Generation method and device of output sequence scene of new energy station and electronic equipment

Abstract

The invention discloses a new energy station output sequence scene generation method and apparatus, and an electronic device. The method comprises the steps of obtaining a target date and a target extreme scene type corresponding to a target new energy station on the target date; deducing daily scene types of each day in the target year based on the target extreme scene type and the target state transition matrix by taking the target date as an anchor point, and determining a daily scene type corresponding to each day from a daily scene type set; based on the daily scene type corresponding to each day, sampling from the joint probability distribution corresponding to the daily scene type to obtain an output daily characteristic index corresponding to each day; determining a target daily output sequence corresponding to each day based on the output day characteristic index corresponding to each day; and generating the target annual output sequence scene of the target new energy station based on the target daily output sequence corresponding to each day, so that the accuracy of the new energy output sequence scene in the face of an extreme weather event can be improved.

Description

Technical Field

[0001] This application relates to the field of power analysis technology, and in particular to a method, apparatus and electronic equipment for generating power output sequence scenarios of new energy power plants. Background Technology

[0002] With the advancement of the "dual carbon" target, the proportion of new energy sources such as wind power and photovoltaics in the power system is increasing. The randomness and volatility of their output pose a serious challenge to the safe and stable operation of the power system. Against this backdrop, conducting accurate power system adequacy assessments is crucial. The reliability of the assessment results directly depends on whether the new energy output sequence scenarios used accurately reflect their uncertainties, especially their performance under extreme weather events.

[0003] In related technologies, the generation of power output sequence scenarios often relies on direct sampling of historical data or simple statistical modeling, resulting in low accuracy of the generated power output sequence scenarios when facing extreme weather events. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0005] The main objective of this disclosure is to propose a method, apparatus, and electronic device for generating power output sequence scenarios for new energy power plants, which can improve the accuracy of power output sequence scenarios in the face of extreme weather events.

[0006] Obtain the target date and the target extreme scenario type corresponding to the target new energy power station on the target date; Using the target date as an anchor point, the daily scene type for each day within the target year is derived based on the target extreme scene type and the target state transition matrix, and the daily scene type corresponding to each day is determined from the set of daily scene types; the target year is the year in which the target date is located, and the target state transition matrix is ​​used to calculate the daily scene type corresponding to the date before or after the current date; Based on the daily scene type corresponding to each day, the daily output characteristic index corresponding to each day is obtained by sampling from the joint probability distribution corresponding to the daily scene type; the joint probability distribution includes the combination relationship between each daily output characteristic index in the daily scene type. Based on the daily output characteristic indicators, the daily target day output sequence is determined. Based on the daily target day power generation sequence, the target annual power generation sequence scenario of the target new energy power station is generated.

[0007] In some embodiments of this application, the set of daytime scene types is determined through the following steps: The power generation of the target renewable energy power plant in each month in history is obtained, and the power generation of each month is determined based on the historical daily power generation sequence of the target renewable energy power plant. With the goal of minimizing the monthly power generation differences in each quarter, 12 consecutive months are divided based on the power generation of each month to determine multiple target quarters; each target quarter includes at least two months, and the months in each target quarter are consecutive. Feature extraction is performed on the historical daily output sequence to obtain the output daily feature index; Based on the aforementioned output day characteristic indicators, the historical daily output sequences in each target quarter are clustered to determine the types of non-extreme scenario days; Historical extreme sample days that meet the preset extreme scenario definition are selected from the historical daily power generation sequence; Clustering is performed based on the historical daily power output sequences corresponding to the aforementioned historical extreme sample days to obtain the extreme day scenario types; Based on the non-extreme scenario day type and the extreme scenario day type, construct the day scenario type set.

[0008] In some embodiments of this application, the target new energy power station includes multiple new energy sub-power stations; The preset extreme scenario definition includes: If the number of new energy sub-stations in the target new energy power station that are experiencing consecutive low-output days is greater than a first threshold, it is determined that the target new energy power station is in an extreme scenario of continuous low new energy output. If the number of new energy sub-stations in the target new energy power station that are on a day of large fluctuations is greater than the first threshold, it is determined that the target new energy power station is in an extreme scenario of large fluctuations in new energy. If the sum of the number of new energy sub-stations in the target new energy power station that are in the continuous low-output days and the number of new energy sub-stations that are in the large-fluctuation days is greater than the first threshold, it is determined that the target new energy power station is in an extreme scenario of continuous low output and large fluctuation of new energy. Wherein, if the average daily output of the new energy sub-station is less than the second threshold, the new energy sub-station is determined to be in the continuous low output day. If the daily power output standard deviation of the new energy sub-station is greater than the third threshold, the new energy sub-station is determined to be on the day of large fluctuation; the second threshold is the difference between the average daily power output of all historical sample days in the target new energy sub-station and twice the standard deviation of the daily average power output sample; the third threshold is the sum of the average daily power output standard deviation of all historical sample days in the target new energy sub-station and twice the standard deviation of the daily power output standard deviation sample.

[0009] In some embodiments of this application, the target state transition matrix includes at least one of a daily state transition matrix and a quarterly state transition matrix; Before deriving the daily scene type for each day within the target year based on the target extreme scene type and the target state transition matrix, using the target date as the anchor point, and determining the target daily scene type corresponding to each day from the target daily scene type set, the method further includes: Within each target quarter, the daily scene types corresponding to each of the historical daily power output sequences are statistically analyzed to obtain the frequency of mutual transfer between the daily scene types within the target quarter; The frequency of transitions between the various daily scene types within the target quarter is determined by the maximum likelihood method to form a daily state transition matrix. This daily state transition matrix is ​​used to calculate the daily scene type corresponding to the previous or next date of the current date within each target quarter. The daily scene type on the first day of each target quarter and the daily scene type corresponding to the previous date of the first day of each target quarter are statistically analyzed to obtain the frequency of daily scene type transitions between the target quarters. Based on the frequency of daily scene type transitions between the target quarters, the quarterly state transition matrix is ​​determined by the maximum likelihood method; the quarterly state transition matrix is ​​used to calculate the daily scene type of the first day of the current quarter.

[0010] In some embodiments of this application, generating the target annual output sequence scenario of the target renewable energy power station based on the target daily output sequence includes: By splicing the daily target day power output sequence in chronological order, the initial annual power output sequence scenario of the target new energy power station is obtained; Based on the initial annual power output sequence scenario, the initial annual electricity consumption of the target renewable energy power station is determined; If the initial annual electricity consumption is within the target range, the ratio of the target boundary value of the target range to the initial annual electricity consumption is determined as the scaling factor; the target boundary value is the minimum or maximum annual electricity consumption of the target renewable energy power station. The target annual output sequence scenario is obtained by multiplying each output point in the initial annual output sequence scenario by the scaling factor.

[0011] In some embodiments of this application, the target boundary value is determined through the following steps: Obtain the historical annual utilization rate data of the target new energy power station; The annual utilization rate data is fitted using the kernel density estimation method to obtain the probability density distribution of the annual utilization rate data; Based on preset probability values, the annual utilization rate in the probability density distribution is filtered to determine the upper and lower limits of the annual utilization rate. Based on the upper and lower limits of the annual utilization rate, the minimum and maximum annual electricity consumption are determined.

