximum value curve of the first clean energy output per month within a year, or a maximum value curve of the first clean energy output per day within a year, etc., and the embodiments of the present application are not limited to this.

[0059] Optionally, for a maximum output curve, the horizontal axis of the curve may be each month or each day within the first period, and the vertical axis may be the maximum power output by the first clean energy under ideal operating conditions.

[0060] Optionally, the historical power generation data of the first clean energy can be integrated according to the corresponding time to obtain multiple initial maximum output curves of the first period, and these multiple initial maximum output curves of the first period can be determined as a set of output curves corresponding to the first period. Alternatively, multiple typical curves can be obtained based on the multiple initial maximum output curves of the first period, and the multiple typical curves can be determined as a set of output curves corresponding to the first period. The embodiments of the present application are not limited to this.

[0061] Optionally, a plurality of typical curves may be obtained according to a plurality of initial maximum output curves of the first cycle, which may be achieved by a cluster analysis method or a principal component analysis method.

[0062] Step 102 : for each maximum output curve of the first period, determine the corresponding short-term flexibility demand according to the maximum output curve of the first period and a preset flexibility analysis model.

[0063] Optionally, the output curve set corresponding to the first period includes multiple maximum output curves of the first period, and for each maximum output curve of the first period, the corresponding short-term flexibility demand can be determined according to the maximum output curve.

[0064] Optionally, short-term flexibility demand can be to cope with changes in power load, volatility in the output of renewable energy (wind power and photovoltaic power), and to maintain stable operation of the power system, and to quickly adjust and adapt to changes in power generation resources within a short period of time. For example, wind power generation and photovoltaic power generation change continuously within a day and have strong volatility. Therefore, the total output of wind power and photovoltaic power stations also changes continuously within a day, and there will be a large gap between peak power and valley power. In order to avoid power shortage or power abandonment, it is a common operating mode to use flexible hydropower in the water-wind-solar hybrid energy system to smooth out the volatility of wind and solar power generation so that the total output of the system remains unchanged within a day. Alternatively, under the premise of smoothing intraday fluctuations, the system's daily power generation may not meet the minimum daily power supply, and additional flexibility resources are needed to supplement it.

[0065] Optionally, the preset flexibility analysis model may be one that takes the maximum output curve of the first cycle as input to obtain short-term flexibility demand. In one possible implementation, the preset flexibility analysis model may be one that is obtained by training a neural network model using historical power generation data and historical short-term flexibility demand as training samples.

[0066] Step 103 , predicting the power generation trend of the second clean energy in the second period according to the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand.

[0067] Optionally, the second clean energy can be wind energy, solar energy, hydropower, etc. The historical power generation data of the second clean energy can include historical wind power generation data, historical photovoltaic power generation data, historical hydropower generation data, etc. Similarly, the second period can be a preset period of time, which is longer, such as one year or one month.

[0068] In one possible implementation, the preset statistical model may be an autoregressive model, wherein the autoregressive model assumes that there is a linear relationship between the power generation trend of the second cycle and the historical power generation data, and the power generation trend of the second cycle may be predicted based on the historical power generation data.

[0069] In another possible implementation, the preset statistical model may also be a vector autoregression model, a state space model, or a recurrent neural network (RNN) and its improved models, such as a long short-term memory network (LSTM) or a gated recurrent unit (GRU).

[0070] The above flexibility demand quantification method determines the output curve set corresponding to the first cycle according to the historical power generation data of the first clean energy, and the output curve set includes multiple maximum output curves of the first cycle; for each maximum output curve of the first cycle, the corresponding short-term flexibility demand is determined according to the maximum output curve of the first cycle and the preset flexibility analysis model; the power generation trend of the second clean energy in the second cycle is predicted according to the historical power generation data of the second clean energy and the preset statistical model to determine the long-term flexibility demand. By determining the maximum output curves corresponding to multiple first cycles from the historical power generation data, and then determining the short-term flexibility demand based on the maximum output curve, and predicting the trend of hydropower, wind and solar power generation according to the historical power generation sequence, the long-term flexibility demand is obtained. In this way, based on the flexibility of the first clean energy and the second clean energy at different time scales, the flexibility demand can be accurately quantified, and the operation of the power system can be better evaluated and planned, which is helpful to formulate effective scheduling strategies and emergency plans, improve the adaptability and response speed of the system, and thus ensure continuous and stable power supply.

[0071] In an exemplary embodiment, Figure 2 As shown, optionally, determining the output curve set corresponding to the first period according to the historical power generation data of the first clean energy includes the following steps 201 to 203. Among them:

[0072] Step 201 : determining a power generation matrix according to historical power generation data of a first clean energy source.

[0073] Among them, the matrix element in the i-th row and j-th column in the power generation matrix is ​​the maximum output value on the j-th day in the i-th month, and the maximum output value represents the maximum power output by the first clean energy under ideal operating conditions.

[0074] Optionally, the first period may be in months.