[0012] In some embodiments of this application, the number of the daily output characteristic indicators is multiple; Before obtaining the daily output characteristic index by sampling from the joint probability distribution corresponding to the daily scene type based on the daily scene type, the method further includes: For each day scenario type, execute: Historical data of the daily characteristic indicators of each output day corresponding to the daily scenario type are extracted from the historical output sequence of the target new energy power station. By fitting the historical data of the characteristic indicators of each output day using the kernel density estimation method, the marginal probability density distribution of the characteristic indicators of each output day is obtained. Based on the marginal probability density distribution of each output day characteristic index, the correlation between each output day characteristic index in the day scene type is analyzed by a preset function to obtain the joint probability distribution corresponding to the day scene type.

[0013] In some embodiments of this application, determining the target day's output sequence based on the daily output characteristic indicators includes: For the corresponding daily output characteristic indicators, execute: The normalized Euclidean distance between the daily output characteristic index and the historical daily output sequence characteristic index of the target new energy power station is calculated. The historical sunrise power sequence with the smallest normalized Euclidean distance is used as the sample sunrise power sequence; The optimization model is solved based on the sample daily power output sequence to obtain the target daily power output sequence; wherein the optimization model takes minimizing the normalized Euclidean distance between the target daily power output sequence and the sample daily power output sequence as the objective function, and takes the characteristic index of the target daily power output sequence being equal to the characteristic index of the power output day as the constraint condition.

[0014] To achieve the above objective, a second aspect of the present invention provides an apparatus for generating a power output sequence scenario for a new energy power station, the apparatus comprising: The acquisition module is used to acquire the target date and the target extreme scenario type of the target new energy power station corresponding to the target date; The derivation module is used to derive the daily scene type of each day within the target year based on the target extreme scene type and the target state transition matrix, using the target date as the anchor point, and to determine the daily scene type corresponding to each day from the daily scene type set; the target year is the year in which the target date is located, and the target state transition matrix is ​​used to calculate the daily scene type corresponding to the previous or next date of the current date; The sampling module is used to sample the daily output characteristic indexes corresponding to the daily scene type from the joint probability distribution corresponding to the daily scene type; the joint probability distribution includes the combination relationship between the daily output characteristic indexes in the daily scene type. The determination module is used to determine the target day output sequence based on the daily output characteristic indicators. The generation module is used to generate the target annual output sequence scenario of the target new energy power station based on the target daily output sequence corresponding to each day.

[0015] To achieve the above objectives, a third aspect of the present invention provides an electronic device, comprising: at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, the instructions being executed by the at least one control processor to enable the at least one control processor to execute the above-described method for generating a power output sequence scenario of a new energy power station.

[0016] To achieve the above objectives, a fourth aspect of the present invention provides a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described method for generating a power output sequence scenario of a new energy power station.

[0017] To achieve the above objectives, a fifth aspect of the present invention provides a computer program product in which the instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to execute the above-described method for generating a power output sequence scenario of a new energy power station.

[0018] This application provides a method for generating a power output sequence scenario for a new energy power station. The method involves obtaining a target date and the target extreme scenario type corresponding to that date. First, using the target date as an anchor point, the daily scenario type for each day within the target year is derived using the target state transition matrix and the target extreme scenario type to determine the corresponding daily scenario type. The daily scenario type for each day within the target year is accurately derived using the target state transition matrix. Then, based on the accurate daily scenario type, sampling is performed from the joint probability distribution corresponding to the daily scenario type to obtain the daily output characteristic index. Based on the daily output characteristic index, the target daily output sequence for each day is determined. Finally, based on the daily target daily output sequence, the target annual output sequence scenario for the target new energy power station is generated. Therefore, through extreme scenario type derivation, joint probability distribution sampling, and output sequence optimization, the output characteristics of new energy power stations under extreme weather conditions are accurately simulated, improving the accuracy of new energy power output sequence scenarios in the face of extreme weather events.

[0019] It is understood that the beneficial effects of the second to fifth aspects compared with the related technologies are the same as the beneficial effects of the first aspect compared with the related technologies. Please refer to the relevant description in the first aspect above, which will not be repeated here. Attached Figure Description

[0020] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating a method for generating a power output sequence scenario for a new energy power station, as provided in an embodiment of this application. Figure 2 This is a schematic diagram illustrating the implementation process of a method for generating a power output sequence scenario for a new energy power station, as provided in an embodiment of this application. Figure 3 This is a schematic diagram of the structure of a device for generating a power output sequence scenario of a new energy power station, provided in an embodiment of this application. Figure 4 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0021] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples.

[0022] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0023] In all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. Additionally, when embodiments of this application require access to sensitive personal information, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments obtained.

[0024] To address the problems of existing technologies, this application provides a method, apparatus, and electronic device for generating power output sequence scenarios for new energy power plants. The method for generating power output sequence scenarios for new energy power plants provided in this application will be described first.

[0025] Figure 1 This illustration shows a flowchart of a method for generating a power output sequence scenario for a new energy power station according to an embodiment of this application. Figure 1 As shown in the embodiment of this application, the method for generating a power output sequence scenario for a new energy power station includes the following steps 101-105, wherein: Step 101: Obtain the target date and the target extreme scenario type of the target new energy power station corresponding to the target date.

[0026] In this step, the target date can be any date specified by the user or any date preset. The target renewable energy plant may include at least one wind farm and / or at least one photovoltaic farm.

[0027] The target extreme scenario type corresponding to the target new energy power station on the target date can be any extreme scenario type in the user-specified daily scenario type set, or any extreme scenario type in the pre-set daily scenario type set.

[0028] Furthermore, the daily scene type set can be obtained by clustering based on the historical daily power generation sequence of the target new energy power station.

[0029] In other implementations, the target date and the daily scenario type of the target new energy power station corresponding to the target date can be obtained. This daily scenario type belongs to the daily scenario type set, so as to realize the generation of output sequence for non-specified extreme scenarios.

[0030] Step 102: Using the target date as the anchor point, deduce the daily scene type for each day within the target year based on the target extreme scene type and the target state transition matrix, and determine the daily scene type corresponding to each day from the daily scene type set; the target year is the year in which the target date is located, and the target state transition matrix is ​​used to calculate the daily scene type corresponding to the previous or next date of the current date.

[0031] In this step, the target year is the year in which the target date falls. For example, if the target date is June 15, 2024, then the target year is 2024.

[0032] The target state transition matrix can be a probability relationship matrix describing the changes of daily scene types over time. It can be obtained by statistically analyzing the historical scene type transition frequency of the target new energy power station and applying the maximum likelihood method. It is used to simulate the dynamic evolution of daily scene types over time.

[0033] The daily scenario type set can be a classification set of daily operating statuses that covers both normal and extreme scenarios. Specifically, it can be constructed by clustering analysis of historical output sequences and combining extreme scenario screening rules to ensure that the scenario types cover the diversity of actual operation.

[0034] In some implementations, the target state transition matrix is ​​used to deduce the day scene type for each day after the target date by working backward from the target date. Then, the target state transition matrix is ​​used to deduce the day scene type for each day before the target date by working backward from the target date. It should be noted that the target state transition matrices used for forward and backward derivations are different; the forward target state transition matrix can be the inverse of the backward target state transition matrix.

[0035] Step 103: Based on the daily scene type corresponding to each day, sample from the joint probability distribution corresponding to the daily scene type to obtain the daily output characteristic index corresponding to each day; the joint probability distribution includes the combination relationship between each output daily characteristic index in the daily scene type.

[0036] In this step, the joint probability distribution can be used to indicate the statistical correlation between multiple output day characteristic indicators under the same day scenario type. This can be achieved by using kernel density estimation to fit the historical data of the target new energy power station and analyzing the correlation between indicators, so as to ensure that the sampling results conform to the actual physical laws.

[0037] Output daily characteristic indicators can be key parameters for quantifying the characteristics of a single day's output sequence. For example, for wind farms, these can be the daily average wind power output, daily standard deviation of wind power output, daily maximum wind power output, daily minimum wind power output, daily wind power output kurtosis, and daily wind power output skewness; for photovoltaic farms, these can be the daily average photovoltaic power output, daily standard deviation of photovoltaic power output, daily maximum photovoltaic power output, daily photovoltaic output kurtosis, and daily photovoltaic output skewness, used to reconstruct an output sequence that conforms to the scenario type.

[0038] Step 104: Based on the daily output characteristic indicators, determine the daily target day output sequence.

[0039] In this step, the target day power generation sequence can be the power generation data curve of the renewable energy power station changing over time within a single day. It can be generated by optimizing the model to match characteristic indicators and retaining typical power generation patterns, ensuring that the target day power generation sequence satisfies both statistical characteristics and physical feasibility.

[0040] Step 105: Based on the daily target day power generation sequence, generate the target annual power generation sequence scenario for the target new energy power station.

[0041] In this step, the target day output sequence for all dates within the target year can be spliced ​​together in chronological order to form a complete output sequence (i.e., the target year output sequence scenario).

[0042] In this embodiment, the target date and the target extreme scenario type corresponding to the target new energy power station are obtained. First, using the target date as an anchor point, the daily scenario type for each day within the target year is derived using the target state transition matrix and the target extreme scenario type to determine the corresponding daily scenario type. The daily scenario type for each day within the target year is accurately derived using the target state transition matrix. Subsequently, based on the accurate daily scenario type, sampling is performed from the joint probability distribution corresponding to the daily scenario type to obtain the daily output characteristic index. Then, based on the daily output characteristic index, the daily target daily output sequence is determined. Finally, based on the daily target daily output sequence, the target annual output sequence scenario for the target new energy power station is generated. Thus, through extreme scenario type derivation, joint probability distribution sampling, and output sequence optimization generation, the output characteristics of new energy power stations under extreme weather conditions are accurately simulated, improving the accuracy of the output sequence scenario when facing extreme weather events.

[0043] In some embodiments, a method is proposed to derive daily scene types from the target state transition matrix to generate power output sequence scenes. However, in determining the set of daily scene types, if the quarterly division method fails to effectively reflect the actual power generation distribution characteristics, it may lead to biased clustering results, thereby affecting the accuracy of extreme scene identification. Therefore, the set of daily scene types can be determined through the following steps: The power generation of the target renewable energy power plant in each month in history is obtained, and the power generation of each month is determined based on the historical daily power generation sequence of the target renewable energy power plant. With the goal of minimizing the monthly power generation differences in each quarter, 12 consecutive months are divided based on the power generation of each month to determine multiple target quarters; each target quarter includes at least two months, and the months in each target quarter are consecutive. Feature extraction is performed on the historical daily output sequence to obtain the output daily feature index; Based on the aforementioned output day characteristic indicators, the historical daily output sequences in each target quarter are clustered to determine the types of non-extreme scenario days; Historical extreme sample days that meet the preset extreme scenario definition are selected from the historical daily power generation sequence; Clustering is performed based on the historical daily power output sequences corresponding to the aforementioned historical extreme sample days to obtain the extreme day scenario types; Based on the non-extreme scenario day type and the extreme scenario day type, construct the day scenario type set.

[0044] In this embodiment, the power generation of each month in the history of the target new energy scenario can be obtained by adding up the daily power generation corresponding to the historical daily power generation sequence of each month.

[0045] Furthermore, this embodiment does not use the conventional four-quarter division or the four-season division, but instead adopts a quarterly division method based on monthly renewable energy power generation (i.e., monthly power generation). Specifically, this can be achieved by obtaining the quarterly division result with the smallest average difference in monthly renewable energy power generation for each quarter, i.e., minimizing the value of the following equation 1: (1); In the formula, This represents the average quarterly difference in new energy power generation. Number of quarters; For quarters The average monthly electricity consumption difference of new energy sources within the region; For quarters The number of months included.

[0046] Furthermore, some constraints can be set for the division of quarters, such as: the months in a quarter should be consecutive, for example, January and March should not be the first quarter and February should not be the second quarter, but it is acceptable for January of the following year and December of the previous year to belong to the same quarter; a quarter should avoid having only one month, etc.

[0047] In some implementation methods, the following steps can be used to divide the quarters and obtain multiple target quarters: Step 1: Set For month variables, For quarterly variables, For the first The number of months included in a quarter. The starting month is used to divide the quarter. , , , .

[0048] Step 2: Calculate the average difference in monthly electricity consumption of new energy sources in the current quarter using the following formula (11)2.

[0049] (2); Step 3: Using variables Temporary record of the current quarterly division scheme The average difference in wind power output per quarter, i.e. .

[0050] Step 4: Order Calculated by equation (11) .

[0051] Step 5: Using variables Temporary record of the current The total number of months included in a quarter, i.e. .

[0052] Step 6: If and If so, return to step 4; and Then The quarterly division for the starting month of the quarterly division has ended; proceed to step 7. and Then let The remaining months will be considered the final quarter, concluding this quarterly division. Proceed to step 7; otherwise, [then proceed to step 7]. and The month as the first Return to step 2 after one quarter.

[0053] Step 7: Calculate and record the average difference of the new energy power generation in the current quarter corresponding to the division of the current quarter using formula (10).

[0054] Step 8: Order , .like Then the quarterly division is complete, and the smallest value is selected. The corresponding quarterly division result is output as the final division result (i.e., the target quarter); otherwise, let... ,Pick , Month as the first The months included in a quarter.

[0055] Furthermore, based on the quarterly division, k-means clustering is performed on the historical daily power output sequence sample data under the same target quarter. For target new energy power plants that include multiple wind farms and multiple photovoltaic power plants, both daily wind power output and daily photovoltaic power output need to be considered simultaneously. Each new energy power plant corresponds to 24-point daily power output sequence data. When the number of included power plants is large, there will be a tendency for the Euclidean distance between samples to be close in high-dimensional cases. To overcome this problem, dimensionality reduction processing is performed on the data in this embodiment.

[0056] In some implementation methods, data dimensionality reduction can be achieved as follows: For wind farms, six data points are selected as the daily power output characteristics indicators of wind farms: average daily wind power output, standard deviation of daily wind power output, maximum daily wind power output, minimum daily wind power output, kurtosis of daily wind power output, and skewness of daily wind power output, to replace the original 24-point daily power output sequence data; for photovoltaic (PV) farms, five data points are selected as the daily power output characteristics indicators of PV farms: average daily PV power output, standard deviation of daily PV power output, maximum daily PV power output, kurtosis of daily PV power output, and skewness of daily PV power output, to replace the original 24-point daily power output sequence data.