[0075] Step 202: pre-processing the power generation matrix.

[0076] Among them, preprocessing includes data cleaning and standardization.

[0077] Optionally, data cleaning may include checking whether there are abnormal values ​​in the power generation matrix (for example, output values ​​that are obviously inconsistent with physical laws). If so, the abnormal value may be replaced with the mean or median of the row of data, or reasonable corrections may be made based on the distribution of the data.

[0078] Optionally, the normalization process may be performed by zero-mean normalization or minimum-maximum normalization.

[0079] Step 203: clustering the processed power generation matrix according to a preset clustering method, and determining a set of output curves corresponding to the first period according to the clustering result.

[0080] Optionally, the clustering method may be a K-means clustering method, a K-medoid clustering method, an agglomerative hierarchical clustering method, or a columnar hierarchical clustering method. The preset clustering method is not limited in the embodiments of the present application.

[0081] Optionally, the processed power generation matrix is ​​clustered by a preset clustering method, and the historical power generation data of the first clean energy is grouped according to similar distribution and fluctuation patterns. At least two groups of typical curves that can represent the main characteristics, trends or patterns of the historical power generation data can be found, and the output curve set corresponding to the first period can be determined based on the typical curves.

[0082] In an exemplary embodiment, Figure 3 As shown, optionally, the clustering method is a hierarchical clustering method, the output curve set corresponding to the first period includes a plurality of typical daily wind and solar maximum output curves within a month, the processed power generation matrix is ​​clustered according to a preset clustering method, and the output curve set corresponding to the first period is determined according to the clustering result, including the following steps 301 to 303. Among them:

[0083] Step 301, determining a plurality of output curves according to the processed row data of the power generation matrix.

[0084] Optionally, for each row of data in the processed power generation matrix, the elements in the row of data are used as the vertical coordinate (maximum output value) and the corresponding time point is used as the horizontal coordinate, and the data points are plotted one by one. After all the data corresponding to a row of data are plotted, these data points can be connected by interpolation method to obtain an output curve; repeating this process for each row of the matrix can obtain multiple output curves corresponding to the power generation matrix after output.

[0085] Step 302: for each output curve, determine the characteristic vector of the output curve, where the characteristic vector includes origin moment, center distance, skewness and kurtosis.

[0086] Optionally, origin moment, center distance, skewness and kurtosis can be calculated using the following formulas:

[0087]

[0088]

[0089]

[0090]

[0091] in, is the origin moment, is the center distance, is the kurtosis, is the skewness, r is the order of the sample moment, is a data point in the output curve, i.e., the maximum daily output value of First Clean Energy, and D is the total number of data points in the output curve.

[0092] Optionally, the origin moment can be used to describe the distribution characteristics of data points in the output curve, the center distance can be used to describe the shape characteristics of the distribution of data points in the output curve, the skewness can be used to measure the asymmetry of the probability distribution, and the kurtosis can be used to measure the sharpness or flatness of the distribution; when the order of the sample moment is 1, the first-order origin moment is the mean of the data points in the output curve, and when the order of the sample moment is 2, the second-order center distance is the variance of the data points in the output curve.

[0093] Optionally, for each output curve, a characteristic vector of the output curve may be formed according to the origin moment, center distance, skewness and kurtosis of the output curve.

[0094] Step 303: cluster the output curves based on the characteristic vectors according to a hierarchical clustering method, and determine a plurality of typical maximum wind and solar output curves for each day of the month according to the clustering results.

[0095] Among them, the hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets the preset conditions into one cluster until the target cluster is reached.

[0096] Optionally, compared to directly applying the clustering method to the curve, the embodiment of the present application is a hierarchical clustering method based on feature vectors, which can represent each output curve through several representative features, has high computational efficiency, and has good stability when outliers exist.

[0097] Optional, such as Figure 4 As shown in the figure, it is a dendrogram obtained by the hierarchical clustering method. The internal hierarchical structure of each cluster can be observed through the dendrogram. At the initial node of the hierarchical clustering, each output curve is regarded as an independent cluster structure, and two output curves whose similarities meet the preset conditions are merged into one cluster. Then, clusters whose similarities meet the preset conditions are found for merging until the current number of clusters reaches the target cluster, and then the merging is stopped.

[0098] Optionally, the similarity of two output curves can be measured by the Euclidean distance of the characteristic vectors of the two output curves, as shown in the following formula:

[0099]

[0100] Where, d(a,b) is the distance between the two output curves, SF a and SF a are the characteristic vectors of the two output curves, The similarity of the output curves can also be measured by squared Euclidean distance or Manhattan distance, etc., which is not limited in the present embodiment of the application.

[0101] Optionally, the two output curves with the greatest similarity may be merged into one cluster. When there are multiple output curves in one cluster, when clusters whose similarities meet preset conditions are merged, the similarity between the clusters may be measured by a similarity measurement method.

[0102] Optionally, similarity measures for hierarchical clustering methods include: nearest neighbor clustering, farthest neighbor clustering, or average clustering;

[0103] Optionally, single-linkage clustering is used in the clustering process. For any two cluster structures, the shortest distance between each sample point in the two clusters is used as a measure of the similarity between the two clusters.

[0104] Optionally, each sample point in a cluster may be each output curve in the cluster. The distance between the output curves in two clusters can be calculated, and the similarity between the two clusters can be determined by the shortest distance, which can be expressed by the following formula:

[0105]

[0106] Wherein, A and B represent any two cluster structures, a is an output curve in cluster structure A, and b is an output curve in cluster structure B.

[0107] Optionally, complete-linkage clustering is used in the clustering process. For any two cluster structures, the farthest distance between each sample point in the two clusters is used as a measure of the similarity between the two clusters.

[0108] Alternatively, the furthest neighbor clustering can be expressed by the following formula:

[0109]

[0110] Alternatively, furthest neighbor clustering can ensure that observations in a cluster are similar to all observations in the same cluster.

[0111] Optionally, unweighted average linkage clustering is used in the clustering process to take the average distance between each sample point in the two cluster structures as a measure of the similarity between the two clusters for any two cluster structures.

[0112] Alternatively, the average clustering can be expressed by the following formula:

[0113]

[0114] Among them, n A represents the number of output curves in cluster A, n B Indicates the number of output curves in cluster B.

[0115] Optionally, different scenarios are obtained through clustering, and different rules are obtained in different scenarios. The output curve with the smallest distance from other output curves in the same cluster structure can be used as the center point of the cluster structure, that is, the output curve can be used as a typical monthly daily wind and solar maximum output curve. Similarly, for other cluster structures, the corresponding typical monthly daily wind and solar maximum output curves can also be determined.

[0116] Optional, such as Figure 4 As shown in Figure 2, when the number of clusters is five, the various types are obviously different, and five typical daily wind and solar maximum output curves within a month can be determined, such as Figure 5 As shown in the figure, it is a statistical box plot of the maximum wind and solar power output types on a daily basis within a month, including 5 typical maximum wind and solar power output curves on a daily basis within a month, Cluster1-Cluster5, as shown in the figure. Figure 6As shown in the figure, it is the curve corresponding to the ratio of the daily maximum wind and solar output curve obtained by clustering to the average output of the month. It can be seen from the figure that the mean values ​​of the curves in types 1 to 4 are almost all greater than 2, and there are some curves with a mean value less than 2 in type 5. The curves in type 5 have relatively low requirements for hourly flexibility. From the perspective of variance characteristics, the curve fluctuation in type 3 is the most obvious, and the fluctuations in types 1 and 5 are relatively low. From the perspective of kurtosis characteristics, the peak coefficients of the curve distribution in types 2 and 3 are basically less than 3, which are smoother than the normal distribution, while the peak coefficients of the curve distribution in type 5 are very large, which are much steeper than the normal distribution, indicating that many curves in this type have extreme values. From the perspective of skewness, the skewness coefficients of most curves in types 1 and 5 are less than 0, and the mean is to the left of the mode, which is a left-skewed distribution. The skewness coefficients of all curves in type 4 are greater than 0, and the mean is to the right of the mode, which is a right-skewed distribution. The typical curve of type 5 does not see extreme values, which is somewhat contradictory to the large peak coefficient, indicating that one curve is not enough to represent the entire type.

[0117] The above method determines multiple output curves based on the row data of the processed power generation matrix; for each output curve, the characteristic vector of the output curve is determined, and the characteristic vector includes the origin moment, the center distance, the skewness and the kurtosis; the output curves are clustered based on the characteristic vectors according to the hierarchical clustering method, and multiple typical monthly daily wind and solar maximum output curves are determined according to the clustering results. The hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets the preset conditions into one cluster until the target cluster is reached. According to the historical power generation data of the First Clean Energy, multiple representative monthly daily wind and solar maximum output curves can be determined, and the short-term flexibility demand can be accurately determined based on the daily wind and solar maximum output curves of each month.

[0118] In an exemplary embodiment, Figure 7 As shown, optionally, the preset flexibility analysis model includes a first sub-model, a second sub-model and a third sub-model, and the short-term flexibility demand includes an intra-day hourly flexibility demand, a daytime flexibility demand and a monthly flexibility demand, and the corresponding short-term flexibility demand is determined according to the maximum output curve of the first cycle and the preset flexibility analysis model, including the following steps 701 to 704. Among them:

[0119] Step 701, determining the daily maximum output value according to the maximum output curve of the first cycle.

[0120] Optionally, the first clean energy may include wind energy and solar energy, the first cycle may be one month, the horizontal axis of the maximum output curve of the first cycle represents each day of the month, and the vertical axis may represent the maximum output value of each day, that is, a data point in the maximum output curve of the first cycle corresponds to the maximum output value of a day in the month.