[0057] For example, taking a specific target renewable energy power station as an example, its daily output characteristic indicators and daily scenario clustering can be obtained through the following steps: Step 1: Assign the current quarter to the first quarter.

[0058] Step 2: Assign the current day to the first day of the current quarter.

[0059] Step 3: For the current day, extract 24-point data for wind power (or photovoltaic) from each station, assuming a total of wind farms and A photovoltaic power station, generating a [missing information] A vector of n elements.

[0060] Step 4: Calculate the average daily wind power (or photovoltaic) output, standard deviation of daily wind power (or photovoltaic) output, maximum daily wind power (or photovoltaic) output, minimum daily wind power output (not required for photovoltaic, which is 0), peak daily wind power (or photovoltaic) output, and skewness of daily wind power (or photovoltaic) output for each station in the current quarter and on the current day according to Equations 3 to 8.

[0061] (3); (4); (5); (6); (7); (8); In the formula, For the first The first wind farm, the first Heaven, the first Average output per hour (wind power hourly output); For the first The first wind farm, the first Average daily output (average daily wind power output); For the first The first wind farm, the first Standard deviation of hourly average output per day (standard deviation of daily wind power output). For the first The first wind farm, the first The maximum daily average hourly output (maximum daily wind power output). For the first The first wind farm, the first Minimum hourly average output per day (minimum daily wind power output). For the first The first wind farm, the first Peak hourly average output per day (peak daily wind power output); For the first The first wind farm, the first Daily wind power output deviation (dayly wind power output deviation).

[0062] It should be noted that equations 3 to 8 above are based on wind farms. For photovoltaic farms, it is only necessary to... Replace with That's all.

[0063] Step 5: Assign the current day to the next day, and repeat steps 3 and 4. If the next day does not exist, proceed to step 7.

[0064] Step 6: Based on the characteristic indicators of the output days, use the k-means method to cluster the output series samples for each day in the current quarter. The number of clusters is selected based on the sum of the squared errors (SSE) within the clusters, obtaining the typical daily renewable energy output series and the number of typical daily types, and saving them. If the input data includes multi-year data, for a specific day, determine the day type (i.e., the non-extreme scenario day type) based on which cluster the samples corresponding to that day are most likely to be clustered into.

[0065] Step 7: Assign the current quarter to the next quarter, and repeat steps 2 through 6 until the traversal is complete.

[0066] Furthermore, the daily scene types obtained by clustering in steps 1-7 above do not cluster the historical daily power output sequences corresponding to extreme scene types, resulting in non-extreme scene daily types. Therefore, sample days that meet preset extreme conditions (such as continuous low power output or large fluctuations) are selected from the historical daily power output sequences and clustered separately to form extreme daily scene types.

[0067] Furthermore, by combining non-extreme scene types with extreme scene types, a complete set of daily scene types can be constructed.

[0068] In some implementations, the target renewable energy power station includes multiple renewable energy sub-stations, which can be wind farms and photovoltaic (PV) farms. For wind farms, the two extreme scenarios affecting power adequacy are continuous low output and short-term large fluctuations. For PV farms, the two extreme scenarios also affecting power adequacy are continuous low output and short-term large fluctuations. Therefore, the preset extreme scenarios can be defined as follows: If the number of new energy sub-stations in the target new energy power station that are experiencing consecutive low-output days is greater than a first threshold, it is determined that the target new energy power station is in an extreme scenario of continuous low new energy output. If the number of new energy sub-stations in the target new energy power station that are on a day of large fluctuations is greater than the first threshold, it is determined that the target new energy power station is in an extreme scenario of large fluctuations in new energy. If the sum of the number of new energy sub-stations in the target new energy power station that are in the continuous low-output days and the number of new energy sub-stations that are in the large-fluctuation days is greater than the first threshold, it is determined that the target new energy power station is in an extreme scenario of continuous low output and large fluctuation of new energy. Wherein, if the average daily output of the new energy sub-station is less than the second threshold, the new energy sub-station is determined to be in the continuous low output day. If the daily power output standard deviation of the new energy sub-station is greater than the third threshold, the new energy sub-station is determined to be on the day of large fluctuation; the second threshold is the difference between the average daily power output of all historical sample days in the target new energy sub-station and twice the standard deviation of the daily average power output sample; the third threshold is the sum of the average daily power output standard deviation of all historical sample days in the target new energy sub-station and twice the standard deviation of the daily power output standard deviation sample.

[0069] In this implementation, a consecutive low-output day for a certain renewable energy power station (i.e., a renewable energy sub-station) is defined as a day when the average daily output of a certain renewable energy power station is lower than the average daily output of all samples minus two sample standard deviations; a large fluctuation day for a certain renewable energy power station is defined as a day when the daily output standard deviation of a certain renewable energy power station is greater than the average of the daily output standard deviations of all samples plus the standard deviation of two sample daily output standard deviations. The threshold can be determined according to the following formula 9.

[0070] (9); In the formula, This represents the second threshold. Indicates the first l The average daily output of each sample day Indicates the first l The daily power sequence of each sample day.

[0071] Furthermore, the first threshold can be set according to the actual situation, and no specific limitation is made here.

[0072] In some implementations, the first threshold can be set to 1 / 3 of the total number of new energy sub-stations in the target new energy power station. That is, if more than 1 / 3 of the new energy sub-stations in the target new energy power station are in a scenario of consecutive low output days, it is determined as a scenario of continuous low output of new energy; if more than 1 / 3 of the new energy power stations are in a scenario of large fluctuation days, it is determined as a scenario of large fluctuation of new energy. Taking into account the combination of the two extreme scenarios (i.e., the extreme scenario of continuous low output and large fluctuation of new energy), there are a total of 3 extreme scenarios.

[0073] For example, if a certain renewable energy power station comprises 21 sub-stations, then the first threshold is 7. When the daily average output of 12 sub-stations is lower than the historical average minus two standard deviations on a certain day, the extreme scenario of continuous low output of renewable energy is triggered. At the same time, if on another day 7 sub-stations experience large fluctuations and 7 sub-stations experience low output, the total number reaches 15, exceeding the threshold, triggering the extreme scenario of continuous low output and large fluctuations of renewable energy.

[0074] In this implementation, by seasonally segmenting and clustering historical data, the seasonal characteristics and extreme conditions of renewable energy output can be captured more accurately, thereby improving the comprehensiveness and representativeness of the generated scenarios. Furthermore, by employing a data-driven clustering method, different types of daily scenarios can be adaptively identified without being limited by preset types. This allows the generated daily scenario types to better reflect actual output characteristics, especially in the face of complex and variable weather conditions. Consequently, this provides more reliable input data for subsequent power system adequacy assessments, contributing to improved accuracy and credibility of the assessment results.