[0121] Step 702: Determine the intra-day hourly flexibility demand according to the first sub-model.

[0122] Among them, in the first sub-model, the hourly flexibility demand within the day is determined according to the daily maximum output value and the preset daily power generation of the first clean energy.

[0123] Optionally, wind power generation and photovoltaic power generation change constantly within a day and have strong volatility. Therefore, the total output of wind power generation and photovoltaic power stations also changes constantly within a day, and there will be a large gap between peak power and valley power. In order to avoid power shortage or power abandonment, flexible hydropower is used in the hydro-wind-solar hybrid energy system to smooth out the volatility of wind and solar power generation, so that the total output of the system remains unchanged within a day, such as Figure 8 As shown, in order to smooth out the volatility of wind and solar power generation, the wind and solar power output during non-peak hours needs to be compensated by hydropower to the peak power of wind and solar power, and the hydropower generation needs to be greater than Figure 8 The area shaded by the middle diagonal line represents the amount of electricity. Therefore, in the first sub-model, the relationship between the intra-day hourly flexibility demand, the daily power generation of the first clean energy, and the daily maximum output value is as follows:

[0124]

[0125] in, represents the intra-day hourly flexibility demand (also called hourly flexibility demand), represents the daily power generation of the First Clean Energy, Indicates the maximum daily output value; Indicates the time period hours.

[0126] Optionally, the intra-day hourly flexibility demand may be determined based on the maximum output curve of the first period and the first sub-model.

[0127] Step 703: Determine the daytime flexibility requirement according to the second sub-model.

[0128] Among them, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy and the preset lower limit of the daily power generation.

[0129] Optionally, in the power system, a part of the power demand is indispensable, so the water-wind-solar hybrid energy system needs to ensure the minimum daily power supply. Due to the daily volatility of renewable energy, the system's monthly average output is greater than the average minimum output demand and cannot guarantee the minimum daily power supply. It is worth mentioning that the system's daily power supply only needs to meet the minimum power requirement and does not need to be the same every day. Fig. 9As shown in the figure, under the premise of smoothing intraday fluctuations, the system's daily power generation may not meet the daily minimum power supply, and additional flexibility resources are needed to supplement it. Therefore, in the second sub-model, the relationship between the comprehensive intraday hourly flexibility demand, the daytime flexibility demand, the first clean energy daily power generation, the daily maximum output value and the lower limit of the daily power generation is as follows:

[0130]

[0131] in, It represents the daily flexibility demand excluding the intra-day hourly flexibility demand, i.e. the daytime flexibility demand. Indicates the lower boundary of the system's daily power generation.

[0132] Step 704 , determining the monthly flexibility demand according to the third sub-model.

[0133] Among them, in the third sub-model, the monthly flexibility demand is determined based on the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy, the preset lower limit of daily power generation and the preset lower limit of monthly power generation.

[0134] Optionally, accumulation of intra-day hourly flexibility requirements on a monthly scale , the accumulation of daily flexibility needs on a monthly scale and the monthly accumulation of First Clean Energy's daily power generation They are shown in the following formulas respectively:

[0135]

[0136]

[0137]

[0138] Among them, D m Indicates the day of the month.

[0139] Optionally, the union of intra-day hourly flexibility demand and daytime flexibility demand is accumulated monthly as shown in the following formula:

[0140]

[0141] Optionally, the relationship between monthly power generation and short-term flexibility demand can be coupled through the maximum wind and solar output curve within the month. In addition, the hourly, daily and monthly flexibility demand in a year can be Fig.10 As shown, the system also needs to meet the minimum monthly power generation constraint, combining the hourly and daily flexibility requirements, and the hourly-daily-monthly flexibility requirements are shown in the following formula:

[0142]

[0143] in, represents the monthly flexibility demand (it can also be called the monthly flexibility demand excluding the daily and hourly flexibility demand), It is the lower boundary of monthly power generation.

[0144] Optional, based on the above analysis, flexibility requirement Eh m The following formula needs to be satisfied:

[0145]

[0146] The above method determines the daily maximum output value according to the maximum output curve of the first cycle, determines the hourly flexibility demand according to the first sub-model, determines the daytime flexibility demand according to the second sub-model, and determines the monthly flexibility demand according to the third sub-model. In this way, the corresponding short-term flexibility demand can be accurately determined based on the maximum output curve of the first cycle and the preset flexibility analysis model.

[0147] In an exemplary embodiment, Fig.11 As shown, optionally, the statistical model is an autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and the power generation trend of the second clean energy in the second cycle is predicted according to the historical power generation data of the second clean energy and the preset statistical model, including the following steps 1101 to 1102. Among them:

[0148] Step 1101, determining a historical power generation time series according to the historical power generation data of the second clean energy, removing the periodic component from the historical power generation time series, and obtaining a time series after periodic adjustment.

[0149] Optionally, the second clean energy may include wind energy, water energy and solar energy.