[0075] In some embodiments, a set of daily scene types is determined by dividing target quarters and clustering historical data based on output day characteristic indicators. However, when generating the daily scene type sequence, the time transition patterns of daily scene types within and between different quarters are not considered, which may result in the generated daily scene type sequence not conforming to the time continuity and seasonal variation characteristics in actual operation. Therefore, the target state transition matrix of this application includes at least one of a daily state transition matrix and a quarterly state transition matrix. Specifically, before step 102, the method for generating the output sequence scene of the new energy power station provided in this application embodiment may further include: Within each target quarter, the daily scene types corresponding to each of the historical daily power output sequences are statistically analyzed to obtain the frequency of mutual transfer between the daily scene types within the target quarter; The frequency of transitions between the various daily scene types within the target quarter is determined by the maximum likelihood method to form a daily state transition matrix. This daily state transition matrix is ​​used to calculate the daily scene type corresponding to the previous or next date of the current date within each target quarter. The daily scene type on the first day of each target quarter and the daily scene type corresponding to the previous date of the first day of each target quarter are statistically analyzed to obtain the frequency of daily scene type transitions between the target quarters. Based on the frequency of daily scene type transitions between the target quarters, the quarterly state transition matrix is ​​determined by the maximum likelihood method; the quarterly state transition matrix is ​​used to calculate the daily scene type of the first day of the current quarter.

[0076] In this implementation, the daily state transition matrix is ​​constructed as follows: for each target quarter, the number of transitions between daily scene types on two adjacent days within that quarter is counted to form a transition frequency matrix; the frequency data in the transition frequency matrix is ​​input into the maximum likelihood estimation algorithm to calculate the transition probability between each scene type, thus forming the daily state transition matrix.

[0077] The quarterly state transition matrix is ​​constructed as follows: For each target quarter, the daily scene type corresponding to the first day and the daily scene type of the previous day are extracted, and the transition frequency of the scene type on the first day of the quarter and the scene type on the previous day is counted; the transition frequency is input into the maximum likelihood estimation algorithm to calculate the scene type transition probability between quarters, thus forming the quarterly state transition matrix.

[0078] Specifically, when generating daily scene type sequences, the daily state transition matrix is ​​used to control the scene type transitions between adjacent dates within the same target quarter. For example, if the current date is in the third quarter, the daily scene type is calculated based on the daily state transition matrix for the third quarter. The quarterly state transition matrix is ​​used to control the scene type transitions on the first day of the quarter. For example, when the date transitions from the last day of the second quarter to the first day of the third quarter, the scene type for the first day of the third quarter is calculated based on the quarterly state transition matrix. Through the synergistic effect of the dual transition matrices, both the continuity of daily scene type changes within a quarter and the seasonality of scene type transitions between quarters are maintained. In practical applications, the calculation process of the maximum likelihood method can be specifically represented as follows: dividing each element in the transition frequency matrix by the sum of the frequencies of the corresponding row yields the transition probability value under the maximum likelihood estimate.

[0079] Furthermore, let the number of daily scene types in each target quarter within the aforementioned set of daily scene types be respectively... .

[0080] Changes in daytime power output can be described by a Markov process. Therefore, a state transition matrix (in total) between different daily scenario types within the target quarter can be generated based on the historical daily power output sequence of the target renewable energy plant. (a total of matrices), and the state transition matrix in the quarterly transition (a total of matrices). (Multiple matrices) to establish a daily state transition model for the new energy output sequence based on Markov chains.

[0081] For the target quarter Counting day scenario type Arrival Scene Type The transition probability is Then the target quarter The corresponding state transition matrix (i.e., the daily state transition matrix) can be represented by the following equation 10: (10); in, Satisfy the following equation 11: (11); In the formula, Indicates the first t Daytime scene type, Indicates the first t -1 Daytime scene type.

[0082] Equation 12 can be obtained through the maximum likelihood estimation method: (12); In the formula, The daily scene type in the sample dataset is determined by... Transfer to The number of times.

[0083] For quarter Similarly, the state transition matrix for the first day of the quarter (such as the quarterly state transition matrix) can be obtained, as shown in Equation 13 below. It should be noted that the matrix in Equation 13 includes... and A column is not necessarily a square.

[0084] (13); In this implementation, the daily scene type transition probability of adjacent dates within each target quarter is calculated using the daily state transition matrix, and the daily scene type transition probability of the first and last dates between quarters is calculated using the quarterly state transition matrix, thereby enabling continuous derivation of the daily scene type throughout the year.

[0085] In some embodiments, after generating the initial annual power output sequence scenario, the initial annual electricity consumption may exceed a reasonable range, causing the scenario to fail to accurately reflect the actual annual utilization rate and affecting the reliability of the assessment results. Therefore, step 105 may include: By splicing the daily target day power output sequence in chronological order, the initial annual power output sequence scenario of the target new energy power station is obtained; Based on the initial annual power output sequence scenario, the initial annual electricity consumption of the target renewable energy power station is determined; If the initial annual electricity consumption is within the target range, the ratio of the target boundary value of the target range to the initial annual electricity consumption is determined as the scaling factor; the target boundary value is the minimum or maximum annual electricity consumption of the target renewable energy power station. The target annual output sequence scenario is obtained by multiplying each output point in the initial annual output sequence scenario by the scaling factor.

[0086] In this implementation, the target boundary value can be determined by obtaining the historical annual utilization rate data of the target renewable energy power station, fitting the probability density distribution based on the kernel density estimation method, and filtering the upper and lower limits of the annual utilization rate based on the preset probability values, thereby determining the minimum and maximum annual electricity consumption.

[0087] The scaling factor can be determined based on the ratio of the target boundary value to the initial annual electricity consumption. If the initial annual electricity consumption exceeds the target range, the output sequence is adjusted by scaling to make it conform to the historical annual utilization rate distribution.

[0088] Specifically, the initial annual power output sequence scenario is generated by concatenating the daily target day power output sequences in chronological order. The initial annual power consumption is obtained by accumulating the values ​​of all power output points in the initial annual power output sequence scenario. If the initial annual power consumption is not within the target range, the ratio of the target boundary value to the initial annual power consumption is selected as a scaling factor. Each power output point in the initial power output sequence scenario is multiplied by this factor, making the adjusted annual power consumption equal to the target boundary value. This process ensures that the annual power consumption is within a reasonable range of historical annual utilization rates while maintaining the time correlation and fluctuation characteristics of the power output sequence. For example, when the initial annual power consumption exceeds the maximum annual power consumption, the scaling factor is the ratio of the maximum value to the initial power consumption, and each power output point is scaled down proportionally to avoid artificially inflated power consumption. Furthermore, the target boundary value determined by the kernel density estimation method can cover the annual utilization rate distribution under the preset probability, ensuring the statistical rationality of the generated scenario.

[0089] In this implementation, while ensuring that the total power generation of the generated annual power output sequence scenario meets expectations, the temporal characteristics and fluctuations of the daily power output sequence are preserved. This improves the accuracy and representativeness of the generated annual power output sequence scenario, providing more reliable input data for subsequent power system adequacy assessments.

[0090] Furthermore, the target boundary values ​​can be determined through the following steps: Obtain the historical annual utilization rate data of the target new energy power station; The annual utilization rate data is fitted using the kernel density estimation method to obtain the probability density distribution of the annual utilization rate data; Based on preset probability values, the annual utilization rate in the probability density distribution is filtered to determine the upper and lower limits of the annual utilization rate. Based on the upper and lower limits of the annual utilization rate, the minimum and maximum annual electricity consumption are determined.