[0150] Optionally, the historical power generation time series corresponding to wind energy can be determined based on the historical power generation data of wind power, the historical power generation time series corresponding to hydropower can be determined based on the historical power generation data of hydropower, and the historical power generation time series corresponding to solar energy can be determined based on the historical power generation data of photovoltaic power generation.

[0151] Optionally, the historical power generation time series can be composed of three parts: deterministic trend component, periodic component and random component, which can be expressed by the following formula:

[0152]

[0153] in, is the historical power generation time series, M represents the length of the time series, K represents the power generation variable, which can be any one of the second clean energy sources, represents the deterministic trend component, represents the periodic component, Represents a random component.

[0154] Optionally, the periodic components may be removed from the historical power generation time series by a seasonal decomposition method to obtain a periodically adjusted time series. For example, the seasonal decomposition method may include an X-12 method or an X-11 method.

[0155] Optionally, in the embodiment of the present application, the specific process of removing the periodic component from the historical power generation time series by the X-12 method can be as follows:

[0156] 1. Use the moving average method (MA) to estimate the initial value of the trend component .

[0157] Optionally, the sliding average method may include a simple sliding average method and a weighted sliding average method, which is not limited in the embodiments of the present application.

[0158] 2. From the original data Remove the trend get , denoted as .

[0159] 3. Application Periodic filter from sequence Get the initial value of the periodic component .

[0160] 4. From the original data Eliminate cyclical components get , denoted as .

[0161] 5. Apply the Henderson filter to the sequence Get the trend component .

[0162] 6. From the original data Remove trend components get , denoted as .

[0163] 7. Application Periodic filter from sequence Get the periodic component .

[0164] 8. From the original data Eliminate cyclical components Get the final time series after periodic adjustment , denoted as .

[0165] Step 1102: predict the power generation trend of the second clean energy in the second period according to the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model.

[0166] Optionally, the parameters in the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model can be estimated based on the adjusted time series to determine the specific form of the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and then the period-adjusted time series is substituted into the model for fitting. Through model fitting, an estimate of the time series trend can be obtained.

[0167] Optionally, the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model may include an autoregressive moving average part and a generalized autoregressive conditional heteroskedasticity part.

[0168] Optionally, the autoregressive moving average part is mainly used to model the linear part of the time series, which can be expressed by the following formula:

[0169]

[0170] Among them, k represents the power generation variable, m represents the time point, represents the hydropower, wind and solar power generation sequence after period adjustment, l represents the time index backshift operator, p represents the autoregressive order, q represents the sliding average order, The parameters of the corresponding items can be estimated by adjusting the time series and maximum likelihood estimation or least squares estimation. represents a constant term, represents the forecast residual, that is, the impact of past forecast errors on the current value, = , represents the marginal standardized residual; Represents the standard deviation of the variable.

[0171] Optionally, the generalized autoregressive conditional heteroskedasticity part can be used to describe the dynamic changes in the conditional variance (i.e. volatility) of the time series, which can be expressed by the following formula:

[0172]

[0173] Among them, gp and gq represent the order of the conditional heteroskedastic part of the generalized autoregression, , It can be estimated through the adjusted time series.

[0174] Optionally, when determining the parameters of the autoregressive moving average part and the generalized autoregressive conditional heteroskedasticity part , and Afterwards, the power generation trend of the second clean energy in the second cycle can be predicted based on the adjusted time series.

[0175] Optionally, the long-term flexibility demand may be a monthly flexibility demand. For example, the changing trend and fluctuation of the second clean energy in January may be predicted based on the historical power generation time series of the second clean energy in the previous year or several months before that.

[0176] The above method determines the historical power generation time series based on the historical power generation data of the second clean energy, removes the periodic components from the historical power generation time series, and obtains the time series after periodic adjustment. Then, based on the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, the power generation trend of the second clean energy in the second period is predicted. The autoregressive moving average-generalized autoregressive conditional heteroskedasticity model can well adapt to data with certain linear laws and obvious fluctuations. In this way, the power generation trend and fluctuation of the second clean energy in the second period can be accurately predicted.

[0177] In an exemplary embodiment, optionally, the flexibility requirement quantification method further includes:

[0178] The short-term flexibility demand and the long-term flexibility demand are input into the preset scheduling model, and the scheduling strategy is determined according to the output of the scheduling model.

[0179] Optionally, the preset scheduling model can be an optimization scheduling model pre-established based on an optimization algorithm. After determining the optimization goal and constraints, the scheduling strategy is determined based on the flexibility requirements at different time levels (short-term, medium-term and long-term).

[0180] For example, various constraints such as the hydropower station reservoir groups in the upstream areas of large river basins can be comprehensively considered, and a preset scheduling model can be established from three aspects: the gradient reservoir groups of the river, the reservoir groups of some power grids, and the reservoir groups of the entire river basin, the flexibility needs at different time levels (short-term, medium-term and long-term), and the total power generation of hydropower stations in the upstream areas of long and large river basins, with the optimization goals of maximizing the power generation generated by joint scheduling, maximizing the benefits generated by joint scheduling, and maximizing the amount of hydropower peak regulation.

[0181] The above-mentioned flexibility of coupling clean energy at different time scales can better evaluate and plan the operation of the power system, help formulate effective scheduling strategies and emergency plans, improve the system's adaptability and response speed, and thus ensure a continuous and stable power supply.

[0182] As an optional implementation, Fig.12 As shown, the flexibility requirement quantification method provided in the embodiment of the present application may include the following specific steps:

[0183] Step 1201, determining a power generation matrix according to historical power generation data of a first clean energy source.

[0184] Among them, the matrix element in the i-th row and j-th column in the power generation matrix is ​​the maximum output value on the j-th day in the i-th month, and the maximum output value represents the maximum power output by the first clean energy under ideal operating conditions.

[0185] Step 1202, preprocessing the power generation matrix.

[0186] Among them, preprocessing includes data cleaning and standardization.

[0187] Step 1203, determining a plurality of output curves according to the processed row data of the power generation matrix.

[0188] Step 1204: for each output curve, determine the characteristic vector of the output curve, the characteristic vector includes the origin moment, center distance, skewness and kurtosis.

[0189] Step 1205 , clustering each output curve based on each characteristic vector according to a hierarchical clustering method, and determining a plurality of typical daily maximum wind and solar output curves within a month according to the clustering results.

[0190] Among them, the hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets the preset conditions into one cluster until the target cluster is reached.

[0191] Step 1206: for the daily maximum wind and solar power output curves in each month, determine the daily maximum output value according to the maximum output curve of the first cycle.

[0192] Step 1207, for the daily maximum wind and solar power output curve in each month, determine the hourly flexibility demand within the day according to the first sub-model.

[0193] Among them, in the first sub-model, the hourly flexibility demand within the day is determined according to the daily maximum output value and the preset daily power generation of the first clean energy.

[0194] Step 1208: for the daily maximum wind and solar output curves in each month, determine the daytime flexibility demand according to the second sub-model.

[0195] Among them, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy and the preset lower limit of the daily power generation.

[0196] Step 1209, for the daily maximum wind and solar power output curve in each month, determine the monthly flexibility demand according to the third sub-model.

[0197] Among them, in the third sub-model, the monthly flexibility demand is determined based on the daily maximum output value, the hourly flexibility demand within the day, the preset daily power generation of the first clean energy, the preset lower limit of daily power generation and the preset lower limit of monthly power generation.

[0198] Step 1210, determining a historical power generation time series based on the historical power generation data of the second clean energy, removing periodic components from the historical power generation time series, and obtaining a time series after periodic adjustment.

[0199] Step 1211, predicting the power generation trend of the second clean energy in the second period according to the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model.

[0200] Step 1212, input the short-term flexibility demand and the long-term flexibility demand into a preset scheduling model, and determine the scheduling strategy according to the output of the scheduling model.

[0201] Among them, short-term flexibility demand includes intra-day hourly flexibility demand, daytime flexibility demand and monthly flexibility demand, and long-term flexibility demand includes the power generation trend of the second clean energy in the second cycle.

[0202] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0203] Based on the same inventive concept, the embodiment of the present application also provides a flexibility demand quantification device for implementing the flexibility demand quantification method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more flexibility demand quantification device embodiments provided below can refer to the limitations of the flexibility demand quantification method above, and will not be repeated here.

[0204] In an exemplary embodiment, Fig.13 As shown, a flexibility requirement quantification device 1300 is provided, comprising: a first determination module 1301, a quantification module 1302, and a second determination module 1303, wherein:

[0205] A first determining module 1301 is used to determine an output curve set corresponding to a first period according to historical power generation data of a first clean energy source, where the output curve set includes multiple maximum output curves of the first period;

[0206] A quantification module 1302 is used to determine, for each maximum output curve of the first period, a corresponding short-term flexibility demand according to the maximum output curve of the first period and a preset flexibility analysis model;

[0207] The second determination module 1303 is used to predict the power generation trend of the second clean energy in the second period according to the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand.

[0208] In an exemplary embodiment, the first determination module 1301 is specifically used to determine the power generation matrix based on the historical power generation data of the first clean energy, where the matrix element in the i-th row and j-th column in the power generation matrix is ​​the maximum output value on the j-th day in the i-th month, and the maximum output value represents the maximum power output of the first clean energy under ideal operating conditions; preprocessing the power generation matrix, which includes data cleaning and standardization; clustering the processed power generation matrix according to a preset clustering method, and determining the output curve set corresponding to the first period according to the clustering results.