[0091] In this implementation, the total annual output of new energy sources fluctuates due to changes in climate, policies, and installed capacity throughout the year. Based on historical daily output data of the target new energy power plants, a probability density function for the annual utilization rate of wind and solar power plants can be established using kernel density estimation. A 95% confidence level can be used as the preset probability value to obtain the annual electricity output. When the spatial distance between new energy power plants is not too far, their climate patterns can be considered similar. When the equipment level and the proportion of wind and solar curtailment do not change significantly over the years, the annual utilization rates of each solar and wind power plant are approximately identically distributed.

[0092] Furthermore, assuming that the annual utilization rate data of each wind farm and photovoltaic power station are available, as shown in Equations 14 and 15 below: (14); (15); In the formula, refer to The No. 1 wind farm is the first recorded wind farm. Annual utilization rate; refer to The No. 1 photovoltaic power station is the first one on record. Annual utilization rate; , They are respectively The number of years with annual utilization rate records for wind farms and photovoltaic power stations; , These are the number of wind farms and the number of photovoltaic farms, respectively.

[0093] The kernel density estimation method approximates the continuous probability density function of a random variable by fitting historical data. Its expression is shown in Equation 16 below: (16); In the formula, It is the density function; For optimal bandwidth, it is generally determined by Sure; The number of samples; Let be the random variable to be fitted; For the first Each sample value; For kernel functions; This represents the sample standard deviation.

[0094] Furthermore, a Gaussian kernel function can be used, the expression of which is shown in Equation 17 below: (17); Kernel density estimation can easily lead to distributions that do not satisfy the natural constraints. Specifically, the probability density function using hours has non-zero values ​​in intervals less than 0 or greater than 1, requiring adjustment, i.e., solving... , and then (18); Let the probability density functions of the annual utilization rates of wind farms and photovoltaic power plants in the region obtained from this be respectively... and Solve equations 19 to 22 below to obtain the upper limit of the annual utilization rate of the wind farm. The minimum annual utilization rate of wind farms The annual utilization rate ceiling of photovoltaic power stations The minimum annual utilization rate of photovoltaic power stations .

[0095] (19); (20); (twenty one); (twenty two); Furthermore, the minimum annual electricity consumption can be obtained by multiplying the total annual installed capacity of the new energy power station by the lower limit of the annual utilization rate, and the maximum annual electricity consumption can be obtained by multiplying the total installed capacity by the upper limit of the annual utilization rate.

[0096] In this implementation, a reasonable range for the annual electricity generation of renewable energy power plants is scientifically determined based on historical data, avoiding the subjectivity and inaccuracy that may result from manually setting the range. Simultaneously, kernel density estimation and probability distribution analysis fully consider the statistical characteristics of historical data, ensuring the high reliability and representativeness of the determined electricity range. This provides more accurate constraints for the subsequently generated power output sequence scenarios, thereby improving the overall quality and credibility of the generated scenarios.

[0097] In some embodiments, there are multiple daily output characteristic indicators, and there are complex nonlinear relationships between these indicators. If only independent probability distributions are used for sampling, the combined relationships between the characteristic indicators will deviate significantly from the actual historical data, thus affecting the accuracy of the output sequence scenario. Therefore, before step 103, the method for generating the output sequence scenario of a new energy power station provided in this application embodiment may further include: For each day scenario type, execute: Historical data of the daily characteristic indicators of each output day corresponding to the daily scenario type are extracted from the historical output sequence of the target new energy power station. By fitting the historical data of the characteristic indicators of each output day using the kernel density estimation method, the marginal probability density distribution of the characteristic indicators of each output day is obtained. Based on the marginal probability density distribution of each output day characteristic index, the correlation between each output day characteristic index in the day scene type is analyzed by a preset function to obtain the joint probability distribution corresponding to the day scene type.

[0098] In this implementation, the kernel density estimation method can use a Gaussian kernel function to perform non-parametric fitting on the historical data of the characteristic indicators of each output day, so as to adapt to the non-normal distribution characteristics of the characteristic indicators of the output day.

[0099] The preset function can be the D-Vine-Copula function, which determines the correlation coefficient matrix between different output day characteristic indicators and captures the correlation between each output day characteristic indicator.

[0100] Specifically, based on the kernel density estimation method, Equations 16 and 17 above are used to construct the probability density function of the intra-day characteristic indicators within each day's scenario type using the statistical results of the daily output characteristic indicators. Analogous to Equation 18 above, the daily output characteristic indicators are corrected; for example, the average, maximum, and minimum output values ​​should all be no less than 0, and the probability density function values ​​on the corresponding intervals should be equal to 0. This yields the marginal probability density function of the daily characteristic indicators.

[0101] Furthermore, the D-Vine-Copula function is used to describe the correlation between the characteristic indicators of the output day. The D-Vine-Copula function can decompose the multivariate Copula function into a product of a series of binary conditional Copula probability density functions, as shown in Equation 23 below: (twenty three); In the formula, Indicates the first One to the first Given that all variables are known, the first The and the first Conditional Copula probability density function among variables.

[0102] In this implementation, the statistical information contained in historical data is fully utilized, and the complex correlation between feature indicators is captured by constructing a joint probability distribution, thereby generating a more realistic output sequence scenario.

[0103] In some embodiments, the target day's output sequence can be determined based on output day characteristic indicators. However, directly matching the characteristic indicators may lead to a deviation between the selected sample day's output sequence and the target characteristic indicators, thereby affecting the accuracy of the generated sequence. Therefore, step 104 may include: For the corresponding daily output characteristic indicators, execute: The normalized Euclidean distance between the daily output characteristic index and the historical daily output sequence characteristic index of the target new energy power station is calculated. The historical sunrise power sequence with the smallest normalized Euclidean distance is used as the sample sunrise power sequence; The optimization model is solved based on the sample daily power output sequence to obtain the target daily power output sequence; wherein the optimization model takes minimizing the normalized Euclidean distance between the target daily power output sequence and the sample daily power output sequence as the objective function, and takes the characteristic index of the target daily power output sequence being equal to the characteristic index of the power output day as the constraint condition.

[0104] In this implementation, the normalized Euclidean distance can be calculated by standardizing each feature index to eliminate dimensional differences and ensure the fairness of the distance measurement.

[0105] The selection of the sample daily output sequence can be based on the principle of minimum distance, prioritizing the retention of samples in historical data that are closest to the target features.

[0106] The optimization model can be solved using a quadratic programming method. Under the premise of satisfying the characteristic index constraints, the morphological deviation from the sample sequence is minimized to achieve a smooth transition of the output sequence.

[0107] In some implementations, the first step is to find the sample data in the historical sample database of each station that has the smallest Euclidean distance to the generated daily output characteristic index after normalization. + The optimization model is then used to obtain the output curves (i.e., the daily output sequence of the sample stations). The optimization model is based on finding an output curve for each station that perfectly matches the daily characteristic index obtained from the sampling, and minimizes the Euclidean distance between the curve and the selected sample (i.e., the optimization model uses minimizing the normalized Euclidean distance between the target daily output sequence and the sample daily output sequence as the objective function, and uses the characteristic index of the target daily output sequence being equal to the daily output characteristic index as a constraint).

[0108] (twenty four); In the formula, To find the average function, To find the standard deviation function, To find the skewness function, To find the kurtosis function; The output curve of the station to be optimized. The power output curves of the selected samples from the same field station.