[0209] In an exemplary embodiment, the clustering method is a hierarchical clustering method, and the set of output curves corresponding to the first period includes multiple typical daily wind and solar maximum output curves within a month. The first determination module 1301 is specifically used to determine multiple output curves based on the processed row data of the power generation matrix; for each output curve, determine the characteristic vector of the output curve, and the characteristic vector includes the origin moment, center distance, skewness and kurtosis; cluster each output curve based on each characteristic vector according to the hierarchical clustering method, and determine multiple typical daily wind and solar maximum output curves within a month according to the clustering results. The hierarchical clustering method is used to regard each output curve as an independent cluster structure, and merge two output curves whose similarity meets preset conditions into one cluster until the target cluster is reached.

[0210] In an exemplary embodiment, the similarity measurement methods of the hierarchical clustering method include: nearest neighbor clustering, farthest neighbor clustering or average clustering; the nearest neighbor clustering is used in the clustering process, for any two cluster structures, to use the nearest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the farthest neighbor clustering is used in the clustering process, for any two cluster structures, to use the farthest distance between each sample point in the two clusters as the measure of the similarity between the two clusters; the average clustering is used in the clustering process, for any two cluster structures, to use the average distance between each sample point in the two cluster structures as the measure of the similarity between the two clusters.

[0211] In an exemplary embodiment, the preset flexibility analysis model includes a first sub-model, a second sub-model and a third sub-model, and the short-term flexibility demand includes intraday hourly flexibility demand, daytime flexibility demand and monthly flexibility demand. The quantification module 1302 is specifically used to determine the daily maximum output value according to the maximum output curve of the first cycle; determine the intraday hourly flexibility demand according to the first sub-model, wherein, in the first sub-model, the intraday hourly flexibility demand is determined according to the daily maximum output value and the preset first clean energy daily power generation; determine the daytime flexibility demand according to the second sub-model, wherein, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the intraday hourly flexibility demand, the preset first clean energy daily power generation and the preset daily power generation lower boundary; determine the monthly flexibility demand according to the third sub-model, wherein, in the third sub-model, the monthly flexibility demand is determined according to the daily maximum output value, the intraday hourly flexibility demand, the preset first clean energy daily power generation, the preset daily power generation lower boundary and the preset monthly power generation lower boundary.

[0212] In an exemplary embodiment, the statistical model is an autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and the second determination module 1303 is specifically used to determine the historical power generation time series based on the historical power generation data of the second clean energy, remove the periodic components from the historical power generation time series, and obtain a time series after periodic adjustment; based on the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, predict the power generation trend of the second clean energy in the second period.

[0213] In an exemplary embodiment, the flexibility demand quantification device also includes a scheduling module, which is used to input short-term flexibility demand and long-term flexibility demand into a preset scheduling model and determine a scheduling strategy according to the output of the scheduling model.

[0214] Each module in the above flexibility demand quantification device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0215] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Fig.14As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a flexibility demand quantification method is implemented.

[0216] Those skilled in the art will understand that Fig.14 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0217] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps described in any of the above method embodiments when executing the computer program.

[0218] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps described in any of the above method embodiments are implemented.

[0219] In one embodiment, a computer program product is provided, including a computer program, which implements the steps described in any of the above method embodiments when executed by a processor.

[0220] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0221] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0222] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A flexibility requirement quantification method, characterized in that: The method comprises: Determine an output curve set corresponding to a first period according to historical power generation data of a first clean energy source, wherein the output curve set includes a plurality of maximum output curves of the first period, and the first period is one month; For each maximum output curve of the first period, determining a corresponding short-term flexibility demand according to the maximum output curve of the first period and a preset flexibility analysis model; Predicting the generation trend of the second clean energy source in the second period based on the historical generation data of the second clean energy source and a preset statistical model to determine the long-term flexibility demand; The preset flexibility analysis model includes a first sub-model, a second sub-model and a third sub-model, the short-term flexibility demand includes an intra-day hourly flexibility demand, a daytime flexibility demand and a monthly flexibility demand, and the corresponding short-term flexibility demand is determined according to the maximum output curve of the first cycle and the preset flexibility analysis model, including: Determine the daily maximum output value according to the maximum output curve of the first cycle; Determining the intra-day hourly flexibility demand according to the first sub-model, wherein, in the first sub-model, the intra-day hourly flexibility demand is determined according to the daily maximum output value and a preset first clean energy daily power generation; The first sub-model is: ,in, represents the hourly flexibility requirement within the day, represents the daily power generation of the first clean energy, Indicates the maximum daily output value; Indicates the hours of the time period; Determining the daytime flexibility demand according to the second sub-model, wherein, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the intra-day hourly flexibility demand, the preset first clean energy daily power generation and a preset daily power generation lower limit; The second sub-model is: ,in, represents the daytime flexibility requirement, represents the lower boundary of the daily power generation; Determining the monthly flexibility demand according to a third sub-model, wherein, in the third sub-model, the monthly flexibility demand is determined according to the daily maximum output value, the intra-day hourly flexibility demand, the preset daily power generation of the first clean energy, the preset daily power generation lower boundary, and the preset monthly power generation lower boundary; The third sub-model is: ,in, represents the monthly flexibility requirement, is the lower boundary of the monthly power generation, , , , D m Indicates the day of the month.

2. The method according to claim 1, characterized in that: The step of determining a set of output curves corresponding to the first period according to the historical power generation data of the first clean energy source includes: Determine a power generation matrix according to the historical power generation data of the first clean energy source, wherein the matrix element in the i-th row and the j-th column in the power generation matrix is ​​the maximum output value on the j-th day in the i-th month, and the maximum output value represents the maximum power output by the first clean energy source under ideal operating conditions; Preprocessing the power generation matrix, wherein the preprocessing includes data cleaning and standardization; The processed power generation matrix is ​​clustered according to a preset clustering method, and a set of output curves corresponding to the first period is determined according to the clustering result.

3. The method according to claim 2, characterized in that The clustering method is a hierarchical clustering method, the output curve set corresponding to the first period includes a plurality of typical daily wind and solar maximum output curves within a month, the processed power generation matrix is ​​clustered according to a preset clustering method, and the output curve set corresponding to the first period is determined according to the clustering result, including: Determining a plurality of output curves according to the processed row data of the power generation matrix; For each of the output curves, determining a characteristic vector of the output curve, the characteristic vector comprising an origin moment, a center distance, a skewness and a kurtosis; According to the hierarchical clustering method, each of the output curves is clustered based on each of the characteristic vectors, and a plurality of typical daily maximum wind and solar output curves within a month are determined according to the clustering results. The hierarchical clustering method is used to regard each of the output curves as an independent cluster structure, and merge two output curves whose similarity meets preset conditions into one cluster until the target cluster is reached.

4. The method according to claim 3, characterized in that The similarity measurement methods of the hierarchical clustering method include: nearest neighbor clustering, farthest neighbor clustering or average clustering; The nearest neighbor clustering is used to use the shortest distance between sample points in the two clusters as a measure of similarity between the two clusters for any two cluster structures during the clustering process; The farthest neighbor clustering is used to use, during the clustering process, for any two cluster structures, the farthest distance between the sample points in the two clusters as a measure of the similarity between the two clusters; The average clustering is used to use, during the clustering process, for any two cluster structures, the average distance between the sample points in the two cluster structures as a measure of the similarity between the two clusters.

5. The method according to claim 1, characterized in that The statistical model is an autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, and the power generation trend of the second clean energy in the second period is predicted based on the historical power generation data of the second clean energy and a preset statistical model, including: Determine a historical power generation time series according to the historical power generation data of the second clean energy source, remove periodic components from the historical power generation time series, and obtain a time series after periodic adjustment; According to the adjusted time series and the autoregressive moving average-generalized autoregressive conditional heteroskedasticity model, the power generation trend of the second clean energy in the second period is predicted.

6. The method according to claim 1, characterized in that The method further comprises: The short-term flexibility demand and the long-term flexibility demand are input into a preset scheduling model, and a scheduling strategy is determined according to an output of the scheduling model.

7. A flexibility demand quantification device, characterized in that: The device comprises: A first determination module, configured to determine an output curve set corresponding to a first period according to historical power generation data of a first clean energy source, wherein the output curve set includes a plurality of maximum output curves of the first period; a quantification module, configured to determine, for each maximum output curve of the first period, a corresponding short-term flexibility demand according to the maximum output curve of the first period and a preset flexibility analysis model; A second determination module is used to predict the power generation trend of the second clean energy in the second period according to the historical power generation data of the second clean energy and a preset statistical model to determine the long-term flexibility demand; The preset flexibility analysis model includes a first sub-model, a second sub-model and a third sub-model. The quantification module is specifically used to determine the daily maximum output value according to the maximum output curve of the first period, and the first period is one month; determine the intra-day hourly flexibility demand according to the first sub-model, wherein, in the first sub-model, the intra-day hourly flexibility demand is determined according to the daily maximum output value and the preset first clean energy daily power generation; the first sub-model is: ,in, represents the hourly flexibility requirement within the day, represents the daily power generation of the first clean energy, Indicates the maximum daily output value; represents the number of hours in the time period; determining the daytime flexibility demand according to the second sub-model, wherein, in the second sub-model, the daytime flexibility demand is determined according to the daily maximum output value, the intra-day hourly flexibility demand, the preset first clean energy daily power generation and the preset daily power generation lower limit; the second sub-model is: ,in, represents the daytime flexibility requirement, represents the lower limit of the daily power generation; and determines the monthly flexibility demand according to the third sub-model, wherein, in the third sub-model, the monthly flexibility demand is determined according to the daily maximum output value, the intra-day hourly flexibility demand, the preset first clean energy daily power generation, the preset daily power generation lower limit and the preset monthly power generation lower limit; the third sub-model is: ,in, represents the monthly flexibility requirement, is the lower boundary of the monthly power generation, , , , D m Indicates the day of the month.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of 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 a computer program, 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.

Images