[0109] Specifically, after determining the daily output characteristic indicators, the first step is to select the sample sequence with the closest characteristic indicators from historical data as the initial solution for the optimization model. By constructing an objective function, the morphological characteristics of the sample sequence are used as the optimization direction, while the numerical values ​​of the characteristic indicators are used as hard constraints. During the solution process, constraints are employed to ensure that the generated target sequence not only meets the characteristic indicator requirements but also retains fluctuation characteristics similar to the sample sequence. For example, in extreme continuous low output scenarios, the optimization model can maintain the low fluctuation characteristics of the output curve while ensuring that the daily average output meets the constraints, avoiding the indicator offset problem caused by directly copying historical sequences.

[0110] In this implementation, by introducing an optimization model based on historical data, while ensuring strict matching of daily output characteristic indicators, the time correlation and fluctuation pattern of the output curve are preserved. This makes the generated output sequence not only consistent with the statistical characteristics of extreme scenarios, but also possess the physical rationality of actual operating data, significantly improving the accuracy of extreme scenario modeling in power system adequacy assessment.

[0111] The following describes the implementation process of the method for generating power output sequence scenarios of new energy power plants provided in the embodiments of this application: See also Figure 2 , Figure 2 A schematic diagram illustrating the implementation process of the new energy power station output sequence scenario generation method provided in this application embodiment.

[0112] Step 1: Input historical renewable energy output data, calculate the daily characteristic index (i.e., daily output characteristic index) of the output sequence of each station, and normalize each characteristic index.

[0113] Step 2: Perform cluster analysis on the power output sequence to obtain clusters of power output curves for various typical days (i.e., obtain multiple daily scene types and construct a set of daily scene types), and divide the historical wind power output sequence into several quarters (i.e., the division of the target quarters).

[0114] In step 2, in order to inject extreme scenario types, historical extreme sample days that conform to the preset extreme scenario definition can also be clustered to obtain extreme scenario types.

[0115] Step 3: Calculate the daily state transition matrix (i.e., the daily state transition matrix) and the inter-quarter state transition matrix (the quarterly state transition matrix) for each quarter.

[0116] Step 4: Establish the marginal distribution of each cluster of typical daily characteristic indicators and construct their joint distribution function (i.e., joint probability distribution).

[0117] Step 5: Randomly sample to determine the typical day type (i.e., day scene type) for day 1, or specify the extreme scene type and corresponding date.

[0118] Step 6: Based on the joint distribution function of the typical daily characteristic indicators corresponding to the corresponding quarter, sample to obtain the corresponding daily output characteristic indicators.

[0119] Step 7: Based on the corresponding quarterly daily type state transition matrix (inverse state transition matrix) and the daily type of the previous day (next day), obtain the daily type of the next day (previous day) through rolling sampling.

[0120] Step 8: Repeat steps 6 and 7 until the annual renewable energy output day type is generated.

[0121] Step 9: Obtain the daily 24-hour new energy output time series by solving the optimization model.

[0122] Step 10: Verify the data based on the upper and lower limits of annual electricity consumption. If the current annual renewable energy consumption is greater than the maximum annual renewable energy consumption (or less than the minimum annual renewable energy consumption), multiply all output points by a coefficient to make the annual renewable energy consumption equal to the maximum annual renewable energy consumption (or the minimum annual renewable energy consumption).

[0123] Figure 3 A structural diagram of the device for generating a power output sequence scenario for a new energy power station, as provided in an embodiment of this application, is shown. Figure 3 As shown, the device 400 for generating power generation sequence scenarios of new energy power plants includes: The acquisition module 401 is used to acquire the target date and the target extreme scenario type of the target new energy power station corresponding to the target date; The derivation module 402 is used to derive the daily scene type of each day within the target year based on the target extreme scene type and the target state transition matrix, using the target date as the anchor point, and to determine the daily scene type corresponding to each day from the daily scene type set; the target year is the year in which the target date is located, and the target state transition matrix is ​​used to calculate the daily scene type corresponding to the previous or next date of the current date; The sampling module 403 is used to sample the daily output characteristic index from the joint probability distribution corresponding to the daily scene type based on the daily scene type; the joint probability distribution includes the combination relationship between the daily output characteristic indexes in the daily scene type. The determination module 404 is used to determine the target day output sequence based on the daily output day characteristic indicators. The generation module 405 is used to generate the target annual output sequence scenario of the target new energy power station based on the target daily output sequence corresponding to each day.

[0124] The device 400 for generating power generation sequence scenarios of new energy power stations provided in this application embodiment can realize the various processes implemented in the aforementioned method embodiment for generating power generation sequence scenarios of new energy power stations and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0125] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-mentioned method for generating the power output sequence scenario of a new energy power station.

[0126] like Figure 4 , Figure 4 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. The electronic device includes: At least one battery; At least one memory; At least one processor; At least one program; The program is stored in memory, and the processor executes at least one program to implement the above-described method for generating a power output sequence scenario of a new energy power station.

[0127] This electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0128] The electronic devices according to embodiments of this application will now be described in detail.

[0129] The processor 1600 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this disclosure. The memory 1700 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 1700 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1700 and is called and executed by the processor 1600 to execute a method for generating a power output sequence scenario for a new energy power station according to an embodiment of this disclosure.

[0130] The input / output interface 1800 is used to implement information input and output. The communication interface 1900 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 2000 transmits information between various components of the device (e.g., processor 1600, memory 1700, input / output interface 1800, and communication interface 1900); The processor 1600, memory 1700, input / output interface 1800 and communication interface 1900 are connected to each other within the device via bus 2000.

[0131] This disclosure also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-described method for generating a power output sequence scenario of a new energy power station.

[0132] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0133] This application also provides a computer program product for implementation, wherein when the instructions in the computer program product are executed by the processor of an electronic device, the electronic device implements any of the new energy power station output sequence scenario generation methods in the above embodiments.

[0134] The embodiments described in this disclosure are for the purpose of more clearly illustrating the technical solutions of this disclosure and do not constitute a limitation on the technical solutions provided by this disclosure. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by this disclosure are also applicable to similar technical problems.

[0135] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this disclosure, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0136] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0137] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0138] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any related variations, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0139] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0140] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0141] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0142] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0143] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0144] The above is a detailed description of the preferred embodiments of this application. However, the embodiments of this application are not limited to the above-described implementation methods. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the embodiments of this application. All such equivalent modifications or substitutions are included within the scope defined by the claims of the embodiments of this application.

[0145] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application.

Claims

1. A method for generating a new energy plant output sequence scenario, characterized in that, The method comprises the following steps: obtaining a target date and a target extreme scene type corresponding to the target date of a target new energy power station; taking the target date as an anchor point, deriving a daily scene type of each day in a target year based on the target extreme scene type and a target state transition matrix, determining a daily scene type corresponding to each day from a daily scene type set; the target year is the year in which the target date is located, and the target state transition matrix is used to calculate the daily scene type corresponding to the previous day or the next day of the current date; based on the daily scene type corresponding to each day, sampling the daily output daily feature index corresponding to each day from the joint probability distribution corresponding to the daily scene type; the joint probability distribution includes the combination relationship between each output daily feature index in the daily scene type; based on the daily output daily feature index corresponding to each day, determining the target daily output sequence corresponding to each day; based on the target daily output sequence corresponding to each day, generating a target annual output sequence scene of the target new energy power station.

2. The method of claim 1, wherein, The daily scene type set is determined by the following steps: obtaining the historical monthly power generation of the target new energy power station, which is determined according to the historical daily output sequence of the target new energy power station; based on the monthly power generation, dividing 12 consecutive months to determine a plurality of target quarters, with the objective of minimizing the difference in monthly power generation in each quarter; each target quarter includes at least two months, and the months in each target quarter are consecutive; feature extraction is performed on the historical daily output sequence to obtain the output daily feature index; based on the output daily feature index, clustering the historical daily output sequence in each target quarter to determine a non-extreme scene day type; screening historical extreme sample days that meet the preset extreme scene definition from the historical daily output sequence; based on the historical daily output sequence corresponding to the historical extreme sample day, clustering to obtain an extreme day scene type; based on the non-extreme scene day type and the extreme scene day type, the daily scene type set is constructed.

3. The method of claim 2, wherein, The target new energy power station comprises a plurality of new energy sub-stations; The preset extreme scene definition comprises: in the case that the number of new energy sub-stations in the target new energy power station that are in continuous low output days is greater than a first threshold value, it is determined that the target new energy power station is in a new energy continuous low output extreme scene; in the case that the number of new energy sub-stations in the target new energy power station that are in large fluctuation days is greater than the first threshold value, it is determined that the target new energy power station is in a new energy large fluctuation extreme scene; in the case that the sum of the number of new energy sub-stations in the target new energy power station that are in the continuous low output days and the number of new energy sub-stations that are in the large fluctuation days is greater than the first threshold value, it is determined that the target new energy power station is in a new energy continuous low output and large fluctuation extreme scene; wherein, in the case that the daily average output of the new energy sub-station is less than a second threshold value, it is determined that the new energy sub-station is in the continuous low output day. In a case where a daily output standard deviation of the new energy sub-field station is greater than a third threshold value, it is determined that the new energy sub-field station is in the large fluctuation day; the second threshold value is a difference between an average value of daily average outputs of all historical sample days in the target new energy field station and twice a standard deviation of the daily average output sample; and the third threshold value is a sum of an average value of daily output standard deviations of all historical sample days in the target new energy field station and twice a standard deviation of the daily output standard deviation sample.

4. The method of claim 2, wherein, The target state transition matrix comprises at least one of a daily state transition matrix and a quarterly state transition matrix. Before the target date is taken as an anchor point, the target extreme scene type and the target state transition matrix are used to derive a daily scene type of each day in a target year, and the target daily scene type corresponding to each day is determined from a target daily scene type set, the method further comprises the following steps: In each target quarter, the daily scene types corresponding to each historical daily output sequence are counted to obtain frequencies of mutual transition of the daily scene types in the target quarter; The frequencies of mutual transition of the daily scene types in the target quarter are used to determine a daily state transition matrix by a maximum likelihood method, and the daily state transition matrix is used to calculate daily scene types corresponding to a previous day or a next day of a current day in each target quarter; The daily scene types of the first day of each target quarter and the daily scene types corresponding to a previous day of the first day of each target quarter are counted to obtain transition frequencies of the daily scene types between the target quarters; The transition frequencies of the daily scene types between the target quarters are used to determine a quarterly state transition matrix by a maximum likelihood method, and the quarterly state transition matrix is used to calculate a daily scene type of a first day of a current quarter.

5. The method of claim 1, wherein, The target annual output sequence scene of the target new energy field station is generated based on the target daily output sequence corresponding to each day, and comprises the following steps: The target daily output sequence corresponding to each day is spliced in a date order to obtain an initial annual output sequence scene of the target new energy field station; An initial annual electricity of the target new energy field station is determined based on the initial annual output sequence scene; In a case where the initial annual electricity is within a target range, a ratio of a target boundary value of the target range to the initial annual electricity is determined as a scaling coefficient; the target boundary value is an annual electricity minimum value or an annual electricity maximum value of the target new energy field station; Each output point in the initial annual output sequence scene is multiplied by the scaling coefficient to obtain the target annual output sequence scene.

6. The method of claim 5, wherein, The target boundary value is determined by the following steps: Historical annual utilization rate data of the target new energy field station are obtained; The annual utilization rate data are fitted based on a kernel density estimation method to obtain a probability density distribution of the annual utilization rate data; Annual utilization rates in the probability density distribution are screened based on a preset probability value to determine an upper limit value and a lower limit value of the annual utilization rate; The annual electricity minimum value and the annual electricity maximum value are determined based on the upper limit value and the lower limit value of the annual utilization rate.

7. The method of claim 1, wherein, The number of the output daily characteristic indexes is multiple. Before the sampling of the daily corresponding output daily feature indicator from the joint probability distribution corresponding to the daily scene type based on the daily corresponding daily scene type, the method further comprises: For each daily scene type, the following is performed: Extract historical data of each output daily feature indicator corresponding to the daily scene type from the historical output sequence of the target new energy power station; Fit the historical data of each output daily feature indicator by kernel density estimation method to obtain the marginal probability density distribution of each output daily feature indicator; Based on the marginal probability density distribution of each output daily feature indicator, analyze the correlation between each output daily feature indicator in the daily scene type by a preset function to obtain the joint probability distribution corresponding to the daily scene type.

8. The method of claim 1, wherein, The determination of the daily corresponding target daily output sequence based on the daily corresponding output daily feature indicator comprises: For each daily corresponding output daily feature indicator, the following is performed: Calculate the normalized Euclidean distance between the output daily feature indicator and the feature indicators of the historical daily output sequence of the target new energy power station to obtain the normalized Euclidean distance between the output daily feature indicator and the feature indicators of the historical daily output sequence; The historical daily output sequence with the smallest normalized Euclidean distance is taken as a sample daily output sequence; Solve the optimization model according to the sample daily output sequence to obtain the target daily output sequence; wherein the optimization model takes the minimization of the normalized Euclidean distance between the target daily output sequence and the sample daily output sequence as the objective function, and takes the feature indicators of the target daily output sequence equal to the output daily feature indicators as the constraint condition.

9. A device for generating a new energy plant output sequence scenario, characterized by The device comprises: An acquisition module configured to acquire a target date and a target extreme scene type corresponding to the target date of a target new energy power station; A derivation module configured to derive daily scene types of each day in a target year based on the target extreme scene type and a target state transition matrix with the target date as an anchor point, and determine daily corresponding daily scene types from a set of daily scene types; the target year is the year in which the target date is located, and the target state transition matrix is used to calculate daily scene types corresponding to a previous day or a next day of a current day; A sampling module configured to sample daily corresponding output daily feature indicators from a joint probability distribution corresponding to the daily scene types based on the daily corresponding daily scene types; the joint probability distribution comprises a combination relationship between each output daily feature indicator in the daily scene type; A determination module configured to determine daily corresponding target daily output sequences based on the daily corresponding output daily feature indicators; A generation module configured to generate a target year output sequence scenario of the target new energy power station based on the daily corresponding target daily output sequences.

10. An electronic device, comprising: The new energy power station output sequence scenario generation method comprises at least one control processor and a memory connected in communication with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the new energy power station output sequence scenario generation method in any one of claims 1 to 8.

Images